Hyperspectral Imaging (HSI) Has Emerged As A Powerful Remote Sensing Technique Due To Its Ability To Capture Detailed Spectral Information Across Hundreds Of Narrow And Contiguous Wavelength Bands. This Rich Spectral Resolution Enables Precise Identification And Discrimination Of Surface Materials, Which Is Critical In Applications Such As Agriculture, Mineral Exploration, Environmental Monitoring, And Defense. Traditional Classification Methods Often Face Challenges In Handling The High Dimensionality And Spectral Similarity Inherent In Hyperspectral Data. To Address This, Maximum Abundance Classification (MAC) Is Employed As An Effective Technique For Hyperspectral Image Analysis. MAC Is Based On Spectral Unmixing, Where Each Pixel Is Represented As A Mixture Of Multiple Endmembers, And Classification Is Performed By Assigning A Pixel To The Class Corresponding To The Endmember With The Highest Abundance Fraction. This Method Not Only Improves Classification Accuracy But Also Provides Insights Into The Sub-pixel Composition Of Heterogeneous Regions. The Proposed Work Focuses On Implementing MAC For Hyperspectral Image Datasets, Highlighting Its Effectiveness In Resolving Mixed Pixels And Enhancing Material Discrimination Compared To Conventional Pixel-based Classifiers. The Results Demonstrate That Maximum Abundance Classification Offers A Robust Approach For Extracting Meaningful Information From Hyperspectral Images, Making It Highly Suitable For Real-world Remote Sensing Applications.
Apple Production Plays A Vital Role In Global Horticulture, But Its Yield And Quality Are Often Threatened By Various Leaf Diseases Such As Apple Scab, Powdery Mildew, And Rust. Early And Accurate Detection Of These Diseases Is Crucial For Effective Management And Prevention. Traditional Manual Inspection Methods Are Time-consuming, Labor-intensive, And Prone To Human Error, While Conventional Machine Learning Techniques Often Lack Robustness And Generalization Ability In Complex Field Environments. To Address These Challenges, This Study Proposes An Apple Leaf Disease Detection Algorithm Based On An Improved YOLOv8 Model. The Enhanced Framework Integrates Optimized Feature Extraction Modules, Multi-scale Attention Mechanisms, And Lightweight Convolutional Operations To Improve Detection Accuracy While Reducing Computational Complexity. The Proposed Model Is Trained And Validated On A Large Dataset Of Apple Leaf Images Under Varying Conditions Of Illumination, Occlusion, And Background Interference. Experimental Results Demonstrate That The Improved YOLOv8 Achieves Higher Precision, Recall, And Mean Average Precision (mAP) Compared To The Baseline YOLOv8 And Other State-of-the-art Object Detection Models. This Approach Provides An Efficient And Reliable Solution For Real-time Apple Leaf Disease Detection, Offering Significant Potential For Precision Agriculture And Intelligent Crop Management Systems.
Age And Gender Prediction Using Facial Images Has Gained Significant Attention In Recent Years Due To Its Wide Range Of Applications In Fields Such As Human–computer Interaction, Security, Personalized Marketing, And Social Media Analytics. Traditional Computer Vision Techniques Often Struggle To Achieve High Accuracy Because Of Variations In Pose, Illumination, Facial Expressions, And Occlusions. To Overcome These Challenges, This Study Employs Deep Learning Models That Automatically Learn Discriminative Features From Raw Images Without Requiring Handcrafted Descriptors. Convolutional Neural Networks (CNNs) And Advanced Architectures Such As Residual Networks (ResNet) And Vision Transformers (ViT) Are Utilized For Robust Feature Extraction And Classification. The Models Are Trained And Validated On Benchmark Facial Image Datasets, Enabling The System To Predict Both Age Groups And Gender With High Precision. Experimental Results Demonstrate That Deep Learning–based Approaches Outperform Conventional Methods In Terms Of Accuracy, Generalization, And Scalability. The Proposed Framework Can Be Integrated Into Real-world Applications, Offering Reliable And Efficient Solutions For Automated Demographic Analysis.
White Blood Cells (WBCs) Play A Vital Role In The Immune System, And Abnormalities In Their Morphology And Distribution Are Key Indicators Of Blood Disorders Such As Leukemia. Early And Accurate Detection Of Leukemic Cells Is Critical For Effective Treatment And Improved Patient Outcomes. Traditional Manual Examination Of Blood Smears Under A Microscope Is Time-consuming, Prone To Human Error, And Requires Significant Expertise. To Address These Limitations, This Study Proposes An Automated System For Classifying Leukemic Blood Images Using Deep Learning And Image Processing Techniques. The Approach Involves Preprocessing Of Microscopic Blood Images To Enhance Quality, Followed By Segmentation Methods To Isolate WBCs From The Background. Feature Extraction Is Then Performed To Capture Both Morphological And Texture Characteristics. A Convolutional Neural Network (CNN) Model Is Employed To Automatically Learn Discriminative Features And Classify WBCs Into Normal Or Leukemic Categories. The Proposed Method Aims To Achieve High Accuracy, Robustness, And Efficiency In Leukemic Cell Detection, Thereby Supporting Hematologists In Clinical Diagnosis. This Research Highlights The Potential Of Integrating Deep Learning With Image Processing To Develop Computer-aided Diagnostic Tools For Leukemia Screening.
Rice Is One Of The Most Important Staple Crops Worldwide, And Its Productivity Is Often Threatened By Various Leaf Diseases Such As Bacterial Blight, Blast, And Brown Spot. Early And Accurate Identification Of These Diseases Is Crucial For Effective Crop Management And Yield Improvement. Traditional Disease Detection Methods, Which Rely On Manual Field Inspection, Are Time-consuming, Labor-intensive, And Prone To Human Error. To Overcome These Limitations, This Study Proposes An Automated Rice Leaf Disease Prediction System Based On MobileNetV2, A Lightweight Deep Learning Model Optimized For Mobile And Embedded Devices. The Model Leverages Transfer Learning To Extract Discriminative Features From Rice Leaf Images And Classify Them Into Healthy Or Diseased Categories. MobileNetV2’s Efficient Architecture Significantly Reduces Computational Cost While Maintaining High Classification Accuracy, Making It Suitable For Real-time Deployment In Agricultural Environments. Experimental Results Demonstrate That The Proposed System Achieves Robust Performance With High Accuracy, Precision, And Recall, Thereby Offering A Reliable And Scalable Solution For Smart Farming. This Work Highlights The Potential Of Integrating Deep Learning With Mobile Technologies To Support Farmers In Disease Monitoring And Decision-making, Ultimately Contributing To Sustainable Agriculture.
Jute Is An Economically Significant Natural Fiber Crop Widely Cultivated In Many Parts Of The World, Particularly In South Asia. However, Its Productivity And Fiber Quality Are Often Threatened By Various Foliar Diseases, Including Stem Rot, Anthracnose, And Leaf Blight, Which Can Cause Severe Yield Losses If Not Detected At An Early Stage. Traditional Methods Of Disease Identification, Which Rely On Manual Field Inspection, Are Labor-intensive, Time-consuming, And Prone To Human Error. To Address These Challenges, This Study Proposes A Deep Learning–based Automated System For Jute Leaf Disease Detection Using The ResNet50 Architecture. ResNet50, A Powerful Convolutional Neural Network (CNN), Is Employed To Extract High-level Discriminative Features From Leaf Images, Enabling Robust Classification Of Healthy And Diseased Leaves. The Model Is Trained And Validated On A Dataset Of Jute Leaf Images, Incorporating Preprocessing And Data Augmentation Techniques To Improve Generalization. Experimental Results Demonstrate That The Proposed Approach Achieves High Accuracy, Precision, And Recall, Making It Suitable For Real-time Deployment In Precision Agriculture. By Enabling Early Detection And Accurate Classification Of Jute Leaf Diseases, This System Can Assist Farmers In Taking Timely Remedial Actions, Thereby Enhancing Crop Yield, Reducing Economic Losses, And Promoting Sustainable Agricultural Practices.
The Accurate Classification Of Pharmaceutical Drugs And Vitamins Plays A Vital Role In Ensuring Patient Safety, Preventing Medication Errors, And Supporting Effective Healthcare Delivery. Traditional Methods Of Drug Identification Often Rely On Manual Inspection Of Physical Features Such As Shape, Size, And Imprint, Which Are Time-consuming, Prone To Human Error, And Impractical For Large-scale Applications. To Overcome These Limitations, This Study Proposes A Deep Learning–based Automated Classification System That Leverages Convolutional Neural Networks (CNNs) Combined With Transfer Learning Techniques. Pretrained CNN Architectures Such As ResNet, VGG, And Inception Are Utilized To Extract Discriminative Visual Features From Drug And Vitamin Images, Significantly Reducing Training Complexity While Improving Classification Accuracy. The Proposed System Is Trained On A Curated Dataset Of Pharmaceutical Drug And Vitamin Images, With Preprocessing And Augmentation Techniques Applied To Enhance Generalization Across Varying Lighting And Imaging Conditions. Experimental Results Demonstrate That The Transfer Learning–based CNN Models Outperform Traditional Machine Learning Methods By Achieving High Accuracy, Precision, And Recall. The System Has Strong Potential For Real-world Applications In Healthcare And Pharmacy Sectors, Such As Mobile-based Drug Recognition Tools, Automated Pill Dispensers, And Intelligent Healthcare Management Systems. By Ensuring Reliable Classification And Reducing The Chances Of Medication Errors, The Proposed Approach Contributes To Safer, More Efficient, And Technology-driven Healthcare Practices.
Potato Is One Of The Most Widely Cultivated Food Crops Worldwide, Yet Its Productivity Is Severely Threatened By Leaf Diseases Such As Early Blight, Late Blight, And Leaf Spot. Early And Accurate Detection Of These Diseases Is Crucial For Minimizing Yield Loss And Supporting Sustainable Agricultural Practices. Traditional Disease Diagnosis Methods Rely Heavily On Manual Field Inspection, Which Is Time-consuming, Labor-intensive, And Prone To Human Error. Recent Advances In Deep Learning Have Demonstrated Significant Potential In Automating Plant Disease Detection By Extracting Discriminative Features Directly From Image Data. This Study Presents A Comparative Analysis Of Multiple Deep Learning Architectures, Including Convolutional Neural Networks (cnns), Vgg16, Resnet50, And Mobilenetv2, For Potato Leaf Disease Detection. The Models Were Trained And Evaluated On Publicly Available Datasets, And Their Performance Was Compared In Terms Of Accuracy, Precision, Recall, F1-score, And Computational Efficiency. Experimental Results Highlight That Transfer Learning–based Models Outperform Traditional Cnns By Achieving Higher Accuracy And Faster Convergence. Among The Evaluated Models, Resnet50 And Mobilenetv2 Demonstrated Superior Performance, Balancing Both Accuracy And Lightweight Computation, Making Them Suitable For Real-time Field Deployment. The Findings Of This Study Emphasize The Importance Of Model Selection In Building Robust Disease Detection Systems And Provide Valuable Insights For Developing Practical, Ai-driven Tools To Assist Farmers And Agronomists In Precision Agriculture.
Forest Fires Are One Of The Most Destructive Natural Disasters, Causing Significant Ecological Damage, Economic Loss, And Threats To Human Life. Early Detection And Accurate Classification Of Forest Fire Incidents Are Crucial For Effective Fire Management And Mitigation Strategies. Traditional Fire Monitoring Systems, Which Rely On Manual Observation And Sensor-based Techniques, Are Often Time-consuming, Expensive, And Limited In Scalability. Recent Advancements In Machine Learning (ML) And Deep Learning (DL) Have Opened New Opportunities For Automated Forest Fire Detection Using Image Classification Techniques. In This Study, We Propose A Comparative Framework For Classifying Forest Fire Images Using Both Traditional Machine Learning Algorithms And Deep Learning Architectures Such As Convolutional Neural Networks (CNNs). Machine Learning Models Are Employed With Handcrafted Features, While Deep Learning Approaches Automatically Extract Discriminative Spatial Features From Raw Images. The Performance Of Different Models Is Evaluated Using Accuracy, Precision, Recall, And F1-score Metrics On Benchmark Datasets. Experimental Results Demonstrate That Deep Learning Methods Significantly Outperform Traditional Machine Learning Techniques In Terms Of Classification Accuracy And Robustness, Particularly Under Complex Environmental Conditions. The Findings Highlight The Potential Of Deep Learning–based Image Classification Systems As Effective Tools For Real-time Forest Fire Detection, Supporting Rapid Response And Disaster Management Efforts.
Chili (Capsicum Spp.) Is One Of The Most Widely Cultivated Spice Crops, Playing A Vital Role In Global Agriculture And The Food Industry. However, Its Productivity Is Significantly Threatened By Various Leaf Diseases Such As Leaf Spot, Mosaic, And Curl, Which Lead To Reduced Yield And Economic Losses. Early And Accurate Identification Of These Diseases Is Essential For Effective Crop Management And Disease Control. Traditional Diagnostic Methods, Which Rely On Manual Field Inspection, Are Often Time-consuming, Labor-intensive, And Prone To Human Error. To Overcome These Limitations, This Study Proposes An Automated Chili Leaf Classification System Using Deep Learning Techniques. Convolutional Neural Networks (CNNs) And Transfer Learning Models Are Employed To Automatically Extract Discriminative Features From Chili Leaf Images, Enabling Precise Classification Of Healthy And Diseased Samples. The Proposed Approach Demonstrates High Accuracy And Robustness Compared To Conventional Machine Learning Methods, Making It A Scalable And Efficient Solution For Real-time Disease Monitoring. The Findings Of This Study Highlight The Potential Of Deep Learning–based Classification Systems To Support Farmers And Agricultural Stakeholders In Improving Chili Crop Productivity Through Timely Disease Detection And Management.
Coffee Is One Of The Most Important Commercial Crops Worldwide, Contributing Significantly To The Global Economy And Agricultural Livelihoods. However, Its Productivity Is Severely Threatened By Various Leaf Diseases Such As Coffee Leaf Rust, Cercospora Leaf Spot, And Coffee Berry Disease, Which Lead To Substantial Yield Losses And Reduced Bean Quality. Early And Accurate Detection Of These Diseases Is Crucial For Effective Crop Management And Sustainable Coffee Production. Traditional Disease Diagnosis Methods Rely On Manual Inspection, Which Is Labor-intensive, Time-consuming, And Often Prone To Human Error. To Address These Challenges, This Study Proposes An Automated Coffee Leaf Disease Detection System Using Deep Learning Algorithms. Convolutional Neural Networks (CNNs) And Transfer Learning Models Such As VGG16, ResNet50, And MobileNetV2 Are Employed To Extract Discriminative Features From Leaf Images And Classify Them Into Healthy And Diseased Categories. The Proposed Approach Enhances Accuracy, Reduces Dependency On Expert Knowledge, And Enables Real-time Disease Monitoring. Experimental Results Demonstrate That Deep Learning–based Models Outperform Conventional Methods, Achieving High Precision, Recall, And Overall Classification Accuracy. This Study Highlights The Potential Of AI-driven Solutions In Supporting Precision Agriculture And Ensuring The Sustainable Cultivation Of Coffee.
Nail Diseases, Including Fungal Infections, Psoriasis, And Onycholysis, Are Common Health Concerns That Affect Both Aesthetics And Overall Well-being. Traditional Diagnostic Methods Rely Heavily On Visual Inspection And Clinical Expertise, Which Are Often Subjective, Time-consuming, And Prone To Human Error. With The Rapid Advancements In Artificial Intelligence, Deep Learning Has Emerged As A Powerful Tool For Automated Medical Image Analysis. This Study Proposes SMART Diagnosis, A Deep Learning–based System Designed To Accurately Identify And Classify Various Nail Diseases From Digital Images. By Leveraging Convolutional Neural Networks (CNNs) And Transfer Learning Techniques, The System Can Extract Discriminative Features From Nail Images, Enabling Precise Disease Classification. The Proposed Model Aims To Assist Dermatologists In Early And Reliable Diagnosis, Reduce Diagnostic Errors, And Provide Scalable Solutions For Telemedicine Applications. Experimental Results Demonstrate High Accuracy And Robustness, Highlighting The Potential Of SMART Diagnosis To Transform Nail Disease Identification Into A Faster, More Accurate, And Accessible Process, Ultimately Improving Patient Care And Health Outcomes.
Livestock Health Plays A Critical Role In Ensuring Sustainable Dairy And Meat Production, Yet Early Detection Of Sickness In Cows Remains A Major Challenge For Farmers. Traditional Methods Of Disease Monitoring Rely On Manual Observation Of Behavioral And Physical Changes, Which Are Often Subjective, Time-consuming, And Prone To Human Error. With Advancements In Artificial Intelligence, Machine Learning (ML) And Deep Learning (DL) Techniques Provide Promising Solutions For Automated Health Monitoring. This Study Proposes An Intelligent System For The Early Detection Of Sickness In Cows By Analyzing Behavioral Patterns, Physiological Parameters, And Visual Features Extracted From Images And Sensor Data. Various ML Algorithms, Including Support Vector Machines (SVM) And Random Forest, Are Compared With State-of-the-art DL Models Such As Convolutional Neural Networks (CNNs) And Long Short-Term Memory (LSTM) Networks For Time-series Data. The Results Demonstrate That Deep Learning Models Achieve Superior Accuracy In Identifying Subtle Patterns Of Illness, Enabling Timely Intervention And Reducing Economic Losses. The Proposed Approach Highlights The Potential Of AI-driven Livestock Monitoring Systems To Support Farmers In Enhancing Animal Welfare, Optimizing Productivity, And Ensuring Food Security.
Osteoarthritis (OA) Is One Of The Most Common Degenerative Joint Diseases, Primarily Affecting The Knee Joint And Leading To Pain, Stiffness, And Reduced Mobility. Early And Accurate Prediction Of OA Progression Is Crucial For Timely Intervention, Improved Patient Outcomes, And Reduced Healthcare Costs. Traditional Diagnostic Methods, Such As Clinical Examination And Radiographic Analysis, Are Often Subjective, Time-consuming, And Limited In Their Ability To Detect Subtle Pathological Changes. With Recent Advancements In Artificial Intelligence, Deep Learning (DL) Techniques Have Emerged As Powerful Tools For Automated Medical Image Analysis. This Study Proposes A Deep Learning–based Framework For The Prediction Of Osteoarthritis In Knee Joints Using Radiographic And MRI Data. Convolutional Neural Networks (CNNs) Are Employed To Automatically Extract Discriminative Features From Knee Images, Enabling Precise Detection And Grading Of OA Severity. The System Is Designed To Assist Clinicians By Providing Objective, Reliable, And Rapid Predictions, Thereby Reducing Diagnostic Errors And Supporting Personalized Treatment Planning. The Proposed Approach Demonstrates The Potential Of Deep Learning In Advancing Computer-aided Diagnosis And Improving The Early Prediction And Management Of Knee Osteoarthritis.
Kidney Anomalies, Including Cysts, Tumors, And Structural Abnormalities, Are Among The Leading Causes Of Renal Dysfunction And Can Progress To Severe Chronic Kidney Disease If Not Diagnosed Early. Traditional Diagnostic Methods, Such As Ultrasound And CT Scan Interpretation, Rely Heavily On Radiologists’ Expertise, Making Them Time-consuming, Subjective, And Prone To Human Error. With Recent Advancements In Artificial Intelligence, Deep Learning (DL) Has Emerged As A Powerful Tool For Automated Medical Image Analysis, Offering High Accuracy In Detecting Subtle Pathological Changes. This Study Proposes An Intelligent Framework For Automated Kidney Anomaly Detection Using Deep Learning Models Integrated With Explainable AI (XAI) Techniques. The DL Models Are Trained On Medical Imaging Datasets To Identify And Classify Kidney Anomalies With Improved Precision, While XAI Methods, Such As Grad-CAM And SHAP, Provide Visual And Interpretable Explanations Of Model Decisions. This Combination Not Only Enhances The Transparency And Reliability Of AI-driven Diagnostics But Also Assists Clinicians In Making Informed Decisions. The Proposed System Demonstrates Significant Potential In Early Detection, Reduced Diagnostic Workload, And Improved Patient Outcomes, Paving The Way For Trustworthy And Clinically Applicable AI Solutions In Nephrology.
Monkeypox, An Emerging Zoonotic Viral Disease, Has Recently Re-emerged As A Global Health Concern Due To Its Rapid Transmission And Clinical Similarity To Other Skin-related Infections. Early And Accurate Diagnosis Is Essential For Effective Containment And Treatment, Yet Traditional Diagnostic Methods Such As Polymerase Chain Reaction (PCR) Testing And Clinical Evaluation Are Time-consuming, Costly, And Require Expert Resources That Are Often Unavailable In Low-resource Settings. With The Advancement Of Artificial Intelligence, Deep Learning (DL) Has Shown Remarkable Potential In The Automated Diagnosis Of Infectious Diseases From Medical Images, Particularly Skin Lesion Analysis. This Survey Provides A Comprehensive Overview Of Recent Research Efforts In Applying Deep Learning Techniques For Monkeypox Detection And Classification. It Discusses Commonly Used Datasets, Image Preprocessing Methods, Deep Neural Network Architectures, Evaluation Metrics, And Comparative Performance Outcomes. Furthermore, The Paper Highlights Existing Challenges, Including Limited Datasets, Class Imbalance, And The Need For Explainable AI In Clinical Settings. Finally, Potential Future Directions Such As Transfer Learning, Federated Learning, And Multimodal Approaches Are Outlined To Enhance Diagnostic Accuracy And Real-world Applicability. This Survey Aims To Serve As A Reference For Researchers And Healthcare Professionals Seeking To Leverage Deep Learning In Improving The Timely And Reliable Diagnosis Of Monkeypox.
Human–wildlife Conflict Is A Growing Concern, Particularly In Regions Where Wild Animals Intrude Into Agricultural Fields, Residential Areas, Or Roadways, Leading To Significant Crop Losses, Property Damage, And Threats To Human Safety. Traditional Animal Monitoring Methods, Such As Manual Patrolling And Camera Trapping, Are Time-consuming, Labor-intensive, And Often Fail To Provide Real-time Alerts. With Advancements In Computer Vision And Deep Learning, Object Detection Models Have Shown Great Potential In Addressing These Challenges. This Study Proposes An Advanced Wild Animal Detection And Alert System Using The YOLOv5 Model, A State-of-the-art Deep Learning Framework Known For Its High Accuracy And Fast Detection Speed. The System Is Designed To Identify And Classify Various Wild Animal Species From Live Video Streams Or Camera Trap Images In Real Time. Once Detection Occurs, An Automatic Alert Mechanism Is Triggered Through IoT-based Communication Channels (e.g., SMS, Alarms, Or Mobile Notifications) To Warn Farmers, Forest Officials, Or Nearby Residents. The Integration Of YOLOv5 With IoT Technologies Ensures Timely Response, Reduces Human–animal Conflict, And Enhances Both Community Safety And Wildlife Conservation Efforts. Experimental Evaluations Demonstrate That The Proposed System Achieves High Precision And Recall Rates While Maintaining Low Latency, Making It Suitable For Deployment In Real-world Environments.
With The Rapid Advancement Of Digital Editing Tools, Image Manipulation Has Become Increasingly Sophisticated And Widespread, Raising Serious Concerns In Areas Such As Journalism, Law Enforcement, Medical Imaging, And Social Media. Traditional Image Forgery Detection Techniques, Such As Error Level Analysis And Metadata Inspection, Often Fail To Detect Subtle Manipulations Or Perform Poorly On Large-scale Datasets. Deep Learning Has Recently Emerged As A Powerful Solution, Offering Automated Feature Extraction And High Accuracy In Identifying Complex Patterns Of Tampering. This Study Explores The Application Of Deep Learning Models—such As Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), And Hybrid Architectures—for Detecting Common Forgery Types Including Copy-move, Splicing, And Deepfake Images. The Proposed Framework Leverages Spatial And Frequency Domain Analysis, Combined With Transfer Learning, To Enhance Detection Robustness Across Diverse Datasets. Experimental Evaluations Demonstrate That Deep Learning–based Methods Outperform Conventional Approaches In Terms Of Precision, Recall, And Generalization To Unseen Manipulations. The Findings Highlight The Potential Of Deep Learning Models As Reliable And Scalable Tools For Real-world Image Forgery Detection, Paving The Way For Stronger Digital Content Authentication Systems.
With The Rapid Growth Of E-commerce And The Fashion Retail Industry, Automated Product Classification And Personalized Recommendations Have Become Essential For Improving User Experience And Increasing Sales. Bags, Being A Diverse Product Category With Variations In Design, Size, Material, And Style, Pose Significant Challenges For Accurate Classification And Effective Product Suggestions. Traditional Machine Learning Approaches Often Rely On Handcrafted Features, Which Are Limited In Capturing Complex Visual Patterns. To Address This, This Study Proposes A Chatbot System For Bag Classification And Product Suggestion Using EfficientNet, A State-of-the-art Deep Learning Architecture Known For Its High Accuracy And Computational Efficiency. The Proposed Model Leverages EfficientNet For Robust Feature Extraction And Fine-grained Classification Of Different Bag Categories, While The Chatbot Interface Enables Interactive Communication With Users To Provide Real-time Product Recommendations Based On Preferences And Browsing Behavior. This Integration Of Deep Learning With Conversational AI Not Only Enhances Classification Accuracy But Also Improves Customer Engagement And Decision-making In Online Shopping.
Skin Cancer Is One Of The Most Prevalent Forms Of Cancer Worldwide, And Its Early Detection Is Crucial For Effective Treatment And Improved Survival Rates. Traditional Diagnostic Methods Rely Heavily On Clinical Expertise And Invasive Biopsy Procedures, Which Are Often Time-consuming, Costly, And Subject To Human Error. With Advancements In Artificial Intelligence, Image Processing Combined With Machine Learning Has Emerged As A Powerful Tool For Automated Skin Cancer Classification. This Study Focuses On Developing A System That Leverages Image Preprocessing Techniques—such As Noise Removal, Contrast Enhancement, And Segmentation—to Accurately Extract Relevant Features From Dermoscopic Images. Machine Learning Algorithms Are Then Applied To Classify Skin Lesions Into Benign Or Malignant Categories. The Proposed Approach Aims To Improve Diagnostic Accuracy, Reduce Dependency On Manual Interpretation, And Provide A Cost-effective, Non-invasive Solution For Early Skin Cancer Detection. By Integrating Image Processing And Intelligent Classification Models, This Work Contributes To The Advancement Of Computer-aided Diagnosis In Dermatology And Supports Timely Clinical Decision-making.
Plant Diseases Significantly Affect Agricultural Productivity, Leading To Severe Economic Losses And Threats To Food Security Worldwide. Traditional Disease Identification Methods Rely Heavily On Expert Knowledge And Manual Inspection, Which Are Time-consuming, Error-prone, And Impractical For Large-scale Monitoring. With The Advancement Of Artificial Intelligence, Deep Learning Has Emerged As A Powerful Tool For Automated Plant Disease Recognition Through Image Analysis. This Study Focuses On Developing An Intelligent System For Plant Leaf Disease Recognition Using Deep Learning Algorithms, Particularly Convolutional Neural Networks (CNNs). The Proposed Model Leverages Image Preprocessing, Feature Extraction, And Classification Techniques To Accurately Identify And Categorize Different Types Of Leaf Diseases. By Training On Large-scale Plant Image Datasets, The System Demonstrates High Accuracy In Distinguishing Between Healthy And Diseased Leaves, Even In Complex Environments. Such An Approach Not Only Reduces Dependency On Manual Expertise But Also Enables Real-time And Cost-effective Disease Detection, Empowering Farmers With Early Diagnosis And Effective Crop Management Strategies. The Results Highlight The Potential Of Deep Learning-based Solutions In Precision Agriculture And Sustainable Farming Practices.
The Rapid Advancement Of Artificial Intelligence And Generative Models Has Led To The Rise Of Deepfakes—synthetically Altered Or Fabricated Audio, Video, And Images That Are Increasingly Indistinguishable From Authentic Content. While Deepfake Technology Has Promising Applications In Entertainment, Education, And Creative Industries, It Also Poses Severe Threats To Privacy, Security, Politics, And Digital Trust When Misused For Disinformation, Identity Theft, Or Fraud. Traditional Detection Methods, Such As Manual Inspection And Handcrafted Feature-based Techniques, Are Insufficient Against The Sophisticated Manipulations Generated By Modern Deep Learning Architectures Like Generative Adversarial Networks (GANs) And Autoencoders. To Address These Challenges, Deep Learning-based Detection Techniques Have Emerged As A Powerful Solution Due To Their Ability To Automatically Learn Discriminative Features From Large-scale Datasets. Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), And Transformer-based Architectures Have Been Widely Employed To Capture Spatial, Temporal, And Frequency Inconsistencies Present In Forged Media. Additionally, Hybrid Models That Integrate Multimodal Analysis—combining Visual, Audio, And Physiological Cues—have Shown Significant Improvements In Detection Accuracy. This Study Explores Recent Advancements In Deepfake Detection Using Deep Learning, Highlighting Key Methodologies, Benchmark Datasets, Performance Metrics, And Ongoing Challenges Such As Generalizability, Adversarial Attacks, And Real-time Implementation. The Findings Emphasize The Critical Role Of Robust And Adaptive Detection Systems In Safeguarding Digital Media Integrity And Mitigating The Societal Risks Associated With Deepfakes.
The Increasing Demand For Health Awareness And Personalized Dietary Management Has Driven The Need For Automated Food Recognition And Nutritional Analysis Systems. Traditional Methods Of Food Logging, Such As Manual Entry Or Barcode Scanning, Are Often Inconvenient, Time-consuming, And Prone To User Error. Recent Advancements In Deep Learning, Particularly Convolutional Neural Networks (CNNs), Have Demonstrated Remarkable Success In Image Classification Tasks, Providing A Foundation For Intelligent Food Recognition. This Study Proposes An Automatic Food Image Classification System That Leverages Deep Learning Techniques To Accurately Identify Food Categories From Images And Estimate Their Corresponding Nutritional Values, Such As Calories, Protein, Carbohydrates, And Fat Content. The Framework Involves Preprocessing Of Food Images, Feature Extraction Through CNN-based Models, And Classification Into Predefined Food Categories. A Nutritional Database Is Integrated To Map Recognized Food Items To Their Nutritional Information, Enabling Precise Prediction Of Dietary Intake. Experimental Evaluations On Benchmark Food Image Datasets Demonstrate The Effectiveness Of The Proposed Approach In Achieving High Classification Accuracy And Reliable Nutrient Estimation. This System Has Significant Applications In Healthcare, Fitness Tracking, And Personalized Diet Management, Ultimately Contributing To Improved Lifestyle And Well-being.
Dog Breed Classification Is A Challenging Computer Vision Task Due To High Intra-class Variation, Inter-class Similarity, And The Large Number Of Breeds With Subtle Visual Differences. Traditional Methods Relying On Handcrafted Features And Shallow Classifiers Often Fail To Achieve Satisfactory Performance In Real-world Scenarios. Recent Advancements In Deep Learning, Particularly Convolutional Neural Networks (CNNs), Have Demonstrated Remarkable Success In Image Recognition Tasks By Automatically Learning Discriminative Features From Data. This Study Proposes A Breakthrough Conventional-based Approach For Dog Breed Classification That Integrates CNNs With Transfer Learning Techniques. Pre-trained Deep Learning Models Such As VGG16, ResNet, And Inception Are Fine-tuned On A Curated Dog Breed Dataset To Leverage Their Feature Extraction Capabilities While Reducing Computational Cost And Training Time. The Proposed Method Enhances Classification Accuracy, Robustness, And Generalization Compared To Training Models From Scratch. Experimental Results Highlight The Effectiveness Of Transfer Learning In Achieving Superior Performance Across Multiple Dog Breeds, Offering A Scalable And Efficient Solution For Applications In Veterinary Research, Animal Identification, And Intelligent Pet-care Systems.
Handwritten Text Recognition Remains A Challenging Problem In The Field Of Computer Vision And Pattern Recognition Due To The High Variability In Individual Writing Styles, Distortions, Overlapping Characters, And Background Noise. Traditional Recognition Systems Based On Handcrafted Features And Rule-based Methods Often Fail To Achieve High Accuracy And Generalization In Real-world Scenarios. Recent Advancements In Deep Learning Have Introduced Powerful Hybrid Models That Combine Convolutional Neural Networks (CNNs) And Recurrent Neural Networks (RNNs), Forming Convolutional Recurrent Neural Networks (CRNNs). CNNs Are Effective In Extracting Robust Spatial Features From Handwritten Images, While RNNs—particularly Long Short-Term Memory (LSTM) Or Gated Recurrent Units (GRUs)—capture The Sequential Dependencies Inherent In Handwriting. This Study Focuses On Developing A CRNN-based Framework For End-to-end Handwritten Text Recognition, Eliminating The Need For Manual Feature Engineering. The Model Is Trained On Large-scale Handwritten Datasets And Evaluated Using Sequence-to-sequence Mapping With Connectionist Temporal Classification (CTC) Loss To Align Input Images With Output Text. Experimental Results Demonstrate That The Proposed CRNN Architecture Achieves High Recognition Accuracy And Robustness Against Diverse Handwriting Variations, Making It A Promising Solution For Applications Such As Digitizing Historical Documents, Automated Form Processing, And Intelligent Handwriting-based Interfaces.
Fish Detection And Tracking In Aquaculture Environments Is A Crucial Task For Monitoring Fish Behavior, Estimating Population Density, And Optimizing Farm Management Practices. Traditional Manual Observation Methods Are Labor-intensive, Prone To Error, And Lack Scalability In Large-scale Fish Farms. Recent Advancements In Computer Vision And Deep Learning Provide Automated Solutions For Real-time Monitoring. This Study Proposes An Efficient Fish Detection And Tracking Framework That Integrates The YOLO (You Only Look Once) Object Detection Algorithm With Euclidean Distance-based Tracking. YOLO Is Employed To Accurately Detect Fish In Underwater Video Streams, Leveraging Its High-speed Inference And Robustness Against Complex Backgrounds, Water Turbidity, And Varying Illumination. Subsequently, A Euclidean Distance Tracker Associates Detections Across Consecutive Frames, Enabling Reliable Multi-fish Tracking Without The Computational Complexity Of Deep Tracking Models. The Proposed System Ensures Real-time Performance With High Detection Accuracy And Low Identity-switch Rates, Making It Suitable For Practical Deployment In Aquaculture Farms. The Approach Can Support Applications Such As Fish Counting, Growth Monitoring, And Behavioral Analysis, Contributing To Sustainable And Intelligent Aquaculture Management.
Cotton Is One Of The Most Important Commercial Crops Worldwide, And Its Productivity Is Severely Affected By Various Leaf Diseases Such As Bacterial Blight, Grey Mildew, Fusarium Wilt, And Alternaria Leaf Spot. Early And Accurate Detection Of These Diseases Is Essential For Minimizing Yield Losses And Ensuring Sustainable Crop Management. Traditional Diagnostic Methods Relying On Manual Observation Are Often Time-consuming, Subjective, And Prone To Human Error, Making Them Unsuitable For Large-scale Monitoring. In Recent Years, Advancements In Computer Vision And Machine Learning Have Enabled The Development Of Automated Systems For Disease Detection And Classification. This Survey Provides A Comprehensive Overview Of Existing Techniques For Cotton Leaf Disease Identification, Ranging From Traditional Image Processing Methods To Deep Learning-based Approaches. Various Algorithms Such As Convolutional Neural Networks (CNNs), Support Vector Machines (SVM), Random Forests, And Transfer Learning Models Are Discussed In Terms Of Their Accuracy, Robustness, And Computational Efficiency. Furthermore, The Survey Highlights Publicly Available Datasets, Key Performance Metrics, Challenges Such As Variations In Illumination And Background Noise, And Potential Solutions Through Hybrid And Ensemble Models. Finally, It Outlines Future Research Directions, Emphasizing The Need For Lightweight, Real-time, And Scalable Solutions Deployable In Smart Farming Applications.
