Image recognition : progress, trends and challenges

Description

IMAGE RECOGNITIONPROGRESS, TRENDSAND CHALLENGES IMAGE RECOGNITIONPROGRESS, TRENDSAND CHALLENGES CONTENTS PREFACE Chapter 1VISUAL COMPUTING FOR INTELLIGENTHUMAN-COMPUTER INTERACTIONS:TREND, CHALLENGES AND PROGRESS Abstract 1. INTELLIGENT HUMAN COMPUTER INTERACTIONS 2. INTERACTIVE HCI SYSTEMS 2.1. Virtual Reality 2.2. Augmented Reality 2.3. Mixed Reality 3. VISION-BASED INTERACTION 3.1. Vision-Based Intelligent HCI 3.2. Visual Computing for Intelligent HCI 4. CHALLENGES 4.1. User Experience 4.2. Knowledge to Awareness 4.3. Human-Aware Consciousness 4.4. System-Aware Consciousness 4.5. Visualization and Immersion 4.6. Target of Interest CONCLUSION REFERENCES Chapter 2PRINCIPLE COMPONENT ANALYSIS BASEDCOMPUTING IN IMAGE RECOGNITION Abstract 1. INTRODUCTION 2. BASIC PCA METHODOLOGY 3. PRINCIPLE COMPONENT SELECTION 4. RESIDUALS ASSESSMENT 4.1. Euclidean Distance 4.1.1. Hausdorff Distance 4.2. Stochastic Mean Error 5. DECOMPOSITION COMPUTATION FOR SUBSPACE TRAINING 6. PROBABILISTIC PCA 7. KERNEL PCA 8. MULTI-DIMENSIONAL PCA 9. ROBUST PCA 10. DISCUSSION AND COMPARISONS CONCLUSION REFERENCES Chapter 3RECOGNITION AND AWARENESS MODELINGFOR QUALITY OF EXPERIENCEAND QUALITY OF SERVICES Abstract 1. INTRODUCTION 2. RELATED WORK 2.1. Impact of QoE on Vision-Based Interaction 2.2. QoE-QoS Recognition 3. DATA AWARE MODELING 4. QOE-QOS MANAGEMENT 4.1. Optimal QoE Mangement 4.2. Optimal QoS Mangement 4.3. QoE-QoS Balance 5. EXPERIMENT 6. DISCUSSION CONCLUSION REFERENCES Chapter 4PERFORMANCE ANALYSIS OF LOCAL BINARYPATTERNS FOR IMAGE TEXTURECLASSIFICATION METHODS ABSTRACT 1. INTRODUCTION 2. SPATIAL DOMAIN LOCAL BINARY PATTERN 2.1. Local Binary Pattern 2.2. Local Concave and Convex Microstructure Patterns 2.4. Other LBP Variants 3. WAVELET DOMAIN LOCAL BINARY PATTERN 3.1. Discrete Wavelet Transform 3.2. Methods with DWT 3.3. Methods with Both LBP and DWT 4. RESULTS AND DISCUSSIONS 4.1. Experiment on Brodatz Database 4.1.1. Performance Comparison of Various LBP Methods in Termsof Precision and Recall 4.1.2. Performance Comparison of Various LBP Methods in Termsof ARR 4.1.3. Performance Comparison of Various LBP Methods in Termsof F-Measure 4.2. Experiment on Outex Database 4.2.1. Performance Comparison of Various LBP Methods in Termsof Precision and Recall 4.2.2. Performance Comparison of Various LBP Methods in Termsof ARR 4.2.3. Performance Comparison of Various LBP Methods in Termsof F-Measure CONCLUSION ACKNOWLEDGMENTS REFERENCES Chapter 5GENERATIVE ADVERSARIAL NETWORKS -AN INTRODUCTION Abstract 1. INTRODUCTION 2. GENERATIVE ADVERSARIAL NETWORKS (GANS) 2.1. GAN Fundamentals 2.2. Objective Functions 2.2.1. f-divergence 2.2.2. Integral ProbabilityMetric 2.2.3. Auxiliary Object Functions 2.3. The Latent Space 2.3.1. Latent Space Decomposition 2.3.2. With an Autoencoder 3. GANS’ VARIANTS 3.1. Fully Connected GANs 3.2. Conditional GANs (CGAN) 3.3. Laplacian Pyramid of Adversarial Networks (LAPGAN) 3.4. Deep Convolutional Generative Adversarial Networks(DCGAN) 3.5. Adversarial Autoencoders (AAE) 3.6. Generative Recurrent Adversarial Networks (GRAN) 3.7. Information Maximizing Generative Adversarial Networks(InfoGAN) 3.8. Bidirectional Generative Adversarial Networks (BiGAN) 4. DISCUSSION 4.1. Advantages 4.2. Disadvantages 4.3. Future Challenges CONCLUSION REFERENCES Chapter 6KNOWLEDGE BASED ADAPTIVE FUZZYSTRATEGY FOR TARGET OF INTERESTDIFFERENTIATION Abstract 1. INTRODUCTION 2. PRELIMINARY 3. PROBLEM FORMULATION 4. OPTIMAL SOLUTION 5. ADAPTIVE FUZZY LEARNING 5.1. The Knowledge Modeling 5.2. The Fuzzy System 5.3. The Adaptive Learning 