Zheng / Xue Statistical Learning and Pattern Analysis for Image and Video Processing

1. Auflage 2009
ISBN: 978-1-84882-312-9
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 365 Seiten

Reihe: Advances in Computer Vision and Pattern Recognition

Why are We Writing This Book? Visual data (graphical, image, video, and visualized data) affect every aspect of modern society. The cheap collection, storage, and transmission of vast amounts of visual data have revolutionized the practice of science, technology, and business. Innovations from various disciplines have been developed and applied to the task of designing intelligent machines that can automatically detect and exploit useful regularities (patterns) in visual data. One such approach to machine intelligence is statistical learning and pattern analysis for visual data. Over the past two decades, rapid advances have been made throughout the ?eld of visual pattern analysis. Some fundamental problems, including perceptual gro- ing,imagesegmentation, stereomatching, objectdetectionandrecognition,and- tion analysis and visual tracking, have become hot research topics and test beds in multiple areas of specialization, including mathematics, neuron-biometry, and c- nition. A great diversity of models and algorithms stemming from these disciplines has been proposed. To address the issues of ill-posed problems and uncertainties in visual pattern modeling and computing, researchers have developed rich toolkits based on pattern analysis theory, harmonic analysis and partial differential eq- tions, geometry and group theory, graph matching, and graph grammars. Among these technologies involved in intelligent visual information processing, statistical learning and pattern analysis is undoubtedly the most popular and imp- tant approach, and it is also one of the most rapidly developing ?elds, with many achievements in recent years. Above all, it provides a unifying theoretical fra- work for intelligent visual information processing applications.

Autoren/Hrsg.