Handwritten Signature Verification Is One Of The Most Widely Used Biometric Authentication Methods Due To Its Simplicity, Uniqueness, And Acceptance In Legal And Financial Transactions. However, Traditional Manual Verification Methods Are Prone To Human Error, Time-consuming, And Susceptible To Forgery. With The Rapid Advancement Of Computer Vision And Pattern Recognition, Automated Signature Verification Has Become A Promising Solution For Enhancing Security And Reliability. This Study Presents A Handwritten Signature Verification System Utilizing OpenCV Techniques For Image Preprocessing, Feature Extraction, And Classification. The System Employs Image Enhancement Methods Such As Grayscale Conversion, Thresholding, Edge Detection, And Contour Analysis To Extract Discriminative Features From Signatures. These Features Are Then Compared Using Similarity Measures Or Machine Learning Models To Distinguish Between Genuine And Forged Signatures. The Proposed Approach Demonstrates The Effectiveness Of OpenCV-based Techniques In Achieving Accurate, Efficient, And Scalable Signature Verification, Offering Potential Applications In Banking, Legal Documentation, And Secure Identity Management.
The Choice Of A Suitable Hairstyle Plays A Significant Role In Enhancing Personal Appearance And Confidence. However, Selecting The Most Appropriate Hairstyle Often Depends On Factors Such As Face Shape, Facial Features, And Individual Preferences, Making It A Challenging And Subjective Process. Traditional Methods Of Hairstyle Selection Rely On Manual Consultation With Experts, Which Can Be Time-consuming, Inconsistent, And Costly. To Address This Challenge, This Study Presents The Design And Implementation Of A Hair Recommendation System Based On Face Recognition. The System Employs Computer Vision Techniques For Face Detection And Feature Extraction, Classifying Face Shapes Such As Oval, Round, Square, And Heart-shaped. Using These Classifications, A Recommendation Engine Suggests Suitable Hairstyles Tailored To The User’s Facial Attributes. The Proposed Framework Integrates Image Preprocessing, Facial Landmark Detection, Feature Analysis, And Machine Learning Models To Provide Personalized Hairstyle Suggestions In Real Time. Experimental Results Demonstrate That The System Offers Accurate Face Shape Classification And Effective Hairstyle Recommendations, Thereby Enhancing User Satisfaction. This Approach Highlights The Potential Of Artificial Intelligence And Face Recognition Technologies In Personal Grooming, Fashion, And Virtual Try-on Applications.
Wild Animal Detection And Segmentation Play A Vital Role In Wildlife Monitoring, Biodiversity Conservation, And Prevention Of Human–animal Conflicts. Traditional Methods Such As Manual Observation And Camera Traps Are Often Time-consuming, Labor-intensive, And Prone To Inaccuracies In Large-scale Or Dense Environments. With Recent Advancements In Deep Learning, Automated Detection Systems Have Become Increasingly Effective For Real-time Monitoring Of Wildlife. This Study Proposes An Efficient Framework For Wild Animal Detection And Segmentation Using The YOLOv7 (You Only Look Once Version 7) Object Detection Algorithm. YOLOv7 Is Utilized For Its Superior Speed And Accuracy In Identifying Multiple Animal Species Within Complex Natural Backgrounds. Furthermore, Segmentation Is Integrated To Provide Precise Boundary Information, Enabling Detailed Analysis Of Animal Size, Movement Patterns, And Habitat Utilization. The Proposed System Can Assist Researchers, Forest Authorities, And Conservationists In Automating Wildlife Surveillance, Reducing Manual Effort, And Enhancing Decision-making For Ecological Management.
Rice Is One Of The Most Important Staple Crops Worldwide, And Its Productivity Is Significantly Affected By Various Leaf Diseases Such As Bacterial Blight, Brown Spot, And Leaf Smut. Early And Accurate Detection Of These Diseases Is Crucial For Effective Crop Management And Yield Improvement. Traditional Manual Identification Methods Are Time-consuming, Subjective, And Prone To Error, Making Them Unsuitable For Large-scale Monitoring. Recent Advances In Deep Learning Have Provided Efficient Solutions For Automated Plant Disease Recognition. This Study Focuses On The Classification Of Rice Leaf Diseases Using Convolutional Neural Networks (CNNs) And Transfer Learning Approaches. A CNN Model Is Developed For Feature Extraction And Classification, While Pre-trained Models Such As VGG16, ResNet50, And InceptionV3 Are Fine-tuned To Improve Accuracy And Reduce Training Time. Experimental Results Demonstrate That Transfer Learning-based Models Outperform Standard CNNs By Achieving Higher Classification Accuracy, Robustness, And Generalization. The Proposed Framework Provides An Efficient And Scalable Solution For Real-time Rice Disease Detection, Thereby Supporting Precision Agriculture And Sustainable Farming Practices.
Potato Is One Of The Most Widely Cultivated And Consumed Crops Worldwide, But Its Productivity Is Severely Threatened By Various Leaf Diseases Such As Early Blight, Late Blight, And Bacterial Infections. Timely And Accurate Detection Of These Diseases Is Essential To Reduce Yield Losses And Ensure Sustainable Agricultural Practices. Traditional Manual Identification Methods Are Often Labor-intensive, Time-consuming, And Prone To Human Error, Making Them Unsuitable For Large-scale Monitoring. Recent Advancements In Computer Vision And Deep Learning Provide Promising Solutions For Automated Plant Disease Recognition. This Study Focuses On Potato Disease Detection Using Convolutional Neural Networks (CNNs) Integrated With Image Processing Techniques. The Proposed Approach Involves Preprocessing Potato Leaf Images, Feature Extraction, And Classification Of Healthy And Diseased Leaves. CNNs Are Employed Due To Their Superior Capability In Learning Hierarchical Features Directly From Image Data Without The Need For Handcrafted Features. Experimental Results Demonstrate High Accuracy And Robustness Of The Model In Detecting And Classifying Potato Leaf Diseases. The Findings Highlight The Potential Of CNN-based Frameworks As An Effective Decision-support Tool For Precision Agriculture, Enabling Farmers To Take Timely Preventive Measures And Improve Crop Yield.
Fruits Are An Essential Component Of Global Agriculture And Food Supply, Yet Accurate Identification Of Fruit Species Remains A Significant Challenge Due To Variations In Size, Shape, Color, Texture, And Environmental Conditions During Image Acquisition. Traditional Manual Classification Methods Are Time-consuming, Error-prone, And Impractical For Large-scale Applications. Recent Advancements In Computer Vision And Deep Learning Have Opened New Opportunities For Automated And Highly Accurate Fruit Species Detection. This Study Proposes An Improved Fruit Species Detection System Using Image Processing Techniques Combined With Deep Learning Models Such As Convolutional Neural Networks (CNNs). The System Utilizes Preprocessing Methods Including Noise Reduction, Image Enhancement, And Segmentation To Extract Discriminative Features. These Features Are Then Fed Into Deep Learning Architectures To Achieve Robust Classification Across Diverse Fruit Categories. Experimental Results Demonstrate That The Proposed Approach Significantly Enhances Detection Accuracy And Efficiency Compared To Conventional Machine Learning Techniques. The System Has Potential Applications In Smart Agriculture, Food Quality Control, Automated Harvesting, And Supply Chain Management, Thereby Contributing To Increased Productivity And Reduced Labor Dependency.
Handwritten Digit Recognition Is One Of The Most Widely Studied Problems In The Field Of Computer Vision And Pattern Recognition, With Applications Ranging From Automated Postal Sorting And Bank Check Processing To Digital Form Analysis And Authentication Systems. Traditional Manual Recognition Methods Are Time-consuming, Error-prone, And Unsuitable For Large-scale Applications. With Recent Advancements In Deep Learning, Significant Progress Has Been Made In Achieving High Accuracy For Digit Classification Tasks. This Study Presents A Handwritten Digit Recognition System That Integrates Deep Learning Networks With OpenCV For Efficient Image Preprocessing, Feature Extraction, And Classification. The Proposed System Employs Techniques Such As Noise Reduction, Thresholding, And Contour Detection To Enhance Image Quality Before Feeding It Into A Deep Neural Network For Accurate Classification. The Model Is Trained And Evaluated Using Benchmark Datasets Such As MNIST To Demonstrate Its Effectiveness. Experimental Results Indicate That The Integration Of OpenCV With Deep Learning Significantly Improves Recognition Accuracy And Computational Efficiency, Making The System Robust And Applicable For Real-world Scenarios.
Mango Is One Of The Most Economically Significant Fruit Crops Cultivated Worldwide, But Its Yield And Quality Are Severely Affected By Various Leaf Diseases Such As Anthracnose, Powdery Mildew, And Bacterial Canker. Accurate And Timely Detection Of These Diseases Is Crucial For Effective Crop Management, Prevention Of Yield Loss, And Ensuring Sustainable Agricultural Practices. Traditional Disease Identification Methods Relying On Visual Inspection Are Labor-intensive, Time-consuming, And Prone To Human Error. In Recent Years, Numerous Computational Techniques, Ranging From Conventional Image Processing And Machine Learning To Advanced Deep Learning Models, Have Been Developed To Improve The Efficiency And Accuracy Of Mango Leaf Disease Detection. This Study Presents A Systematic Analysis Of Existing Techniques Used For Mango Leaf Disease Detection, Highlighting Their Methodologies, Strengths, Limitations, And Application Scope. The Comparative Review Emphasizes The Transition From Handcrafted Feature-based Approaches To Data-driven Deep Learning Architectures, Which Have Demonstrated Superior Performance In Terms Of Robustness And Scalability. Furthermore, The Analysis Discusses Challenges Such As Variability In Imaging Conditions, Dataset Limitations, And Computational Complexity, While Also Outlining Future Research Directions For Developing More Generalized, Real-time, And Farmer-friendly Detection Systems.
Potholes Are A Major Cause Of Road Accidents, Vehicle Damage, And Traffic Disruptions, Making Their Timely Detection And Repair A Critical Aspect Of Modern Transportation Management. Traditional Manual Inspection Methods Are Labor-intensive, Time-consuming, And Prone To Human Error, Highlighting The Need For Automated, Reliable, And Scalable Solutions. Recent Advancements In Computer Vision And Deep Learning Have Opened New Opportunities For Accurate Pothole Detection And Measurement. This Study Proposes A Modern Pothole Detection And Dimension Estimation System That Integrates The YOLO (You Only Look Once) Deep Learning Model With Advanced Image Processing Techniques. The YOLO Architecture Is Employed For Real-time Pothole Localization And Classification, While Image Processing Methods Are Utilized To Estimate The Dimensions Of Detected Potholes, Such As Width, Depth, And Surface Area. The Proposed System Ensures High Accuracy, Robustness Under Varying Lighting And Environmental Conditions, And The Ability To Process Large-scale Road Datasets Efficiently. Experimental Results Demonstrate That This Hybrid Approach Outperforms Traditional Methods, Offering A Practical And Cost-effective Solution For Smart City Infrastructure, Automated Road Condition Monitoring, And Predictive Maintenance Systems.
Medicinal Plants Play A Vital Role In Traditional Healthcare Systems And Modern Pharmacology, Yet Their Accurate Identification In The Wild Remains A Challenging Task Due To Variations In Plant Morphology, Environmental Conditions, And Similarities Between Species. Traditional Identification Methods Rely On Expert Knowledge And Manual Inspection, Which Are Often Time-consuming, Subjective, And Impractical For Large-scale Applications. Recent Advancements In Computer Vision And Deep Learning Provide New Opportunities For Automated And Reliable Plant Recognition. This Study Presents A Medicinal Plant Identification System Based On Convolutional Neural Networks (CNNs), Designed To Classify And Recognize Plant Species Directly From Leaf And Plant Images Captured In Natural Environments. The Proposed System Leverages Image Preprocessing, Feature Extraction, And Classification Through CNN Models To Achieve High Accuracy Under Varying Lighting And Background Conditions. Such An Approach Not Only Facilitates Faster And More Accurate Plant Recognition But Also Contributes To The Preservation Of Traditional Knowledge, Supports Biodiversity Research, And Enhances Applications In Healthcare And Drug Discovery.
Leather Is One Of The Most Widely Used Natural Materials In Industries Such As Fashion, Furniture, And Automotive Manufacturing. However, The Presence Of Defects Such As Scratches, Cuts, Insect Bites, Wrinkles, And Stains Significantly Reduces Its Commercial Value And Usability. Traditional Defect Inspection Relies Heavily On Human Expertise, Which Is Time-consuming, Subjective, And Prone To Error, Especially When Applied To Large-scale Production. To Overcome These Limitations, Automated Leather Defect Detection And Classification Systems Have Gained Increasing Attention. This Study Presents A Deep Learning–based Framework For Accurate Detection, Classification, And Segmentation Of Leather Defects. Convolutional Neural Networks (CNNs) And Advanced Architectures Such As U-Net And YOLO Are Employed To Localize And Categorize Defects At Pixel And Region Levels. The Proposed System Ensures Both High Accuracy In Defect Classification And Precise Segmentation For Defect Localization, Enabling Manufacturers To Automate Quality Control Processes Effectively. Experimental Results Demonstrate The System’s Robustness In Handling Diverse Defect Types Under Varying Illumination And Texture Conditions.
Natural Language Processing (NLP) Has Witnessed Rapid Advancements In Recent Years, Largely Driven By The Integration Of Deep Learning Techniques. This Survey Explores The Diverse Applications Of Deep Learning In NLP, Highlighting Key Models, Architectures, And Methodologies That Have Transformed The Field. Beginning With Foundational Neural Approaches Such As Word Embeddings And Recurrent Neural Networks, The Survey Traces The Evolution Toward More Advanced Architectures, Including Convolutional Neural Networks, Sequence-to-sequence Models, Attention Mechanisms, And Transformer-based Frameworks. The Applications Span A Wide Range Of Tasks, Such As Machine Translation, Sentiment Analysis, Question Answering, Text Summarization, Speech Recognition, And Dialogue Systems. Moreover, We Examine The Advantages Of Deep Learning In Capturing Semantic Meaning, Context, And Linguistic Nuances Compared To Traditional Statistical Methods. Challenges Such As Interpretability, Data Requirements, Computational Costs, And Ethical Concerns Are Also Discussed. This Survey Provides An Overview Of The Current Landscape Of Deep Learning In NLP While Pointing To Promising Directions For Future Research And Applications.
The Increasing Volume Of Digital And Textual Evidence In Forensic Investigations Necessitates Advanced Techniques For Efficient Information Extraction And Analysis. This Study Explores The Application Of Natural Language Processing (NLP) Methods, Particularly Transformer-based Models, To Automatically Extract Relevant Entities, Relationships, And Events From Forensic Documents. Leveraging The Contextual Understanding Of Transformers, The System Identifies Key Forensic Information With High Accuracy And Reduces Manual Effort In Case Analysis. Extracted Data Is Further Structured Into Knowledge Graphs, Enabling Intuitive Visualization Of Complex Relationships Between Entities Such As Suspects, Locations, Incidents, And Evidence. This Approach Not Only Enhances The Speed And Precision Of Forensic Investigations But Also Facilitates Pattern Recognition, Trend Analysis, And Decision-making. The Integration Of Transformer-based NLP With Graph Visualization Represents A Promising Paradigm For Modernizing Forensic Intelligence And Improving Investigative Outcomes.
In Software Engineering, The Clarity And Consistency Of Requirements Are Crucial For Successful Project Outcomes. However, Natural Language Requirements Are Often Ambiguous, Incomplete, Or Inconsistent, Leading To Costly Errors In Later Development Stages. To Address This, Requirements Templates Are Used To Standardize The Structure And Language Of Requirements. Despite Their Usefulness, Manually Verifying Conformance To These Templates Remains A Time-consuming And Error-prone Task. This Paper Presents An Automated Approach That Leverages Natural Language Processing (NLP) Techniques To Check The Conformance Of Requirements Documents Against Predefined Templates. The Proposed System Employs Syntactic And Semantic Analysis To Identify Structural Inconsistencies, Missing Elements, And Deviations From Standard Phrasing. By Integrating Machine Learning Models And Rule-based Methods, The System Can Adapt To Varying Template Styles And Domain-specific Vocabularies. Experimental Results On Real-world Datasets Demonstrate The Effectiveness Of The Approach In Improving Requirements Quality And Reducing The Manual Effort Required For Validation. This Work Contributes To Enhancing Automation In Requirements Engineering And Supports The Development Of More Reliable Software Systems.
The Rapid Advancement Of Deep Learning In Natural Language Processing (NLP) Has Enabled Recurrent Neural Networks (RNNs), Particularly Long Short-Term Memory (LSTM) Architectures, To Generate Human-like Text Sequences. Despite Their Impressive Fluency, The Statistical Properties Of LSTM-generated Texts Often Diverge From Those Found In Natural Human Language. This Study Investigates The Statistical Features Of LSTM-generated Texts By Examining Linguistic Distributions, Such As Word Frequency, Sentence Length Variability, Entropy Measures, And Zipf’s Law Conformity. Comparative Analysis With Human-authored Corpora Highlights Areas Where LSTM Models Successfully Capture Natural Language Regularities And Where They Fall Short, Such As Long-range Dependencies And Higher-order Semantic Coherence. The Findings Provide Insights Into The Strengths And Limitations Of LSTM-based Text Generation, Offering A Deeper Understanding Of How Statistical Patterns Emerge In Synthetic Language. This Contributes To The Broader Evaluation Of Generative Models And Informs The Development Of More Linguistically Grounded NLP Systems.
The Rapid Progress Of Generative Models Has Enabled The Creation Of Highly Realistic Synthetic Voices, Commonly Known As Audio Deep Fakes. While These Technologies Have Beneficial Applications In Entertainment And Assistive Systems, They Also Pose Significant Risks In Misinformation, Fraud, And Security Breaches. Detecting Audio Deep Fakes Remains A Challenging Task Due To The Increasingly Natural Prosody, Timbre, And Linguistic Coherence Of Synthesized Speech. This Paper Proposes A Multi-stage Framework For Audio Deep Fake Detection That Integrates Complementary Strategies Across Acoustic, Linguistic, And Deep Feature Domains. In The First Stage, Handcrafted Acoustic Features Such As Mel-frequency Cepstral Coefficients (MFCCs) And Spectral Distortions Are Extracted To Capture Low-level Signal Artifacts. The Second Stage Leverages Linguistic Consistency Analysis To Identify Irregularities In Phoneme Duration And Speech Rhythm. Finally, Deep Learning–based Embeddings From Pre Trained Models Are Employed To Capture High-level Semantic And Prosodic Patterns. By Combining These Heterogeneous Feature Spaces Through Ensemble Classification, The Proposed Framework Achieves Robust Performance Against State-of-the-art Synthesis And Voice Conversion Systems. Experimental Results On Benchmark Datasets Demonstrate Improved Generalization Across Multiple Attack Scenarios, Highlighting The Potential Of The Framework As A Practical Tool For Safeguarding Digital Communications Against Audio Forgery.
The Rapid Growth Of Digital Data Across Industries Has Created A Pressing Need For Efficient Methods Of Organizing And Extracting Meaningful Information. Document Classification And Data Extraction Have Emerged As Critical Techniques To Address This Challenge. Document Classification Leverages Machine Learning And Natural Language Processing (NLP) To Automatically Categorize Documents Based On Content, Structure, Or Intent, Enabling Streamlined Information Retrieval And Management. Complementarily, Data Extraction Focuses On Identifying And Retrieving Relevant Entities, Attributes, Or Patterns From Unstructured Or Semi-structured Documents, Transforming Raw Text Into Structured, Usable Datasets. Together, These Processes Enhance Decision-making, Improve Workflow Automation, And Reduce Manual Effort In Domains Such As Healthcare, Finance, Legal Systems, And Enterprise Operations. This Study Explores State-of-the-art Methodologies, Including Deep Learning Models, Rule-based Systems, And Hybrid Approaches, To Improve The Accuracy, Scalability, And Adaptability Of Document Classification And Data Extraction Systems. The Findings Highlight The Potential Of These Techniques To Unlock Hidden Insights, Reduce Information Overload, And Support Intelligent Information Systems In A Data-driven World.
With The Exponential Growth Of User-generated Health-related Content On Online Platforms, Patient Drug Reviews Have Become A Valuable Source Of Real-world Insights Into Drug Effectiveness And Side Effects. Traditional Drug Recommendation Systems Often Rely On Structured Clinical Data, Which May Not Fully Capture Patient Experiences And Sentiments. This Study Proposes A Drug Recommendation System Based On Sentiment Analysis Of Drug Reviews Using Machine Learning (ML) Techniques. The System Leverages Natural Language Processing (NLP) To Analyze Patient Reviews, Extracting Both Sentiment Polarity (positive, Negative, Neutral) And Key Opinion Aspects Related To Drug Efficacy, Safety, And Tolerability. Machine Learning Models Are Trained To Classify Sentiments And Predict Drug Suitability For Specific Conditions. By Integrating Sentiment-driven Insights With Recommendation Algorithms, The System Provides More Personalized And Patient-centric Drug Suggestions. Experimental Results On Benchmark Drug Review Datasets Demonstrate The Effectiveness Of The Proposed Approach In Improving Recommendation Accuracy Compared To Conventional Methods. This Work Highlights The Potential Of Combining Sentiment Analysis And ML To Support Informed Decision-making In Healthcare And Enhance Patient Outcomes.
In The Era Of E-commerce, Vast Amounts Of User-generated Product Reviews Provide Valuable Insights Into Customer Satisfaction And Product Quality. However, Extracting Meaningful Information From Such Unstructured Textual Data Remains A Major Challenge. This Study Focuses On Amazon Product Review Classification Using Natural Language Processing (NLP) Techniques And Logistic Regression As The Classification Model. The Proposed System Preprocesses Raw Textual Reviews Through Tokenization, Stop-word Removal, Stemming/lemmatization, And Feature Extraction Methods Such As Bag-of-Words And TF-IDF. Logistic Regression Is Then Employed To Classify Reviews Into Sentiment Categories (positive Or Negative), Enabling Automated Opinion Mining With High Interpretability. Furthermore, This Work Provides A Review Of Machine Learning-based Performance Prediction Systems, Highlighting The Role Of Various Algorithms In Improving Sentiment Analysis Tasks. By Comparing Logistic Regression With Other ML Models, The Study Emphasizes Its Simplicity, Efficiency, And Robustness In Handling Large-scale Review Data. The Findings Contribute To The Development Of More Reliable Recommendation Systems, Assisting Businesses In Decision-making While Enhancing Customer Experience.
Stone Inscriptions Are A Vital Source Of Historical Knowledge, Particularly In The Tamil Civilization, Which Has Preserved Cultural, Linguistic, And Political Information Through Ancient Scripts. However, These Inscriptions Face Challenges Such As Erosion, Complex Letter Forms, And Limited Accessibility For Modern Readers. This Study Proposes A System For Automatic Recognition And Speech Synthesis Of Ancient Tamil Characters From Stone Inscriptions Using Advanced Computational Techniques. The Methodology Involves Image Preprocessing To Enhance Inscription Clarity, Segmentation To Isolate Characters, And Recognition Using Deep Learning Models Such As Convolutional Neural Networks (CNNs) Trained On Ancient Tamil Script Datasets. Once The Characters Are Identified, They Are Mapped To Their Modern Tamil Equivalents And Converted Into Speechable Audio Using Text-to-Speech (TTS) Technology. This Approach Not Only Preserves Ancient Tamil Heritage But Also Makes Inscriptions More Accessible To Historians, Researchers, And The General Public. The Proposed System Contributes To The Fields Of Optical Character Recognition (OCR), Digital Preservation, And Cultural Informatics, Ensuring That Ancient Tamil Inscriptions Are Both Digitally Archived And Audibly Experienced By Future Generations.
Visual Impairment Poses Significant Challenges To Independent Mobility And Safe Navigation In Daily Life. Traditional Aids Such As White Canes And Guide Dogs Provide Partial Assistance But Lack The Ability To Perceive And Interpret Complex Environments. This Project Proposes A Guidance System Virtual Assistance Device That Integrates Sensors, Computer Vision, And Voice-based Interaction To Assist Visually Impaired Individuals In Real Time. The System Employs Ultrasonic Sensors For Obstacle Detection, A Camera With Deep Learning Models For Object And Path Recognition, And GPS For Outdoor Navigation. A Speech Synthesis Module Provides Audio Feedback To The User, Offering Instructions, Warnings, And Navigation Guidance. The Device Is Designed To Be Lightweight, Portable, And User-friendly, Ensuring Accessibility In Both Indoor And Outdoor Environments. By Combining Smart Sensing, AI-driven Decision-making, And Virtual Assistance, The Proposed System Aims To Enhance Mobility, Safety, And Independence For Visually Impaired Individuals, Ultimately Improving Their Quality Of Life.
Text Extraction From Images Is A Crucial Task In The Field Of Document Digitization, Heritage Preservation, And Information Retrieval. With The Growing Availability Of Optical Character Recognition (OCR) Technologies, It Is Now Possible To Efficiently Convert Printed Or Handwritten Text Into Editable And Searchable Digital Formats. This Project Focuses On Developing A Web-based Application For Extracting Tamil Text From Images Using The Tesseract OCR Engine Integrated With The Flask Framework. The System Preprocesses Input Images Through Techniques Such As Grayscale Conversion, Noise Removal, And Thresholding To Improve Recognition Accuracy. Tesseract OCR Is Then Employed To Recognize And Extract Tamil Characters, Which Are Displayed In A User-friendly Interface Powered By Flask. The Proposed System Enables Users To Upload Images And Retrieve The Extracted Tamil Text In Real Time, Thereby Offering An Effective Solution For Digitizing Documents, Preserving Ancient Scripts, And Improving Accessibility Of Tamil Content. This Approach Demonstrates The Potential Of Combining Open-source OCR Tools With Lightweight Web Frameworks To Create Efficient, Language-specific Text Recognition Systems.
With The Rapid Growth Of Digital Information Processing, Extracting And Understanding Text From Images Has Become An Essential Task In Various Domains Such As Document Digitization, Education, And Multilingual Communication. Optical Character Recognition (OCR) Technology Enables The Automatic Conversion Of Printed Or Handwritten Text In Images Into Machine-readable Formats. This Project Focuses On Implementing An OCR-based System Using The Tesseract Engine To Detect And Extract Text From Images, Followed By Automatic Translation Into The Desired Language. The Methodology Involves Preprocessing Input Images Through Techniques Such As Grayscale Conversion, Noise Removal, And Thresholding To Enhance Accuracy. Once Text Is Extracted Using Tesseract OCR, A Translation Module Is Applied To Convert The Recognized Text Into The Target Language, Enabling Cross-lingual Accessibility. The Proposed System Provides An Efficient And Scalable Solution For Bridging Language Barriers And Making Text-based Information More Accessible For Diverse Users. Applications Of This Work Include Real-time Translation For Travelers, Digitization Of Multilingual Documents, And Assistance For Individuals With Limited Language Proficiency.
The Integration Of Artificial Intelligence (AI) Into Healthcare Has Opened New Opportunities For Delivering Personalized And Accessible Medical Guidance. Ayurveda, One Of The Oldest Traditional Systems Of Medicine, Emphasizes Holistic Well-being Through Natural Remedies, Diet, And Lifestyle Management. This Project Proposes The Development Of An AI-based Ayurvedic Chatbot Designed To Provide Healthcare Support And Personalized Diet Plans Rooted In Ayurvedic Principles. The Chatbot Leverages Natural Language Processing (NLP) To Interact With Users, Analyze Their Health Conditions, And Recommend Ayurvedic Remedies, Lifestyle Changes, And Diet Plans Based On Body Constitution (Prakriti) And Reported Symptoms. Machine Learning Models Are Employed To Enhance Recommendation Accuracy By Learning From User Interactions And Expert-verified Data. The System Also Integrates A Knowledge Base Of Ayurvedic Herbs, Treatments, And Dietary Guidelines, Ensuring Both Accessibility And Reliability. This Approach Enables Users To Receive Instant, Personalized, And Preventive Healthcare Support While Promoting The Traditional Wisdom Of Ayurveda Through Modern Technology. The Proposed Solution Aims To Bridge The Gap Between Ancient Health Practices And Contemporary Digital Healthcare Needs, Offering A Cost-effective, Scalable, And User-friendly Wellness Tool.
The Increasing Demand For Personalized Healthcare Solutions Has Accelerated The Integration Of Artificial Intelligence (AI) And Machine Learning (ML) Into Medical Assistance Systems. This Project Proposes The Development Of A Chatbot For Healthcare And Diet Planning That Leverages Natural Language Processing (NLP) And Machine Learning Techniques To Provide Intelligent, Real-time, And User-friendly Health Support. The Chatbot Is Designed To Interact With Users Through Conversational Interfaces, Collect Basic Health Parameters, Lifestyle Information, And Dietary Preferences, And Subsequently Generate Personalized Recommendations. Machine Learning Models Are Employed To Analyze User Inputs, Classify Health Conditions, And Predict Suitable Diet Plans Tailored To Individual Needs. The System Also Incorporates A Knowledge Base Of Medical Guidelines And Nutritional Data To Ensure Evidence-based Recommendations. By Offering 24/7 Accessibility, Scalability, And Cost-effectiveness, This Chatbot Has The Potential To Support Preventive Healthcare, Promote Healthy Lifestyles, And Reduce The Burden On Medical Professionals. The Proposed Solution Demonstrates The Role Of AI-driven Chatbots As A Valuable Digital Healthcare Assistant For Personalized Wellness Management.
Identity Verification Is A Critical Requirement In Sectors Such As Banking, Government Services, Security, And Digital Onboarding. Manual Extraction Of Information From Identity Cards Is Time-consuming, Error-prone, And Inefficient. To Address This Challenge, Optical Character Recognition (OCR) Integrated With Image Processing Techniques Offers An Automated And Reliable Solution. This Project Proposes The Development Of An Identity Card Recognition System That Captures And Processes Images Of ID Cards To Extract Textual Information Such As Name, Date Of Birth, Address, And Identification Number. The Methodology Involves Preprocessing The Image Using Techniques Like Grayscale Conversion, Noise Reduction, Edge Detection, And Contrast Enhancement To Improve Recognition Accuracy. The Processed Image Is Then Fed Into An OCR Engine, Such As Tesseract, To Convert Printed Or Handwritten Text Into A Machine-readable Format. Post-processing Steps, Including Text Segmentation And Error Correction, Further Refine The Extracted Data. The System Is Designed To Provide Fast, Accurate, And Secure Extraction Of Identity Information, Enabling Seamless Integration With Authentication And Verification Systems. This Approach Enhances Efficiency In Applications Like KYC (Know Your Customer), E-governance, And Digital Record Management While Reducing Manual Effort And Minimizing Fraud Risks.
The Rapid Growth Of Social Media And Online Communication Platforms Has Led To An Increase In Cyberbullying Incidents, Negatively Impacting The Mental Health And Well-being Of Individuals, Especially Adolescents. Detecting Cyberbullying Manually Is Challenging Due To The Massive Volume Of User-generated Content And The Subtlety Of Abusive Language. This Project Proposes An Automated System For Cyberbullying Detection Using Machine Learning Integrated With The Flask Framework For Web-based Deployment. The System Leverages Natural Language Processing (NLP) Techniques To Preprocess Textual Data, Including Tokenization, Stopword Removal, And Vectorization, And Applies Machine Learning Algorithms Such As Logistic Regression, Support Vector Machines, And Random Forest To Classify Content As Bullying Or Non-bullying. The Flask Framework Provides A User-friendly Web Interface For Real-time Input And Detection, Enabling Users To Monitor Social Media Posts Or Messages Efficiently. Experimental Results Demonstrate That The Proposed System Can Accurately Identify Cyberbullying Content, Offering A Scalable And Practical Solution To Mitigate Online Harassment And Promote Safer Digital Interactions.
The Rapid Growth Of Social Media Platforms, Particularly Twitter, Has Resulted In The Generation Of Vast Amounts Of User-generated Content That Reflects Public Opinions, Emotions, And Sentiments On Diverse Topics. Analyzing This Data Provides Valuable Insights For Businesses, Governments, And Researchers In Areas Such As Market Analysis, Policy-making, And Public Opinion Monitoring. This Project Focuses On Sentiment Analysis Of Twitter Data Using A Combination Of Machine Learning Approaches And Semantic Analysis Techniques. The Proposed System Involves Data Collection Through Twitter APIs, Preprocessing Of Textual Data Using Natural Language Processing (NLP) Methods, And The Application Of Supervised Learning Algorithms Such As Naïve Bayes, Support Vector Machines (SVM), And Logistic Regression For Sentiment Classification. Additionally, Semantic Analysis Techniques Are Integrated To Improve Context Understanding And Handle Challenges Such As Sarcasm, Ambiguity, And Slang, Which Are Common In Social Media Text. The Performance Of The Models Is Evaluated Using Metrics Like Accuracy, Precision, Recall, And F1-score To Determine Their Effectiveness. The Integration Of Machine Learning With Semantic Analysis Is Expected To Enhance Sentiment Detection Accuracy And Provide A Deeper Understanding Of Public Sentiment Trends On Twitter.
Scene Text Recognition (STR) Has Emerged As A Challenging Task In Computer Vision Due To Variations In Font Styles, Illumination, Perspective Distortions, Occlusions, And Complex Backgrounds. Traditional Recognition Methods Often Struggle To Maintain Robustness In Such Unconstrained Environments. To Address These Challenges, We Propose A Novel Approach That Integrates Structure-guided Character Detection With Linguistic Knowledge Modeling For Improved Recognition Accuracy. The System First Employs A Structure-aware Character Detection Mechanism That Leverages Spatial Relationships Between Characters To Generate Reliable Candidate Regions, Reducing The Effect Of Noisy Backgrounds And Distortions. Subsequently, Linguistic Knowledge Is Incorporated Through Lexicon Constraints And Language Models To Refine Recognition Outputs And Enforce Semantic Consistency. This Joint Utilization Of Structural Cues And Linguistic Priors Enables The System To Not Only Detect Characters More Precisely But Also Correct Misclassifications In Ambiguous Scenarios. Experimental Results On Benchmark Scene Text Datasets Demonstrate That The Proposed Method Significantly Outperforms Conventional STR Approaches, Achieving Higher Accuracy And Robustness Under Real-world Conditions.
HUMAN ATTENTION PREDICTION IN NATURAL DAILY LIFE WITH FINE-GRAINED HUMAN-ENVIRONMENT-OBJECT INTERACTION MODEL
Copy-move Forgery Is One Of The Most Prevalent Image Tampering Techniques, Where A Region Of An Image Is Copied And Pasted Within The Same Image To Conceal Or Duplicate Content. Detecting Such Manipulations Is Highly Challenging Due To Post-processing Operations Such As Scaling, Rotation, And Compression. In This Work, We Propose A Novel Framework For Copy-move Forgery Detection That Integrates Deep PatchMatch With Pairwise Ranking Learning. The Deep PatchMatch Module Leverages Deep Feature Representations To Establish Reliable Correspondences Between Image Patches, Overcoming Limitations Of Handcrafted Descriptors. Subsequently, A Pairwise Ranking Learning Strategy Is Employed To Differentiate Authentic Patch Correspondences From Forged Ones, Enabling Robust Detection Even Under Complex Transformations. The Proposed Approach Achieves Precise Localization Of Forged Regions While Maintaining Resilience Against Common Post-processing Attacks. Extensive Experiments On Publicly Available Benchmark Datasets Demonstrate That Our Method Outperforms Existing State-of-the-art Techniques In Both Detection Accuracy And Localization Quality. This Work Highlights The Potential Of Combining Deep Patch Similarity Search With Learning-based Ranking For Advancing Image Forensics.
