Pattern recognition and artificial intelligence : 5th Mediterranean Conference, MedPRAI 2021, Istanbul, Turkey, December 17-18, 2021, Proceedings / / Chawki Djeddi [and four others], editors

• Record Nr.: UNINA9910561293603321
• Pubbl/distr/stampa: Cham, Switzerland : , : Springer Nature Switzerland AG, , [2022]; ©2022
• ISBN: 3-031-04112-7
• Descrizione fisica: 1 online resource (388 pages)
• Collana: Communications in computer and information science ; ; 1543
• Disciplina: 006.4
• Soggetti: Artificial intelligence; Pattern recognition systems; Education - Data processing
• Lingua di pubblicazione: Inglese
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Nota di bibliografia: Includes bibliographical references and index.
• Nota di contenuto: Intro -- Preface -- Organization -- Contents -- Computer Vision and Image Processing -- Human Action Recognition Using Attention Mechanism and Gaze Information -- 1 Introduction -- 2 Related Works -- 2.1 HOI Detection -- 2.2 Gaze Following -- 3 The Proposed Idea -- 3.1 The Basic Model -- 3.2 The Proposed Structure -- 3.3 Gaze Heatmap and the Improved iCAN Module -- 3.4 Depth-Wise Separable Convolution -- 4 Experimental Results -- 4.1 Dataset -- 4.2 Evaluation Metrics -- 4.3 Implementation Details -- 4.4 Quantitative Evaluation -- 4.5 Qualitative Evaluation -- 4.6 Ablation Study -- 5 Conclusion -- References -- Face Recognition Under Partial Occlusion: A Detection and Exclusion of Occluded Face Regions Approach -- 1 Introduction -- 2 Literature Review -- 3 The Proposed Framework -- 3.1 Training Phase -- 3.2 Testing Phase -- 4 Experiments -- 4.1 Implementation Details -- 4.2 Evaluation Metrics -- 4.3 Non-occluded Region Retrieval Using Skin Detection -- 4.4 Non-occluded Region Retrieval Using Haar Cascade Classifiers -- 4.5 Effect of the Number of Images in Learning a Class -- 5 Conclusion -- References -- Color Vision Deficiency and Live Recoloring -- 1 Introduction -- 2 Background and Related Work -- 3 Proposed System -- 4 Experiments and Results -- 5 Conclusion -- References -- Real Time Handling Occlusion in Augmented Reality ; Based on Photogrammetry -- 1 Introduction -- 2 Related Work -- 2.1 Contour-Based Methods -- 2.2 Depth-Based Method -- 2.3 3D Reconstruction Based Method -- 3 System Overview -- 3.1 Photogrammertic 3D Modeling -- 3.2 AR Application Development with Handling Occlusion -- 4 Results and Discussion -- 4.1 Experimental Results -- 4.2 Discussion -- 5 Conclusion -- References -- Colour Image Encryption Based on Fisher-Yates Algorithm and Chaotic Maps -- 1 Introduction -- 2 Materials and Methods -- 2.1 2D Logistic-Tent Map.; 2.2 Fisher-Yates Algorithm -- 2.3 Diffusion Using 2D LTM -- 2.4 Proposed Method for Encrypting Colour Image -- 3 Results and Discussion -- 3.1 Statistical Attack Analysis -- 4 Conclusion -- References -- Robust SLAM System by Incorporating UWB Positioning and Landmark Localization -- 1 Introduction -- 2 Related Work -- 3 Proposed Approach -- 4 Experiments -- 5 Conclusion -- References -- Intelligent Systems for Medical Diagnosis -- Deep Learning for Liver Disease Prediction -- 1 Introduction -- 2 Related Works -- 3 Materials and Methods -- 3.1 Datasets -- 3.2 Preprocessing -- 3.3 Model Building -- 4 Experimental Results -- 4.1 Model Evaluation -- 5 Results and Discussion -- 5.1 Experiments on BUPA Dataset -- 5.2 Experiments on ILPD Dataset -- 6 Conclusion -- References -- Pre-trained Backbones Effect on Nuclei Segmentation Performance -- 1 Introduction -- 2 Background -- 3 Methods -- 3.1 VGG16 -- 3.2 MobileNet -- 3.3 EfficientNet B3 -- 4 Experiments and Results -- 4.1 Triple Negative Breast Cancer Dataset -- 4.2 Pre-processing -- 4.3 Methodology -- 5 Conclusion -- References -- Organ Detection in US Scanning by Non-expert Operator -- 1 Introduction -- 2 Principle of the Method -- 2.1 Detection of Seeds -- 2.2 Connectivity of Seeds with Nearby Frames -- 2.3 Point Detection -- 2.4 Coherence of the Contour -- 2.5 Study Population -- 2.6 Experimental Protocol -- 3 Results -- 4 Discussion -- 4.1 Limitations -- 5 Conclusion -- References -- Deep Transfer Learning Techniques for Automated Classification of Histological Childhood Medulloblastoma Images -- 1 Introduction -- 2 Related Works -- 3 Material and Methods -- 3.1 Childhood MB Dataset -- 3.2 Convolutional Neural Networks (CNN) -- 3.3 Transfer Learning -- 3.4 Conducted Comparative Study -- 3.5 Classification Process -- 4 Experiments and Results -- 5 Conclusion -- References.; A New Medical Image Encryption Algorithm for IoMT Applications -- 1 Introduction -- 2 Materials and Methods -- 2.1 Bit-level Permutation -- 2.2 2D SIM Model -- 2.3 Diffusion Using 2D SIM -- 2.4 Proposed Encryption Algorithm -- 3 Results and Discussion -- 3.1 Key Sensitivity Analysis -- 3.2 Key-Space Analysis -- 3.3 Histogram Based Analysis -- 3.4 Information Entropy Analysis -- 3.5 UACI and NPCR Analysis -- 4 Conclusion -- References -- Feature Relevance Analysis for Handwriting Based Identification of Parkinson's Disease -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Dataset -- 3.2 Input -- 3.3 Feature Extraction -- 3.4 Feature Relevance -- 3.5 Classification and Decision Fusion -- 4 Experimental Results and Analysis -- 4.1 Results of Offline and Online Features -- 4.2 Impact of Feature Relevance on Task-wise Accuracies -- 4.3 Impact of Feature Fusion Versus Decision Fusion -- 5 Conclusion -- References -- .26em plus .1em minus .1emEEG Based Major Depressive Disorder (MDD) Detection Using Machine Learning -- 1 Introduction -- 2 Material and Methods -- 2.1 Participants -- 2.2 Data Acquisition and Preprocessing -- 2.3 New Feature Generation -- 3 Proposed Machine Learning Methods -- 3.1 K-Nearest Neighbors -- 3.2 Decision Tree -- 3.3 Support Vector Machine -- 3.4 Naïve Bayes -- 4 Proposed Deep ; Learning Method -- 5 Validation of Classification Models -- 6 Results -- 7 Discussion -- 8 Conclusion -- References -- Document Analysis and Understanding -- Islamophobic Hate Speech Detection from Electronic Media Using Deep Learning -- 1 Introduction -- 2 Related Work -- 3 Methods -- 3.1 Dataset -- 3.2 Data Pre-processing -- 3.3 Feature Extraction and Classification -- 4 Experimental Results and Analysis -- 5 Conclusion -- References -- A Semi-automatic Document Screening System for Computer Science Systematic Reviews -- 1 Introduction.; 2 Related Work -- 2.1 Research Issues -- 2.2 Research Contributions -- 3 Proposed System -- 3.1 Adopted Features -- 3.2 Classification Process -- 4 Experimentation -- 5 Results and Analysis -- 6 Conclusion -- References -- Handwriting Based Personality Traits Identification Using Adaptive Boosting and Textural Features -- 1 Introduction -- 2 Data Corpus -- 3 Proposed Method -- 3.1 Feature Extraction -- 3.2 Combined Classification and Experimentation -- 4 Results and Discussion -- 5 Conclusions and Future Works -- References -- An Automated Approach for Analysing Students Feedback Using Sentiment Analysis Techniques -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 4 Results -- 4.1 Evaluation Metrics -- 4.2 Polarity Prediction Using Textblob and MonkeyLearn -- 4.3 Emotion Identification Using MonkeyLearn -- 5 Conclusion -- References -- Keyword Spotting in Modern Handwritten Documents Using oBIFs -- 1 Introduction -- 2 Related Work -- 3 Proposed Methods -- 3.1 Feature Extraction -- 3.2 Word Matching -- 4 Experimental Results -- 5 Conclusions and Future Works -- References -- Artificial Intelligence and Intelligent Systems -- Real-Time Detection of Traffic Anomalies Near Roundabouts -- 1 Introduction -- 2 Literature Review -- 3 Proposed Methodology -- 3.1 Foreground Object Detection -- 3.2 Roundabout Detection -- 3.3 Traffic Anomalies Near the Roundabout -- 4 Results and Discussions -- 4.1 Experimental Setup -- 4.2 Datasets Used -- 4.3 Experimental Results -- 5 Conclusions -- References -- A Two-Stage Framework for Diverse Traffic Light Recognition Based on Individual Signal Detection -- 1 Introduction -- 2 Related Works -- 3 Method -- 3.1 Light Candidate Detection -- 3.2 Traffic Light Classification -- 3.3 Dataset -- 4 Experiments -- 5 Conclusion -- References.; Enhancing Vehicle Networks Performance by Cache-Aided Non-Orthogonal Multiple Access (CA-NOMA) -- 1 Introduction -- 2 Related Works -- 2.1 Caching -- 2.2 Non-Orthogonal Multiple Access (NOMA) -- 2.3 Cache-Aided NOMA -- 3 System Model -- 3.1 Caching -- 3.2 Optimal Power Allocation -- 4 Simulation Result -- 5 Conclusion and Future Work -- References -- Performance Analysis of Hard Decision and Soft Decision Algorithms Over In VivoRadio Channel -- 1 Introduction -- 1.1 An Overview for In-VivoRadio Channel -- 1.2 In VivoCommunication System -- 1.3 Viterbi Algorithm -- 2 Soft and Hard Decisions -- 2.1 Hard Decision Decoding -- 2.2 Soft Viterbi Decoding -- 3 Simulations and Results -- 4 Conclusion -- References -- The Thermal Modeling for Underground Cable Based on ANN Prediction -- 1 Introduction -- 2 A Geometric Model for Underground Cable Used -- 3 Modeling of Thermal Resistance in the Internal and External Environments -- 4 A Mathematical Model for Underground Cables -- 5 Proposed of ANN-Based Underground Cable Rating -- 6 Results and Discussion -- 7 Conclusion -- References -- Layer-Wise Relevance Propagation for Smart-Grid Stability Prediction -- 1 Introduction -- 2 Related Works -- 2.1 Big Data Analytics in Smart Grids and Smart Grid Stability -- 2.2 XAI and Interpretability -- 3 XAI for Smart-Grid Stability Prediction -- 3.1 Optimized Deep Models -- 3.2 Layer-Wise Relevance ; Propagation -- 4 Experimental Analysis -- 4.1 Building Dataset -- 4.2 Experimental Setup and Experiment -- 5 Conclusion and Future Works -- References -- Abalone Age Prediction Using Machine Learning -- 1 Introduction -- 2 Materials and Methodology -- 2.1 Dataset -- 2.2 Classification -- 3 Experimental Results -- 4 Conclusion -- References -- A Novel Optimized Method for Feature Selection Using Non-linear Kernel-Free Twin Quadratic Surface Support Vector Machine.; 1 Introduction.