Zheng, Nanning

Xue, Jianru

Inhaltsverzeichnis

1;Preface;5
1.1;Why are We Writing This Book?;5
2;Acknowledgments;8
3;Contents;9
4;Chapter 1 Pattern Analysis and Statistical Learning;15
4.1;1.1 Introduction;15
4.2;1.1.1 Statistical Pattern Recognition;16
4.3;1.1.2 Pattern Theory;18
4.4;1.2 Statistical Classification;20
4.5;1.2.1 Feature Extraction and Selection;20
4.6;1.2.2 Classifier;21
4.7;1.3 Visual Pattern Representation;22
4.8;1.3.1 The Curse of Dimensionality;23
4.9;1.3.2 Dimensionality Reduction Techniques;23
4.10;1.4 Statistical Learning;24
4.11;1.4.1 Prediction Risk;25
4.12;1.4.2 Supervised, Unsupervised, and Others;26
4.13;1.5 Summary;28
4.14;References;28
5;Chapter 2 Unsupervised Learning for Visual Pattern Analysis;29
5.1;2.1 Introduction 2.1.1 Unsupervised Learning;29
5.2;2.1.2 Visual Pattern Analysis;30
5.3;2.1.3 Outline;31
5.4;2.2 Cluster Analysis;31
5.5;2.3 Clustering Algorithms;35
5.6;2.3.1 Partitional Clustering;35
5.7;2.3.2 Hierarchical Clustering;44
5.8;2.4 Perceptual Grouping;47
5.9;2.4.1 Hierarchical Perceptual Grouping;47
5.10;2.4.2 Gestalt Grouping Principles;49
5.11;2.4.3 Contour Grouping;53
5.12;2.4.4 Region Grouping;59
5.13;2.5 Learning Representational Models for Visual Patterns;61
5.14;2.6 Summary;62
5.15;Appendix;62
5.16;References;62
6;Chapter 3 Component Analysis;64
6.1;3.1 Introduction;64
6.2;3.2 Overview of Component Analysis;67
6.3;3.3 Generative Models;68
6.4;3.3.1 Principal Component Analysis;68
6.5;3.3.2 Nonnegative Matrix Factorization;79
6.6;3.3.3 Independent Component Analysis;85
6.7;3.4 Discriminative Models;89
6.8;3.4.1 Linear Discriminative Analysis;89
6.9;3.4.2 Oriented Component Analysis;92
6.10;3.4.3 Canonical Correlation Analysis;92
6.11;3.4.4 Relevant Component Analysis;94
6.12;3.5 Standard Extensions of the Linear Model 3.5.1 Latent Variable Analysis;96
6.13;3.5.2 Kernel Method;96
6.14;3.6 Summary;96
6.15;References;97
7;Chapter 4 Manifold Learning;99
7.1;4.1 Introduction;99
7.2;4.2 Mathematical Preliminaries;103
7.3;4.2.1 Manifold Related Terminologies;103
7.4;4.2.2 Graph Related Terminologies;104
7.5;4.3 Global Methods;106
7.6;4.3.1 Multidimensional Scaling;106
7.7;4.3.2 Isometric Feature Mapping;107
7.8;4.3.3 Variants of the Isomap;108
7.9;4.4 Local Methods;112
7.10;4.4.1 Locally Linear Embedding;112
7.11;4.4.2 Laplacian Eigenmaps;115
7.12;4.4.3 Hessian Eigenmaps;119
7.13;4.4.4 Diffusion Maps;121
7.14;4.5 Hybrid Methods: Global Alignment of Local Models;125
7.15;4.5.1 Global Coordination of Local Linear Models;125
7.16;4.5.2 Charting a Manifold;127
7.17;4.5.3 Local Tangent Space Alignment;129
7.18;4.6 Summary;129
7.19;Appendix;130
7.20;References;130
8;Chapter 5 Functional Approximation;132
8.1;5.1 Introduction;132
8.2;5.2 Modeling and Approximating the Visual Data;135
8.3;5.2.1 On Statistical Analysis;136
8.4;5.2.2 On Harmonic Analysis;137
8.5;5.2.3 Issues of Approximation and Compression;138
8.6;5.3 Wavelet Transform and Lifting Scheme 5.3.1 Wavelet Transform;140
8.7;5.3.2 Constructing a Wavelet Filter Bank;141
8.8;5.3.3 Lifting Scheme;143
8.9;5.3.4 Lifting-Based Integer Wavelet Transform;144
8.10;5.4 Optimal IntegerWavelet Transform;145
8.11;5.5 Introducing Adaptability into the Wavelet Transform;147
8.12;5.5.1 Curve Singularities in an Image;148
8.13;5.5.2 Anisotropic Basis;148
8.14;5.5.3 Adaptive Lifting-Based Wavelet;150
8.15;5.6 Adaptive Lifting Structure;151
8.16;5.6.1 Adaptive Prediction Filters;151
8.17;5.6.2 Adaptive Update Filters;153
8.18;5.7 Adaptive Directional Lifting Scheme;154
8.19;5.7.1 ADL Framework;155
8.20;5.7.2 Implementation of ADL;156
8.21;5.8 Motion Compensation Temporal Filtering in Video Coding;159
8.22;5.8.1 Overview of MCTF;159
8.23;5.8.2 MC in MCTF;162
8.24;5.8.3 Adaptive Lifting-Based Wavelets in MCTF;163
8.25;5.9 Summary and Discussions;164
8.26;References;165
9;Chapter 6 Supervised Learning for Visual Pattern Classification;170
9.1;6.1 Introduction;170
9.2;6.2 An Example of Supervised Learning;171
9.3;6.3 Support Vector Machine;174
9.4;6.3.1 Optimal Separating Hyper-plane;174
9.5;6.3.2 Realization of SVM;178
9.6;6.3.3 Kernel Function;180
9.7;6.4 Boosting Algorithm;182
9.8;6.4.1 AdaBoost Algorithm;183
9.9;6.4.2 Theoretical Analysis of AdaBoost;184
9.10;6.4.3 AdaBoost Algorithm as an Additive Model;187
9.11;6.5 Summary;189
9.12;Appendix;189
9.13;References;190
10;Chapter 7 Statistical Motion Analysis;191
10.1;7.1 Introduction;191
10.2;7.1.1 Problem Formulation;191
10.3;7.1.2 Overview of Computing Techniques;193
10.4;7.2 Bayesian Estimation of Optical Flow;196
10.5;7.2.1 Problem Formulation;196
10.6;7.2.2 MAP Estimation;200
10.7;7.2.3 Occlusion;202
10.8;7.3 Model-Based Motion Analysis;203
10.9;7.3.1 Motion Models;204
10.10;7.3.2 Statistical Model Selection;205
10.11;7.3.3 Learning Parameterized Models;206