The Rapid Rise Of Video Content Has Sparked A Pressing Need For Robust Video-language Understanding (VLU) Systems Capable Of Linking Visual Dynamics With Natural Language Reasoning. However, Progress Is Constrained By What We Call The Impossible Data Trinity: The Simultaneous Requirement For Large-scale Data Quantity, Fine-grained Annotation Quality, And Broad Domain Diversity. Traditional Dataset Construction Approaches Struggle To Balance All Three Dimensions, Inevitably Sacrificing One To Achieve The Others. In This Work, We Introduce The Video Dataflywheel, A Self-reinforcing Paradigm That Leverages Multimodal Foundation Models, Synthetic Data Generation, And Human-in-the-loop Refinement To Progressively Resolve The Trinity Challenge. The Dataflywheel Operates In Iterative Cycles: Models Generate And Refine Pseudo-labeled Data, Which Is Then Validated And Expanded By Scalable Curation Strategies, Feeding Back Into Stronger Models And Richer Datasets. We Demonstrate How This Framework Accelerates The Creation Of Diverse, High-quality, And Large-scale Video-language Resources, Reducing Reliance On Costly Manual Annotation While Maintaining Semantic Fidelity. Beyond Dataset Construction, The Video Dataflywheel Provides A Blueprint For Sustainable VLU Research, Enabling The Community To Move Closer To Generalizable, Efficient, And Context-aware Video-language Understanding.
Biometric Authentication Has Become A Cornerstone Of Modern Security Systems Due To Its Uniqueness And Resistance To Forgery. However, Traditional Biometric Verification Methods Often Raise Concerns Regarding Privacy Leakage, Data Misuse, And Template Theft. This Work Proposes A Privacy-preserving Biometric Verification Framework Using Handwritten Random Digit Strings As An Authentication Factor. In The Proposed System, Users Are Prompted To Write A Randomly Generated Digit Sequence, Combining The Inherent Individuality Of Handwriting Dynamics With The Unpredictability Of Random Strings. This Approach Prevents Replay Attacks, Minimizes The Risk Of Stolen Static Templates, And Enhances Resilience Against Impersonation Attempts. The Verification Process Leverages Machine Learning-based Handwriting Recognition And Feature Extraction Techniques, While Privacy-preserving Transformations Ensure That The Raw Biometric Data Is Never Stored Or Transmitted In Its Original Form. Experimental Evaluation Demonstrates That The Method Achieves A Balance Between Robust Verification Accuracy, User Privacy, And System Security, Making It A Promising Solution For Next-generation Secure Authentication Systems.
This Project Presents A Python-based Application That Converts Text Embedded In Images Into Editable, Translatable Text And Delivers Fluent Outputs In A Target Language. The System Couples Image Preprocessing (noise Removal, Binarization, Skew Correction) With Optical Character Recognition (OCR) To Extract Text From Varied Inputs Such As Documents, Signboards, And Screenshots. Language Identification Triggers A Neural Machine Translation Pipeline To Produce The Translated Text, While Confidence Scores Guide Optional Human Review. A Lightweight GUI Enables Drag-and-drop Images, Batch Processing, And Export To TXT/PDF. The Implementation Leverages Open Libraries For Computer Vision And OCR, Supports On-device Processing For Privacy, And Can Fall Back To Online Translation Services For Higher Quality. Experiments On Multilingual Datasets Evaluate OCR Accuracy, Translation Quality (BLEU/chrF), And Latency Across Device Profiles. Results Show That Careful Preprocessing And Model Selection Substantially Improve End-to-end Quality, Making The Tool Practical For Education, Travel, And Accessibility Use Cases (e.g., Assisting Low-vision Users). The System’s Modular Design Facilitates Future Upgrades, Including Domain-specific Glossaries And Fully Offline Neural Translation.
The Aadhaar Card Is One Of The Most Widely Used Identification Documents In India, Containing Essential Demographic Details And A Unique 12-digit Identification Number. With The Rapid Digitalization Of Services, Automating The Extraction Of Aadhaar Card Information Has Become A Critical Requirement For Various Applications Such As E-KYC, Digital Onboarding, And Identity Verification. This Project Aims To Develop A Web-based Application That Extracts Aadhaar Card Details And The Profile Image Using Optical Character Recognition (OCR) And Haarcascade-based Face Detection Techniques. The System Utilizes OCR To Identify And Extract Textual Information Such As The Aadhaar Number, Name, Date Of Birth, Gender, And Address Directly From The Scanned Or Uploaded Aadhaar Card Image. Simultaneously, Haarcascade, A Machine Learning-based Object Detection Algorithm, Is Employed To Detect And Extract The Profile Image From The Card. The Extracted Details Are Then Structured And Displayed On The Webpage For Further Use In Authentication Or Record Management. This Approach Minimizes Manual Data Entry, Reduces Human Error, And Accelerates Digital Onboarding Processes. The Integration Of OCR And Haarcascade Ensures Efficient Extraction Of Both Textual And Image Components, Making The System Reliable, Scalable, And Suitable For Real-world Identity Verification Applications.
Stone Inscriptions Are One Of The Most Significant Sources Of Historical, Cultural, And Linguistic Knowledge, Especially For Understanding Ancient Civilizations. In Maharashtra, Stone Inscriptions Written In Early Forms Of The Marathi Script Provide Invaluable Insights Into Socio-political, Religious, And Cultural Developments Of Their Time. However, The Manual Study Of Such Inscriptions Is Challenging Due To Natural Weathering, Erosion, Script Variations, And The Complexity Of Ancient Writing Styles. In Recent Years, Optical Character Recognition (OCR) Has Emerged As A Powerful Tool To Digitize And Analyze Ancient Scripts, Enabling Automated Recognition And Preservation Of Historical Texts. This Paper Presents A Comprehensive Survey Of Existing Methods And Approaches For Ancient Marathi Script Recognition From Stone Inscriptions. The Study Reviews Key Preprocessing Techniques Such As Image Enhancement, Noise Reduction, And Segmentation; Feature Extraction Methods Including Structural, Statistical, And Deep Learning-based Approaches; And Classification Models Ranging From Traditional Machine Learning To Modern Convolutional Neural Networks. Challenges Such As Degraded Surfaces, Broken Characters, And Script Variability Are Discussed Alongside Potential Solutions. The Survey Also Highlights Future Research Directions, Particularly The Integration Of Deep Learning, Transfer Learning, And Multimodal Analysis For Improved Accuracy. By Compiling And Analyzing Existing Research, This Work Aims To Provide A Foundation For Developing Robust OCR Systems Tailored To Ancient Marathi Stone Inscriptions, Thereby Contributing To Digital Preservation And Historical Scholarship.
Machine And Deep Learning Applications Play A Dominant Role In The Current Scenario In The Agriculture Sector. To Date, The Classification Of Fruits Using Image Features Has Attained The Researcher’s Attraction Very Much From The Last Few Years. Fruit Recognition And Classification Is An Ill-posed Problem Due To The Heterogeneous Nature Of Fruits. In The Proposed Work, Convolution Neural Network (CNN), Recurrent Neural Network (RNN), And Long-short Term Memory (LSTM) Deep Learning Methods Are Used To Extract The Optimal Image Features, And To Select Features After Extraction, And Finally, Use Extracted Image Features To Classify The Fruits. To Evaluate The Performance Of The Proposed Approach, The Support Vector Machine (SVM) Unsupervised Learning Method, Artificial Neuro-fuzzy Inference System (ANFIS), And Feed-forward Neural Network (FFNN) Classification Results Are Compared, And Observed That The Proposed Fruit Classification Approach Results Are Quite Efficient And Promising.
Machine Learning Is An Important Decision Support Tool For Crop Yield Prediction, Including Supporting Decisions On What Crops To Grow And What To Do During The Growing Season Of The Crops. Several Machine Learning Algorithms Have Been Applied To Support Crop Yield Prediction Research. In This Study, We Performed A Systematic Literature Review (SLR) To Extract And Synthesize The Algorithms And Features That Have Been Used In Crop Yield Prediction Studies. Based On Our Search Criteria, We Retrieved 567 Relevant Studies From Six Electronic Databases, Of Which We Have Selected 50 Studies For Further Analysis Using Inclusion And Exclusion Criteria. We Investigated These Selected Studies Carefully, Analyzed The Methods And Features Used, And Provided Suggestions For Further Research. According To Our Analysis, The Most Used Features Are Temperature, Rainfall, And Soil Type, And The Most Applied Algorithm Is Artificial Neural Networks In These Models. After This Observation Based On The Analysis Of Machine Learning-based 50 Papers, We Performed An Additional Search In Electronic Databases To Identify Deep Learning-based Studies, Reached 30 Deep Learning-based Papers, And Extracted The Applied Deep Learning Algorithms. According To This Additional Analysis, Convolutional Neural Networks (CNN) Is The Most Widely Used Deep Learning Algorithm In These Studies, And The Other Widely Used Deep Learning Algorithms Are Long-Short Term Memory (LSTM) And Deep Neural Networks (DNN).
This Paper Presents A Unique Method To Classify Food Items And Estimate The Calorie Content Based On Photos By Combining Convolutional Neural Networks (CNNs) With Image Processing Techniques. To Improve The Quality Of The Dataset, We First Curate A Wide Range Of Food Photographs From Different Presentation Styles And Cuisines. We Do This By Applying Pre-processing Techniques Including Image Segmentation And Feature Extraction. Next, Using The CNN's Capacity To Extract Hierarchical Features From Raw Pixel Data, A Custom Deep CNN Architecture Is Trained On This Dataset To Efficiently Classify Diverse Food Items And Achieve High Accuracy In Differentiating Between Different Dishes. Furthermore, We Tackle The Problem Of Calorie Prediction By Employing Regression Models That Include Features Extracted From The CNN, Which Allows One To Forecast The Calorie Content Of A Food Item Based On Its Visual Attributes. Our Technique Intends To Enable Users To Make More Educated Nutritional Decisions And Better Control Their Caloric Intake By Fusing Image Categorization With Calorie Prediction. The Suggested Approach Shows Encouraging Results In Terms Of Food Item Classification Accuracy And Precision In Calorie Estimate. It’s Possible Uses Include Promoting Better Eating Practices And Assisting Individuals, Dietitians, And The Food Industry With Their Dietary Monitoring. In Addition, Our Technology Gives Customers Access To Weekly Calorie Intake Information, Which Can Help Them Avoid Obesity-related Illnesses Like Diabetes.
Object Detection Is One Of The Key Areas For All The Researchers In The Field Of Computer Science. The Research Is To Find The Types Of Objects In The Image And Provide Their Temporal And Spatial Characteristics. In The Recent Times There Have Been A Lot Of Improvements In The Field Of Satellite Image Detection With Varying Data Sets Being Available And Has Left High Impact On The Performance Of Such Analysis. A Number Of Algorithms Have Evolved Over A Period Of Time In Object Detection And Analysis Namely Different Versions Of YOLO, CNN, DETR Etc. There Is A Need To Deploy A Study Which Enables The Performance Comparison Of Different Versions Of These Algorithms On Specific Data Set To Understand Their Efficacy. The Study In The Paper Contributes To Understanding, Evaluating And Analyzing The Performance Characteristics Of YOLO V7 Algorithm With Varying Parameters.
Agriculture Is The Backbone Of Indian Economy And Livelihood To Many People. The Use Of Computer Science In The Field Of Agriculture Will Potentially Solve Many Problems Faced By Farmers. Farmers Often Choose Crops For Their Field Based On Their Own Experience And Instinct. This Sometimes Leads To Loss And Less Yield. If The Selection Of Crops Is Done With Productivity Data Of The Entire Region, It May Lead To Better Results.However All The Crops Cannot Be Cultivated In A Particular Soil. So The Soil Must Be Analysed And Crops Must Be Suggested Based On The Type Of Soil. Many Soil Classification Techniques Involve Testing In Laboratories Whichmight Not Be Affordable And Available To All The Farmers. This Work Suggests An Idea That Is Useful And Easily Accessible To All The Farmers In India Without Any Need Of Hardware. A List Of Crops With Their Success Rate Will Be Suggested To The Farmer When The Region Of Agriculture And Soil Image (used For Agriculture) Are Given As Inputs. This List Of Crops Are Both Profitable And Produce More Yield In That Region. The Results Obtained Are Promising. An Accuracy Of 94% Is Achieved In The Soil Classification Module.The Success Rate For The Crops Obtained Are Realistic With The Agricultural Practices In The Region. The Web Application Developed Is Extremely User Friendly And Easy To Use By The Farmers.
The Widespread Adoption Of QR Codes Has Revolutionized Various Industries, Streamlined Transactions And Improved Inventory Management. However, This Increased Reliance On QR Code Technology Also Exposes It To Potential Security Risks That Malicious Actors Can Exploit. QR Code Phishing, Or “Quishing”, Is A Type Of Phishing Attack That Leverages QR Codes To Deceive Individuals Into Visiting Malicious Websites Or Downloading Harmful Software. These Attacks Can Be Particularly Effective Due To The Growing Popularity And Trust In QR Codes. This Paper Examines The Importance Of Enhancing The Security Of QR Codes Through The Utilization Of Artificial Intelligence (AI). The Abstract Investigates The Integration Of AI Methods For Identifying And Mitigating Security Threats Associated With QR Code Usage. By Assessing The Current State Of QR Code Security And Evaluating The Effectiveness Of AI-driven Solutions, This Research Aims To Propose Comprehensive Strategies For Strengthening QR Code Technology’s Resilience. The Study Contributes To Discussions On Secure Data Encoding And Retrieval, Providing Valuable Insights Into The Evolving Synergy Between QR Codes And AI For The Advancement Of Secure Digital Communication.
Phishing Is An Internet Scam In Which An Attacker Sends Out Fake Messages That Look To Come From A Trusted Source. A URL Or File Will Be Included In The Mail, Which When Clicked Will Steal Personal Information Or Infect A Computer With A Virus. Traditionally, Phishing Attempts Were Carried Out Through Wide-scale Spam Campaigns That Targeted Broad Groups Of People Indiscriminately. The Goal Was To Get As Many People To Click On A Link Or Open An Infected File As Possible. There Are Various Approaches To Detect This Type Of Attack. One Of The Approaches Is Machine Learning. The URL’s Received By The User Will Be Given Input To The Machine Learning Model Then The Algorithm Will Process The Input And Display The Output Whether It Is Phishing Or Legitimate. There Are Various ML Algorithms Like SVM, Neural Networks, Random Forest, Decision Tree, XG Boost Etc. That Can Be Used To Classify These URLs. The Proposed Approach Deals With The Random Forest, Decision Tree Classifiers. The Proposed Approach Effectively Classified The Phishing And Legitimate URLs With An Accuracy Of 87.0% And 82.4% For Random Forest And Decision Tree Classifiers Respectively.
As Seen In The Past Few Decades, It Is Very Common To Observe The Patient’s Paper Work At The Hospital. Even Though The Same Personal Information Is Used, An Unusual Way To Actually Decrement The Amount Of These Paper Works Does Not Exist. The Development Of Mobile Web Provides Development Direction For Medical Industry And A New Service Mode. In This Paper, We Introduce QR Code Based E-health Authentication System To Obtain Patient’s Health Record Easily And Securely In The Local Hospital And Also To Reduce The Redundant Paper Work. One Of The Aims Of This Project Is To Use The Dataset And Machine Learning Techniques To Predict The Type Of Disease Based On The Symptoms. A QR Code Which Includes Predicted Disease And Personal Information Of Patient Is Sent To The Doctor Automatically Via Email. Further The Doctor Sends A QR Code Generated Prescription To The Patient Which Is Scanned By The Pharmacist .Here, We Describe An Integrated System, Developed For Use By The Healthcare Personnel Within Healthcare Facilities, Adapted To All Handheld Devices .With Our Proposed Scheme, We Believe That It Will Improve Efficiency In Terms Of The Cost And Time For The Patient, Hospital And The Doctor And Protect Patient’s Personal Information.
Text-based Password Authentication Is A Common Method Used To Verify The Identity Of Users Who Are Trying To Access A Secure System Or Service. In Order To Use This Authentication Method, The User Must Input A Password Or Other Secret Phrase That Is Then Compared To A Server-side Copy Of The Same Password. Access Is Given If The Password Typed Matches The One Saved. Graphical Password Authentication Is A Type Of User Authentication That Involves Using Images Or Visual Elements Instead Of Alphanumeric Characters To Verify The Identity Of The User. Unlike Traditional Text-based Passwords, Graphical Passwords Offer An Intuitive And User-friendly Way Of Authentication, As They Rely On The User's Ability To Remember Pictures, Shapes, And Patterns. This Technology Has Been Developed To Address The Limitations Of Traditional Text-based Passwords, Such As The Difficulty Of Creating And Remembering Complex Passwords, And The Vulnerability To Brute-force Attacks. Compared To Conventional Text-based Passwords, Graphical Password Authentication Has A Number Of Benefits, Including Better Usability And Higher Security.
Films Are A Significant Form Of Entertainment Form Of Entertainment In Modern Society, With Filmmakers Investing Substantial Resources Into Their Production. However, This Effort Is Undermined By Piracy, Where Individuals Copy And Distribute Film Content Illegally, Often By Recording Movies With Portable Often By Recording Movies With Portable Cameras And Uploading Them To Online Platforms. Camcorder Theft In Particular Has A Significant Impact On The Film Industry. Despite Efforts To Track Pirates, Watermarking In Pirated Movies Is Often Undetectable, Making It Difficult To Deter Piracy Effectively.To Address This Issue, This Paper Proposes Two Innovative Solutions. Firstly, It Suggests Embedding A Secret Key Using Steganography Via MATLAB To Secure Movies Files. Steganography Allows For The Concealment Of Information Within Digital Media, Providing A Covert Means Of Protection. Secondly, This Recommendation Involves Constructing A Screen Fitted With An Infrared Transmitter Which Would Prevent People From Filming Illegally In Cinemas. The Idea Behind This System Is That It Emits Infrared Signals At The Same Time As Films Are Being Shown Thereby Making Recording Impossible. Also GSM Technology Can Be Used To Send Quick SMS Alerts To Authorized Personal Whenever There Is An Attempt At Piracy For Immediate Response. On The Whole, These Measures Are Designed To Combat Theater Piracy By Making Movies More Secure Against Illegal Duplication While At The Same Time Preventing People From Recording Them Without Permission.
It Is Estimated That Globally 425 Million Subjects Have Moderate To Severe Obstructive Sleep Apnea (OSA). The Accurate Prediction Of Sleep Apnea Events Can Offer Insight Into The Development Of Treatment Therapies. However, Research Related To This Prediction Is Currently Limited. We Developed A Covert Framework For The Prediction Of Sleep Apnea Events Based On Low-frequency Breathing-induced Vibrations Obtained From Piezoelectric Sensors. A CNN-transformer Network Was Utilized To Efficiently Extract Local And Global Features From Respiratory Vibration Signals For Accurate Prediction. Our Study Involved Overnight Recordings Of 105 Subjects. In Five-fold Cross-validation, We Achieved An Accuracy Of 85.9% And An F1 Score Of 85.8%, Which Are 3.5% And 5.3% Higher Than The Best-performed Classical Model, Respectively. Additionally, In Leave-one-out Cross-validation, 2.3% And 3.8% Improvements Are Observed, Respectively. Our Proposed CNN-transformer Model Is Effective In The Prediction Of Sleep Apnea Events. Our Framework Can Thus Provide A New Perspective For Improving OSA Treatment Modes And Clinical Management.
As Quantum Computing Evolves From Theoretical Promise To Emerging Reality, The Urgency To Develop Quantum-resilient Data Protection Mechanisms Becomes Increasingly Paramount-particularly Within Critical Infrastructure Systems Dependent On Multi-cloud Architectures. This Study Explores The Design And Deployment Of Quantum-resilient Encryption Protocols Tailored To Secure Sensitive Data Flows Across Heterogeneous And Decentralized Cloud Environments Supporting Energy, Transportation, Defense, And Healthcare Infrastructures. Beginning From A Broader Analysis Of Cryptographic Vulnerabilities Posed By Quantum Adversaries-especially Those Exploiting Shor's And Grover's Algorithms-the Paper Highlights The Limitations Of Current Asymmetric Key Systems And Symmetric Encryption Practices In Multi-cloud Data Orchestration. Building On This Foundation, The Research Narrows In On Post-quantum Cryptographic (PQC) Frameworks, Including Lattice-based, Code-based, And Multivariate Polynomial Schemes, Evaluating Their Performance And Adaptability For Dynamic Cloud-native Systems. A Key Focus Is Placed On Designing Lightweight, Interoperable Encryption Protocols That Can Seamlessly Integrate With Federated Identity Management, Zero-trust Security Models, And Real-time Data Streams Without Introducing Prohibitive Latency Or Computational Burden. The Study Also Presents An Architectural Model That Allows For Real-time Key Negotiation, Distributed Trust Management, And Algorithm Agility, Ensuring Compliance With Both Current And Forward-looking Regulatory Standards (e.g., NIST PQC Guidelines). Simulation And Benchmarking Conducted Across Hybrid Cloud Environments Demonstrate That Carefully Optimized Quantum-resilient Protocols Can Be Implemented Without Compromising System Availability Or Scalability. The Results Validate The Feasibility Of Transitioning From Conventional Cryptography To Quantum-safe Models In Mission-critical Multi-cloud Operations. The Paper Concludes By Offering A Strategic Roadmap For Organizations Seeking To Future-proof Their Cloud Infrastructures Against Quantum-era Threats.
Nowadays, Distributed Data Processing In Cloud Computing Has Gained Increasing Attention From Many Researchers. The Intense Transfer Of Data Has Made The Network An Attractive And Vulnerable Target For Attackers To Exploit And Experiment With Different Types Of Attacks. Therefore, Many Intrusion Detection Techniques Have Been Evolving To Protect Cloud Distributed Services By Detecting The Different Attack Types On The Network. Machine Learning Techniques Have Been Heavily Applied In Intrusion Detection Systems With Different Algorithms. This Paper Surveys Recent Research Advances Linked To Machine Learning Techniques. We Review Some Representative Algorithms And Discuss Their Proprieties In Detail. We Compare Them In Terms Of Intrusion Accuracy And Detection Rate Using Different Data Sets.
Road Safety For Two-wheeler Riders Remains A Critical Issue Due To Frequent Non-compliance With Helmet-wearing Regulations, Leading To Numerous Road Fatalities, This Report Presents An Advanced Traffic Monitoring System That Utilizes Machine Learning To Automatically Detect Helmet Violations, Recognize Vehicle Number Plates, And Calculate Fines. Real-time Helmet Detection Is Achieved Using YOLOv5, While CNN And OCR Are Employed For Reading Number Plates. Upon Detecting A Violation, The System Captures The Vehicle's Plate, Calculates Fines According To Predefined Rules, And Stores The Data For Further Action. Designed To Operate Under Various Lighting And Environmental Conditions, The System Aims To Enhance Traffic Management, Reduce Accidents, And Ensure Compliance With Minimal Human Involvement.
A Crime Is A Deliberate Act That Can Cause Physical Or Psychological Harm, As Well As Property Damage Or Loss, And Can Lead To Punishment By A State Or Other Authority According To The Severity Of The Crime. The Number And Forms Of Criminal Activities Are Increasing At An Alarming Rate, Forcing Agencies To Develop Efficient Methods To Take Preventive Measures. In The Current Scenario Of Rapidly Increasing Crime, Traditional Crime-solving Techniques Are Unable To Deliver Results, Being Slow Paced And Less Efficient. Thus, If We Can Come Up With Ways To Predict Crime, In Detail, Before It Occurs, Or Come Up With A “machine” That Can Assist Police Officers, It Would Lift The Burden Of Police And Help In Preventing Crimes. To Achieve This, We Suggest Including Machine Learning (ML) And Computer Vision Algorithms And Techniques. In This Paper, We Describe The Results Of Certain Cases Where Such Approaches Were Used, And Which Motivated Us To Pursue Further Research In This Field. The Main Reason For The Change In Crime Detection And Prevention Lies In The Before And After Statistical Observations Of The Authorities Using Such Techniques. The Sole Purpose Of This Study Is To Determine How A Combination Of ML And Computer Vision Can Be Used By Law Agencies Or Authorities To Detect, Prevent, And Solve Crimes At A Much More Accurate And Faster Rate. In Summary, ML And Computer Vision Techniques Can Bring About An Evolution In Law Agencies.
Assistive Technologies That Utilize Object Detection Have The Potential To Significantly Improve The Independence And Safety Of Visually Impaired Individuals. Object Detection Is The Process Of Identifying And Localizing Objects Within An Image Or Video Sequence, Typically Using Deep Learning Algorithms Trained On Annotated Datasets. For The Blind, Object Detection Systems Can Be Used To Provide Real-time Feedback About The Presence And Location Of Objects In The Environment, Using Sensors Or Cameras Mounted On A Wearable Device. This Technology Has Been Implemented In Various Systems, Such As The Seeing AI App Developed By Microsoft, The Horus System Developed By The University Of Michigan, And The NAVIS System Developed By The University Of California. Ongoing Research In This Field Is Focused On Improving The Accuracy And Speed Of Object Detection Algorithms And Developing New Wearable Devices And Sensor Technologies To Make Object Detection Systems More Practical And Accessible For Everyday Use. To Ensure Real-time Performance, The System Is Optimized For Deployment On Edge Devices With Limited Computational Resources. The Architecture Is Designed To Balance Accuracy And Efficiency, Considering The Constraints Imposed By Mobile And Wearable Devices Commonly Used By Visually Impaired Individuals. Additionally, The Proposed Solution Aims To Be Robust To Varying Environmental Conditions, Including Changes In Lighting, Occlusions And Object Orientations. In This Report Paper, We Provide An Overview Of The Current State Of Object Detection For The Blind And Highlight The Potential Impact Of This Technology In Improving The Quality Of Life For Visually Impaired Individuals.
Emotion Recognition And Generation Have Emerged As Crucial Topics In Artificial Intelligence Research, Playing A Significant Role In Enhancing Human-computer Interaction Within Healthcare, Customer Service, And Other Fields. Although Several Reviews Have Been Conducted On Emotion Recognition And Generation As Separate Entities, Many Of These Works Are Either Fragmented Or Limited To Specific Methodologies, Lacking A Comprehensive Overview Of Recent Developments And Trends Across Different Modalities. In This Survey, We Provide A Holistic Review Aimed At Researchers Beginning Their Exploration In Emotion Recognition And Generation. We Introduce The Fundamental Principles Underlying Emotion Recognition And Generation Across Facial, Vocal, And Textual Modalities. This Work Categorises Recent State-of-the-art Research Into Distinct Technical Approaches And Explains The Theoretical Foundations And Motivations Behind These Methodologies, Offering A Clearer Understanding Of Their Application. Moreover, We Discuss Evaluation Metrics, Comparative Analyses, And Current Limitations, Shedding Light On The Challenges Faced By Researchers In The Field. Finally, We Propose Future Research Directions To Address These Challenges And Encourage Further Exploration Into Developing Robust, Effective, And Ethically Responsible Emotion Recognition And Generation Systems.
Accidents Involving Pedestrians Is One Of The Leading Causes Of Death And Injury Around The World. Intelligent Driver Support Systems Hold A Promise To Minimize Accidents And Save Many Lives. Such A System Would Detect The Pedestrian, Predict The Possibility Of Collision, And Then Warn The Driver Or Engage Automatic Braking Or Other Safety Devices. This Chapter Describes The Framework And Issues Involved In Developing A Pedestrian Protection System. It Is Emphasized That The Knowledge Of The State Of The Environment, Vehicle, And Driver Are Important For Enhancing Safety. Classification, Clustering, And Machine Learning Techniques For Effectively Detecting Pedestrians Are Discussed, Including The Application Of Algorithms Such As SVM, Neural Networks, And AdaBoost For The Purpose Of Distinguishing Pedestrians From Background. Pedestrians Unlike Vehicles Are Capable Of Sharp Turns And Speed Changes, Therefore Their Future Paths Are Difficult To Predict. In Order To Estimate The Possibility Of Collision, A Probabilistic Framework For Pedestrian Path Prediction Is Described Along With Related Research. It Is Noted That Sensors In Vehicle Are Not Always Sufficient To Detect All The Pedestrians And Other Obstacles.
Object Detection Is One Of The Key Areas For All The Researchers In The Field Of Computer Science. The Research Is To Find The Types Of Objects In The Image And Provide Their Temporal And Spatial Characteristics. In The Recent Times There Have Been A Lot Of Improvements In The Field Of Satellite Image Detection With Varying Data Sets Being Available And Has Left High Impact On The Performance Of Such Analysis. A Number Of Algorithms Have Evolved Over A Period Of Time In Object Detection And Analysis Namely Different Versions Of YOLO, CNN, DETR Etc. There Is A Need To Deploy A Study Which Enables The Performance Comparison Of Different Versions Of These Algorithms On Specific Data Set To Understand Their Efficacy. The Study In The Paper Contributes To Understanding, Evaluating And Analyzing The Performance Characteristics Of YOLO V7 Algorithm With Varying Parameters.
Animal Recognition And Tracking Systems Have The Potential To Provide Accurate And Cost-effective Monitoring Of Animal Populations And Their Behaviour In The Wild, Which Can Be Crucial For Conservation Efforts And Ecological Research. This Study Explores The Utilization Of The YOLO Version 3 Object Detection And Recognition Algorithm, In Conjunction With Firebase Technology, To Trigger An Application That Sends Real-time Data To Nearby Safari Vehicles For Animal Tracking Purposes. However, It Is Important To Consider The Ethical And Privacy Concerns Of Both The Animals And The People Involved When Developing And Using Animal Recognition And Tracking Systems And To Ensure That Adequate Training And Support Are Provided For Users, Particularly In Cases Where It Involves Interaction With Wild Animals.
Student Attendance Plays Significant Role In Order To Justify Academic Outcome Of A Student And School As Overall. Unfortunately, There Is No Automated Attendance Record Keeping Application Available In Malaysia’s Secondary Schools. A Preliminary Study Has Been Conducted In One Of Secondary Schools In Selangor, Malaysia In Order To Understand The Manual Attendance Record Keeping Process. Through Interview Session, Student Attendance System (SAS) Development Team, Have Identified That Teachers And School Management Face Problems In Recording And Managing Attendance Of Their Students. Therefore, SAS Has Been Proposed And Developed. Need For A Tool To Systematically Keep The Students Attendance Record Increased Due To Increasing Number Of School Students. Upon Completion Of SAS, User Acceptance Testing Conducted Among Potential End Users. Result Of UAT Shows Most Of The User Satisfied With The System With Some Minor Changes Required.
This Paper Proposes A Method To Measure A Person's Heart Rate From A Frontal Face Video With The Presence Of Motion Using A Webcam. Histogram Of Oriented Gradients (HOG) Facial Detection Was Used To Detect Face And Regions Of Interest (ROIs). A Photoplethysmography (PPG) Signal Was Extracted From The Green Colour Channel From The ROIs. A Detrending Filter, Hamming Window, And Bandpass Filter Were Applied Onto The PPG Signal To Remove Noise. Fast Fourier Transform (FFT) Was Applied To The Filtered PPG Signal, And The Heart Rate Measurement Was Extracted From The PPG Signal Using Power Spectral Analysis. The Results From This Method Were Compared To An Orange Theory Fitbit Monitor. The Results Were Very Similar.
A Blood Bank Is A Bank Of Blood Or Blood Components, Gathered As A Result Of Blood Donation, Stored And Preserved For Later Use In Blood Transfusion..To Provide Web Based Communication There Are Numbers Of Online Web Based Blood Bank Management System Exists For Communicating Between Department Of Blood Centers And Hospitals, To Satisfy Blood Necessity, To Buy, Sale And Stock The Blood, To Give Information About This Blood. Manual Systems As Compared To Computer Based Information Systems Are Time Consuming, Laborious, And Costly. This Paper Introduces The Review Of The Main Features, Merits And Demerits Provided By The Existing Web-Based Information System For Blood Banks. This Study Shows The Comparison Of Various Existing System And Provide Some More Idea For Improve The Existing System. First I Will Give Some Basic Introduction About Blood Banks Then I Will Try To Provide Comparative Study Of Some Existing Web Based Blood Bank System. After That I Will Introduce Some New Idea For Improving The Existing Techniques Used In Web Based Blood Bank System And At End I Will Conclude This Paper.
Agriculture Has Long Been The Backbone Of The Indian Economy, Defining The Country’s Social And Cultural Milieu. Some Of The Most Common Difficulties That Farmers Face Includes Selecting Suitable Crops For Their Area And Utilizing The Necessary Fertilizers, Which Can Lead To A Drop In Production. To Address These Challenges, Precision Agriculture Is Employed. Precision Agriculture Is A Smart And Modern Farming Methodology That Uses IoT-based Research To Provide Information On Soil Qualities, Types Of Soil, And Crop Production Statistics. This Assists Farmers In Choosing The Optimum Crops To Plant And Provides Fertilizer Recommendations Based On Specific Location Characteristics, Also Predicting The Likelihood Of Plant Diseases. In This Project, We Introduce A Recommendation System By Applying Machine Learning Models Such As Support Vector Machine (SVM), Logistic Regression, Decision Tree, Random Forest, And Naïve Bayes On A Crop Dataset, Which Is Entirely IoT-based. The System Recommends The Best Crop For The Location’s Particular Parameters With The Highest Accuracy. The System Achieved An Accuracy Of 99% For Crop Recommendation And 92% For Plant Disease Prediction. This Research Utilizes Soil Nutrients And PH Data As Inputs To Create A Website That Predicts The Most Suitable Crops For A Particular Soil Type And Suggests Appropriate Fertilizers. Additionally, The System Predicts Plant Diseases And Provides Information About The Diseases, Including Suggested Cures. Thousands Of Farmers Will Be Able To Access The System Through A Flask-based Web Interface.
The Aspiration Of The “Employee Stress Management System” Is To Identify The Employees Under Stress Within Companies Of Various Work Environments And Embedded Remote Work Adaptations To Help Raise The Balance In Their Work, Life, And Health. Since The COVID-19 Epidemic, Most Companies Transformed Their Working Styles Into Unusual Spaces Such As Working From Home Leaving Employees With Ambiguity And Stress In Managing Their Working Goals. So, It Is Need Of The Hour For The Company Executives To Undertake This Prediction As Assistance To Conduct Appropriate Remediation To Help Employees Balance Their Work And Manage Performance Outcomes. Thus, This Project Is Pivotal In The Work-life Which Uses Machine Learning Algorithms To Analyze The Database And To Perform Prediction Analysis In Determining Stressed Employees. The Main Framework Of The Project Relies On Python With Provided Graphical User Interface Including Visual Graphs And Heatmap For Scrutiny By The Company’s Management Along With Prediction Results.