5.4. Discussion 6. EXPERIMENT 6.1. Qualitative Analysis 6.2. QuantitativeAnalysis CONCLUSION REFERENCES Chapter 7PCA-BASED IMAGE RECOGNITIONAPPLICATIONS ON VISION BASEDCOMPUTING Abstract 1. INTRODUCTION 2. IMAGE COMPRESSION 3. VISUAL TRACKING 4. VISUAL RECOGNITION 5. SUPER-RESOLUTION IMAGE RECONSTRUCTION 6. DISCUSSION 7. ISSUES ON THE IMPLEMENTATIONOF COMPUTING CORE 8. ISSUES ON THE IMPLEMENTATIONOF COMPUTING STRATEGY CONCLUSION REFERENCES Chapter 8REVIEW OF FEATURE EXTRACTIONAND CLASSIFICATION TECHNIQUESFOR EPILEPTIC SEIZURE DETECTION ABSTRACT 1. INTRODUCTION 2. TECHNIQUES AND MATERIALS FOR EEG CLASSIFICATION 2.1. Experimental Benchmark Dataset 3. FEATURE EXTRACTION, SELECTIONAND CLASSIFICATION TECHNIQUES 3.1. Feature Extraction Techniques 3.1.1. Time-Domain Features 3.1.2. Frequency-Domain Features 3.1.2.1. Power Spectral Density 3.1.3. Time-Frequency Distributions 3.1.3.1. Gabor Transform (GT) 3.1.3.2. Wigner-Ville Distribution (WVD) 3.1.3.3. Wavelet Transformation (WT) 3.1.4. Time-Frequency Domain Features 3.1.4.1. Approximate Entropy (AE) 3.1.4.2. Largest Lyapunov Exponent (LLE) 3.1.4.3. Correlation Dimension (CD) 3.2. Feature Selection Techniques 3.2.1. Fuzzy Logic Based Features Selection 3.3. Classification Techniques 3.3.1. Artificial Neural Networks 3.3.2. Support Vector Machine (SVM) 3.3.3. Directed Acyclic Graph Support Vector Machine (DAG SVM) 4. COMPARATIVE PERFORMANCE ANALYSISOF VARIOUS TECHNIQUES 4.1. Performance Comparison of Various Domain Features 4.2. Performance Comparison of Classifiers Based on VariousFeature Extraction Techniques 4.3. Performance Comparison of Classifier Based on FeatureSelection 4.4. Performance Comparison of Various Classifiers CONCLUSION ACKNOWLEDGMENTS REFERENCES Chapter 9GENERATIVE ADVERSARIAL NETWORKS -APPLICATION DOMAINS Abstract Generative adversarial 2. GENERATIVE ADVERSARIAL NETWORKS (GANS) 3. GANS’ APPLICATIONS 3.1.Based Applications 3.1.1. Generation of High-Quality Images 3.1.2. Image Inpainting 3.1.3. Super-Resolution 3.1.4. Person Re-Identification 3.1.5. Object Detection 3.1.6. Video Prediction and Generation 3.1.7. Facial Attribute Manipulation 3.1.8. Anime Character Generation 3.1.9. Image to Image Translation 3.1.10. Text to Image Translation 3.1.11. Face Aging 3.1.12. Human Pose Estimation 3.1.13. De-Occlusion 3.1.14. Image Blending 3.2. Domain Adaptation 3.3. Sequential Data Based Applications 3.3.1. Speech 3.3.2. Music 3.4. Improving Classification and Recognition 3.5. Miscellaneous Applications 3.5.1. Drug Discovery Development in OncologyInsilico Medicine CONCLUSION REFERENCES Chapter 10VISION-BASED INTELLIGENTHUMAN-COMPUTER INTERACTIONSWITH MULTIPLE AGENT COLLABORATION Abstract 1. INTRODUCTION 2. FUNCTIONAL FRAMEWORK 3. AN AGENT-BASED MIXED REALITY SYSTEM 4. AGENT-BASED COLLABORATIVE INFORMATIONPROCESSING 5. AGENT-BASED COLLABORATION 6. AN EXAMPLE IN MIXED REALITY 6.1. Application in Intelligent Driving System CONCLUSION REFERENCES Chapter 11A COMPREHENSIVEVISION-AWARE-COMPUTING-BASEDINTERACTIVE SYSTEMWITH AGENT-BASED COLLABORATIVEINFORMATION PROCESSING Abstract 1. INTRODUCTION 2. RELATED WORKS 3. AGENT-AWARE COMPUTING 3.1. Methodological Functions 3.2. Implementation 4. SYSTEM DESIGN 5. AGENT-BASED MR SYSTEM DESIGN 6. SYSTEM ARCHITECTURE 7. CAMERA SYSTEM 8. QOE-QOS MANAGEMENT 9. CONFIDENTIALITY 10. CONTEXT PATTERN ANALYSIS 11. DATA MANAGEMENT FOR USER-AWARENESS 12. SCENARIO FUSION CONCLUSION REFERENCES Chapter 12 SUMMARY AND FUTURE WORKS Abstract 1. SUMMARY 2. FUTURE WORKS 2.1. Fusion of Awareness 2.2. Psycho-Physiological Signals 2.3. Ambient HCI inWide Area Mixed Reality ABOUT THE EDITORS INDEX Blank Page