It's No Secret That The Microblogging Service Twitter (X) Has Quickly Risen To Prominence As One Of The Most Reliable Places To Get The Latest Updates On Breaking Events. Tweets, Twitter's Information Streams, Are Sent Out Voluntarily By Registered Users And Can Reach Even Non-registered Users, Often Before More Conventional Sources Of Mass News. In This Research, We Use Machine Learning To Create Models That Can Find Helpful Tweets On Disasters Automatically. Social Media Users Provide Massive Amounts Of Data During Natural Catastrophe Situations, Some Of Which Are Useful For Relief Operations And Emergency Management. In This Work, We Analyze The Material Shared On Social Media During Two Hurricanes And One Earthquake. This Research Has Shown A Machine-learning Approach To Categorizing Tweets In Relation To Disasters And Labeling Twitter Data. This Study Has Applied Five Machine Learning Algorithms To Predict Disaster
Water Pollution Is One Of The Biggest Fears For The Green Globalization. In Order To Ensure The Safe Supply Of The Drinking Water The Quality Needs To Be Monitor In Real Time. In This Paper We Present A Design And Development Of A Low Cost System For Real Time Monitoring Of The Water Quality In IOT(internet Of Things).The System Consist Of Several Sensors Is Used To Measuring Physical And Chemical Parameters Of The Water. The Parameters Such As Temperature, PH, Turbidity, Flow Sensor Of The Water Can Be Measured. The Measured Values From The Sensors Can Be Processed By The Core Controller. The Arduino Model Can Be Used As A Core Controller. Finally, The Sensor Data Can Be Viewed On Internet Using WI-FI System.
Resume Screening Is The Process Of Analyzing The Resumes Where The Candidates Apply For The Different Types Of Jobs Where The Company Feel The Tedious Job To Find The Appropriate Candidate Due To The Complexity In Resumes Formats Since It Has Different Styles. As A Result, Selecting Applicants For The Appropriate Job Within A Company Is A Difficult Task For Recruiters. We Can Extract The Key Information From The CV Using NLTK, Natural Language Processing (NLP) Techniques To Save Time And Effort. This System Could Work With A Large Number Of Resumes For Classifying The Right Categories Using Different Classifiers Like KNN, SVM, MLP, LR. Furthermore, This System Attempts To Find The Accuracy And Performance Of The Proposed Methodology And Incorporate It In The IT Firms And Other Regulations For The Prevention Of Manual Screening And Establish A Safe Allocation Of Resources For The Companies.
With The Advancement In Technology, There Are So Many Enhancements In The Banking Sector Also. The Number Of Applications Is Increasing Every Day For Loan Approval. There Are Some Bank Policies That They Have To Consider While Selecting An Applicant For Loan Approval. Based On Some Parameters, The Bank Has To Decide Which One Is Best For Approval. It Is Tough And Risky To Check Out Manually Every Person And Then Recommended For Loan Approval. In This Work, We Use A Machine Learning Technique That Will Predict The Person Who Is Reliable For A Loan, Based On The Previous Record Of The Person Whom The Loan Amount Is Accredited Before. This Work’s Primary Objective Is To Predict Whether The Loan Approval To A Specific Individual Is Safe Or Not.
Autism Spectrum Disorder (ASD) Represents A Multifaceted Neuro-developmental State That Presents Significant Difficulties In Its Early Identification And Intervention. This Survey Explores The Recent Advancements And Methodologies In ASD Detection Leveraging Machine Learning (ML), Deep Learning (DL), And Neuroimaging Techniques. An Extensive Survey Of Literature Between 2018 And 2023 Reveals A Paradigm Shift In Diagnostic Approaches, Emphasizing The Integration Of ML Algorithms, Like Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), And Decision-making Models, In Conjunction With Various Neuro-imaging Modalities Like Magnetic Resonance Imaging (MRI), Electroencephalography (EEG), And Functional Near-Infrared Spectroscopy (fNIRS). These Modalities Facilitate The Identification Of Distinctive Biomarkers, Behavioral Patterns, And Neural Correlates Associated With ASD. The Survey Also Looks At Potential Ethical Issues, The Importance Of Early Detection Using ML-driven Methodologies, And The Changing Diagnostic Tool Landscape That Aims To Offer Timely And Individualized Interventions For People With ASD. The Combination Of These Data Demonstrates The Revolutionary Effect Of ML, DL, And Neuro-imaging In Improving The Accuracy Of ASD Detection, Allowing Access To Additional Potent Intervention Methods And A More Thorough Understanding Of The Neurobiology Underlying The Condition.
Over The Years, Researchers Have Developed Several Expert Systems To Help Cardiologists Improve The Diagnostic Process By Predicting Heart Diseases Early On. Most Of The Available Machine Learning Approaches Are Complicated And They Were Generally Created For Use With Large Data. Unfortunately, These Approaches Can’t Be Effectively Used In The Scenarios With Small Data To Train The Model. In This View, This Paper Proposes A Simple And Effective Diagnostic System That Uses Extreme Gradient Boosting (XGBoost) With Feature Selection Algorithm To Predict Heart Disease In Case Of Dataset With Less Records. Proper Hyperparameter Tuning Is Vital For The Effective Deployment Of Any Classifier. To Improve The Hyperparameters Of XGBoost, Grid Search Is Employed, Which Is An Optimal Method For Hyperparameter Optimization. Also, The One-Hot (OH) Encoding Approach Is Employed To Encode Categorical Information In Cleveland Heart Disease Dataset. To Evaluate The Proposed Work, The Suggested Model Is Assessed And Compared To Other Classifiers. The Proposed Model Achieved An Area Under Curve (AUC) Of 0.853 And Prediction Accuracy Of 85.96%. From The Experimental Results, The Proposed Model Achieved Higher Accuracy When Compared To The Other Models.
The Paper Presents Development Of A Smart Infant Monitoring System Using Multiple Non-invasive Sensors To Detect Various Physiological Functions. The System Can Evaluate Different Physiological Activities Such As Respiration, Movement, Noise, Position, As Well As Ambient Temperature, And Humidity. By Processing The Acquired Data From Different Sensor Modules, The System Can Generate Alarm Signals For Adverse Situations Such As The Occurrence Of Apnea, Seizure, Or Noisy And Uncomfortable Environmental Conditions. The System Will Also Be Able To Detect Critical Respiratory Conditions By Analyzing Breathing Data And Saturated Blood Oxygen Level (SpO2) Using Machine Learning (ML) Models Such As Neural Networks. The Proposed System Allows The Caregiver To Monitor The Condition Of The Patient From A Remote Location By Implementing Wireless Communication With A Remote Computer Or A Cell Phone.
Accurate Diagnosis Of Parkinson's Disease (PD) At An Early Stage Is Challenging For Clinicians As Its Progression Is Very Slow. Currently Many Machine Learning And Deep Learning Approaches Are Used For Detection Of PD And They Are Popular Too. This Study Proposes Four Deep Learning Models And A Hybrid Model For The Early Detection Of PD. Further To Improve The Performance Of The Models, Grey Wolf Optimization (GWO) Is Used To Automatically Fine-tune The Hyperparameters Of The Models. The Simulation Study Is Carried Out Using Two Standard Datasets, T1,T2-weighted And SPECT DaTscan. The Metaherustic Enhanced Deep Learning Models Used Are GWO-VGG16, GWO-DenseNet, GWO-DenseNet + LSTM, GWO-InceptionV3 And GWO-VGG16 + InceptionV3. Simulation Results Demonstrated That All The Models Perform Well And Obtained Near Above 99% Of Accuracy. The AUC-ROC Score Of 99.99 Is Achieved By The GWO-VGG16 + InceptionV3 And GWO-DenseNet Models For T1, T2-weighted Dataset. Similarly, The GWO-DenseNet, GWO-InceptionV3 And GWO-VGG16 + InceptionV3 Models Result An AUC-ROC Score Of 100 For SPECT DaTscan Dataset.
Alzheimer's Disease (AD) Is A Progressive Neurological Disease Considered The Most Common Form Of Late-stage Dementia. Usually, AD Leads To A Reduction In Brain Volume, Impacting Various Functions. This Article Comprehensively Analyzes The AD Context In Fivefold Main Topic. Firstly, It Reviews The Main Imaging Techniques Used In Diagnosing AD Disease. Secondly, It Explores The Most Proposed Deep Learning (DL) Algorithms For Detecting The Disease. Thirdly, The Article Investigates The Commonly Used Datasets To Develop DL Techniques. Fourthly, We Conducted A Systematic Review And Selected 45 Papers Published In Highly Ranked Publishers (Science Direct, IEEE, Springer, And MDPI). We Analyzed Them Thoroughly By Delving Into The Stages Of AD Diagnosis And Emphasizing The Role Of Preprocessing Techniques. Lastly, The Paper Addresses The Remaining Practical Implications And Challenges In The AD Context. Building On The Analysis, This Survey Contributes To Covering Several Aspects Related To AD Disease That Have Not Been Studied Thoroughly.
Assessment And Outcome Monitoring Are Critical For The Effective Detection And Treatment Of Mental Illness. Traditional Methods Of Capturing Social, Functional, And Behavioral Data Are Limited To The Information That Patients Report Back To Their Health Care Provider At Selected Points In Time. As A Result, These Data Are Not Accurate Accounts Of Day-to-day Functioning, As They Are Often Influenced By Biases In Self-report. Mobile Technology (mobile Applications On Smartphones, Activity Bracelets) Has The Potential To Overcome Such Problems With Traditional Assessment And Provide Information About Patient Symptoms, Behavior, And Functioning In Real Time. Although The Use Of Sensors And Apps Are Widespread, Several Questions Remain In The Field Regarding The Reliability Of Off-the-shelf Apps And Sensors, Use Of These Tools By Consumers, And Provider Use Of These Data In Clinical Decision-making.
Online Recommender Systems Are Being Used Increasingly Often For Hospitals, Medical Professionals, And Drugs. Today, The Great Majority Of Consumers Look Online Before Asking Their Doctors For Prescription Suggestions For A Range Of Health Conditions. The Medical Suggestion System Can Be Valuable When Pandemics, Floods, Or Cyclones Hit. In The Age Of Machine Learning (ML), Recommender Systems Give More Accurate, Precise, And Reliable Clinical Predictions While Using Less Resources. The Medicine Recommendation System Gives The Patient Reliable Information About The Medication, The Dosage, And Any Possible Adverse Effects. Medication Is Given Based On The Patient's Symptoms, Blood Pressure, Diabetes, Temperature, And Other Parameters. Drug Recommendation Systems Provide Precise Information At Any Time While Improving The Performance, Integrity, And Privacy Of Patient Data In The Decision-making Process. Recommender System, The Decision Tree Produces The Most Accurate Results. In Times Of Medical Emergency, A Drug Recommendation System Is Helpful For Giving Patients Recommendations For Safe Medications.
Resume Screening Is The Process Of Analysing The Resumes Where The Candidates Apply For The Different Types Of Jobs Where The Company Feel The Tedious Job To Find The Appropriate Candidate Due To The Complexity In Resumes Formats Since It Has Different Styles. As A Result, Selecting Applicants For The Appropriate Job Within A Company Is A Difficult Task For Recruiters. We Can Extract The Key Information From The CV Using NLTK, Natural Language Processing (NLP) Techniques To Save Time And Effort. This System Could Work With A Large Number Of Resumes For Classifying The Right Categories Using KNN Algorithm. Furthermore, This System Attempts To Find The Accuracy And Performance Of The Proposed Methodology And Incorporate It In The IT Firms And Other Regulations For The Prevention Of Manual Screening And Establish A Safe Allocation Of Resources For The Companies. As Such Companies Emerge Even For Them Manually Going Through All Of The Resume Of Candidates Is Very Time Consuming And Tedious So These Talent Acquisition Companies Use Various Machine Learning Models To Filter Out Top Resumes According To The Job Roles, Which Reduces The Efforts For The Human Resource Team.
In This Project, We Present A Novel Web Application That Harnesses The Power Of Natural Language Processing And Deep Learning To Analyze Emotions In Textual Input And Provide Personalized Music Recommendations Based On The Detected Mood. The Main Goal Of This Project Is To Create An Interactive And Engaging Platform That Enhances User Experience By Understanding Their Emotions And Offering Appropriate Music Selections In Real-time. The Core Component Of Our System Is An Emotion-driven Chatbot, Which Utilizes The Long Short-Term Memory (LSTM) Algorithm, A Type Of Recurrent Neural Network, To Process And Interpret Textual Inputs. The LSTM Model Is Trained On A Diverse Dataset Of Text-emotion Pairs To Learn Patterns And Correlations Between Specific Emotions And Corresponding Linguistic Cues. By Employing LSTM, The Chatbot Can Accurately Identify The Underlying Emotions Expressed In User-provided Text. Upon Receiving A Textual Input From The User, The Chatbot Processes The Text And Extracts The Associated Emotions, Which Are Then Used To Recommend Songs That Align With The User's Current Mood. These Song Suggestions Are Retrieved From An Extensive Music Database, Curated To Cater To Various Emotions And Moods. The Music Recommendation Process Leverages Content-based Filtering Techniques To Ensure That The Offered Songs Resonate With The User's Emotional State.
In Today's World, People Are Having Very Tight Schedules Due To The Changes In Their Lifestyles And Work Commitments. But It Requires Regular Physical Activity To Stay Fit And Healthy. People Do Not Concentrate On Their Food Habits, Leading To Obesity. Obesity Is Becoming A Major And Common Problem In Today’s Lifestyle. This Leads People To Choose Their Diet And Do An Equal Amount Of Exercise To Stay Fit And Healthy. The Main Part Here Is People Should Have Adequate Knowledge About Their Calorie Intake And Burn, Keeping A Track Of Their Calorie Intake Is Easy As It's Available On The Product Label Or On The Internet. Keeping Track Of Calories Burnt Is A Difficult Part As There Are Very Few Devices For That. Calories Burned By An Individual Are Based On MET Charts And Formulas. The Main Agenda Of This Study Is A Prediction Of The Burnt Calories With The Help Of An XG Boost Regression Model As The ML (machine Learning) Algorithm To Show Accurate Results. The Model Is Fed With More Than 15,000 Data And Its Mean Absolute Error Is 2.7 Which Will Become Better Over Time By Feeding The XG Boost Regression Model With More Data.
Crop Yield Prediction And Crop Recommendation As Plays A Crucial Role In Agricultural Decision-making Processes, Enabling Farmers To Optimize Resource Allocation And Plan For Potential Risks. In Recent Years, Machine Learning Algorithms Have Emerged As Powerful Tools For Predicting Crop Yields Accurately. This Abstract Focuses On The Application Of The Decision Tree Algorithm To Train For Crop Yield Prediction. Once The Decision Tree Model Is Constructed, It Can Be Used To Predict Crop Yields For Unseen Data. New Input Variables, Such As Weather Forecasts Or Soil Measurements, Can Be Fed Into The Model To Obtain Yield Predictions And Crop Recommend. The Interpretability Of Decision Trees Allows Farmers To Understand Which Factors Contribute Most Significantly To Crop Yield Variations And Make Informed Decisions Accordingly. User Can Interact With The Chat Bot To Give Details Then Model Is Predicted Crop And Reply By Chat Bot.
Credit Scoring Is A Way Of Analyzing Statistical Data Used In Financial Organizations And Banks To Acquire A Persons Creditworthiness. The Best Owers Generally Manipulate It To Decide To Widen Or Retract Credit. The Score Plays A Significant Role In Determining The Creditworthiness Of A Person And If He/she Can Be Sanctioned A Loan Or Not. Machine Learning Techniques Help Us To Predict The Credit Score More Accurately Using Classification Algorithms. Few Base And Ensemble Classification Algorithms Were Used In This Research To Perform A Comparative Analysis. To Achieve Better Results. The Objective Of This Paper Is To Predict The Credit Score Based On Different Classifier Models And Evaluate The Performance Of Each Model Based On The Metrics. A Comparative Analysis Is Done To Identify The Best Classifier To Predict The Credit Score. The Evaluation Metrics Used For Evaluating The Model Are Recall, Precision, Fmeasure, And Accuracy. This Helps Us To Improve The Decision In Identifying The More Accurate Classifier Model. The Dataset Used For This Analysis Is The Credit Dataset From The Machine Learning Repository. Experimental Results Prove That The K-nearest Neighbor And Extratree Classifier Model Produces Better Accuracy In Ensemble SMOTE Classifiers And The 95% Better Accuracy In The Base Classifier.
In The Fast-paced And Demanding World Of Information Technology (IT), The Well-being Of Employees Is Of Paramount Importance. Recognizing And Addressing Stress In IT Employees Is Crucial For Both Individual Health And Overall Organizational Performance. This Project Endeavours To Provide A Reliable And Efficient Solution For The Early Detection Of Stress In IT Employees Through The Analysis Of Text Data. The Primary Objective Of This Study Is To Develop A Stress Detection System Using The Random Forest Algorithm, Which Has Demonstrated Exceptional Performance In The Field Of Machine Learning. These Features Encompass A Wide Range Of Text-based Attributes, Including Sentiment Analysis, Topic Modelling, And Linguistic Markers Associated With Stress And Well-being. Through A Rigorous Process Of Data Pre-processing, Feature Engineering, And Model Training, We Achieve An Impressive 99% Accuracy In Stress Detection.
Water Quality Prediction Is A Research Hotspot In The Field Of Ecological Environment, Which Is Of Great Significance To The Prevention Of Water Pollution And The Construction Of Automatic Water Quality Monitoring Network. The Accuracy Of Prediction Model Results Will Affect The Scientificity And Correctness Of Applied Engineering Projects, As Well As The Accuracy Of Water Pollution Control Measures. Firstly, The Background Of Water Quality Prediction And The Development And Research Trends Of Water Quality Models At Home And Abroad Are Systematically Introduced. Then, The Water Quality Prediction Method Based On Machine Learning Is Mainly Introduced, Focusing On Time Series Prediction Method, Regression Analysis Method, Neural Network Method And Combination Prediction Method. The Applicability And Limitations Of The Model Are Analyze Respectively. Finally, According To The Research History And Present Situation Of Water Quality Prediction Model, The Development Trend Of Water Quality Prediction Model Is Prospected.
Customer Segmentation Is A Crucial Strategy For Businesses That Want To Better Understand Their Customers And Tailor Their Marketing Efforts To Meet Their Specific Needs And Preferences. In The Case Of Product Segmentation, Businesses Can Use Customer Segmentation Techniques To Identify Groups Of Customers Will Separate 4 Category How Is Pay And What He/she Can Buy The Product To Suggest. The K-means Algorithm Works By Partitioning The Customer Data Into K Clusters, Where Each Cluster Represents A Unique Group Of Customers With Similar Attributes For A Specific Product. The Algorithm Iteratively Assigns Each Customer To A Cluster Based On The Distance Between Their Attributes And The Cluster Centroid. The Centroid Is The Average Value Of All Attributes In The Cluster, Which Represents The Center Of The Group. Customer Segmentation Using K-means Clustering For Product Segmentation Has Several Benefits, Including Improved Product Targeting, Personalized Marketing, And Better Customer Experience. By Dividing The Customer Base Into Distinct Groups, Businesses Can Tailor Their Product Offerings And Marketing Strategies To Meet The Specific Needs And Preferences Of Each Group
Demand Forecasting For Products Is A Critical Aspect Of Supply Chain Management And Business Planning. It Involves Predicting The Future Demand For A Product Based On Historical Sales Data, Market Trends, And Other Relevant Factors. Accurate Demand Forecasting Can Help Businesses Optimize Their Inventory Levels, Production Schedules, And Pricing Strategies, While Reducing Waste And Improving Customer Satisfaction. The Main Concepts And Techniques Involved In Demand Forecasting For Products. We Will Discuss The Importance Of Demand Forecasting, The Different Types Of Forecasting Methods, And The Factors That Influence Demand. We Will Also Explore The Challenges And Limitations Of Demand Forecasting, As Well As Best Practices For Improving The Accuracy And Reliability Of Forecasts.
Blood Donation Saves Lives Every Day At Various Situations. A Blood Transfusion May Give Them The Energy To Spend Time With Family And Friends. Blood Cannot Be Manufactured It Can Only Come As A Gift From People. One Person Only Allowed 6 Pints Of Blood Donation. One Pint Of Blood Can Save Upto 33 Lives The Number Of Blood Donor Is Very Less When Compared With Other Countries. Here We Propose A New And Efficient Way To Overcome Such Outline. When We Just Touch The Button Donor The App Will Be Ask You To Enter An Individual's Details Like Name, Phone Number, Age, Date Of Birth, Blood Group, Address Etc. At The Emergency Time Of Blood Needed We Can Check For Blood Donor. Once The App User Enter The Blood Group Which He/she Needed It Will Automatically Show The Donor And You Will Find Nearby Donor. Blood Donation App Provider List Of Donor In Your City/area.
In This Research, An Inference System Of Fuzzy Mamdani System Have Been Designed For The Analysis Of Teaching Skills Of Faculties In Academic Or Educational Institutes. Nowadays, Various Academic Institutions Have Started Web-based Techniques To Collect Students’ Feedback Of Faculty Teaching Performance. The Performance Evaluation Of Faculty In Teaching Activities Performance Is A Major Key To Build A Fairer Academic Institute. Major Purpose Of Faculty Performance Is To Identify Strength & Weakness Of Professional Development Of A Teacher. This Soft Computing Technique By Using Fuzzy Mamdani Inference System (FMIS) For Evaluating Faculty Teaching Performance Will Be Useful For Management Of Organization For Evaluate Faculty Abilities With Student Outcomes.
Agriculture Plays A Pivotal Role In Ensuring Food Security And Sustaining Livelihoods Globally. Soil Quality Is A Fundamental Determinant Of Agricultural Productivity, And Precise Soil Analysis Is Essential For Optimizing Crop Selection And Cultivation Practices. This Paper Presents An Innovative Approach To Soil Analysis And Crop Recommendation Using Long Short-Term Memory (LSTM) Algorithms With Soil Image Data. Traditionally, Soil Analysis Involves Time-consuming And Costly Laboratory Tests, Making It Challenging For Farmers To Access Real-time Information About Their Soil Quality. In This Study, We Propose A Non-invasive And Efficient Method That Leverages Soil Image Data Collected Through Remote Sensing And Drone Technology. These Images Capture Crucial Information About Soil Properties, Such As Texture, Moisture, And Nutrient Levels.
Technology Has Boosted The Existence Of Humankind The Quality Of Life They Live. Every Day We Are Planning To Create Something New And Different. We Have A Solution For Every Other Problem We Have Machines To Support Our Lives And Make Us Somewhat Complete In The Banking Sector Candidate Gets Proofs/ Backup Before Approval Of The Loan Amount. The Application Approved Or Not Approved Depends Upon The Historical Data Of The Candidate By The System. Every Day Lots Of People Applying For The Loan In The Banking Sector But Bank Would Have Limited Funds. In This Case, The Right Prediction Would Be Very Beneficial Using Some Classes-function Algorithm. An Example The Logistic Regression, Random Forest Classifier, Support Vector Machine Classifier, Etc. A Bank's Profit And Loss Depend On The Amount Of The Loans That Is Whether The Client Or Customer Is Paying Back The Loan. Recovery Of Loans Is The Most Important For The Banking Sector. The Improvement Process Plays An Important Role In The Banking Sector. The Historical Data Of Candidates Was Used To Build A Machine Learning Model Using Different Classification Algorithms. The Main Objective Of This Paper Is To Predict Whether A New Applicant Granted The Loan Or Not Using Machine Learning Models Trained On The Historical Data Set.
We Propose To Implement A House Price Prediction Model. It’s A Machine Learning Model Which Integrates Data Science And Web Development. Housing Prices Fluctuate On A Daily Basis And Are Sometimes Exaggerated Rather Than Based On Worth. The Major Focus Of This Project Is On Predicting Home Prices Using Genuine Factors. Here, We Intend To Base An Evaluation On Every Basic Criterion That Is Taken Into Account When Establishing The Pricing. The Goal Of This Project Is To Learn Python And Get Experience In Data Analytics, Machine Learning, And AI.
Customer Segmentation Is A Crucial Strategy For Businesses That Want To Better Understand Their Customers And Tailor Their Marketing Efforts To Meet Their Specific Needs And Preferences. In The Case Of Product Segmentation, Businesses Can Use Customer Segmentation Techniques To Identify Groups Of Customers Will Separate 4 Category How Is Pay And What He/she Can Buy The Product To Suggest. The K-means Algorithm Works By Partitioning The Customer Data Into K Clusters, Where Each Cluster Represents A Unique Group Of Customers With Similar Attributes For A Specific Product. The Algorithm Iteratively Assigns Each Customer To A Cluster Based On The Distance Between Their Attributes And The Cluster Centroid. The Centroid Is The Average Value Of All Attributes In The Cluster, Which Represents The Center Of The Group. Customer Segmentation Using K-means Clustering For Product Segmentation Has Several Benefits, Including Improved Product Targeting, Personalized Marketing, And Better Customer Experience. By Dividing The Customer Base Into Distinct Groups, Businesses Can Tailor Their Product Offerings And Marketing Strategies To Meet The Specific Needs And Preferences Of Each Group.
Crop Yield Prediction And Crop Recommendation As Plays A Crucial Role In Agricultural Decision-making Processes, Enabling Farmers To Optimize Resource Allocation And Plan For Potential Risks. In Recent Years, Machine Learning Algorithms Have Emerged As Powerful Tools For Predicting Crop Yields Accurately. This Abstract Focuses On The Application Of The Decision Tree Algorithm To Train For Crop Yield Prediction. Once The Decision Tree Model Is Constructed, It Can Be Used To Predict Crop Yields For Unseen Data. New Input Variables, Such As Weather Forecasts Or Soil Measurements, Can Be Fed Into The Model To Obtain Yield Predictions And Crop Recommend. The Interpretability Of Decision Trees Allows Farmers To Understand Which Factors Contribute Most Significantly To Crop Yield Variations And Make Informed Decisions Accordingly. User To Give Details Then Model Is Predicted Crop .
Cryptocurrencies Are A Digital Way Of Money In Which All Transactions Are Held Electronically. It Is A Soft Currency Which Doesn’t Exist In The Form Of Hard Notes Physically. Here, We Are Emphasizing The Difference Of Fiat Currency Which Is Decentralized That Without Any Third-party Intervention All Virtual Currency Users Can Get The Services. However, Getting Services Of These Cryptocurrencies Impacts On International Relations And Trade, Due To Its High Price Volatility. There Are Several Virtual Currencies Such As Bitcoin, Ripple, Ethereum, Ethereum Classic, Lite Coin, Etc. In Our Study, We Especially Focused On A Popular Cryptocurrency, I.e., Bitcoin. From Many Types Of Virtual Currencies, Bitcoin Has A Great Acceptance By Different Bodies Such As Investors, Researchers, Traders, And Policy-makers. To The Best Of Our Knowledge, Our Target Is To Implement The Efficient Deep Learning-based Prediction Models Specifically Long Short-term Memory (LSTM) And Gated Recurrent Unit (GRU) To Handle The Price Volatility Of Bitcoin And To Obtain High Accuracy. Our Study Involves Comparing These Two Time Series Deep Learning Techniques And Proved The Efficacy In Forecasting The Price Of Bitcoin.
At Present Social Network Sites Are Part Of The Life For Most Of The People. Every Day Several People Are Creating Their Profiles On The Social Network Platforms And They Are Interacting With Others Independent Of The User’s Location And Time. The Social Network Sites Not Only Providing Advantages To The Users And Also Provide Security Issues To The Users As Well Their Information. To Analyze, Who Are Encouraging Threats In Social Network We Need To Classify The Social Networks Profiles Of The Users. From The Classification, We Can Get The Genuine Profiles And Fake Profiles On The Social Networks. Traditionally, We Have Different Classification Methods For Detecting The Fake Profiles On The Social Networks. But, We Need To Improve The Accuracy Rate Of The Fake Profile Detection In The Social Networks. In This Paper We Are Proposing Machine Learning And Natural Language Processing (NLP) Techniques To Improve The Accuracy Rate Of The Fake Profiles Detection. We Can Use The Support Vector Machine (SVM) And Naïve Bayes Algorithm.
Crop Yield Prediction And Crop Recommendation As Plays A Crucial Role In Agricultural Decision-making Processes, Enabling Farmers To Optimize Resource Allocation And Plan For Potential Risks. In Recent Years, Machine Learning Algorithms Have Emerged As Powerful Tools For Predicting Crop Yields Accurately. This Abstract Focuses On The Application Of The Decision Tree Algorithm To Train For Crop Yield Prediction. Once The Decision Tree Model Is Constructed, It Can Be Used To Predict Crop Yields For Unseen Data. New Input Variables, Such As Weather Forecasts Or Soil Measurements, Can Be Fed Into The Model To Obtain Yield Predictions And Crop Recommend. The Interpretability Of Decision Trees Allows Farmers To Understand Which Factors Contribute Most Significantly To Crop Yield Variations And Make Informed Decisions Accordingly. User To Give Details Then Model Is Predicted Crop .
Floods Are Very Harmful For Nature, Which Are Very Complex To Model. The Flood Prediction Model Will Give Risk Reduction & It Minimizes The Future Loss Of Human Life. On 18 May 2016 A South Indian State Kerala Was Affected By Flood. Machine Learning Is A Method Which Provides Intelligence To Predict The Result In Future. The Performance Comparison Of ML Models Is Based On The Speed, Time And Accuracy Of The Result. There Exist A Lot Of Machine Algorithms Which Generate Models With More Accuracy. For Flood Prediction Classification Algorithms Like Decision Tree And Linear Regression Are Used In This Research. This Paper Will Present The Dataset Of Kerala Flood 2016 Which Is Provided By Government.
The Aim Of This Project Is To Develop A Web Application That Can Extract Text From An Image And Translate It To A Desired Language. The Application Is Built Using The Tesseract OCR Engine And The Flask Web Framework In Python. The Tesseract OCR Engine Is Used To Extract Text From The Image And The Flask Web Framework Is Used To Build The Web Application. The User Can Upload An Image Containing Text In Any Language. The Image Is Processed Using Tesseract OCR Engine To Extract The Text. The Extracted Text Is Then Translated To The Desired Language Using A Translation API. The Translated Text Is Displayed On The Web Page. The Application Also Provides An Option For The User To Select The Language They Want To Translate The Text To. The User Can Select The Desired Language From A Dropdown Menu On The Web Page. The Application Supports A Wide Range Of Languages For Translation. The Application Is Designed To Be User-friendly And Easy To Use. The User Interface Is Simple And Intuitive. The User Can Upload The Image And Select The Language They Want To Translate To With Just A Few Clicks. The Application Is Also Scalable And Can Handle Large Volumes Of Image And Text Data.
The Development Of Information Technology Has Been Increasingly Changing The Means Of Information Exchange Leading To The Need Of Digitizing Print Documents. In The Present Era, There Is A Lot Of Fraud That Often Occurs. For Example, Is Account Fraud, To Avoid Account Fraud There Was Verification Using ID Card Extraction Using OCR And NLP. Optical Character Recognition (OCR) Is A Technology That Used To Generate Text From Images. With OCR We Can Extract Aadhar Card Into Text Using Pytesseract. To Improve The Accuracy We Made Text Corrections Using Natural Language Processing (NLP) Basic Tools To Fixing The Text. With 5 Aadhar Card Image, We Compared The Performance With Three Different OCR Libraries. The Result Of Our Experiment Shows That Pytesseract Had The Best Performance.The Resultant Edge Image Contains The Broken Characters. To Fill These Gaps, We Apply The Dilation Operator That Increases The Thickness Of The Characters. Dilation Fills The Broken Characters, However, Also Add Extra Thickness That Is Then Removed Through Applying The Morphological Thinning. Finally, Dilation And Thinning Are Applied In Combination To Optical Character Recognition (OCR) To Segment And Recognize The Characters Including The Name, ID, DOB, Gender And Photo Of Person.
Because Of The Fast Expansion Of Internet Users, Phishing Attacks Have Become A Significant Menace Where The Attacker Poses As A Trusted Entity In Order To Steal Sensitive Data, Causing Reputational Damage, Loss Of Money, Ransomware, Or Other Malware Infections. Intelligent Techniques Mainly Machine Learning (ML) And Deep Learning (D L) Are Increasingly Applied In The Field Of Cyber Security Due To Their Ability To Learn From Available Data In Order To Extract Useful Insight And Predict Future Events. The Effectiveness Of Applying Such Intelligent Approaches In Detecting Phishing Web Sites Is Investigated In This Paper. We Used Two Separate Datasets And Selected The Highest Correlated Features Which Comprised Of A Combination Of Content-based, URL Lexical-based, And Domain-based Features. A Set Of ML Models Were Then Applied, And A Comparative Performance Evaluation Was Conducted. Results Proved The Importance Of Features Selection In Improving The Models' Performance. Furthermore, The Results Also Aimed To Identify The Best Features That Influence The Model In Identifying Phishing Websites. For Classification Performance, Random Forest (RF) Algorithm Achieved The Highest Accuracy For Both Datasets.
Credit Card Fraud Detection Is Presently The Most Frequently Occurring Problem In The Present World. This Is Due To The Rise In Both Online Transactions And E-commerce Platforms. Credit Card Fraud Generally Happens When The Card Was Stolen For Any Of The Unauthorized Purposes Or Even When The Fraudster Uses The Credit Card Information For His Use. In The Present World, We Are Facing A Lot Of Credit Card Problems. To Detect The Fraudulent Activities The Credit Card Fraud Detection System Was Introduced. This Project Aims To Focus Mainly On Machine Learning Algorithms. The Algorithms Used Are Random Forest Algorithm And The ExtraTreesClassifier Algorithm. The Results Of The Two Algorithms Are Based On Accuracy, Precision, Recall, And F1-score. The ROC Curve Is Plotted Based On The Confusion Matrix. The Random Forest And The Extra-Trees Classifier Algorithms Are Compared And The Algorithm That Has The Greatest Accuracy, Extra-Trees Classifier Is Considered As The Best Algorithm That Is Used To Detect The Fraud.
The Security Of Any Public Key Cryptosystem Depends On The Private Key Thus, It Is Important That Only An Authorized Person Can Have Access To The Private Key. The Paper Presents A New Algorithm That Protects The Private Key Using The Transposition Cipher Technique. The Performance Of The Proposed Technique Is Evaluated By Applying It In The Random Forest Algorithm’s Generated Private Keys Using 512-bit, 1024-bit, And 2048-bit, Respectively. The Result Shows That The Technique Is Practical And Efficient In Securing Private Keys While In Storage As It Produced High Avalanche Effect. Key Generator Is Part Of The Stream Cipher System That Is Responsible For Generating A Long Random Sequence Of Binary Bits Key That Used In Ciphering And Deciphering Processes In Everyday Life, Image Security Is Important These Days As Data Is Increasing A Lot. These Data Can Be Images, Videos, Text, Audio, Etc. So To Protect These Images From Attackers Who Can Destroy The Image Quality Or Modify The Images, Some Technologies Like AES, DES, RSA, Etc. Have Been Invented. With The Generation, Data Security Has Also Become An Essential Issue. Considering These Issues, The Proposed Technique Ensures Confidentiality, Integrity, And Authentication. Using These Techniques, The Host Can Encrypt And Decrypt The Image ,text ,video ,audio. The Digital Technology Was Completely Different From Today And The Scale Of Challenges Was Smaller, So With Recent Advanced Technology And The Emergence Of New Applications Such As Big Data Applications, In Addition To Applications Running With 64-bit And Many Other Applications Have Become Necessary To Design A New Current Algorithm For Current Requirements. Advanced Encryption Algorithm (AES) Is A Symmetric Algorithm, Which We Will Further And In Addition To New Recommendations For Future Work, A List Of Shortcomings And Vulnerabilities Of The Internal Structure Of The AES Algorithm Will Be Diagnosed.
For Secure Data Transmission Over Internet, It Is Important To Transfer Data In High Security And High Confidentiality, Information Security Is The Most Important Issue Of Data Communication In Networks And Internet. Either Image Or Video To Secure Transferred Information From Intruders, It Is Important To Convert The Information Into Cryptic Format The Image And Video Work On The Same Process. Different Methods Used To Ensure Data Security And Confidentiality During Transmission Like Steganography And Cryptography. This Paper We Convert Plaintext To Cipher Text For Doing So We Have Used RC6 Encryption Algorithm The Proposed Algorithm Ensure The Encryption And Decryption Using RC6 Stream Cipher And RGB Pixel Shuffling With Steganography By Using Hash-least Significant Bit (HLSB) That Make Use Of Hash Function To Developed Significant Way To Insert Data Bits In LSB Bits Of RGB Pixels Of Cover Image. The Security Evaluations For The Steganography Part We Will Be Using Modified LSB Algorithm Where We Overwrite The LSB Bits Of The Selected Frame (given By The User) From The Cover Video, With The Bit Of Text Message Character With Help Of Secret Key And Using KSA And PRGA.
Student Attendance System Is Used To Measure Student Participation In A Classroom. Before Pandemic Attendance Was Taken Manually Like In Sheets Or Registers. But When The Pandemic Hit, Everything Was Online, So Even The Classes. The Attendance Count Is A Very Important Problem That The Administrator Needs To Be More Careful About Taking During The Online Classes As There Are Many Chances Of A Proxy Happening. So, We Came Up With This Proposed System “Student Attendance Using QR Code” This Paper Proposes An Attendance System That Is Based On The QR Code-based Attendance System. The Students Need To Scan The QR In The Class According To The Professor Instruction. The Paper Explains The High Level Implementation Details Of The Proposed System. It Also Discusses How The System Verifies Student Identity To Eliminate False Registrations.
Nowadays, Dependency On Banking In The Virtual World Has Been Increased To The Peak Position. To Make It Consistent Advanced Technologies Should Be Used. As OTP Is Currently Used Worldwide For Security Purposes, It Can Be Overruled By QR Code. Main Advantage Of QR Code Over OTP Data Storage. OTP Can Only Confirm That The User Is Authorised User And Not Some Third Party Is Involved In This Transaction While QR Code Not Only Confirms The Authorised User But QR Code Itself Can Store Information Such As Transaction Id, Transaction Date, Time And Also Amount Of Transaction. So, There Is No Need Of Explicitly Keeping Track Of Transaction Every Transaction. Aim Of This Paper To Enhance The Functionality Of ATM Machine Using Android Application. Proposed System Is Combining The ATM And Mobile Banking And Minimizes The Time Of Withdrawing Cash From ATM. This Will Increase The Speed Of Transaction Almost Three Times Fast; Could Have Excellent Impact On Customer's Satisfaction. With The Help Of QR Code Information Get Encrypted So It Also Increases Security. As The Population Increasing ATM Queues Will Be Longer Day By Day. By Implementing Proposed System Current System Will Not Hampered, By Doing Some Minor Changes In Existing System It Will Be Possible To Get Cash Within Seconds. According To Analyst Report, Cost Of Transaction Using Mobile Application Is Almost Ten Times Less Than ATM And About Fifty Times Less, If Physical Bank Branch Used.
In Order To Prevent Health Risks And Provide A Better Service To The Patients That Have Visited The Hospital, There Is A Need For Monitoring The Patients After Being Released And Providing The Data Submitted By The Patient E-Health Enablers To The Medical Personnel. This Article Proposes Architecture For Providing The Secure Exchange Of Data Between The Patient And The Hospital Infrastructure. The Implemented Solution Is Validated On A Laboratory Tested. When It Comes To Exchanging Health Data Between Departments Or Across Institutions, There Are So Many Variables At Play That Additional Rules And Descriptions Are Absolutely Necessary. There Can’t Be Any Ambiguity When Transferring And Interpreting Information About The Patient's Allergies Or The Procedures, Materials, And Medications Required.
Nowadays, Quick Response (QR) Codes Seem To Be Present Everywhere. They Can Be Found On Advertisements In Magazines, Websites, Product Packaging, And Other Places. Since Mobile Phones Have Become A Basic Necessity For Everyone, Using QR Codes Is One Of The Most Fascinating Ways To Link Patients To The Internet Digitally. QR Codes Consist Of Black Squares Arranged In A Grid (matrix) On A White Background And Are Read By Specialized Software That Is Able To Extract Data From The Patterns That Are Present In The Matrix. Now Days It Is Used Widely In Many Organizations. In This Project, We Proposed QR Code-based For Hospital Management System. The Emergence Of QR Has Opened A Vast Variety Of Possibilities In The Technology Sector Which Made Accessing, Retrieving And Viewing Information And Data From Anywhere With Great Speed And Low Fault. It Is A Captivating Way Of Accessing Anything From A Website. Nowadays Due To The Ample Use Of Mobile Devices, Using QR Code Technology We Can Easily Establish Connections And Communicate With People And Share Information. It Is Also A Secure Way To Share Or Secure The Data Because Without The Correct Tool Retrieving Of Data For Someone Else Who Is Not Intended To View Is Impossible. Introducing QR Code Will Increase This Security One More Level Further. In This Paper, This Is A Patient Management App That Uses Both Quick Response (QR) Code Technology Hospital And Accesses Those Data In A Secure And Fast Manner. It Also Can Be Used By Patients To Recollect The Doctor Consultation Data And Retrieving Their Medical Records And Doctor-prescribed Medicines.
Graphical Password Is One Of Technic For Authentication Of Computer Security. The Most Crucial Aspect Of Computer Science Nowadays Is Digital/computer Security, Which Protects User Or Customer Data. And One Of The Hazards Is Shoulder-surfing, In Which A Criminal Can Acquire A Password By Watching Directly Or By Recording The Authentication Session. There Are A Number Of Methods For This Authentication, But The Most Popular And Straightforward Is The Graphic Password Method. A Bank Is Essential To People's Daily Lives. The Bank's Top Priority Is The Security Of Its Customers. To Safeguard User Accounts, The Authentication Process Must Be Secure. Textual Passwords Are A Frequently Used Method. The System Uses The Graphical Password To Demonstrate The Banking Website's Security In Order To Offer A Possible Substitute For The Traditional Alphanumeric Password Techniques To Prevent Shoulder Surfing Techniques.
The Security Of Any Public Key Cryptosystem Depends On The Private Key Thus, It Is Important That Only An Authorized Person Can Have Access To The Private Key. The Paper Presents A New Algorithm That Protects The Private Key Using The Transposition Cipher Technique. The Performance Of The Proposed Technique Is Evaluated By Applying It In The Random Forest Algorithm’s Generated Private Keys Using 512-bit, 1024-bit, And 2048-bit, Respectively. The Result Shows That The Technique Is Practical And Efficient In Securing Private Keys While In Storage As It Produced High Avalanche Effect.
With The Evolution In Wireless Communication, There Are Many Security Threats Over The Internet. The Intrusion Detection System (IDS) Helps To Find The Attacks On The System And The Intruders Are Detected. Previously Various Machine Learning (ML) Techniques Are Applied On The IDS And Tried To Improve The Results On The Detection Of Intruders And To Increase The Accuracy Of The IDS. This Paper Has Proposed An Approach To Develop Efficient IDS By Using The Principal Component Analysis (PCA) And The Random Forest Classification Algorithm. Where The PCA Will Help To Organise The Dataset By Reducing The Dimensionality Of The Dataset And The Random Forest Will Help In Classification. Results Obtained States That The Proposed Approach Works More Efficiently In Terms Of Accuracy As Compared To Other Techniques Like SVM, Naive Bayes, And Decision Tree. The Results Obtained By Proposed Method Are Having The Values For Performance Time (min) Is 3.24 Minutes, Accuracy Rate (%) Is 96.78 %, And The Error Rate (%) Is 0.21 %.
The Phenomenon Of Fake News Is Experiencing A Rapid And Growing Progress With The Evolution Of The Means Of Communication And Social Media. Fake News Detection Is An Emerging Research Area Which Is Gaining Big Interest. It Faces However Some Challenges Due To The Limited Resources Such As Datasets And Processing And Analyzing Techniques. In This Work, We Propose A System For Fake News Detection That Uses Machine Learning Techniques. We Used Term Frequency Inverse Document Frequency Of Bag Of Words And N-grams As Feature Extraction Technique, And Naïve Bayes As A Classifier. We Propose Also A Dataset Of Fake And True News To Train The Proposed System. Obtained Results Show The Efficiency Of The System
With Daily Installs, Third-party Apps Can Be A Important Cause For The Popularity And Attractiveness Of Facebook Or Any Online Social Media. Sadly, Cyber Criminals Get Came To The Realization That The Capability Of Using Apps For Spreading Spam And Malware. We Realize That At The Least 13% Of Facebook Apps In The Dataset Are Usually Malevolent. However With Their Findings , Several Issues Like Faux Profiles, Malicious Application Have Conjointly Full-grown. There Aren't Any Possible Method Exist To Regulate These Issues. During This Project, We Tend To Came Up With A Framework With That Automatic Detection Of Malicious Applications Is Feasible And Is Efficient. Suppose There's Facebook Application, Will The Facebook User Verify That The App Is Malicious Or Not. First We Identify A Set Of Features That Help Us To Analyze Malicious From Benign Ones. Second, Leveraging These Distinguishing Features ,where We Show That Post Of Application As Malicious With 95.9% Accuracy. Finally, We Explore The Ecosystems Of Malicious Facebook Apps And Identify Mechanisms That These Apps Use To Spread.
There Is An Abnormal Increase In The Crime Rate And Also The Number Of Criminals Are Increasing, This Leads Towards A Great Concern About The Security Issues. Crime Preventions And Criminal Identification Are The Primary Issues Before The Police Personnel, Since Property And Lives Protection Are The Basic Concerns Of The Police But To Combat The Crime, The Availability Of Police Personnel Is Limited. With The Advent Of Security Technology, Cameras Especially CCTV Have Been Installed In Many Public And Private Areas To Provide Surveillance Activities. The Footage Of The CCTV Can Be Used To Identify Suspects On Scene. This Real Time Criminal Identification System Based On Face Recognition Works With A Fully Automated Facial Recognition System. Here Feature-based Cascade Classifier And OpenCV LBPH (Local Binary Pattern Histograms) Algorithms Are Used For Face Detection And Recognition. This System Will Be Able To Detect Face And Recognize Face Automatically In Real Time. An Accurate Location Of The Face Is Still A Challenging Task. Viola-Jones Framework Has Been Widely Used By Researchers In Order To Detect The Location Of Faces And Objects In A Given Image. Face Detection Classifiers Are Shared By Public Communities, Such As OpenCV.
An Intelligent Licence Plate Detection Method Can Make The Travel More Convenient And Efficient. However, Traditional Methods Are Reasonably Effective Under The Specific Circumstances Or Strong Assumptions Only,. Therefore, A Novel Real-time Car Plate Detection Method Based On Improved Yolov3 Has Been Proposed. In Order To Select The More Precise Number Of Candidate Anchor Boxed And Aspect Ratio Dimensions, The Deep Learning Object Detection Algorithm Is Utilized. As Shown In The Experimental Results, The Method Which Is Proposed By This Paper Is Better Than Original Yolov3. However, Good Performing Models Such As YOLOv3 In More General Object Detection And Recognition Tasks Could Be Effectively Transferred To The License Plate Detection Application With A Small Effort In Model Tuning. This Paper Focuses On The Design Of Experiment (DOE) Of Training Parameters In Transferring YOLOv3 Model Design And Optimising The Training Specifically For License Plate Detection Tasks. The Parameters Are Categorised To Reduce The DOE Run Requirements While Gaining Insights On The YOLOv3 Parameter Interactions Other Than Seeking Optimised Train Settings. The Result Shows That The DOE Effectively Improve The YOLOv3 Model To Fit The Vehicle License Plate Detection Task.
The Vision-Based Live Vehicle Detection And Distance Calculation Project Leverages The State-of-the-art YOLOv5 Algorithm To Provide Real-time Vehicle Detection And Accurate Distance Measurement. This Innovative Solution Is Designed To Seamlessly Analyze Both Live Video Streams And Static Images, Offering A Versatile Platform For Monitoring And Assessing Vehicle Movements. The YOLOv5 Architecture, Known For Its High Accuracy And Speed, Has Been Employed To Detect Vehicles Within The Input Data. The Project Accommodates Two Main Modes Of Operation: Live Video Input And Image Upload. For Live Video Input, Users Can Receive Instant, Real-time Feedback On Vehicle Detection And Distance Calculation.
Weed Identification In Vegetable Plantation Is More Challenging Than Crop Weed Identification Due To Their Random Plant Spacing. So Far, Little Work Has Been Found On Identifying Weeds In Vegetable Plantation. Traditional Methods Of Crop Weed Identification Used To Be Mainly Focused On Identifying Weed Directly; However, There Is A Large Variation In Weed Species. This Paper Proposes A New Method In A Contrary Way, Which Combines Deep Learning And Image Processing Technology. Firstly, A Trained CenterNet Model Was Used To Detect Vegetables And Draw Bounding Boxes Around Them. Afterwards, The Remaining Green Objects Falling Out Of Bounding Boxes Were Considered As Weeds. In This Way, The Model Focuses On Identifying Only The Vegetables And Thus Avoid Handling Various Weed Species. Furthermore, This Strategy Can Largely Reduce The Size Of Training Image Dataset As Well As The Complexity Of Weed Detection, Thereby Enhancing The Weed Identification Performance And Accuracy. To Extract Weeds From The Background, A Color Index-based Segmentation Was Performed Utilizing Image Processing. The Employed Color Index Was Determined And Evaluated Through Genetic Algorithms (GAs) According To Bayesian Classification Error. During The Field Test, The Trained CenterNet Model Achieved A Precision Of 95.6%, A Recall Of 95.0%, And A F1 Score Of 0.953, Respectively. The Proposed Index −19R + 24G −2B ≥ 862 Yields High Segmentation Quality With A Much Lower Computational Cost Compared To The Wildly Used ExG Index. These Experiment Results Demonstrate The Feasibility Of Using The Proposed Method For The Ground-based Weed Identification In Vegetable Plantation.
Face Recognition Play A Vital Role In Variety Of Applications From Biometrics, Surveillance, Security, Identification To The Authentication. In This Paper We Design And Implement A Smart Security System For Restricted Area Where Access Is Limited To People Whose Faces Are Available In The Training Database. First We Are Going To Detect The Face By Detecting The Human Motion. Then Face Recognition Is Performed To Determine The Authority Of The Person To Enter The Sensitive Area. At The Same Time, We Track The Coordinate Of Detected Motion. Failing To Recognize The Face Finally Passes The Estimated Coordinate To Detect The Intruder Automatically. Experimental Results Demonstrate The Effectiveness Of Proposed Security System In Order To Restrict The Unauthorized Access And Enhanced Reliability By Use Of Face Recognition. Although This Reduces The Complexity Of Face Recognition, There Is Still A Concern Regarding The Real-time Protection Of Sensitive Portion. It Should Be Noted That This Problem Is Somewhat A Hard Task And Can Be Solved By Automatically Send Mail The Admin Person Attempting To Trespass The Sensitive Area.
Depression Is A Common And Serious Mental Health Disorder That Affects Millions Of People Worldwide. Early Detection And Diagnosis Of Depression Are Crucial For Effective Treatment And Management Of The Condition. The Use Of Machine Learning Algorithms, Such As Convolutional Neural Networks (CNNs), Has Shown Promising Results In Detecting Depression From Various Data Sources, Including Audio, Text, And Images. This Paper Proposes A CNN-based Approach For Depression Detection Using Facial Images. The Proposed System Involves The Use Of A Pre-trained CNN Model, Such As VGG-16, To Extract Features From Facial Images. The Extracted Features Are Then Used To Train A Support Vector Machine (SVM) Classifier To Detect Depression.
The Worst Possible Situation Faced By Humanity, COVID-19, Is Proliferating Across More Than 180 Countries And About 37,000,000 Confirmed Cases, Along With 1,000,000 Deaths Worldwide As Of October 2020. The Absence Of Any Medical And Strategic Expertise Is A Colossal Problem, And Lack Of Immunity Against It Increases The Risk Of Being Affected By The Virus. Since The Absence Of A Vaccine Is An Issue, Social Spacing And Face Covering Are Primary Precautionary Methods Apt In This Situation. This Study Proposes Automation With A Deep Learning Framework For Monitoring Social Distancing Using Surveillance Video Footage And Face Mask Detection In Public And Crowded Places As A Mandatory Rule Set For Pandemic Terms Using Computer Vision. The Paper Proposes A Framework Is Based On Object Detection Model To Define The Background And Human Beings With Bounding Boxes And Assigned Identifications. In The Same Framework, A Trained Module(mobilenet_v2) Checks For Any Unmasked Individual. And Segment The Human Face Then Show Whether “Mask” Or “Unmask”.
Online Reviews Have Become An Important Source Of Information For Users Before Making An Informed Purchase Decision. Early Reviews Of A Product Tend To Have A High Impact On The Subsequent Product Sales. In This Paper, We Take The Initiative To Study The Behaviour Characteristics Of Early Reviewers Through Their Posted Reviews On Two Real-world Large E-commerce Platforms, I.e., Amazon And Yelp. In Specific, We Divide Product Lifetime Into Three Consecutive Stages, Namely Early, Majority And Laggards. A User Who Has Posted A Review In The Early Stage Is Considered As An Early Reviewer. We Quantitatively Characterize Early Reviewers Based On Their Rating Behaviours, The Helpfulness Scores Received From Others And The Correlation Of Their Reviews With Product Popularity. We Have Found That (1) An Early Reviewer Tends To Assign A Higher Average Rating Score; And (2) An Early Reviewer Tends To Post More Helpful Reviews. Our Analysis Of Product Reviews Also Indicates That Early Reviewers’ Ratings And Their Received Helpfulness Scores Are Likely To Influence Product Popularity. By Viewing Review Posting Process As A Multiplayer Competition Game, We Propose A Novel Margin-based Embedding Model For Early Reviewer Prediction. Extensive Experiments On Two Different E-commerce Datasets Have Shown That Our Proposed Approach Outperforms A Number Of Competitive Baselines.
Recognizing Ancient Tamil Inscription Characters Enable Archeologists To Reveal Historical Events In Ancient Sri Lanka. Currently, This Is Done By The Archaeology Experts With A Huge Effort. The Inefficiency Of This Manual Procedure Will Negatively Impact On The Future Research In Field Of Archaeology. This Research Involves In Developing An Application With Optical Character Recognition (OCR) Functionality To Recognize Ancient Tamil Inscription. This Paper Focus On The OCR Module Of The Application. OCR Module Comprises Of The Technologies Of Artificial Neural Network (ANN) And Convolutional Neural Network (CNN). Experiments Were Carried Out To Evaluate The Recognition Rate Of The Two OCR Technologies Which Performs On Train Data, Test Data (preprocessed) And Test Data (real Images). After Evaluating Each OCR Solution, CNN Was Selected As The Best Resulted OCR Solution. Lack Of Data Is The Main Limitation Of This Research And It Will Be Highly Impacted On The OCR Accuracy.
Sign Language Is A Crucial Means Of Communication For The Deaf And Hard Of Hearing Community. This Project Presents A Comprehensive Approach To Real-time Sign Language Detection Using Convolutional Neural Networks (CNN) And The You Only Look Once (YOLO) Object Detection Framework. The Primary Objective Of The Project Is To Bridge The Communication Gap Between Individuals Who Use Sign Language And Those Who May Not Understand It. Firstly, The Implementation Involves Real-time Voice Recognition To Detect Spoken Words And Translate Them Into Corresponding Alphabet Letter Sign Language Gestures. A CNN Model Is Trained On A Custom Dataset Of Sign Language Gestures Associated With Various Spoken Words. This Model Achieves Accurate Word Recognition, Enabling The Translation Of Spoken Language Into Visual Sign Representations. Secondly, The Project Integrates Live Camera Input To Detect Sign Language Gestures Directly From Hand Movements. The YOLO Framework Is Employed To Identify And Localize Individual Signs Corresponding To Letters Of The Alphabet. By Training The YOLO Model On An Annotated Dataset Of Hand Signs, The System Can Accurately Recognize And Display The Appropriate Letter For Each Detected Gesture. The Experimental Results Demonstrate The Effectiveness Of The Proposed Approach. The CNN Model Achieves High Accuracy In Recognizing Spoken Words, Facilitating Accurate Translation Into Sign Language. Additionally, The YOLO-based Sign Language Detection Exhibits Robustness In Identifying Different Sign Gestures From Live Camera Feed, With Real-time Performance Suitable For Practical Applications. The Implementation Of The System Is By Using OpenCV-Python. The System Uses Various Libraries.
In The Modern, Highly Technological World, It Is Acknowledged That People With Visual Impairments, Whose Main Issue Is Social Isolation, Need To Be Able To Live Independently. Visually Impaired People Experience Difficulties And Are At A Disadvantage Because They Lack The Most Visual Information, Which Is The Information They Miss The Most, In Their Environment. Visually Challenged People Can Be Aided With The Use Of Cutting-edge Technology. They Suffer In Strange Circumstances With No Manual Assistance. Since Most Tasks Are Based On Visual Information, People Who Are Blind Are At A Disadvantage Because They Lack The Necessary Information About The Environment. It Is Possible To Extend The Assistance Provided To People With Visual Impairment With The Most Recent Advancements In Inclusive Technology. Using Image And Text Recognition, Machine Learning, And Artificial Intelligence, This Project Aims To Assist Visually Impaired Individuals. The Concept Is Implemented By Means Of A Desktop Application That Focuses On Voice Assistant, Image Recognition, Google Search, To-do List, Weather, Screen Shots, Directions On A Map Chat Bot, Capture Photo, And Other Features. The App Can Assist With Object Recognition And Distance Estimation Using Voice Commands Using YOLOV4 Algorithm. Blind People Will Be Able To Use The Technology's Features And Interact With The Environment In A More Effective Manner Thanks To This Method.
The Modern Keyboard For Personal Computers Has Been Developed From A Similar One Used In Typewriters. The Layout Has Remained The Same, But The Computer Keyboard Uses The Making And Breaking Of An Electrical Contact To Detect The Key-press. The Major Disadvantage Of This Concept Is That A Large Amount Of Physical Space Is Needed To Accommodate The Keyboard, Making It Unsuitable For Applications Such As Mobile Phones Where It Placed Limitations On The Screen Size. To Overcome This Drawback, The Touchscreens Were Developed, Which Integrated The Input Mechanism In The Screen Itself. However, Typing On Touchscreens Is Inconvenient For Most Users, Due To The Small Size Of Buttons. Also, As The Touchscreen Typing Keypads Are Integrated In The Computer Software, There Are Some Security Issues.
In This Project, We Propose Using YOLOv7 To Recognize Sign Language Gestures. We Will Use A Dataset Of Images And Videos Of Individuals Performing Various Sign Language Gestures, And Train A YOLOv7 Model To Recognize These Gestures In Real-time. To Do This, We Will First Preprocess The Dataset By Extracting The Relevant Frames And Labeling The Sign Language Gestures. We Will Then Use YOLOv7 To Train A Neural Network To Recognize These Gestures. The Model Will Be Trained Using A Combination Of Image Augmentation Techniques And Transfer Learning To Improve Its Accuracy And Reduce Overfitting. Once The Model Is Trained, We Will Use It To Detect And Classify Sign Language Gestures In Real-time Video Streams. We Will Use A Webcam To Capture Video Input And Apply The YOLOv7 Model To Recognize The Gestures. The Output Will Be Displayed On The Screen, Allowing Individuals Who Are Deaf Or Hard Of Hearing To Communicate More Easily.
In This Paper We Describe A Methodology And An Algorithm To Estimate The Real-time Age, Gender, And Emotion Of A Human By Analysing Of Face Images On A Webcam. Here We Discuss The CNN Based Architecture To Design A Real-time Model. Emotion, Gender And Age Detection Of Facial Images In Webcam Play An Important Role In Many Applications Like Forensics, Security Control, Data Analysis, Video Observation And Humancomputer Interaction. In This Paper We Present Some Method & Techniques Such As PCA,LBP, SVM, VIOLA-JONES, HOG Which Will Directly Or Indirectly Used To Recognize Human Emotion, Gender And Age Detection In Various Conditions.
In Multimodal Emotion Recognition (SER), Emotional Characteristics Often Appear In Diverse Forms Of Energy Patterns In Spectrograms. Typical Attention Neural Network Classifiers Of SER Are Usually Optimized On A Fixed Attention Granularity. In This Paper, We Apply Multiscale Area Attention In A Deep Convolutional Neural Network To Attend Emotional Characteristics With Varied Granularities And Therefore The Classifier Can Benefit From An Ensemble Of Attentions With Different Scales. To Deal With Data Sparsity,we Conduct Data Augmentation With Vocal Tract Length Perturbation (VTLP) To Improve The Generalization Capability Of The Classifier. We Can Classified Three Various Emotion Detection In Real-time (speech,face,text) Experiments Are Carried Out On The Interactive Emotional Dyadic Motion Capture (IEMOCAP) Dataset. Which, To The Best Of Our Knowledge, Is The State Of The Art On This Dataset.
In This Paper We Describe A Methodology And An Algorithm To Estimate The Real-time Age, Gender, And Emotion Of A Human By Analysing Of Face Images On A Webcam. Here We Discuss The CNN Based Architecture To Design A Real-time Model. Emotion, Gender And Age Detection Of Facial Images In Webcam Play An Important Role In Many Applications Like Forensics, Security Control, Data Analysis, Video Observation And Humancomputer Interaction. In This Paper We Present Some Method & Techniques Such As PCA,LBP, SVM, VIOLA-JONES, HOG Which Will Directly Or Indirectly Used To Recognize Human Emotion, Gender And Age Detection In Various Conditions.
The Use Of Doctor-computer Interaction Devices In The Operation Room (OR) Requires New Modalities That Support Medical Imaging Manipulation While Allowing Doctors’ Hands To Remain Sterile, Supporting Their Focus Of Attention, And Providing Fast Response Times. This Paper Presents “Gestix,” A Vision-based Hand Gesture Capture And Recognition System That Interprets In Real-time The User’s Gestures For Navigation And Manipulation Of Images In An Electronic Medical Record (EMR) Database. Navigation And Other Gestures Are Translated To Commands Based On Their Temporal Trajectories, Through Video Capture. “Gestix” Access The Image Then Control Image What Ever We Give The Command To Processing The Image. A Simple Way To Store The Information Is Image Capturing Of The Handwritten Document And Save It In Image Format. The Method To Transform Handwritten Data Into Electronic Format Is ‘Optical Character Recognition’. It Involves Several Steps Including Pre-processing, Segmentation, Feature Extraction And Post-processing. Many Researchers Have Been Used OCR For Recognizing Character. This System Uses The Android Phone To Capture The Image Of The Document And Further Steps Are Done By OCR. The Main Challenge Is To Recognize The Characters From Different Styles Of Handwriting. Thus, A System Is Designed That Recognizes The Handwritten Data To Obtain An Editable Text. The Output Of This System Depends Upon The Data That Has To Be Written By The Writer. Our System Offers 90% Accuracy For Handwritten Documents And Gives The Easiest Way To Edit Or Share The Recognized Data.
The Goal Of The Paper Is To Improve The Recognition Of The Human Hand Postures In A Human Computer Interaction Application, The Reducing Of The Time Computing And To Improve The User Comfort Regarding The Used Human Hand Postures. The Authors Developed An Application For Computer Mouse Control. The Application Based On The Proposed Algorithm, Hand Pad Color And On The Selected Hand Feature Presents Good Behavior Regarding The Time Computing. The User Has An Increased Comfort In Use Of The System Due To The Proposed Hand Postures. Also, The System Works Well Having The Same Behavior Under Very Low Illuminance Level And High Illuminance Level.
This Study Is An Attempt To Understand And Address The Mental Health Issue, Of Working Professionals Through Facial Expression Recognition. As A Society, We Are All Currently Talking About Ways As To How A Person Who Is Suffering From Any Emotional Issue Can Adopt Certain Ways To Come Out Of A Specific Circumstance And How We As A Society Can Support Such People In These Situations. Our Endeavor Is To Work On A Way Where The Identification Of Such Persons Who Are Going Through A Difficult Phase In Their Life Can Be Performed.
Early Disease Detection And Pets Are Important For Better Yield And Quality Of Crops. With Reduction In Quality Of The Agricultural Product, Disease Plant Can Lead To The Huge Economic Losses To The Individual Farmers. In Country Like India Whose Major Population Is Involved In Agriculture It Is Very Important To Find The Disease At Early Stages. Faster And Precise Prediction Of Plant Disease Could Help Reducing The Losses. With The Significant Advancement And Developments In Deep Learning Have Given The Opportunity To Improve The Performance And Accuracy Of Detection Of Object And Recognition System. This Paper Focuses On Finding The Plant Diseases And Reducing The Economic Losses. We Have Proposed The Deep Leaning Based Approach For Image Recognition. We Have Examined The Three Main Architecture Of The Neural Network: Faster Region-based Convolution Neural Network (Faster R-CNN), Region-based Fully CNN(R-CNN) And Single Shot Multibook Detector (SSD). System Proposed In The Paper Can Detect The Different Types Of Disease Efficiently And Have The Ability To Deal With Complex Scenarios. Validation Results Show The Accuracy Of 94.6% Which Depicts The Feasibility Of Convolution Neural Network And Present The Path For AI Based Deep Learning Solution To This Complex Problem.
Traffic Sign Recognition System (TSRS) Is A Significant Portion Of Intelligent Transportation System (ITS). Being Able To Identify Traffic Signs Accurately And Effectively Can Improve The Driving Safety. Mainly Aims At The Detection And Classification Of Circular Signs. Firstly, An Image Is Preprocessed To Highlight Important Information. Secondly, Hough Transform Is Used For Detecting And Locating Areas. This Paper Brings Forward A Traffic Sign Recognition Technique On The Strength Of Deep Learning, Which Finally, The Detected Road Traffic Signs Are Classified Based On Deep Learning. In This Article, A Traffic Sign Detection And Identification Method On Account Of The Image Processing Is Proposed, Which Is Combined With Convolutional Neural Network (CNN) To Sort Traffic Signs. On Account Of Its High Recognition Rate, CNN Can Be Used To Realize Various Computer Vision Tasks. TensorFlow Is Used To Implement CNN. In The German Data Sets, We Are Able To Identify The Circular Symbol With Best Accuracy.
The Increasing Crime Rate In Crowded Events Or Isolated Areas Has Heightened The Importance Of Security Across All Domains. Computer Vision Has Significant Applications In Addressing A Range Of Problems Through The Identification And Surveillance Of Abnormalities. The Increasing Need To Protect Safety, Security, And Personal Belongings Has Led To A Significant Demand For Video Surveillance Systems Capable Of Recognizing And Interpreting Scenes As Well As Detecting Unusual Events. These Systems Play A Crucial Role In Intelligent Monitoring. This Research Paper Applies Convolutional Neural Network (CNN) Based SSD And Faster RCNN Algorithms To Achieve Automatic Detection Of Guns Or Weapons. The Suggested Approach Entails The Utilization Of Two Distinct Categories Of Datasets. One Dataset Containing Pre-labeled Images.
Heart Disease Is One Of The Most Significant Causes Of Mortality In The World Today. Prediction Of Cardio Vascular Disease Is A Critical Challenge In The Area Of Clinical Data Analysis.Machine Learning(ML)has Been Shown To Be Effective In Assisting In Making Decisions And Predictions From The Large Quantity Of Data Produced By The Health Care Industry.We Have Also Seen ML Techniques Being Used In Recent Developments In Different Areas Of The Internet Of Things (IoT). Various Studies Give Only A Glimpse Into Predicting Heart Disease With ML Techniques. In This Paper, We Propose A Novel Method That Aims At finding Significant Features By Applying Machine Learning Techniques Resulting In Improving The Accuracy In The Prediction Of Cardiovascular Disease. The Prediction Model Is Introduced With Different Combinations Of Features And Several Known Classification Techniques.
The Bone Fracture Detection And Classification Project Is A Pioneering Endeavor In The Field Of Medical Imaging And Diagnostics. Leveraging Advanced Image Processing And Deep Learning Techniques, This Project Is Designed To Detect And Classify Bone Fractures Based On Uploaded Medical Images. Specifically, It Focuses On The Identification Of Fractures In The Elbow, Hand, And Shoulder Regions. The Bone Fracture Detection And Classification Project Represents A Significant Advancement In The Realm Of Medical Imaging, Offering A Cost-effective And Efficient Solution For The Early Detection And Accurate Classification Of Bone Fractures. With Its Focus On The Elbow, Hand, And Shoulder Regions, It Serves As A Valuable Tool For Medical Practitioners, Enabling Timely And Informed Decisions In Patient Care.
Poor Nutrition Can Lead To Reduced Immunity, Increased Susceptibility To Disease, Impaired Physical And Mental Development, And Reduced Productivity. A Conversational Agent Can Support People As A Virtual Coach, However Building Such Systems Still Have Its Associated Challenges And Limitations. This Paper Describes The Background And Motivation For Chatbot Systems In The Context Of Healthy Nutrition Recommendation. We Discuss Current Challenges Associated With Chatbotapplication, We Tackled Technical, Theoretical, Behavioural, And Social Aspects Of The Challenges. We Then Propose A Pipeline To Be Used As Guidelines By Developers To Implement Theoretically And Technically Robust Chatbot Systems
Hospital Readmissions Pose Additional Costs And Discomfort For The Patient And Their Occurrences Are Indicative Of Deficient Health Service Quality, Hence Efforts Are Generally Made By Medical Professionals In Order To Prevent Them. These Endeavors Are Especially Critical In The Case Of Chronic Conditions, Such As Diabetes. Recent Developments In Machine Learning Have Been Successful At Predicting Readmissions From The Medical History Of The Diabetic Patient. However, These Approaches Rely On A Large Number Of Clinical Variables Thereby Requiring Deep Learning Techniques. This Article Presents The Application Of Simpler Machine Learning Models Achieving Superior Prediction Performance While Making Computations More Tractable. Index Terms—diabetes, Hospital Readmission, Neural Network, Random Forest, Logistic Regression.
Poor Nutrition Can Lead To Reduced Immunity, Increased Susceptibility To Disease, Impaired Physical And Mental Development, And Reduced Productivity. A Conversational Agent Can Support People As A Virtual Coach, However Building Such Systems Still Have Its Associated Challenges And Limitations. This Paper Describes The Background And Motivation For Chatbot Systems In The Context Of Healthy Nutrition Recommendation. We Discuss Current Challenges Associated With Chatbotapplication, We Tackled Technical, Theoretical, Behavioural, And Social Aspects Of The Challenges. We Then Propose A Pipeline To Be Used As Guidelines By Developers To Implement Theoretically And Technically Robust Chatbot Systems.
Is Human Eye Illness Which Occurs In Individuals Who Have Diabetics Which Harms Their Retina And In The Long Run, May Lead Visual Deficiency. Till Now DR Is Being Screened Manually By Ophthalmologist Which Is A Very Time Consuming Procedure. And Henceforth This Task (project) Focuses On Analysis Of Different DR Stages, Which Is Done With Deep Learning (DL) And It Is A Subset Of Artificial Intelligence (AI). We Trained A Model Called DenseNet On An Enormous Dataset Including Around 3662 Train Images To Automatically Detect The DR Stage And These Are Classified Into High Resolution Fundus Images. The Dataset Which Are Using Is Available On Kaggle (APTOS). There Are Five DR Stages, Which Are 0, 1, 2, 3, And 4. In This Paper Patient’s Fundus Eye Images Are Used As The Input Parameters. A Trained Model (DenseNet Architecture) Will Further Extract The Feature Of Fundus Images Of Eye And After That Activation Function Gives The Output. This Architecture Gave An Accuracy Of 0.9611 (quadratic Weighted Kappa Score Of 0.8981) To DR Detection. And In The End, We Are Comparing The Two CNN Architectures, Which Are VGG16 Architecture And DenseNet121 Architecture
Exponential Growth Of Fake ID Cards Generation Leads To Increased Tendency Of Forgery With Severe Security And Privacy Threats. University ID Cards Are Used To Authenticate Actual Employees And Students Of The University. Manual Examination Of ID Cards Is A Laborious Activity, Therefore, In This Paper, We Propose An Effective Automated Method For Employee/student Authentication Based On Analyzing The Cards. Additionally, Our Method Also Identifies The Department Of Concerned Employee/student. For This Purpose, We Employ Different Image Enhancement And Morphological Operators To Improve The Appearance Of Input Image Better Suitable For Recognition. More Specifically, We Employ Median Filtering To Remove Noise From The Given Input Image.
Thyroid Disease Is A Major Cause Of Formation In Medical Diagnosis And In Theprediction, Onset To Which It Is A Difficult Axiomin The Medical Research. Thyroid Gland Is One Of The Most Important Organs In Our Body. The Secretions Of Thyroid Hormones Are Culpable In Controlling The Metabolism.Hyperthyroidism And Hypothyroidism Are One Of The Two Common Diseases Of The Thyroid That Releases Thyroid Hormones In Regulating The Rate Of Body’s Metabolism. Data Cleansing Techniques Were Applied To Make The Data Primitive Enough For Performing Analytics To Show The Risk Of Patients Obtaining Thyroid. The Machine Learning Plays A Decisive Role In The Process Of Disease Prediction And This Paper Handles The Analysis Andclassificationmodels That Are Being Used In The Thyroid Disease Based On The Information Gathered From The Dataset Taken From UCI Machine Learning Repository. It Is Important To Ensure A Decent Knowledge Base That Can Be Entrenched And Used As A Hybrid Model In Solving Complex Learning Task, Such As In Medical Diagnosis And Prognostic Tasks. In This Paper, We Also Proposed Different Machine Learning Techniques And Diagnosis For The Prevention Of Thyroid. Machine Learning Algorithms, Support Vector Machine (SVM), K-NN, Decision Trees Were Used To Predict The Estimated Risk On A Patient’s Chance Of Obtaining Thyroid Disease.
Social Connections Evolved Within Local Cultural Boundaries Suchs As Geosspatials Areas Prior To The Invention Of Informations Communicationstechnologys (ICT). The Recent Advancements In Communication Technologies Have Significantly Surpassed The Old Communications' Time And Spatial Limits. These Social Technologiess Have Ushereds In A News Eras Of User-generated Content, Online Human Networks, And Rich Data On Human Behaviour. However, The Misuses Of Socials Technologiess Such As Socialsmedias (SM) Stages Has Resulted In A New Type Of Online Anger And Violence. This Research Highlights A Novels Ways Of Exhibiting Hostile Conduct On Social Media Websites. The Reasons For Developing Prediction Models To Combat Aggressive Behaviour In SM Are Also Discussed. We Examine Cyberbullyings Predictionsmodelss In Depth And Identify The Major Challenges That Arise While Building Cyberbullying Predictionsmodelssin SM. This Paper Gives A Summarys Of The General Procedure For Detecting Cyberbullying And, More Crucially, The Technique. Despites The Facts That The Data Collecting And Feature Engineering Processes Have Been Detailed, The Focus Is Mostly On Feature Selection Methods And Then The Application Of Various Machine Learning Algorithms To Anticipate Cyberbullyingbehaviours. Finally, The Concerns And Obstacles Have Been Identified, Presenting New Study Directions For Scholars To Investigate.
Hospital Readmissions Pose Additional Costs And Discomfort For The Patient And Their Occurrences Are Indicative Of Deficient Health Service Quality, Hence Efforts Are Generally Made By Medical Professionals In Order To Prevent Them. These Endeavors Are Especially Critical In The Case Of Chronic Conditions, Such As Diabetes. Recent Developments In Machine Learning Have Been Successful At Predicting Readmissions From The Medical History Of The Diabetic Patient. However, These Approaches Rely On A Large Number Of Clinical Variables Thereby Requiring Deep Learning Techniques. This Article Presents The Application Of Simpler Machine Learning Models Achieving Superior Prediction Performance While Making Computations More Tractable. Index Terms—diabetes, Hospital Readmission, Neural Network, Random Forest, Logistic Regression.
Prediction Of Academic Performance Of Students Beforehand Provides Scope To Universities To Lower Their Dropout Rate And Help The Students In Improving Their Performance. In This Field, Research Is Being Done To Find Out Which Algorithm Is Best To Use And Which Features Should Be Considered While Predicting The Academic Performance Of Students. This Kind Of Research Work Has Been Increasing Over The Years. This Paper Performs A Survey On The Techniques Used In Various Research Papers For Academic Performance Prediction And Also Point Out The Limitations If Any, In The Methodology Used.
Liver, A Crucial Interior Organ Of The Human Body Whose Principal Tasks Are To Eliminate Generated Waste Produced By Our Organism, Digest Food, And Preserve Vitamins And Energy Materials. The Liver Disease Can Cause Various Fatal Diseases, Including Liver Cancer. Early Diagnosis, And Treating The Patients Are Compulsory To Reduce The Risk Of Those Lethal Diseases. As The Diagnosis Of Liver Disease Is Expensive And Sophisticated, Numerous Researches Have Been Performed Using Machine Learning (ML) Methods For Classifying Liver Disorder Cases. In This Paper, We Have Compared Four Different ML Algorithms Such As Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), And Nearest Neighbour Classifier (NN) , Support Vector Classifier (SVM), Gaussian Naïve Bayes (GNB) For Classifying Indian Liver Patient Dataset (ILPD). Pearson Correlation Coefficient Based Feature Selection (PCC-FS) Is Applied To Eliminate Irrelevant Features From The Dataset. The Comparative Analysis Is Evaluated In Terms Of Accuracy, ROC, F-1 Score, Precision, And Recall. After Comparing Experimental Results, We Have Found That Logistic Regression On ET Provides The Highest Accuracy Of 71.24 %..
Prediction Of Academic Performance Of Students Beforehand Provides Scope To Universities To Lower Their Dropout Rate And Help The Students In Improving Their Performance. In This Field, Research Is Being Done To Find Out Which Algorithm Is Best To Use And Which Features Should Be Considered While Predicting The Academic Performance Of Students. This Kind Of Research Work Has Been Increasing Over The Years. This Paper Performs A Survey On The Techniques Used In Various Research Papers For Academic Performance Prediction And Also Point Out The Limitations If Any, In The Methodology Used.
Cervical Cancer Is One Of The Main Reasons Of Death From Cancer In Women. The Complication Of This Cancer Can Be Limited If It Is Diagnosed And Treated At An Early Stage. In This Paper, We Propose A Cervical Cancer Cell Detection And Classification System Based On Convolutional Neural Networks (CNNs). The Cell Images Are Fed Into A CNNs Model To Extract Deep-learned Features. Cervical Cancer Is Screened Using Visual Inspection After Application Of Acetic Acid (VIA), Papanicolaou (Pap) Test, Human Papillomavirus (HPV) Test And Histopathology Test. Inter- And Intra-observer Variability May Occur During The Manual Diagnosis Procedure, Resulting In Misdiagnosis. The Purpose Of This Study Was To Develop An Integrated And Robust System For Automatic Cervix Type And Cervical Cancer Classification Using Deep Learning Techniques.
Poor Nutrition Can Lead To Reduced Immunity, Increased Susceptibility To Disease, Impaired Physical And Mental Development, And Reduced Productivity. A Conversational Agent Can Support People As A Virtual Coach, However Building Such Systems Still Have Its Associated Challenges And Limitations. This Paper Describes The Background And Motivation For Chatbot Systems In The Context Of Healthy Nutrition Recommendation. We Discuss Current Challenges Associated With Chatbotapplication, We Tackled Technical, Theoretical, Behavioural, And Social Aspects Of The Challenges. We Then Propose A Pipeline To Be Used As Guidelines By Developers To Implement Theoretically And Technically Robust Chatbot Systems.
Pneumonia Is A Prevalent And Life-threatening Respiratory Disease That Affects Individuals Of All Age Groups Worldwide. Timely And Accurate Diagnosis Is Crucial For Effective Patient Management And Treatment. In Recent Years, Deep Learning Techniques Have Shown Great Promise In Automating The Detection Of Pneumonia From Medical Images, Particularly X-ray Radiographs. This Study Focuses On The Development Of A Deep Learning-based System For Pneumonia Detection Using X-ray Images. The Primary Objective Of This Research Is To Design, Implement, And Evaluate A Robust And Accurate Deep Learning Model That Can Aid Medical Professionals In The Rapid And Reliable Identification Of Pneumonia. The Proposed Model Leverages Convolutional Neural Networks (CNNs) And State-of-the-art Deep Learning Architectures To Extract Relevant Features From X-ray Images. These Features Are Then Used For Binary Classification, Distinguishing Between Normal And Pneumonia-affected Cases. In Conclusion, This Study Presents A Comprehensive Exploration Of The Application Of Deep Learning In Pneumonia Detection Using X-ray Images, With The Aim Of Advancing The State Of The Art In Automated Diagnostic Tools For Pneumonia In The Medical Field.
Prediction Of Academic Performance Of Students Beforehand Provides Scope To Universities To Lower Their Dropout Rate And Help The Students In Improving Their Performance. In This Field, Research Is Being Done To Find Out Which Algorithm Is Best To Use And Which Features Should Be Considered While Predicting The Academic Performance Of Students. This Kind Of Research Work Has Been Increasing Over The Years. This Paper Performs A Survey On The Techniques Used In Various Research Papers For Academic Performance Prediction And Also Point Out The Limitations If Any, In The Methodology Used. Prediction Of Academic Performance Of Students Beforehand Provides Scope To Universities To Lower Their Dropout Rate And Help The Students In Improving Their Performance. In This Field, Research Is Being Done To Find Out Which Algorithm Is Best To Use And Which Features Should Be Considered While Predicting The Academic Performance Of Students. This Kind Of Research Work Has Been Increasing Over The Years. This Paper Performs A Survey On The Techniques Used In Various Research Papers For Academic Performance Prediction And Also Point Out The Limitations If Any, In The Methodology Used.
Chatbot For Bag Classification And Product Suggestion Is An Emerging Technology That Can Revolutionize The E-commerce Industry. The Chatbot Interacts With The User In A Conversational Manner And Uses Natural Language Processing And Machine Learning Algorithms To Classify Bags Based On Various Attributes Such As Color, Size, Material, And Style. The Chatbot Can Also Suggest Products Based On The User's Preferences, Purchase History, And Browsing Behavior. The Chatbot Can Enhance The User's Shopping Experience By Providing Personalized Product Recommendations, Reducing The Time And Effort Required To Search For Products, And Improving Customer Engagement And Satisfaction. The Chatbot Can Also Help E-commerce Businesses Streamline Their Inventory Management By Providing Insights Into User Preferences And Behaviour.
One Of The Most Rapidly Spreading Cancers Among Various Other Types Of Cancers Known To Humans Is Skin Cancer. Melanoma Is The Worst And The Most Dangerous Type Of Skin Cancer That Appears Usually On The Skin Surface And Then Extends Deeper Into The Layers Of Skin. However, If Diagnosed At An Early Stage; The Survival Rate Of Melanoma Patients Is 96% With Simple And Economical Treatments. The Conventional Method Of Diagnosing Melanoma Involves Expert Dermatologists, Equipment, And Biopsies. To Avoid The Expensive Diagnosis, And To Assist Dermatologists, The Field Of Machine Learning Has Proven To Provide State Of The Art Solutions For Skin Cancer Detection At An Earlier Stage With High Accuracy. In This Paper, A Method For Skin Lesion Classification And Segmentation As Benign Or Malignant Is Proposed Using Image Processing And Machine Learning. A Novel Method Of Contrast Stretching Of Dermoscopic Images Based On The Methods Of Mean Values And Standard Deviation Of Pixels Is Proposed. Then The OTSU Thresholding Algorithm Is Applied For Image Segmentation. After The Segmentation, Features Including Gray Level Co-occurrence Matrix (GLCM) Features For Texture Identification, The Histogram Of Oriented Gradients (HOG) Object, And Color Identification Features Are Extracted From The Segmented Images. Principal Component Analysis (PCA) Reduction Of HOG Features Is Performed For Dimensionality Reduction. Synthetic Minority Oversampling Technique (SMOTE) Sampling Is Performed To Deal With The Class Imbalance Problem. The Feature Vector Is Then Standardized And Scaled.
Machine Learning Is A Branch Of Artificial Intelligence. In Recent Years, With The Advantages Of Automatic Learning And Feature Extraction, It Has Been Widely Concerned By Academic And Industrial Circles. It Has Been Widely Used In Image And Video Processing, Voice Processing, And Natural Language Processing. At The Same Time, It Has Also Become A Research Hotspot In The field Of Agricultural Plant Protection, Such As Plant Disease Recognition And Pest Range Assessment, Etc. The Application Of Machine Learningin Plant Disease Recognition Can Avoid The Disadvantages Caused By Artificial Selection Of Disease Spot Features, Make Plant Disease Feature Extraction More Objective,and Improve There Search Efficiency And Technology Transformationspeed. This Review Provides There Search Progress Of Machine Learning Technology In The field Of Crop Leaf Disease Identification In Recent Years. In This Paper, We Present The Current Trends And Challenges For The Detection Of Plant Leaf Disease Using Machine Learning And Advanced Imaging Techniques. We Hope That This Work Will Be A Valuable Resource For Researchers Who Study The Detection Of Plant Diseases And Insect Pests. At The Same Time, We Also Discussed Some Of The Current Challenges And Problems That Need To Be Resolved.
Deep Learning Is An Effective And Useful Technique That Has Been Widely Applied In A Variety Of Fields, Including Computer Vision, Machine Vision, And Natural Language Processing. Deepfakes Uses Deep Learning Technology To Manipulate Images And Videos Of A Person That Humans Cannot Differentiate Them From The Real One. In Recent Years, Many Studies Have Been Conducted To Understand How Deepfakes Work And Many Approaches Based On Deep Learning Have Been Introduced To Detect Deepfakes Videos Or Images. In This Paper, We Conduct A Comprehensive Review Of Deepfakes Creation And Detection Technologies Using Deep Learning Approaches. In Addition, We Give A Thorough Analysis Of Various Technologies And Their Application In Deepfakes Detection. Our Study Will Be Beneficial For Researchers In This Field As It Will Cover The Recent State-of-art Methods That Discover Deepfakes Videos Or Images In Social Contents. In Addition, It Will Help Comparison With The Existing Works Because Of The Detailed Description Of The Latest Methods And Dataset Used In This Domain.
Classification Is Now Simpler Thanks To Deep Learning, Bigger Datasets, And More Powerful Computing Power. Convolutional Neural Networks Have Emerged As The Most Popular And Widely Used Method For Categorising Images In Recent Years. Several Transfer Learning Techniques Are Employed In This Article To Categorise Photos From A Dataset Of Indian Food. Everyone Needs To Keep Track Of Their Eating Habits Because Food Is Essential To Life Because It Provides Us With A Range Of Nutrients. Food Classification Is Therefore Essential For A Better Way Of Life. In This Project, Pre-trained Models Are Used Instead Of Conventional Methods For Building A Model From Scratch, Which Reduces Processing Costs And Time While Simultaneously Generating Better Outcomes.
In The Present World, We Have Wide Varieties Of Species And Organisms. This Brings Into Light, The Criticality Of Classification Of Various Tangible Objects. Also, Keeping In Mind, The On-going Research On Genetics And Evolution By Various Scientists Across The World, Discerning The Resemblance Among Different Classes Also Becomes Very Crucial. This Paper Is Based On A Project Which Builds A CNN (Convolutional Neural Network) To Classify Different Dog Breeds. If The Image Of A Dog Is Found, This Algorithm Would Find The Estimate Of The Breed. The Resembling Dog Breed Is Identified If The Image Of A Human Is Supplied. We Have Built A Pipeline To Process Real World Images.
Underwater Fish Detection Plays A Crucial Role In Various Domains, Such As Marine Biology, Environmental Monitoring, And Fisheries Management. Traditional Methods For Fish Detection In Underwater Environments Often Suffer From Limitations In Accuracy And Efficiency. In Recent Years, Deep Learning-based Object Detection Models Have Shown Remarkable Success In Various Computer Vision Tasks. This Study Presents An Underwater Fish Detection System Utilizing The YOLOv8 Architecture, A State-of-the-art Deep Learning Framework
Deep Learning Is A Branch Of Artificial Intelligence. In Recent Years, With The Advantages Of Automatic Learning And Feature Extraction, It Has Been Widely Concerned By Academic And Industrial Circles. It Has Been Widely Used In Image And Video Processing, Voice Processing, And Natural Language Processing. At The Same Time, It Has Also Become A Research Hotspot In The field Of Agricultural Plant Protection, Such As Plant Disease Recognition And Pest Range Assessment, Etc. The Application Of Deep Learning In Plant Disease Recognition Can Avoid The Disadvantages Caused By Artificial Selection Of Disease Spot Features, Make Plant Disease Feature Extraction More Objective, And Improve Their Search Efficiency And Technology Transformation Speed. This Review Provides The Research Progress Of Deep Learning Technology In The field Of Crop Leaf Disease Identification In Recent Years. In This Paper, We Present The Current Trends And Challenges For The Detection Of Plant Leaf Disease Using Deep Learning And Advanced Imaging Techniques. We Hope That This Work Will Be A Valuable Resource For Researchers Who Study The Detection Of Plant Diseases And Insect Pests. At The Same Time, We Also Discussed Some Of The Current Challenges And Problems That Need To Be Resolved.
Hairstyling Is An Art Of Fashion Transformed Since Ancient Era, With The Influences From Many Diverse Factors. It Has Been A Primary Aspect Of Human Lifestyle And Society In Various Different Ways With The Growth Of Research Fields Like Modeling Human, Visual Searching, Visual Matching, Facial Verification For Security Measures And Etc. Perfect Hairstyle Improves Specially A Woman’s Self-confidence. This Paper Presents A Hairstyle Recommendation System Based On Face Shapes And Suitable Hairstyle With Expert’s Knowledge For The Face Shape Derived From Face Shape Classification Algorithm. Different Noise Types And Lighten, Which Have Given Less Accuracy Of Features Landmark Detection. The Images Whose Values Are Confined To Some Specific Range Of Values Only. For Brighter Image Will Have All Pixels Confined High Values. But A Good Image Will Have Pixels From All Regions Of The Image. Therefore, We Applied Histogram Equalization Improves The Contrast Of The Image. The Median Filter Is Used To Remove The Different Types Of Noises.
Underwater Fish Detection Plays A Crucial Role In Various Domains, Such As Marine Biology, Environmental Monitoring, And Fisheries Management. Traditional Methods For Fish Detection In Underwater Environments Often Suffer From Limitations In Accuracy And Efficiency. In Recent Years, Deep Learning-based Object Detection Models Have Shown Remarkable Success In Various Computer Vision Tasks. This Study Presents An Underwater Fish Detection System Utilizing The YOLO Architecture, A State-of-the-art Deep Learning Framework.
This Study Is An Attempt To Understand And Address The Cancer Disease Prediction ,—Since, Cancer Is Curable When Diagnosed At An Early Stage, Lung Cancer Screening Plays An Important Role In Preventive Care. Although Both Low Dose Computed Tomography (LDCT) And Computed Tomography (CT) Scans Provide Greater Medical Information Than Normal Chest X-rays, Access To These Technologies In Rural Areas Is Very Limited. There Is A Recent Trend Toward Using Computer-aided Diagnosis (CADx) To Assist In The Screening And Diagnosis Of Cancer From Biomedical Images. In This Study, The 121-layer Convolutional Neural Networkalong With The Transfer Learning Scheme Is Explored As A Means Of Classifying Lung Cancer Using Chest Xray Images. The Model Was Trained On A Lung Nodule Dataset Before Training On The Lung Cancer Dataset To Alleviate The Problem Of Using A Small Dataset. The Proposed Model Yields 74.43±6.01% Of Mean Accuracy, 74.96±9.85% Of Mean Specificity, And 74.68±15.33% Of Mean Sensitivity. The Proposed Model Also Provides A Heatmap For Identifying The Location Of The Lung Nodule. These Findings Are Promising For Further Development Of Chest X-ray-based Lung Cancer Diagnosis Using The Deep Learning Approach. Moreover, They Solve The Problem Of A Small Dataset..
One Of The Most Difficult Issues Is Conflict Between Humans And Animals. The Main Reason For Animal Detection, Particularly Wild Animal Infiltration Into Human Habitation, Has Been The Expansion Of Human Settlement Into Forest Areas. The Conflict Between Humans And Animals Is Largely A Result Of The Loss Of Habitat, Food, And Other Necessities For The Animal Kingdom. Many Catastrophic Events In The Future Can Be Prevented By The Early Recognition Of These Conflicts. The Scope Of Earlier Automated Image Detection Processes Aids In Finding A Remedy For These Occurrences. The Detection Can Utilise An Algorithm For Earlier Detection Because The Manual, Traditional Method Is More Difficult. Different Deep Learning Algorithms Help To Identify Invaders Sooner To Prevent Confrontation.
Pneumonia Is A Prevalent And Life-threatening Respiratory Disease That Affects Individuals Of All Age Groups Worldwide. Timely And Accurate Diagnosis Is Crucial For Effective Patient Management And Treatment. In Recent Years, Deep Learning Techniques Have Shown Great Promise In Automating The Detection Of Pneumonia From Medical Images, Particularly X-ray Radiographs. This Study Focuses On The Development Of A Deep Learning-based System For Pneumonia Detection Using X-ray Images. The Primary Objective Of This Research Is To Design, Implement, And Evaluate A Robust And Accurate Deep Learning Model That Can Aid Medical Professionals In The Rapid And Reliable Identification Of Pneumonia. The Proposed Model Leverages Convolutional Neural Networks (CNNs) And State-of-the-art Deep Learning Architectures To Extract Relevant Features From X-ray Images. These Features Are Then Used For Binary Classification, Distinguishing Between Normal And Pneumonia-affected Cases.
Rice Is One Of The World's Most Important Staple Crops, And The Health Of Rice Plants Is Crucial For Global Food Security. Detecting And Classifying Diseases In Rice Leaves Is Essential To Ensure A Healthy Crop Yield. In This Project, We Propose A System For The Classification Of Rice Leaf Diseases Using Convolutional Neural Networks (CNN) And Transfer Learning. We Aim To Leverage The Power Of Deep Learning And Pre-trained Models To Accurately Identify And Categorize Various Rice Leaf Diseases, Providing Farmers And Agronomists With An Efficient Tool For Early Disease Detection. This Paper Discusses The Existing System's Limitations And Presents The Advantages Of Our Proposed Approach.
Potato Is One Of The Prominent Food Crops All Over The World. In Some Places Potato Cultivation Has Been Getting Remarkable Popularity Over The Last Decades. Many Diseases Affect The Proper Growth Of Potato Plants. Noticeable Diseases Are Seen In The Leaf Region Of This Plant. Two Common And Popular Leaf Diseases Of The Potato Plants Are Early Blight (EB) And Late Blight (LB). However, If These Diseases Were Identified At An Early Stage It Would Be Very Helpful For Better Production Of This Crop. To Solve This Problem By Detecting And Analyzing These Diseases Image Processing Is The Best Option. This Paper Proposes An Image Processing And Deep Learning-based Automatic System That Will Identify And Classify Potato Leaf Diseases. In This Paper, Image Segmentation Is Done Over 152 Images Of Healthy And Diseased Potato Leaf, Which Is Taken From Publicly Available Plant Village Database And Seven Classifier Algorithms Are Used For Recognition And Classification Of Diseased And Healthy Leaves. Among Them, The Convolutional Neural Network Gives An Accuracy Of 97%. In This Manner, Our Proposed Approach Leads To A Path Of Automatic Plant Leaf Disease Detection.
The Process Of Discovering Or Mining Information From A Huge Volume Of Data Is Known As Data Mining Technology. Today Data Mining Has Lots Of Application In Every Aspects Of Human Life. Applications Of Data Mining Are Wide And Diverse. Among This Health Care Is A Major Application Of Data Mining. Medical Field Has Get Benefited More From Data Mining. Heart Disease Is The Most Dangerous Life-threatening Chronic Disease Globally. The Objective Of The Work Is To Predicts The Occurrence Of Heart Disease Of A Patient Using Random Forest Algorithm. The Dataset Was Accessed From Kaggle Site. The Dataset Contains 303 Samples And 14 Attributes Are Taken For Features Of The Dataset. Then It Was Processed Using Python Open Access Software In Jupyter Notebook. The Datasets Are Classified And Processed Using Machine Learning Algorithm Random Forest. The Outcomes Of The Dataset Are Expressed In Terms Of Accuracy, Sensitivity And Specificity In Percentage. Using Random Forest Algorithm, We Obtained Accuracy Of 86.9% For Prediction Of Heart Disease With Sensitivity Value 90.6% And Specificity Value 82.7%. From The Receiver Operating Characteristics, We Obtained The Diagnosis Rate For Prediction Of Heart Disease Using Random Forest Is 93.3%. The Random Forest Algorithm Has Proven To Be The Most Efficient Algorithm For Classification Of Heart Disease And Therefore It Is Used In The Proposed System.
Developing An Android Application For Character Recognition To Read The Text From An Image Is A Big Area Of Research. Nowadays, There Is A Trend Of Storing Information From The Handwritten Documents For Future Use. A Simple Way To Store The Information Is Image Capturing Of The Handwritten Document And Save It In Image Format. The Method To Transform Handwritten Data Into Electronic Format Is ‘Optical Character Recognition’. It Involves Several Steps Including Pre-processing, Segmentation, Feature Extraction And Post-processing. Many Researchers Have Been Used OCR For Recognizing Character. This System Uses The Android Phone To Capture The Image Of The Document And Further Steps Are Done By OCR. The Main Challenge Is To Recognize The Characters From Different Styles Of Handwriting. Thus, A System Is Designed That Recognizes The Handwritten Data To Obtain An Editable Text. The Output Of This System Depends Upon The Data That Has To Be Written By The Writer. Our System Offers 90% Accuracy For Handwritten Documents And Gives The Easiest Way To Edit Or Share The Recognized Data.
Falling Is A Significant Health Problem. Fall Detection, To Alert For Medical Attention, Has Been Gaining Increasing Attention. Still, Most Of The Existing Studies Use Falls Simulated In A Laboratory Environment To Test The Obtained Performance. We Analyzed The Acceleration Signals Recorded By An Inertial Sensor On The Lower Back During 143 Real-world Falls (the Most Extensive Collection To Date) From The FARSEEING Repository. Such Data Were Obtained From Continuous Real-world Monitoring Of Subjects With A Moderate-to-high Risk Of Falling. We Designed And Tested Fall Detection Algorithms Using Features Inspired By A Multiphase Fall Model And A Machine Learning Approach Such As SVM And Decision Tree. The Obtained Results Suggest That Algorithms Can Learn Effectively From Features Extracted From A Multiphase Fall Model, Consistently Overperforming More Conventional Features. The Most Promising Method (support Vector Machines And Features From The Multiphase Fall Model) Obtained A Sensitivity Higher Than 80%, A False Alarm Rate Per Hour Of 0.56, And An F-measure Of 64.6%. The Reported Results And Methodologies Represent An Advancement Of Knowledge On Real-world Fall Detection And Suggest Useful Metrics For Characterizing Fall Detection Systems For Real-world Use. The SVM And Decision Tree Has Implemented For The Classification Of The Fall.
Malaria Is The Deadliest Disease In The Earth And Big Hectic Work For The Health Department. The Traditional Way Of Diagnosing Malaria Is By Schematic Examining Blood Smears Of Human Beings For Parasite-infected Red Blood Cells Under The Microscope By Lab Or Qualified Technicians. This Process Is Inefficient And The Diagnosis Depends On The Experience And Well Knowledgeable Person Needed For The Examination. Deep Learning Algorithms Have Been Applied To Malaria Blood Smears For Diagnosis Before. However, Practical Performance Has Not Been Sufficient So Far. This Paper Proposes A New And Highly Robust Machine Learning Model Based On A Convolutional Neural Network (CNN) Which Automatically Classifies And Predicts Infected Cells In Thin Blood Smears On Standard Microscope Slides. A Ten-fold Cross-validation Layer Of The Convolutional Neural Network On 27,558 Single-cell Images Is Used To Understand The Parameter Of The Cell. Three Types Of CNN Models Are Compared Based On Their Accuracy And Select The Precise Accurate - Basic CNN, VGG-19 Frozen CNN, And VGG-19 Fine Tuned CNN. Then By Comparing The Accuracy Of The Three Models, The Model With A Higher Rate Of Accuracy Is Acquired.
The World Is Advancing Towards An Autonomous Environment At A Great Pace And It Has Become A Need Of An Hour, Especially During The Current Pandemic Situation. The Pandemic Has Hindered The Functioning Of Many Sectors, One Of Them Being Road Development And Maintenance. Creating A Safe Working Environment For Workers Is A Major Concern Of Road Maintenance During Such Difficult Times. This Can Be Achieved To Some Extent With The Help Of An Autonomous System That Will Aim At Reducing Human Dependency. In This Paper, One Of Such Systems, A Pothole Detection And Dimension Estimation, Is Proposed. The Proposed System Uses A Deep Learning Based Algorithm YOLO (You Only Look Once) For Pothole Detection. Further, An Image Processing Based Triangular Similarity Measure Is Used For Pothole Dimension Estimation. The Proposed System Provides Reasonably Accurate Results Of Both Pothole Detection And Dimension Estimation. The Proposed System Also Helps In Reducing The Time Required For Road Maintenance. The System Uses A Custom Made Dataset Consisting Of Images Of Water-logged And Dry Potholes Of Various Shapes And Sizes.
In This Paper, Our Main Aim Is To Create A Medicinal Plant Identification System Using Deep Learning Concept. This System Will Classify The Medicinal Plant Species With High Accuracy. Identification And Classification Of Medicinal Plants Are Essential For Better Treatment. In This System We Are Going To Use Five Different Indian Medicinal Plant Species Namely Alstonia, Chinar, And Jamun (Naval)l. We Utilize Dataset Contains 58,280 Images, Includes Approximately 10,000 Images For Each Species. We Use Leaf Texture, Shape, And Color, Physiological Or Morphological As The Features Set Of The Data. The Data Are Collected By Us. We Use CNN Architecture To Train Our Data And Develop The System With High Accuracy. Several Model Architectures Were Trained, With The Best Performance Reaching A 95.67% Success Rate In Identifying The Corresponding Medicinal Plant. The Significantly High Success Rate Makes The Model A Very Useful Advisory Or Early Warning Tool.
The Leather Industry Plays A Crucial Role In Fashion, Manufacturing, And Luxury Goods, Where The Quality Of Leather Products Is Of Paramount Importance. Ensuring The Absence Of Defects In Leather Is A Fundamental Aspect Of Maintaining Product Quality. In This Context, We Present A Cutting-edge Approach To Leather Defect Detection Using Deep Learning Techniques. Our Research Leverages The Power Of Deep Neural Networks, Specifically Convolutional Neural Networks (CNNs), To Develop A Robust And Efficient System For Identifying And Classifying Defects In Leather Hides. We Employ State-of-the-art Image Analysis Methods To Process High-resolution Images Of Leather Surfaces, Enabling The Automated Detection Of Various Types Of Defects, Including Scars, Blemishes, Wrinkles, And Discolorations. This Paper Outlines The Design, Training, And Evaluation Of The Deep Learning Model, Highlighting Its Ability To Accurately And Rapidly Detect Defects With High Precision And Recall. The Proposed System Holds The Potential To Revolutionize Quality Control Processes Within The Leather Industry, Improving Product Quality, Reducing Waste, And Enhancing Overall Efficiency. As The Demand For High-quality Leather Products Continues To Grow, Our Work Demonstrates The Potential Of Deep Learning To Advance Quality Control And Defect Detection In This Industry, Ultimately Contributing To The Production Of Flawless Leather Goods.
Smartphones Are Becoming More Preferred Companions To Users Than Desktops Or Notebooks. Knowing That Smartphones Are Most Popular With Users At The Age Around 26, Using Smartphones To Speed Up The Process Of Taking Attendance By University Instructors Would Save Lecturing Time And Hence Enhance The Educational Process. This Paper Proposes A System That Is Based On A BAR Code, Which Is Being Displayed For Students During Or At The Beginning Of Each Lecture. The Students Will Need To Scan The Code In Order To Confirm Their Attendance. The Paper Explains The High Level Implementation Details Of The Proposed System. It Also Discusses How The System Verifies Student Identity To Eliminate False Registrations.
Biometrics With Facial Recognition Is Now Widely Used. A Face Identification System Should Identify Not Only Someone's Faces But Also Detect Spoofing Attempts With Printed Face Or Digital Presentations. A Sincere Spoofing Prevention Approach Is To Examine Face Liveness As Eye Blinking. Nevertheless, This Approach Is Helpless When Dealing With Video-based Replay Attacks. For This Reason, This Paper Proposes A Combined Method Of Face Liveness Detection And CNN (Convolutional Neural Network) Classifier. The Test Results Show That The Module Created Can Recognize Various Kinds Of Facial Spoof Attacks, Such As Using Posters, Masks, Or Smartphones And Second Step Verification. Then Develops A Model To Classify Each Character's Face From A Captured Image Using A Collection Of Rules I.e., HOG Algorithm To Record The Student Attendance. The Proposed ASAS (Automated Smart Attendance System) Will Capture The Image And Will Be Compared To The Image Stored In The Database. ASAS Marks Individual Attendance, If They Image Matches The Image In The Database. The Proposed Algorithm Reduces Effort And Captures Day-to-day Actions Of Managing Each Student And Also Makes It Simple To Mark The Presence.
This Study Is An Attempt To Understand And Address The Mental Health Issue, Of Working Professionals Through Facial Expression Recognition And Suggest A List Of Appropriate Songs That Can Improve His Mood. A Brief Search Was Conducted On How Music Can Affect The User Mood In Short-term To Gain Knowledge And Enable Us To Provide The Users With A List Of Music Tracks That Work Well On Improving The User Moods. As A Society, We Are All Currently Talking About Ways As To How A Person Who Is Suffering From Any Emotional Issue Can Adopt Certain Ways To Come Out Of A Specific Circumstance And How We As A Society Can Support Such People In These Situations. Our Endeavor Is To Work On A Way Where The Identification Of Such Persons Who Are Going Through A Difficult Phase In Their Life Can Be Performed. Our Endeavor Is To Work On A Way Where The Identification Of Such Persons Who Are Going Through A Difficult Phase In Their Life Can Be Performed. It Is Not Always Evident That A Person Going Through A Tough Phase May Open Up About Their Feelings To People Around Them And Hence Making Use Of AI/ML To Identify A Person’s Emotion Through Their Facial Expressions Captured Over A Span Of Time Thereby Recommending Them Some Activities, Thoughts Which Can Help Them In Getting Over Their Emotions When They Are Sad, Fearful It Suggestion The Music.
A System That Takes Down Students Attendance Using Qr Code. Every Student Is Provided With A Card Containing A Unique Qr Code. Students Just Have To Scan Their Cards In Front Of Webcam And The System Notes Down Their Attendance As Per Dates. Each Qr Code Contains A Unique Id For Students. System Then Stores All The Students’ Attendance Records And Generates Defaulter List. It Also Generates An Overall Report In Excel Sheet For Admin. Such Type Of Application Is Very Useful In School As Well As In College For Daily Attendance.
Women And Girls Have Been Experiencing A Lot Of Violence And Harassment In Public Places In Various Cities Starting From Stalking And Leading To Sexual Harassment Or Sexual Assault. This Research Paper Basically Focuses On The Role Of Social Media In Promoting The Safety Of Women In Indian Cities With Special Reference To The Role Of Social Media Websites And Applications Including Twitter Platform Facebook And Instagram. This Paper Also Focuses On How A Sense Of Responsibility On Part Of Indian Society Can Be Developed The Common Indian People So That We Should Focus On The Safety Of Women Surrounding Them. Tweets On Twitter Which Usually Contains Images And Text And Also Written Messages And Quotes Which Focus On The Safety Of Women In Indian Cities Can Be Used To Read A Message Amongst The Indian Youth Culture And Educate People To Take Strict Action And Punish Those Who Harass The Women. Twitter And Other Twitter Handles Which Include Hash Tag Messages That Are Widely Spread Across The Whole Globe Sir As A Platform For Women To Express Their Views About How They Feel While We Go Out For Work Or Travel In A Public Transport And What Is The State Of Their Mind When They Are Surrounded By Unknown Men And Whether These Women Feel Safe Or Not?
The Interactive Chess Board Game Is Unlike Games In Its Ordinary Way. This Board Game Together With Tangible Movements Of All Pieces Is Considered To Be Users Attraction. Therefore, The New Chessboard With An Automatic Moving Mechanism For Every Piece Is Chosen. Initially, We Have Designed And Developed An Aluminum Core Structure For Positioning X And Y-axis. Furthermore, A Controllable Magnet Is Deliberated For Holding And Moving An Individual Chess Piece According To Player Manipulations. Purpose Of This Interactive Chess Board Is Applying Technology To Board Game For Excitement, Interest, Amazement, And Attraction. Arduino Microcontroller Is Used For Controlling Every Step Of Piece Movement. The Microcontroller Receives Control Information Through The User Interface And Then Moves The Chess Piece To The Destination On The Board. The Position Calculation Is Brought To Identify The Chess Piece And Drive Accurately The Stepper Motors In X And Y-axis.
In This Paper We Introduce An AI Bot To Enhance The Skills Of The Player And The AI Bot Uses The Algorithms Further Discussed In This Paper. Player Can Follow The Simultaneously Running AI Bot To Play The Game Effectively. In This We Use The Classic Snake Game, For That We Present Different Algorithms Or Methods For AI Bot. It Includes Three Searching Algorithms Related To Artificial Intelligence, Best First Search, A* Search And Improved A* Search With Forward Checking, And Two Baseline Methods Random Move And Almighty Move.
In A Generation Led By Millennials, Technologies Are Becoming Redundant Each Year. The Organizations Are Competing On A Global Scale And Newer And Innovative Strategies Are Introduced In The Field Of Marketing To Reach Out To The Potential Buyers. Real Estate, Being One Of The Biggest Business Sectors, Needs More Efficiently Targeted Marketing Campaigns As This Is A Very Nice And Unexplored Field In The Indian Scenario. Real Estate Projects Are Highly Priced Products Which Cannot Be Sold Efficiently Without A Well Strategized Marketing Campaign So As To Reach Out To The Exact Targeted Market. The Unsold Inventories In Various Metro Cities Range From 15-60%. The Stipulated Real Estate Sector Growth Trends By Government Do Not Go Hand In Hand With The On-ground Realities Of The Piled-up Inventories. The Marketing Strategies Have Not Evolved With The Digitization Boom, And Still The Real Estate Marketing Techniques Are Conventional And Financially Heavy. Through This Study, Efforts Have Been Made To Pin Point The Existing Supply-demand Problems In The Cities Of Ahmedabad And Mumbai In The Affordable And HIG Housing Sector Specifically. Also, Suitable Solutions For Marketing Campaigns Have Been Proposed Considering The Current Market Realities For Both Cities.
The Basic Nonverbal Interaction That Is Now Evolving In The Upcoming Generation Is Eye Gaze. This Eye Blink System Builds A Bridge For Communication Of People Affected With Disabilities. The Operation Is So Simple That With The Eyes Blinking At The Control Keys That Are Built In The Screen . This Type Of System Can Synthesize Speech, Control His Environment, And Give A Major Development Of Confidence In The Individual . Our Paper Mainly Enforces The Virtual Keyboard That Not Only Has The Built In Phrases But Also Can Provide The Voice Notification/ Speech Assistance For The People Who Are Speech Disabled. To Achieve This We Have Used Our Pc/laptop Camera Which Is Built In And It Recognizes The Face And Parts Of The Face. This Makes The Process Of Detecting The Face Much Easier Than Anything. The Eye Blink Serves As The Alternative For A Mouse Click On The Virtual Interface. As Already Mentioned, Our Ultimate Achievement Is To Provide A Nonverbal Communication And Hence The Physically Disabled People Should Get A Mode Of Communication Along With A Voice Assistant. This Type Of Innovation Is A Golden Fortune For The People Who Lost Their Voice And Affected To Paralytic Disorders. We Have Further Explained With The Respective Flowcharts And With Each Juncture
The Present-day World Has Become All Dependent On Cyberspace For Every Aspect Of Daily Living. The Use Of Cyberspace Is Rising With Each Passing Day. The World Is Spending More Time On The Internet Than Ever Before. As A Result, The Risks Of Cyber Threats And Cybercrimes Are Increasing. The Term 'cyber Threat' Is Referred To As The Illegal Activity Performed Using The Internet. Cybercriminals Are Changing Their Techniques With Time To Pass Through The Wall Of Protection. Conventional Techniques Are Not Capable Of Detecting Zero-day Attacks And Sophisticated Attacks. Thus Far, Heaps Of Machine Learning Techniques Have Been Developed To Detect The Cybercrimes And Battle Against Cyber Threats. The Objective Of This Research Work Is To Present The Evaluation Of Some Of The Widely Used Machine Learning Techniques Used To Detect Some Of The Most Threatening Cyber Threats To The Cyberspace. Three Primary Machine Learning Techniques Are Mainly Investigated, Including Deep Belief Network, Decision Tree And Support Vector Machine. We Have Presented A Brief Exploration To Gauge The Performance Of These Machine Learning Techniques In The Spam Detection, Intrusion Detection And Malware Detection Based On Frequently Used And Benchmark Datasets.
Machine Learning, A Branch Of Artificial Intelligence, Can Be Described Simply As Systems That Learn From Data In Order To Make Predictions Or To Act, Autonomously Or Semi-autonomously, In Response To What It Has Learned. Unlike Pre-programmed Solutions Or Business-rules-engines, Machine Learning Can Eliminate The Need For Someone To Continuously Code Or Analyze Data Themselves To Solve A Problem. While There Are A Variety Of Applications Of Machine Learning, And The More Advanced “deep Learning”, Most Have Been Focused On Machine Learning That Trains A Computer To Perform Human-like Tasks, Such As Recognizing Speech, Identifying Images (or Objects And Events Portrayed Therein) And In Making Predictions. In This Paper, We Will Explore The Use Of Machine Learning As An Approach To Helping With Upstream Activities In Data Management Including Classification And Feature Identification, As Well As Discuss Implications For Data Quality, Data Governance And Master Data Management.
The Need For A Method To Create A Collaborative Machine Learning Model Which Can Utilize Data From Different Clients, Each With Privacy Constraints, Has Recently Emerged. This Is Due To Privacy Restrictions, Such As General Data Protection Regulation, Together With The Fact That Machine Learning Models In General Needs Large Size Data To Perform Well. Google Introduced Federated Learning In 2016 With The Aim To Address This Problem. Federated Learning Can Further Be Divided Into Horizontal And Vertical Federated Learning, Depending On How The Data Is Structured At The Different Clients. Vertical Federated Learning Is Applicable When Many Different Features Is Obtained On Distributed Computation Nodes, Where They Can Not Be Shared In Between. The Aim Of This Thesis Is To Identify The Current State Of The Art Methods In Vertical Federated Learning, Implement The Most Interesting Ones And Compare The Results In Order To Draw Conclusions Of The Benefits And Drawbacks Of The Different Methods. From The Results Of The Experiments, A Method Called FedBCD Shows Very Promising Results Where It Achieves Massive Improvements In The Number Of Communication Rounds Needed For Convergence, At The Cost Of More Computations At The Clients. A Comparison Between Synchronous And Asynchronous Approaches Shows Slightly Better Results For The Synchronous Approach In Scenarios With No Delay. Delay Refers To Slower Performance In One Of The Workers, Either Due To Lower Computational Resources Or Due To Communication Issues.
The Traffic And Accident Datasets For This Research Are Sourced By Data.gov.uk. The Data Analytics In This Paper Comprises Three Levels Namely: Descriptive Statistical Analysis; Inferential Statistical Analysis; Machine Learning. The Aim Of The Data Analytics Is To Explore The Factors That Could Have Impact On The Number Of Accidents And Their Associated Fatalities. Some Of The Factors Investigated On Are: Time Of The Day, Day Of The Week, Month Of The Year, Speed Limits, Etc… Machine Learning Approaches Have Also Been Employed To Predict The Types Of Accident Severity.
The Advancements In Neural Networks And The On-demand Need For Accurate And Near Real-time Speech Emotion Recognition (SER) In Human–computer Interactions Make It Mandatory To Compare Available Methods And Databases In SER To Achieve Feasible Solutions And A Firmer Understanding Of This Open-ended Problem. The Proposed System Reviews Deep Learning Approaches For SER With Available Datasets, Followed By Conventional Machine Learning Techniques For Speech Emotion Recognition. Ultimately, We Present A Multi-aspect Comparison Between Practical Neural Network Approaches In Speech Emotion Recognition. The Goal Of This Study Is To Provide A Survey Of The Field Of Discrete Speech Emotion Recognition For The Customer Service By Analyzing The Customers Emotion Using Speech Recognition And Provide Rating According To The Emotions.
Speaker Recognition Is A Technique That Automatically Identifies A Speaker From A Recording Of Their Voice. Speaker Recognition Technologies Are Taking A New Trend Due To The Progress In Artificial Intelligence And Machine Learning And Have Been Widely Used In Many Domains. Continuing Research In The Field Of Speaker Recognition Has Now Spanned Over 50 Years. In Over Half A Century, A Great Deal Of Progress Has Been Made Towards Improving The Accuracy Of The System’s Decisions, Through The Use Of More Successful Machine Learning Algorithms. This Paper Presents The Development Of Automatic Speaker Recognition System Based On Optimised Machine Learning Algorithms. The Algorithms Are Optimised For Better And Improved Performance. Four Classifier Models, Namely, Support Vector Machines, ‘KNEAREST NEIGHBORS, RANDOM FOREST’, Logistic Regression, And Artificial Neural Networks Are Trained And Compared. The System Resulted With Artificial Neural Networks Obtaining The State-ofthe- Art Accuracy Of 96.03% Outperforming KNN, SVM, RF And LR Classifiers.
Music Genre Classification Utilizing CNN And RNN Algorithm Has Achieved Some Limited Success In Recent Years. Differences In Song Libraries, Machine Learning Techniques, Input Formats, And Types Of NNs Implemented Have All Had Varying Levels Of Success. This Article Reviews Some Of The Machine Learning Techniques Utilized In This Area. It Also Presents Research Work On Music Genre Classification. The Research Uses Images Of Spectrograms Generated From Timeslices Of Songs As The Input Into An NN To Classify The Songs Into Their Respective Musical Genres.
The Paper Describes Possibilities, Which Are Provided By Open APIs, And How To Use Them For Creating Unified Interfaces Using The Example Of Our Bot Based On Google API. In Last Decade AI Technologies Became Widespread And Easy To Implement And Use. One Of The Most Perspective Technology In The AI Field Is Speech Recognition As Part Of Natural Language Processing. New Speech Recognition Technologies And Methods Will Become A Central Part Of Future Life Because They Save A Lot Of Communication Time, Replacing Common Texting With Voice/audio. In Addition, This Paper Explores The Advantages And Disadvantages Of Well-known Chatbots. The Method Of Their Improvement Is Built. The Algorithms Of Rabin-Karp And Knut-Pratt Are Used. The Time Complexity Of Proposed Algorithm Is Compared With Existed One.
Deep Neural Networks Achieve Best Classification Accuracy On Videos. However, Traditional Methods Or Shallow Architectures Remain Competitive And Combinations Of Different Network Types Are The Usual Chosen Approach. A Reason For This Less Important Impact Of Deep Methods For Video Recognition Is The Motion Representation. The Time Has A Stronger Redundancy, And An Important Elasticity Compared To The Spatial Dimensions. The Temporal Redundancy Is Evident, But The Elasticity Within An Action Class Is Well Less Considered .Several Instances Of The Action Still Widely Differ By Their Style And Speed Of Execution.
Road Accidents Are Man-made Cataclysmic Phenomena And Are Not Generally Predictable. With Increasing Numbers Of Deaths Due To Accidents In The Roadways, A Smart And Fast Detection System For Road Accidents Is The Need Of The Hour. Often, Precious Few Seconds After The Accidents Make The Difference Between Life And Death. To Address This Problem More Efficiently, “A Novel Approach For Road Accident Detection In CCTV Videos Using DETR Algorithm” Has Been Developed To Aid In Notifying Hospitals And The Local Police At Places Where Instant Notification Is Seldom Feasible.
Human Pose Estimation In Images Is Challenging And Important For Many Computer Vision Applications. Large Improvements In Human Pose Estimation Have Been Achieved With The Development Of Convolutional Neural Networks. Even Though, When Encountered Some Difficult Cases Even The State-ofthe-art Models May Fail To Predict All The Body Joints Correctly. Some Recent Works Try To Refine The Pose Estimator. GAN (Generative Adversarial Networks) Has Been Proved To Be Efficient To Improve Human Pose Estimation. However, GAN Can Only Learn Local Body Joints Structural Constrains. In This Paper, We Propose To Apply Self-Attention GAN To Further Improve The Performance Of Human Pose Estimation. With Attention Mechanism In The Framework Of GAN, We Can Learn Long-range Body Joints Dependencies, Therefore Enforce The Entire Body Joints Structural Constrains To Make All The Body Joints To Be Consistent. Our Method Outperforms Other State-of-the-art Methods On Two Standard Benchmark Datasets MPII And LSP For Human Pose Estimation.
Exponential Growth Of Fake ID Cards Generation Leads To Increased Tendency Of Forgery With Severe Security And Privacy Threats. University ID Cards Are Used To Authenticate Actual Employees And Students Of The University. Manual Examination Of ID Cards Is A Laborious Activity, Therefore, In This Paper, We Propose An Effective Automated Method For Employee/student Authentication Based On Analyzing The Cards. Additionally, Our Method Also Identifies The Department Of Concerned Employee/student. For This Purpose, We Employ Different Image Enhancement And Morphological Operators To Improve The Appearance Of Input Image Better Suitable For Recognition. More Specifically, We Employ Median Filtering To Remove Noise From The Given Input Image.
Rigorous Research Has Been Done On Ancient Indian Script Character Recognition. Many Research Articles Are Published In Last Few Decades. Number Of OCR Techniques Is Available In Market, But OCR Techniques Are Not Useful For Ancient Script Recognition. But More Research Work Is Required To Recognize Ancient Marathi Scripts. This Paper Presents Different Techniques Which Are Published By Different Researchers To Recognize Ancient Scripts. Also Challenges In Recognition Of Ancient Marathi Scripts Are Discussed In This Paper.
Handwritten Signature Recognition Is An Important Behavioral Biometric Which Is Used For Numerous Identification And Authentication Applications. There Are Two Fundamental Methods Of Signature Recognition, On-line Or Off-line. On-line Recognition Is A Dynamic Form, Which Uses Parameters Like Writing Pace, Change In Stylus Direction And Number Of Pen Ups And Pen Downs During The Writing Of The Signature. Off-line Signature Recognition Is A Static Form Where A Signature Is Handled As An Image And The Author Of The Signature Is Predicted Based On The Features Of The Signature. The Current Method Of Off-line Signature Recognition Predominantly Employs Template Matching, Where A Test Image Is Compared With Multiple Specimen Images To Speculate The Author Of The Signature. This Takes Up A Lot Of Memory And Has A Higher Time Complexity. This Paper Proposes A Method Of Off-line Signature Recognition Using Convolution Neural Network. The Purpose Of This Paper Is To Obtain High Accuracy Multi-class Classification With A Few Training Signature Samples. Images Are Preprocessed To Isolate The Signature Pixels From The Background/noise Pixels Using A Series Of Image Processing Techniques. Initially, The System Is Trained With 27 Genuine Signatures Of 10 Different Authors Each. A Convolution Neural Network Is Used To Predict A Test Signature Belongs To Which Of The 10 Given Authors. Different Public Datasets Are Used To Demonstrate Effectiveness Of The Proposed Solution.
Our Country, India Is The Largest Democratic Country In The World. So It Is Essential To Make Sure That The Governing Body Is Elected Through A Fair Election. India Has Only Offline Voting System Which Is Not Effective And Up To The Mark As It Requires Large Man Force And It Also Requires More Time To Process And Publish The Results. Therefore, To Be Made Effective, The System Needs A Change, Which Overcomes These Disadvantages. The New Method Does Not Force The Person's Physical Appearance To Vote, Which Makes The Things Easier. This Paper Focuses On A System Where The User Can Vote Remotely From Anywhere Using His/her Computer Or Mobile Phone And Doesn’t Require The Voter To Get To The Polling Station Through Two Step Authentication Of Face Recognition And OTP System. This Project Also Allows The User To Vote Offline As Well If He/she Feels That Is Comfortable. The Face Scanning System Is Used To Record The Voters Face Prior To The Election And Is Useful At The Time Of Voting. The Offline Voting System Is Improvised With The Help Of RFID Tags Instead Of Voter Id. This System Also Enables The User The Citizens To See The Results Anytime Which Can Avoid Situations That Pave Way To Vote Tampering.
In The Human Body, The Face Is The Most Crucial Factor In Identifying Each Person As It Contains Many Vital Details. There Are Different Prevailing Methods To Capture Person's Presence Like Biometrics To Take Attendance Which Is A Time-consuming Process. This Paper Develops A Model To Classify Each Character's Face From A Captured Image Using A Collection Of Rules I.e., HOG Algorithm To Record The Student Attendance. HOG(Histogram Of Oriented Gradients) Is One Among The Methods And Is Popular As Well As Effective Technique Used For The Image Representation And Classification And It Was Chosen For Its Robustness To Pose And Illumination Shifts. The Proposed ASAS (Automated Smart Attendance System) Will Capture The Image And Will Be Compared To The Image Stored In The Database. The Database Is Updated Upon The Enrolment Of The Student Using An Automation Process That Also Includes Name And Rolls Number. ASAS Marks Individual Attendance, If The Captured Image Matches The Image In The Database I.e., If Both Images Are Identical. The Proposed Algorithm Reduces Effort And Captures Day-to-day Actions Of Managing Each Student And Also Makes It Simple To Mark The Presence.
This Study Is An Attempt To Understand And Address The Mental Health Issue, Of Working Professionals Through Facial Expression Recognition. As A Society, We Are All Currently Talking About Ways As To How A Person Who Is Suffering From Any Emotional Issue Can Adopt Certain Ways To Come Out Of A Specific Circumstance And How We As A Society Can Support Such People In These Situations. Our Endeavor Is To Work On A Way Where The Identification Of Such Persons Who Are Going Through A Difficult Phase In Their Life Can Be Performed.
Health Monitoring Is An Important Parameter To Determine The Health Status Of A Person. Measuring The Heart Rate Is An Easy Way To Gauge Our Health. Normal Heart Rate May Vary From Person To Person And A Usually High Or Low Resting Heart Rate Can Be A Sign Of Trouble. There Are Several Methods For The Measurement Of Heart Rate Monitoring Such As Ecg, Ppg Etc. Such Methods Having A Disadvantage That These Are Invasive And Have A Continuous Contact With The Human Body. In Order To Overcome This Problem A New System Is Proposed Using Camera. In This Method A Blind Source Separation Algorithm Is Used For Extracting The Heart Rate Signal From The Face Image. Viola Jones Based Face Detection Algorithm Is Used To Track The Face. FastICA Algorithm Is Exploited To Separate Heart Rate Signal From Noise And Artefacts. Machine Learning Algorithm Is Implemented To Standardize The Signal. The Data Is Successfully Tested With Real Time Video.
Smart Home Is One Application Of The Pervasive Computing Branch Of Science. Three Categories Of Smart Homes, Namely Comfort, Healthcare, And Security. The Security System Is A Part Of Smart Home Technology That Is Very Important Because The Intensity Of Crime Is Increasing, Especially In Residential Areas. The System Will Detect The Face By The Webcam Camera If The User Enters The Correct Password. Face Recognition Will Be Processed By The Raspberry Pi 3 Microcontroller With The Principal Component Analysis Method Using OpenCV And Python Software Which Has Outputs, Namely Actuators In The Form Of A Solenoid Lock Door And Buzzer. The Test Results Show That The Webcam Can Perform Face Detection When The Password Input Is Successful, Then The Buzzer Actuator Can Turn On When The Database Does Not Match The Data Taken By The Webcam Or The Test Data And The Solenoid Door Lock Actuator Can Run If The Database Matches The Test Data Taken By The Sensor. Webcam. The Mean Response Time Of Face Detection Is 1.35 Seconds.
The Real-time Sign Language Recognition System Is Developed For Recognising The Gestures Of Indian Sign Language (ISL). Generally, Sign Languages Consist Of Hand Gestures And Facial Expressions. For Recognising The Signs, The Regions Of Interest (ROI) Are Identified And Tracked Using The Skin Segmentation Feature Of OpenCV. The Training And Prediction Of Hand Gestures Are Performed By Applying Fuzzy C-means Clustering Machine Learning Algorithm. The Gesture Recognition Has Many Applications Such As Gesture Controlled Robots And Automated Homes, Game Control, Human-Computer Interaction (HCI) And Sign Language Interpretation. The Proposed System Is Used To Recognize The Real-time Signs. Hence It Is Very Much Useful For Hearing And Speech Impaired People To Communicate With Normal People.
While Recognizing Any Individual, The Most Important Attribute Is Face. It Serves As An Individual Identity Of Everyone And Therefore Face Recognition Helps In Authenticating Any Person’s Identity Using His Personal Characteristics. The Whole Procedure For Authenticating Any Face Data Is Sub-divided Into Two Phases, In The First Phase, The Face Detection Is Done Quickly Except For Those Cases In Which The Object Is Placed Quite Far, Followed By This The Second Phase Is Initiated In Which The Face Is Recognized As An Individual. Then The Whole Process Is Repeated Thereby Helping In Developing A Face Recognition Model Which Is Considered To Be One Of The Most Extremely Deliberated Biometric Technology. Basically, There Are Two Type Of Techniques That Are Currently Being Followed In Face Recognition Pattern That Is, The Eigenface Method And The Fisherface Method. The Eigenfacemethod Basically Make Use Of The PCA (Principal Component Analysis) To Minimize The Face Dimensional Space Of The Facial Features. The Area Of Concern Of This Paper Is Using The Digital Image Processing To Develop A Face Recognition System.
Communications Is Facial Expression Recognition, As In Non-verbal Communication, Facial Expressions Are Key. In The Field Of Artificial Intelligence, Facial Expression Recognition (FER) Is An Active Research Area, With Several Recent Studies Using Convolutional Neural Networks (Emotions Are A Powerful Tool In Communication And One Way That Humans Show Their Emotions Is Through Their Facial Expressions. One Of The Challenging And Powerful Tasks In Social CNNs). In This Paper, We Demonstrate The Classification Of FER Based On Static Images, Using CNNs, Without Requiring Any Pre-processing Or Feature Extraction Tasks. The Paper Also Illustrates Techniques To Improve Future Accuracy In This Area By Using Preprocessing, Which Includes Face Detection And Illumination Correction. Feature Extraction Is Used To Extract The Most Prominent Parts Of The Face, Including The Jaw, Mouth, Eyes, Nose, And Eyebrows. Furthermore, We Also Discuss The Literature Review And Present Our CNN Architecture, And The Challenges Of Using Max-pooling And Dropout, Which Eventually Aided In Better Performance. We Obtained A Test Accuracy Of 61.7% On FER2013 In A Seven-classes Classification Task Compared To 75.2% In State-of-the-art Classification
Emotion Recognition From Speech Signals Is An Important But Challenging Component Of Human-Computer Interaction (HCI). In The Literature Of Speech Emotion Recognition (SER), Many Techniques Have Been Utilized To Extract Emotions From Signals, Including Many Well-established Speech Analysis And Classification Techniques. Deep Learning Techniques Have Been Recently Proposed As An Alternative To Traditional Techniques In SER. This Paper Presents An Overview Of Deep Learning Techniques And Discusses Some Recent Literature Where These Methods Are Utilized For Speech-based Emotion Recognition. The Review Covers Databases Used, Emotions Extracted, Contributions Made Toward Speech Emotion Recognition And Limitations Related To It.
Blood Cancer (Leukemia) Is One Of The Leading Causes Of Death Among Humans. The Pace Of Healing Depends Mainly On Early Detection And Diagnosis Of A Disease. The Main Reason Behind Occurrence Of Leukemia Is When Bone Marrow Produces A Lot Of Abnormal White Blood Cells This Happens. Microscopic Study On Images Is Done By Hematologists Who Make Use Of Human Blood Samples, From Which It Leads To The Requirement Of Following Methods, Which Are Microscopic Color Imaging, Image Segmentation, Clustering And Classification Which Allows Easy Identification Of Patients Suffering From This Disease. Microscopic Imaging Allows For Various Methods Of Detecting Blood Cancer In Visible And Immature White Blood Cells. Identifying Leukemia Early And Quickly Greatly Helps Practitioners In Providing Appropriate Treatment To Patients. Initially To Start With, Segmentation Stage Is Achieved By Segregating White Blood Cells From Other Blood Components .For Diagnosing Prediction Of Leukemia, Geometrical Features Such As Area, Perimeter Of The White Blood Cell Nucleuses Investigated. In The Proposed Methodology We Make Use Of K-means, For Identifying Cancerous Stages And Its Early Detection. Experimentation And Results Were Found To Be Promising With The Accuracy Of 90% Identification Of The Cancer Cells.
Breast Cancer Is More Common Hence, Identification Of BC And Detection Of Region Of Breast Affected Is More Important. Mammography Screening Images Two Views CC And MLO Are Widely Use In Diagnosis Process. This Paper Presents The Method To Detect Cancer Region And Classify Normal And Cancerous Patient. Pre-processing Operation Perform On The Input Mammogram Image And Undesirable Part Removed From The Image, Tumor Region Segmented From The Image Using Morphological Operation And Highlighted The Region On Original Mammogram Image Or If Mammogram Image Is Normal Case Then It Shows That Patient Is Normal. Random Forest (RF) Classifiers Is Used For Classification Of BC Patient And Normal Patient. Classification Accuracy Of RF Is 95% For Image Of Different Patient. Processing Time Of RF Classifier Is 6.25s.
Computer Aided Diagnosis (CAD) Is Quickly Evolving, Diverse Field Of Study In Medical Analysis. Significant Efforts Have Been Made In Recent Years To Develop Computer-aided Diagnostic Applications, As Failures In Medical Diagnosing Processes Can Result In Medical Therapies That Are Severely Deceptive. Machine Learning (ML) Is Important In Computer Aided Diagnostic Test. Object Such As Body-organs Cannot Be Identified Correctly After Using An Easy Equation. Therefore, Pattern Recognition Essentially Requires Training From Instances. In The Bio Medical Area, Pattern Detection And ML Promises To Improve The Reliability Of Disease Approach And Detection. They Also Respect The Dispassion Of The Method Of Decisions Making. ML Provides A Respectable Approach To Make Superior And Automated Algorithm For The Study Of High Dimension And Multi - Modal Bio Medicals Data. The Relative Study Of Various ML Algorithm For The Detection Of Various Disease Such As Heart Disease, Diabetes Disease Is Given In This Survey Paper. It Calls Focus On The Collection Of Algorithms And Techniques For ML Used For Disease Detection And Decision Making Processes.
The Outbreaks Of COVID-19 Virus Have Crossed The Limit To Our Expectation And It Breaks All Previous Records Of Virus Outbreaks. The Effect Of Corona Virus Causes A Serious Illness May Result In Death As A Consequence Of Substantial Alveolar Damage And Progressive Respiratory Failure. Automatic Detection And Classification Of This Virus From Chest X-ray Image Using Computer Vision Technology Can Be Very Useful Complement With Respect To The Less Sensitive Traditional Process Of Detecting COVID-19 I.e. Reverse Transcription Polymerase Chain Reaction (RT-PCR). This Automated Process Offers A Great Potential To Enhance The Conventional Healthcare Tactic For Tackling COVID-19 And Can Mitigate The Shortage Of Trained Physicians In Remote Communities. Again, The Segmentation Of The Infected Regions From Chest X-ray Image Can Help The Medical Specialists To View Insights Of The Affected Region.
Human Activity Recognition Has Attracted The Attention Of Researchers Around The World. This Is An Interesting Problem That Can Be Addressed In Different Ways. Many Approaches Have Been Presented During The Last Year .These Applications Present Solutions To Recognize Different Kinds Of Activities Such As If The Person Is Walking, Running, Jumping, Jogging, Or Falling, Among Others. Amongst All These Activities, Fall Detection Has Special Importance Because It Is A Common Dangerous Event For People Of All Ages With A More Negative Impact On The Elderly Population. Usually, These Applications Use Sensors To Detect Sudden Changes In The Movement Of The Person. These Kinds Of Sensors Can Be Embedded In Smartphones, Necklaces, Or Smart Wristbands To Make Them ‘‘wearable’’ Devices. The Main Inconvenience Is That These Devices Have To Be Placed On The Subjects’ Bodies. This Might Be Uncomfortable And Is Not Always Feasible Because This Type Of Sensor Must Be Monitored Constantly, And Can Not Be Used In Open Spaces With Unknown People.
Breast Cancer Is More Common Hence, Identification Of BC And Detection Of Region Of Breast Affected Is More Important. Mammography Screening Images Two Views CC And MLO Are Widely Use In Diagnosis Process. This Paper Presents The Method To Detect Cancer Region And Classify Normal And Cancerous Patient. Pre-processing Operation Perform On The Input Mammogram Image And Undesirable Part Removed From The Image, Tumor Region Segmented From The Image Using Morphological Operation And Highlighted The Region On Original Mammogram Image Or If Mammogram Image Is Normal Case Then It Shows That Patient Is Normal. Random Forest (RF) Classifiers Is Used For Classification Of BC Patient And Normal Patient. Classification Accuracy Of RF Is 95% For Image Of Different Patient. Processing Time Of RF Classifier Is 6.25s.
Blood Cancer (Leukemia) Is One Of The Leading Causes Of Death Among Humans. The Pace Of Healing Depends Mainly On Early Detection And Diagnosis Of A Disease. The Main Reason Behind Occurrence Of Leukemia Is When Bone Marrow Produces A Lot Of Abnormal White Blood Cells This Happens. Microscopic Study On Images Is Done By Hematologists Who Make Use Of Human Blood Samples, From Which It Leads To The Requirement Of Following Methods, Which Are Microscopic Color Imaging, Image Segmentation, Clustering And Classification Which Allows Easy Identification Of Patients Suffering From This Disease. Microscopic Imaging Allows For Various Methods Of Detecting Blood Cancer In Visible And Immature White Blood Cells. Identifying Leukemia Early And Quickly Greatly Helps Practitioners In Providing Appropriate Treatment To Patients. Initially To Start With, Segmentation Stage Is Achieved By Segregating White Blood Cells From Other Blood Components .For Diagnosing Prediction Of Leukemia, Geometrical Features Such As Area, Perimeter Of The White Blood Cell Nucleuses Investigated. In The Proposed Methodology We Make Use Of K-means, For Identifying Cancerous Stages And Its Early Detection. Experimentation And Results Were Found To Be Promising With The Accuracy Of 90% Identification Of The Cancer Cells.
Oral Cancer Is A Major Global Health Issue Accounting For 177,384 Deaths In 2018 And It Is Most Prevalent In Low- And Middle-income Countries. Enabling Automation In The Identification Of Potentially Malignant And Malignant Lesions In The Oral Cavity Would Potentially Lead To Low-cost And Early Diagnosis Of The Disease. Building A Large Library Of Well-annotated Oral Lesions Is Key. As Part Of The MeMoSA® (Mobile Mouth Screening Anywhere) Project, Images Are Currently In The Process Of Being Gathered From Clinical Experts From Across The World, Who Have Been Provided With An Annotation Tool To Produce Rich Labels.
Heartdiseaseisoneofthemostsignificantcausesofmortalityintheworldtoday.Predictionof Cardio Vascular Disease Is A Critical Challenge In The Area Of Clinical Data Analysis.Machine Learning(ML)has Been Shown To Be Effective In Assisting In Making Decisions And Predictions From The Large Quantity Of Data Produced By The Health Care Industry.We Have Also Seen ML Techniques Being Used In Recent Developments In Different Areas Of The Internet Of Things (IoT). Various Studies Give Only A Glimpse Into Predicting Heart Disease With ML Techniques. In This Paper, We Propose A Novel Method That Aims At finding Significant Features By Applying Machine Learning Techniques Resulting In Improving The Accuracy In The Prediction Of Cardiovascular Disease. The Prediction Model Is Introduced With Different Combinations Of Features And Several Known Classification Techniques
Prediction Of Academic Performance Of Students Beforehand Provides Scope To Universities To Lower Their Dropout Rate And Help The Students In Improving Their Performance. In This Field, Research Is Being Done To Find Out Which Algorithm Is Best To Use And Which Features Should Be Considered While Predicting The Academic Performance Of Students. This Kind Of Research Work Has Been Increasing Over The Years. This Paper Performs A Survey On The Techniques Used In Various Research Papers For Academic Performance Prediction And Also Point Out The Limitations If Any, In The Methodology Used.
Fake Review Detection And Its Elimination From The Given Dataset Using Different Natural Language Processing (NLP) Techniques Is Important In Several Aspects. In This Article, The Fake Review Dataset Is Trained By Applying Two Different Machine Learning (ML) Models To Predict The Accuracy Of How Genuine Are The Reviews In A Given Dataset. The Rate Of Fake Reviews In Ecommerce Industry And Even Other Platforms Is Increasing When Depend On Product Reviews For The Item Found Online On Different Websites And Applications. The Products Of The Company Were Trusted Before Making A Purchase. So This Fake Review Problem Must Be Addressed So That These Large E-commerce Industries Such As Flipkart, Amazon, Etc. Can Rectify This Issue So That The Fake Reviewers And Spammers Are Eliminated To Prevent Users From Losing Trust On Online Shopping Platforms. This Model Can Be Used By Websites And Applications With Few Thousands Of Users Where It Can Predict The Authenticity Of The Review Based On Which The Website Owners Can Take Necessary Action Towards Them. This Model Is Developed Using ‘NAIVE BAYES’. By Applying These Models One Can Know The Number Of Spam Reviews On A Website Or Application Instantly. To Counter Such Spammers, A Sophisticated Model Is Required In Which A Need To Be Trained On Millions Of Reviews. In This Work ”amazon Yelp Dataset” Is Used To Train The Models And Its Very Small Dataset Is Used For Training On A Very Small Scale And Can Be Scaled To Get High Accuracy And Flexibility.
The Intent Recognition And Natural Language Understanding Of Multi-turn Dialogue Is Key For The Commercialization Of Chatbots.Chatbots Are Mainly Used For The Processing Of Specific Tasks, And Can Introduce Products To Customers Or Solve Related Problems, Thus Saving Human Resources. Text Sentiment Recognition Enables A Chatbot To Know The User’s Emotional State And Select The Best Response, Which Is Important In Medical Care. In This Study, We Combined The Multiturn Dialogue Model And Sentiment Recognition Model To Develop A Chatbot, That Is Designed For Used In Daily Conversations Rather Than For Specific Tasks. Thus, The Chatbot Has The Ability To Provide The Robot’s Emotions As Feedback While Talking With A User. Moreover, It Can Exhibit Different Emotional Reactions Based On The Content Of The User’s Conversation.
Intelligent Personal Assistant (IPA) Is A Software Agent Performing Tasks On Behalf Of An Human Or Individual L Based On Commands Or Questions Which Are Similar To Chat Bots. They Are Also Referred As Intelligent Virtual Assistant Which Interprets Human Speech And Respond Via Synthesized Voices. IPAs And IVAs Finds Their Usage In Various Applications Such As Home Automation, Manage To-do Tasks And Media Playback Through Voice. This Paper Aims To Propose Speech Recognition Systems And Dealing With Creating A Virtual Personal Assistant. The Existing System Serves On The Internet And Is Maintained By The Third Party. This Application Shall Protect Personal Data From Others And Use The Local Database, Speech Recognition And Synthesiser. A Parser Named SURR(Semantic Unification And Reference Resolution) Is Employed To Recognise The Speech. Synthesizer Uses Text To Phoneme. ‘DNN ALGORITHM’ Is Used.
Poor Nutrition Can Lead To Reduced Immunity, Increased Susceptibility To Disease, Impaired Physical And Mental Development, And Reduced Productivity. A Conversational Agent Can Support People As A Virtual Coach, However Building Such Systems Still Have Its Associated Challenges And Limitations. This Paper Describes The Background And Motivation For Chatbot Systems In The Context Of Healthy Nutrition Recommendation. We Discuss Current Challenges Associated With Chatbotapplication, We Tackled Technical, Theoretical, Behavioural, And Social Aspects Of The Challenges. We Then Propose A Pipeline To Be Used As Guidelines By Developers To Implement Theoretically And Technically Robust Chatbot Systems.
Voice Control Is A Major Growing Feature That Change The Way People Can Live. The Voice Assistant Is Commonly Being Used In Smartphones And Laptops. AI-based Voice Assistants Are The Operating Systems That Can Recognize Human Voice And Respond Via Integrated Voices. This Voice Assistant Will Gather The Audio From The Microphone And Then Convert That Into Text, Later It Is Sent Through GTTS (Google Text To Speech). GTTS Engine Will Convert Text Into Audio File In English Language, Then That Audio Is Played Using Play Sound Package Of Python Programming Language.
In Order To Prevent Health Risks And Provide A Better Service To The Patients That Have Visited The Hospital, There Is A Need For Monitoring The Patients After Being Released And Providing The Data Submitted By The Patient EHealth Enablers To The Medical Personnel. This Article Proposes Architecture For Providing The Secure Exchange Of Data Between The Patient Mobile Application And The Hospital Infrastructure. The Implemented Solution Is Validated On A Laboratory Testbed.
Real-time Communication (RTC) Is A New Standard And Industry-wide Effort That Expand The Web Browsing Model, Allowing Access To Information In Areas Like Social Media, Chat, Video Conferencing, And Television Over The Internet, And Unified Communication. These Systems Users Can View, Record, Remark, Or Edit Video And Audio Content Flows Using Time-critical Cloud Infrastructures That Enforce The Quality Of Services. However, There Are Many Proprietary Protocols And Codecs Available That Are Not Easily Interoperable And Scalable To Implement Multipoint Videoconference Systems. WebRTC (Web Real-Time Communication) Is A State-of-the-Art Open Technology That Makes Real-time Communication Capabilities In Audio, Video, And Data Transmission Possible In Real-time Communication Through Web Browsers Using JavaScript APIs (Application Programming Interfaces) Without Plug-ins. This Paper Aims To Introduce The P2P Video Conferencing System Based On Web Real-Time Communication (WebRTC). In This Paper, We Have Proposed A Web-based Peer-to-peer Real-time Communication System Using The Mozilla Firefox Together With The ScaleDrone Service That Enables Users To Communicate With Highspeed Data Transmission Over The Communication Channel Using WebRTC Technology, HTML5 And Use Node.js Server Address. Our Experiments Show That WebRTC Is A Capable Building Block For Scalable Live Video Conferencing Within A Web Browser.
Recent Developments In The Speed Of The Internet And Information Technology Have Made The Rapid Exchange Of Multimedia Information Possible. However, These Developments In Technology Lead To Violations Of Information Security And Private Information. Digital Steganography Provides The Ability To Protect Private Information That Has Become Essential In The Current Internet Age. Among All Digital Media, Digital Video Has Become Of Interest To Many Researchers Due To Its High Capacity For Hiding Sensitive Data. Numerous Video Steganography Methods Have Recently Been Proposed To Prevent Secret Data From Being Stolen. Nevertheless, These Methods Have Multiple Issues Related To Visual Imperceptibly, Robustness, And Embedding Capacity. To Tackle These Issues, This Paper Proposes A New Approach To Video Steganography Based On The Corner Point Principle And LSBs Algorithm. The Proposed Method First Uses Shi-Tomasi Algorithm To Detect Regions Of Corner Points Within The Cover Video Frames. Then, It Uses 4-LSBs Algorithm To Hide Confidential Data Inside The Identified Corner Points. Besides, Before The Embedding Process, The Proposed Method Encrypts Confidential Data Using Arnold’s Cat Map Method To Boost The Security Level.
The Security Of Any Public Key Cryptosystem Depends On The Private Key Thus, It Is Important That Only An Authorized Person Can Have Access To The Private Key. The Paper Presents A New Algorithm That Protects The Private Key Using The Transposition Cipher Technique. The Performance Of The Proposed Technique Is Evaluated By Applying It In The RSA Algorithm’s Generated Private Keys Using 512-bit, 1024-bit, And 2048-bit, Respectively. The Result Shows That The Technique Is Practical And Efficient In Securing Private Keys While In Storage As It Produced High Avalanche Effect.
Initially The Barcodes Have Been Widely Used For The Unique Identification Of The Products. Quick Response I.e. QR Codes Are 2D Representation Of Barcodes That Can Embed Text, Audio, Video, Web URL, Phone Contacts, Credentials ¬¬and Much More. This Paper Primarily Deals With The Generation Of QR Codes For Question Paper. We Have Proposed Encryption Of Question Paper Data Using AES Encryption Algorithm. The Working Of The QR Codes Is Based On Encrypting It To QR Code And Scanning To Decrypt It. Furthermore, We Have Reduced The Memory Storage By Redirecting To A Webpage Through The Transmission And Online Acceptance Of Data.
Communication Technology Has Completely Occupied All The Areas Of Applications. Last Decade Has However Witnessed A Drastic Evolution In Information And Communication Technology Due To The Introduction Of Social Media Network. Business Growth Is Further Achieved Via These Social Media. Nevertheless, Increase In The Usage Of Online Social Networks (OSN) Such As Face Book, Twitter, Instagrametc Has However Led To The Increase In Privacy And Security Concerns. Third Party Applications Are One Of The Many Reasons For Facebook Attractiveness. Regrettably, The Users Are Unaware Of Detail That A Lot Of Malicious Facebook Applications Provide On Their Profile. The Popularity Of These Third Party Applications Is Such That There Are Almost 20 Million Installations Per Day. But Cyber Criminals Have Appreciated The Popularity Of Third Party Applications And The Possibility Of Using These Apps For Distributing The Malware And Spam. This Paper Proposes A Method To Categorize A Given Application As Malicious Or Safe By Using FRAppE (Facebook’s Rigorous Application Evaluator), Possibly One Of The First Tool For Detecting Malicious Apps On The Facebook. To Develop The FRAppE, The Data Is Gathered From MyPagekeeper Application, A Website That Provides Significant Information About Various Third Party Applications And Their Insight Into Their Behavior.
Data Mining Is The Process Of Extracting Useful Unknown Knowledge From Large Datasets. Frequentitemset Mining Is The Fundamental Task Of Data Mining That Aims At Discovering Interesting Itemsets That Frequently Appear Together In A Dataset. However, Mining Infrequent (rare) Itemsets May Be More Interesting In Many Real-life Applications Such As Predicting Telecommunication Equipment Failures, Genetics, Medical Diagnosis, Or Anomaly Detection. In This Paper, We Survey Up-to-date Methods Of Rare Itemset Mining. The Main Goal Of This Survey Is To Provide A Comprehensive Overview Of The State-of-the-art Algorithms Of Rare Itemset Mining And Its Applications. The Main Contributions Of This Survey Can Be Summarized As Follows. In The First Part, We Define The Task Of Rare Itemset Mining By Explaining Key Concepts And Terminology, Motivation Examples, And Comparisons With Underlying Concepts. Then, We Highlight The State-of-art Methods For Rare Itemsets Mining. Furthermore, We Present Variations Of The Task Of Rare Itemset Mining To Discuss Limitations Of Traditional Rare Itemset Mining Algorithms. After That, We Highlight The Fundamental Applications Of Rare Itemset Mining. In The Last, We Point Out Research Opportunities And Challenges For Rare Itemset Mining For Future Research.
Data Streams Are Usually Non-stationary With Continually Changing Their Underlying Structure. Solving Of Predictive Or Classification Tasks On Such Data Must Consider This Aspect. Traditional Machine Learning Models Applied On The Drifting Data May Become Invalid In The Case When A Concept Change Appears. To Tackle This Problem, We Must Utilize Special Adaptive Learning Models, Which Utilize Various Tools Able To Reflect The Drifting Data. One Of The Most Popular Groups Of Such Methods Are Adaptive Ensembles. This Paper Describes The Work Focused On The Design And Implementation Of A Novel Adaptive Ensemble Learning Model, Which Is Based On The Construction Of A Robust Ensemble Consisting Of A Heterogeneous Set Of Its Members. We ‘USED K-NN, NAIVE BAYES’and Hoeffding Trees As Base Learners And Implemented An Update Mechanism, Which Considers Dynamic Class-weighting And Q Statistics Diversity Calculation To Ensure The Diversity Of The Ensemble. The Model Was Experimentally Evaluated On The Streaming Datasets, And The Effects Of The Diversity Calculation Were Analyzed.
The Neural Network Architecture Is Proposed As A Promising Approach To Increase The Accuracy Of The 2m Temperature Forecast Given By The COSMO Regional Model. This Architecture Allows Predicting Errors Of The Atmospheric Model Forecasts With Their Further Corrections. Experiments Are Conducted With Different Histories Of Regional Model Errors. The Number Of Epochs After Which Network Overfitting Happens Is Determined. It Is Shown That The Proposed Architecture Makes It Possible To Achieve An Improvement Of A 2m Temperature Forecast In Approximately 50% Of Cases.
Epidemics Affect People’s Daily Consumption Activities, For Example, By Causing Them To Shop Less, Travel Less, Consume Less And Invest Less. The Reduction Of A Large Number Of Economic Activities Leads To The Suppression Of Social Demand And The Reduction Of Consumption Level, Which Further Affects The GDP Of Various Countries Around The World. It Is Necessary To Investigate And Analyze The Impact Of The Epidemic On GDP In Order To Control And Analyze The Economic Situation Under The Impact Of The Epidemic. In This Paper, We Take The Impact Of COVID-19 On The GDP Of Each Country As A Regression Problem, And Propose To Forecast GDP Through Feature Engineering Combined With Aaboost Model. The Model Was Tested On More Than 50,000 Data Records From More Than 200 Countries Provided By The Kaggle Platform To Prove The Validity. The Experiment Shows That Adaboost Has Stronger Robustness Compared With Other Methods, Such As Random Forest, SVR. Adaboost Improves The MSE Of Random Forest By 2.39 And SVR By 0.38.
Hospital Readmissions Pose Additional Costs And Discomfort For The Patient And Their Occurrences Are Indicative Of Deficient Health Service Quality, Hence Efforts Are Generally Made By Medical Professionals In Order To Prevent Them. These Endeavors Are Especially Critical In The Case Of Chronic Conditions, Such As Diabetes. Recent Developments In Machine Learning Have Been Successful At Predicting Readmissions From The Medical History Of The Diabetic Patient. However, These Approaches Rely On A Large Number Of Clinical Variables Thereby Requiring Deep Learning Techniques. This Article Presents The Application Of Simpler Machine Learning Models Achieving Superior Prediction Performance While Making Computations More Tractable. Index Terms—diabetes, Hospital Readmission, Neural Network, Random Forest, Logistic Regression.
Hotel Booking Cancellation Is Provided A Substantial Effects On Demand Management Decisions In The Hospitality Industry. The Goal Of This Work Is To Investigate The Effects Of Different Machine Learning Methods In Hotel Booking Cancellation Process. In This Work, We Gathered A Hotel Booking Cancellation Dataset FromKaggle Data Repository. Then, Different Feature Transformation Techniques Were Implemented Into Primary Dataset And Generate Transformed Datasets. Further, We Reduced Insignificant Variables Using Feature Selection Methods. Therefore, Various Classifiers Were Employed Into Primary And Generated Subsets. The Effects Of The Machine Learning Methods Were Evaluated And Explored The Best Approaches In This Step. Among All Of These Methods, We Found That XGBoost Is The Most Frequent Method To Analyze These Datasets. Besides, Individual Classifiers Are Generated The Highest Result For Information Gain Feature Selection Method. This Analysis Can Be Used As The Complementary Tool To Investigate Hotel Booking Cancellation Dataset More Effectively.
This Paper Reports On The Smart Automated Irrigation System With Disease Detection. The System Design Includes Soil Moisture Sensors, Temperature Sensors, Leaf Wetness Sensors Deployed In Agriculture Field, The Sensed Data From Sensors Will Be Compared With Pre-determined Threshold Values Of Various Soil And Specific Crops. The Deployed Sensors Data Are Fed To The Arduino Uno Processor Which Is Linked To The Data Center Wirelessly Via GSM Module. The Data Received By The Data Center Is Stored To Perform Data Analysis Using Data Mining Technique Such As Markov Model To Detect The Possible Disease For That Condition. Finally, The Analysis Results And Observed Physical Parameters Are Transmitted To Android Smart Phone And Displayed On User Interface. The User Interface In Smart Phone Allows Remote User To Control Irrigation System By Switching, On And Off, The Motor Pump By The Arduino Based On The Commands From The Android Smart Phone.
Agriculture Is The Key Point For Survival. For Agriculture, Rainfall Is Most Important. These Days Rainfall Prediction Has Become A Major Problem. Prediction Of Rainfall Gives Awareness To People And Know In Advance About Rainfall To Take Certain Precautions To Protect Their Crop From Rainfall. Many Techniques Came Into Existence To Predict Rainfall. Machine Learning Algorithms Are Mostly Useful In Predicting Rainfall. Some Of The Major Machine Learning Algorithms Are ARIMA Model(Auto-Regressive Integrate D Moving Average), Artificial Neural Network, Logistic Regression, Support Vector Machine And Self Organizing Map. Two Commonly Used Models Predict Seasonal Rainfall Such As Linear And Non-Linear Models. The First Models Are ARIMA Model. While Using Artificial Neural Network(ANN) Predicting Rainfall Can Be Done Using Back Propagation NN, Cascade NN Or Layer Recurrent Network. Artificial NN Is Same As Biological Neural Networks.
Abstract—In This Paper, We Propose An Android Based Restaurant Automation System. The Main Aim Of The Project Is To Make The Restaurant Management Easier. Recently In Most Of The Restaurants, The Ordering And Delivery Of Food Items Are Doing Manually, The Disadvantages Are Huge Time Consumption, And In Some Cases The Customers Arent Delivered The Right Item At Right Time These Cause Many Problems. Hence We Thought Of Automating This Procedure Using Modern Electronic Technology. Here The Individual Tables In The Restaurant Are Provided With A Touch Screen, Represent Each Individual Digital Menu, And It Facilitates The Ordering. The Customer Can See All The Available Food Items With Its Cost In The Digital Menu And Can Select The Item. The Order From Each Table Is Received In The Kitchen Wirelessly By Bluetooth. The Electronic Menu System Helps The People To Select The Food From The Rolling Screen Of Android Touch Screen And To See The Cost And Recent Availability Of Food Items, And Showing Table Number Also. By Using A Thermal Printer Taking Bill From The Kitchen And The Hotel Staff Can Read The Items From Each Table. If The Food Is Ready In The Kitchen It Can Be Indicated To Corresponding Customers Table By An LED Glow. Index Terms—Restaurant Automation, Electronic Food Ordering System, Android Based Food Ordering, Touch Technology Based Food Ordering
Online-to-offline (O2O) Commerce Connecting Service Providers And Individuals To Address Daily Human Needs Is Quickly Expanding. In Particular, On-demand Food, Whereby Food Orders Are Placed Online By Customers And Delivered By Couriers, Is Becoming Popular. This Novel Urban Food Application Requires Highly Efficient And Scalable Real-time Delivery Services. However, It Is Difficult To Recruit Enough Couriers And Route Them To Facilitate Such Food Ordering Systems. This Article Presents An Online Crowd Sourced Delivery Approach For On-demand Food. Facilitated By Internet-of-Things And 3G/4G/5G Technologies, Public Riders Can Be Attracted To Act As Crowd Sourced Workers Delivering Food By Means Of Shared Bicycles Or Electric Motorbikes. An Online Dynamic Optimization Framework Comprising Order Collection, Solution Generation, And Sequential Delivery Processes Is Presented. A Hybrid Metaheuristic Solution Process Integrating The Adaptive Large Neighborhood Search And Tabu Search Approaches Is Developed To Assign Food Delivery Tasks And Generate High-quality Delivery Routes In A Real-time Manner. The Crowdsourced Riders Are Dynamically Shared Among Different Food Providers. Simulated Small-scale And Real-world Large-scale On-demand Food Delivery Instances Are Used To Evaluate The Performance Of The Proposed Approach. The Results Indicate That The Presented Crowdsourced Food Delivery Approach Outperforms Traditional Urban Logistics.
For Surveillance Purpose, Lots Of Method Were Used By The Researchers But Computer Vision Based Human Activity Recognition (HAR) Technologies/systems Received The Most In- Terest Because They Automatically Distinguish Human Behaviour And Movements From Video Data Utilizing Recorded Details From Cameras. But The Extraction Of Accurate And Opportune Infor- Mation From Video Of Human’s Activities And Behaviours Is Most Important And Difficult Task In Pervasive Computing Environment. Due To Lots Of Applications Of HAR Systems Like In Medical Field, Security, Visual Monitoring, Video Recovery, Entertainment And Irregular Behaviour Detection, The Accuracy Of System Is Most Important Factors For Researchers. This Review Article Presents A Brief Survey Of The Existing Video Or Vision-based HAR System To Find Out Their Challenges And Applications In Three Aspects Such As Recognition Of Activities, Activity Analysis, And Decision From Visual Content Representation. In Many Applications, System Recognition Time And Accuracy Is Most Important Factor And It Is Affected Due To An Increase In The Usage Of Simple Or Low Quality Type Cameras For Automated Systems. So, To Obtain A Better Accuracy And Fast Responses, The Usage Of Demanding And Computationally Intelligent Classification Techniques Such As Deep Learning And Machine Learning Is A Better Option For Researchers. In This Survey, We Addressed Numerous Computationally Intelligent Classification Techniques-based Research For HAR From 2010 To 2020 For A Better Analysis Of The Benefits And Drawbacks Of Systems, The Challenges Faced And Applications With Future Directions For HAR. We Also Present Some Accessible Problems And Ideas That Should Be Discussed In Future Research For The HAR System Utilizing Machine Learning And Deep Learning Principles Due To Their Strong Relevance.
Writing In Air Has Been One Of The Most Fascinating And Challenging Research Areas In Field Of Image Processing And Pattern Recognition In The Recent Years. It Contributes Immensely To The Advancement Of An Automation Process And Can Improve The Interface Between Man And Machine In Numerous Applications. Several Research Works Have Been Focusing On New Techniques And Methods That Would Reduce The Processing Time While Providing Higher Recognition Accuracy. Object Tracking Is Considered As An Important Task Within The Field Of Computer Vision. The Invention Of Faster Computers, Availability Of Inexpensive And Good Quality Video Cameras And Demands Of Automated Video Analysis Has Given Popularity To Object Tracking Techniques. Generally, Video Analysis Procedure Has Three Major Steps: Firstly, Detecting Of The Object, Secondly Tracking Its Movement From Frame To Frame And Lastly Analysing The Behaviour Of That Object. For Object Tracking, Four Different Issues Are Taken Into Account; Selection Of Suitable Object Representation, Feature Selection For Tracking, Object Detection And Object Tracking. In Real World, Object Tracking Algorithms Are The Primarily Part Of Different Applications Such As: Automatic Surveillance, Video Indexing And Vehicle Navigation Etc. The Project Takes Advantage Of This Gap And Focuses On Developing A Motion-to-text Converter That Can Potentially Serve As Software For Intelligent Wearable Devices For Writing From The Air. This Project Is A Reporter Of Occasional Gestures. It Will Use Computer Vision To Trace The Path Of The Finger. The Generated Text Can Also Be Used For Various Purposes, Such As Sending Messages, Emails, Etc. It Will Be A Powerful Means Of Communication For The Deaf. It Is An Effective Communication Method That Reduces Mobile And Laptop Usage By Eliminating The Need To Write
Visually Impaired People Are Unaware Of The Danger That They Are Facing In Their Life. They May Face Many Challenges While Performing Their Daily Activity Even In Their Familiar Environments. Vision Is The Necessary Human Senses And It Plays The Important Role In Human Perception About Surrounding Environment. Hence, There Are Variety Of Computer Vision Products And Services Which Are Used In The Development Of New Electronic Aids For Those Blind People. In This Paper We Designed To Provide Navigation To Those People. It Guides The People About The Object As Well As Provides The Distance Of The Object. The Algorithm Itself Calculates The Distance Of The Object. Here It Also Provides The Audio Jack To Insist Them About The Object. Here We Are Using SSD Algorithm For Object Detection And Calculating The Distance Of The Object By Using Monodepth Algorithm.
The Worst Possible Situation Faced By Humanity, COVID-19, Is Proliferating Across More Than 180 Countries And About 37,000,000 Confirmed Cases, Along With 1,000,000 Deaths Worldwide As Of October 2020. The Absence Of Any Medical And Strategic Expertise Is A Colossal Problem, And Lack Of Immunity Against It Increases The Risk Of Being Affected By The Virus. Since The Absence Of A Vaccine Is An Issue, Social Spacing And Face Covering Are Primary Precautionary Methods Apt In This Situation. This Study Proposes Automation With A Deep Learning Framework For Monitoring Social Distancing Using Surveillance Video Footage And Face Mask Detection In Public And Crowded Places As A Mandatory Rule Set For Pandemic Terms Using Computer Vision. The Paper Proposes A Framework Is Based On YOLO Object Detection Model To Define The Background And Human Beings With Bounding Boxes And Assigned Identifications. In The Same Framework, A Trained Module Checks For Any Unmasked Individual. The Automation Will Give Useful Data And Understanding For The Pandemic’s Current Evaluation; This Data Will Help Analyse The Individuals Who Do Not Follow Health Protocol Norms.
One Of The Major Reasons Behind Car Accidents Is The Drowsy Nature Acquired By A Driver While Driving Any Vehicle. Owing To The Ongoing Scenario, In This Project, We Aim To Develop A Real Time Driver Drowsiness Detection System In Order To Detect The Drivers’ Fatigue Status, Such As Dozing, Flickering Of Eye Lids And Time Span Of Eye Closure Without Having To Equip Their Bodies With Devices. The Objective Of This Project Is To Build A Drowsiness Detection System That Will Detect That A Person’s Eyes Are Closed For A Few Seconds. This System Will Alert The Driver When Drowsiness Is Detected. Apart From CNN, Computer Vision Also Plays A Major Role To Detect The Drowsiness Pattern Of The Driver. Cloud Architecture Has Also Proved To Be Beneficial In Case Of Capturing And Analyzing Real Time Video Streams.
Drone Is One Of The Latest Drone Technologies That Grows With Multiple Applications; One Of The Critical Applications Is For Fire-fighting Drones Such As Water Hose Carrying For Firefighting. One Of The Main Challenges Of The Drone Technologies Is The Non-linear Dynamic Movement Caused By A Variety Of Fire Conditions. One Solution Is To Use A Nonlinear Controller Such As Reinforcement Learning. In This Paper, Reinforcement Learning Has Been Applied As Their Key Control System To Improve The Conventional Approach, Which Is The Agent (drone) That Will Interact With The Environment Without Need Of The Controller For The Flying Process. This Paper Is Introduced An Optimization Method For The Hyperparameter In Order To Achieve A Better Reward. In Addition, We Only Concentrate On The Learning Rate (alpha) And Potential Reward Factor Discount (gamma) For Optimization In This Paper. From This Optimization, The Better Performance And Response From Our Result By Using Alpha = 0.1 & Gamma = 0.8 With Reward Produced 6100 And It Takes 49 Seconds In The Learning Process.
A Systematic And Exact Detection Of An Object Is A Foremost Point In Computer Vision Technology, With The Unfolding Of Recent Deep Learning Techniques, The Precision Of Detecting An Object Has Increased Greatly Thereby Igniting The Interest In This Area To Large Extent. The Main Aim Is To Integrate The Stateof-the-art Deep Learning Method On Pedestrian Object Detection In Real-time With Improved Accuracy. One Of The Crucial Problems In Deep Learning Is Using Computer Vision Techniques, Which Tend To Slow Down The Process With Trivial Performance. In This Work, An Improved SSD Transfer Learning-based Deep Learning Technique Is Used For Object Detection. It Is Also Shown That This Approach Can Be Used For Solving The Problem Of Object Detection In A Sustained Manner Having The Ability To Further Separate Occluded Objects. Moreover, The Use Of This Approach Has Enhanced The Accuracy Of Object Detection. The Network Used Is Trained On A Challenging Data Set And The Output Obtained Is Fast And Precise Which Is Helpful For The Application That Requires Object Detection.
The Rapid Development Of Artificial Intelligence Has Revolutionized The Area Of Autonomous Vehicles By Incorporating Complex Models And Algorithms. Self-driving Cars Are Always One Of The Biggest Inventions In Computer Science And Robotic Intelligence. Highly Robust Algorithms That Facilitate The Functioning Of These Vehicles Will Reduce Many Problems Associated With Driving Such As The Drunken Driver Problem. In This Paper Our Aim Is To Build A Deep Learning Model That Can Drive The Car Autonomously Which Can Adapt Well To The Real-time Tracks And Does Not Require Any Manual Feature Extraction. This Research Work Proposes A Computer Vision Model That Learns From Video Data. It Involves Image Processing, Image Augmentation, Behavioural Cloning And Convolutional Neural Network Model. The Neural Network Architecture Is Used To Detect Path In A Video Segment, Linings Of Roads, Locations Of Obstacles, And Behavioural Cloning Is Used For The Model To Learn From Human Actions In The Video.
The World Is Advancing Towards An Autonomous Environment At A Great Pace And It Has Become A Need Of An Hour, Especially During The Current Pandemic Situation. The Pandemic Has Hindered The Functioning Of Many Sectors, One Of Them Being Road Development And Maintenance. Creating A Safe Working Environment For Workers Is A Major Concern Of Road Maintenance During Such Difficult Times. This Can Be Achieved To Some Extent With The Help Of An Autonomous System That Will Aim At Reducing Human Dependency. In This Paper, One Of Such Systems, A Pothole Detection And Dimension Estimation, Is Proposed. The Proposed System Uses A Deep Learning Based Algorithm YOLO (You Only Look Once) For Pothole Detection. Further, An Image Processing Based Triangular Similarity Measure Is Used For Pothole Dimension Estimation. The Proposed System Provides Reasonably Accurate Results Of Both Pothole Detection And Dimension Estimation. The Proposed System Also Helps In Reducing The Time Required For Road Maintenance. The System Uses A Custom Made Dataset Consisting Of Images Of Water-logged And Dry Potholes Of Various Shapes And Sizes.
Images Captured In Low-light Conditions Are Often Disturbed By Low-light, Blur And Noise. Most Of The Conventional Image Enhancement Methods Are Less Robust Without Considering The Effectiveness Of The Blur And Noise. To Enhance Image Equality Under The Complex Environment, We Propose A Novel Image Enhancement Method Based On Joint Generative Adversarial Network (GAN) And Image Quality Assessment (IQA) Techniques. GAN Can Be Well Used For Image Enhancement In Low-light Case, But It Is Not Robust In Blur And Noise Case. IQA Method Uses CNN To Evaluate Each Enhanced Image Quality Based On Some Scores That Correlates Well With The Human Perception. The Scores Can Guide The GAN Learning For Further Enhancing The Image Quality. Instead Of L2-term Loss Function, We Define A Multi-term Loss Function For Its Minimization To Create A Good Image Estimate. Experimental Results Demonstrate The Proposed Method Is More Effective Than Current State-of-art Methods In Terms Of The Quantitative And Qualitative Evaluation.
Detection Of Diseases In Plants Is A Significant Task That Has To Be Done In Agriculture. This Is Something On Which The Economy Profoundly Depends. Infection Discovery In Plants Is A Significant Job In The Agribusiness Field, As Having Diseases In Plants Is Very Common. To Recognize The Diseases In Leaves, A Continuous Observation Of The Plants Is Required. This Observation Or Continuous Monitoring Of The Plants Takes A Lot Of Human Effort And It Is Time-consuming Too. To Make It Simply Some Sort Of Programmed Strategy Is Required To Observe The Plants. Program Based Identification Of Diseases In Plants Makes Easier To Detect The Damaged Leaves And Reduces Human Efforts And Time-saving. The Proposed Algorithm Distinguishing Sickness In Plants And Classify Them More Accurately As Compared To Existing Techniques.
Dogs Are One Of The Most Common Domestic Animals. Due To A Large Number Of Dogs, There Are Several Issues Such As Population Control, Decrease Outbreak Such As Rabies, Vaccination Control, And Legal Ownership. At Present, There Are Over 180 Dog Breeds. Each Dog Breed Has Specific Characteristics And Health Conditions. In Order To Provide Appropriate Treatments And Training, It Is Essential To Identify Individuals And Their Breeds. The Paper Presents The Classification Methods For Dog Breed Classification Using Two Image Processing Approaches 1) Conventional Based Approaches By Local Binary Pattern (LBP) And Histogram Of Oriented Gradient (HOG) 2) The Deep Learning Based Approach By Using Convolutional Neural Networks (CNN) With Transfer Learning. The Result Shows That Our Retrained CNN Model Performs Better In Classifying A Dog Breeds. It Achieves 96.75% Accuracy Compared With 79.25% Using The HOG Descriptor.
Detection Of Diseases In Plants Is A Significant Task That Has To Be Done In Agriculture. This Is Something On Which The Economy Profoundly Depends. Infection Discovery In Plants Is A Significant Job In The Agribusiness Field, As Having Diseases In Plants Is Very Common. To Recognize The Diseases In Leaves, A Continuous Observation Of The Plants Is Required. This Observation Or Continuous Monitoring Of The Plants Takes A Lot Of Human Effort And It Is Time-consuming Too. To Make It Simply Some Sort Of Programmed Strategy Is Required To Observe The Plants. Program Based Identification Of Diseases In Plants Makes Easier To Detect The Damaged Leaves And Reduces Human Efforts And Time-saving. The Proposed Algorithm Distinguishing Sickness In Plants And Classify Them More Accurately As Compared To Existing Techniques.
Historical Photographs Record The True Face Of A Moment In The Development Of Human History; They Have Authenticity, Vividness, And Unique Values. However, Due To Various Factors, Aging And Damage Will Occur. With The Development Of Computer Technology, The Restoration Technology Is More Used In Photo Restoration And Virtual Restoration Of Cultural Relics. This Paper First Analyzes The Principle Of Repairing Photo Archives Based On Computer Technology, And Then Uses The Combination Of Statistics And Computer Image Processing Technology To Detect And Repair The Scratches In Historical Photographs. And The Paper Establishes A Model Repair Framework, Which Provides A New Idea For The Repair Of Such Historical Photos. The Experimental Results Show That The Method Has A Significant Repair Effect.
Action Recognition In Videos, Especially For Violence Detection, Is Now A Hot Topic In Computer Vision. The Interest Of This Task Is Related To The Multiplication Of Videos From Surveillance Cameras Or Live Television Content Producing Complex 2D + T Data. State-of-the-art Methods Rely On End-to-end Learning From 3D Neural Network Approaches That Should Be Trained With A Large Amount Of Data To Obtain Discriminating Features. To Face These Limitations, We Present In This Article A Method To Classify Videos For Violence Recognition Purpose, Byusingaclassical 2D Convolutional Neural Network(CNN). The Strategy Of The Method Is Two-fold: We Start By Building Several 2D Spatio-temporal Representations From An Input Video, The New Representations Are Considered To Feed The CNN To The Train/test Process. The Classification Decision Of The Video Is Carried Out By Aggregating The Individual Decisions From Its Different 2D Spatio-temporal Representations. An Experimental Study On Public Datasets Containing Violent Videos Highlights The Interest Of The Presented Method.
One Of The Most Rapidly Spreading Cancers Among Various Other Types Of Cancers Known To Humans Is Skin Cancer. Melanoma Is The Worst And The Most Dangerous Type Of Skin Cancer That Appears Usually On The Skin Surface And Then Extends Deeper Into The Layers Of Skin. However, If Diagnosed At An Early Stage; The Survival Rate Of Melanoma Patients Is 96% With Simple And Economical Treatments. The Conventional Method Of Diagnosing Melanoma Involves Expert Dermatologists, Equipment, And Biopsies. To Avoid The Expensive Diagnosis, And To Assist Dermatologists, The Field Of Machine Learning Has Proven To Provide State Of The Art Solutions For Skin Cancer Detection At An Earlier Stage With High Accuracy. In This Paper, A Method For Skin Lesion Classification And Segmentation As Benign Or Malignant Is Proposed Using Image Processing And Machine Learning. A Novel Method Of Contrast Stretching Of Dermoscopic Images Based On The Methods Of Mean Values And Standard Deviation Of Pixels Is Proposed. Then The OTSU Thresholding Algorithm Is Applied For Image Segmentation.
Over The Last Decades, The Incidence Of Skin Cancer, Melanoma And Non-melanoma, Has Increased At A Continuous Rate. In Particular For Melanoma, The Deadliest Type Of Skin Cancer, Early Detection Is Important To Increase Patient Prognosis. Recently, Deep Neural Networks (DNNs) Have Become Viable To Deal With Skin Cancer Detection. In This Work, We Present A Smartphone-based Application To Assist On Skin Cancer Detection. This Application Is Based On A Convolutional Neural Network (CNN) Trained On Clinical Images And Patients Demographics, Both Collected From Smartphones. Also, As Skin Cancer Datasets Are Imbalanced, We Present An Approach, Based On The Mutation Operator Of Differential Evolution (DE) Algorithm, To Balance Data.In This Sense, Beyond Provides A flexible Tool To Assist Doctors On Skin Cancer Screening Phase, The Method Obtains Promising Results With A Balanced Accuracy Of 85% And A Recall Of 96%. Index Terms Skin Cancer Detection, Smartphone Application, Deep Learning, Convolutional Neural Network.
We do code in matlab for all domains, Image Processing, Simulink Design and Power Electronics.
Our Complete portfolio describes how stuff we are watch our demo @ Youtube Link.
2025 © All Rights Reserved. Site Map