10.12;7.4 Motion Segmentation;211
10.13;7.4.1 Layered Model: Multiple Motion Models;212
10.14;7.4.2 Clustering Optical Flow Field into Layers;214
10.15;7.4.3 Mixture Estimation for Layer Extraction;215
10.16;7.5 Statistics of Optical Flow;218
10.17;7.5.1 Statistics of Optical Flow;218
10.18;7.5.2 Motion Prior Modeling;220
10.19;7.5.3 Contrastive Divergence Learning;221
10.20;7.6 Summary;222
10.21;Appendix Graph and Neighborhood;222
10.22;Steerable Filter Design;223
10.23;References;224
11;Chapter 8 Bayesian Tracking of Visual Objects;226
11.1;8.1 Introduction;226
11.2;8.2 Sequential Bayesian Estimation;228
11.3;8.2.1 Problem Formulation of Bayesian Tracking;229
11.4;8.2.2 Kalman Filter;230
11.5;8.2.3 Grid-Based Methods;231
11.6;8.2.4 Sub-optimal Filter;231
11.7;8.3 Monte Carlo Filtering;233
11.8;8.3.1 Problem Formulation;233
11.9;8.3.2 Sequential Importance Sampling;235
11.10;8.3.3 Sequential Monte Carlo Filtering;240
11.11;8.3.4 Particle Filter;241
11.12;8.4 Object Representation Model;244
11.13;8.4.1 Visual Learning for Object Representation;245
11.14;8.4.2 Active Contour;246
11.15;8.4.3 Appearance Model;250
11.16;8.5 Summary;252
11.17;References;253
12;Chapter 9 Probabilistic Data Fusion for Robust Visual Tracking;254
12.1;9.1 Introduction;254
12.2;9.2 Earlier Work on Robust Visual Tracking;257
12.3;9.3 Data Fusion-Based Visual Tracker;260
12.4;9.3.1 Sequential Bayesian Estimator;260
12.5;9.3.2 The Four-Layer Data Fusion Visual Tracker;262
12.6;9.4 Layer 1: Visual Cue Fusion;264
12.7;9.4.1 Fusion Rules: Product Versus Weighted Sum;264
12.8;9.4.2 Adaptive Fusion Rule;266
12.9;9.4.3 Online Approach to Determining the Reliability of a Visual cue;267
12.10;9.5 Layer 2: Model Fusion;269
12.11;9.5.1 Pseudo-Measurement-Based Multiple Model Method;270
12.12;9.5.2 Likelihood Function;272
12.13;9.6 Layer 3: Tracker Fusion;273
12.14;9.6.1 Problem Formulation;274
12.15;9.6.2 Interactive Multiple Trackers;275
12.16;9.6.3 Practical Issues;276
12.17;9.7 Sensor Fusion;278
12.18;9.8 Implementation Issues and Empirical Results;280
12.19;9.8.1 Visual Cue Fusion Layer;280
12.20;9.8.2 Model Fusion Layer;283
12.21;9.8.3 Tracker Fusion Layer;285
12.22;9.8.4 Bottom-Up Fusion with a Three-Layer Structure;290
12.23;9.8.5 Multi-Sensor Fusion Tracking System Validation;290
12.24;9.9 Summary;292
12.25;References;293
13;Chapter 10 Multitarget Tracking in Video-Part I;295
13.1;10.1 Introduction;295
13.2;10.2 Overview of MTTV Methods;298
13.3;10.3 Static Model for Multitarget;300
13.4;10.3.1 Problem formulation;300
13.5;10.3.2 Observation Likelihood Function;302
13.6;10.3.3 Prior Model;303
13.7;10.4 Approximate Inference;304
13.8;10.4.1 Model Approximation;304
13.9;10.4.2 Algorithm Approximation;307
13.10;10.5 Fusing Information from Temporal and Bottom-Up Detectors;310
13.11;10.6 Experiments and Discussions;312
13.12;10.6.1 Proof-of-Concept;313
13.13;10.6.2 Comparison with Other Trackers;316
13.14;10.6.3 The Efficiency of the Gibbs Sampler;323
13.15;10.7 Summary;323
13.16;References;323
14;Chapter 11 Multi-Target Tracking in Video – Part II;326
14.1;11.1 Introduction;326
14.2;11.2 Overview of the MTTV Data Association Mechanism;329
14.3;11.2.1 Handing Data Association Explicitly;329
14.4;11.2.2 Handing Data Association Implicitly;331
14.5;11.2.3 Detection and Tracking;332
14.6;11.3 The Generative Model for MTT;333
14.7;11.3.1 Problem Formulation;333
14.8;11.3.2 The Generative Model;334
14.9;11.4 Approximating The Marginal Term;336
14.10;11.4.1 The State Prediction;337
14.11;11.4.2 Existence and Association Posterior;339
14.12;11.5 Approximating the Interactive Term;341
14.13;11.6 Hybrid Measurement Process;342
14.14;11.7 Experiments and Discussion;342
14.15;11.7.1 Tracking Soccer Players;343
14.16;11.7.2 Tracking Pedestrians in a Dynamic Scene;344
14.17;11.7.3 Discussion;344
14.18;11.8 Summary;347
14.19;References;347
15;Chapter 12 Information Processing in Cognition Process and New Artificial Intelligent Systems;349
15.1;12.1 Introduction;349
15.2;12.2 Cognitive Model: A Prototype of Intelligent System;351
15.3;12.3 Issues in Theories and Methodologies of Current Brain Research and Vision Science;353
15.4;12.4 Interactive Behaviors and Selective Attention in the Process of Visual Cognition;357
15.5;12.5 Intelligent Information Processing and Modeling Based on Cognitive Mechanisms 12.5.1 Cognitive Modeling and Behavioral Control in Complex Systems in an Information Environment;359
15.6;12.5.2 Distributed Cognition;362
15.7;12.5.3 Neurophysiological Mechanism of Learning and Memory and Information Processing Model;364
15.8;12.6 Cognitive Neurosciences and Computational Neuroscience;365
15.9;12.6.1 Consciousness and Intention Reading;366
15.10;12.6.2 The Core of Computational Neuroscience Is to Compute and Interpret the States of Nervous System;366
15.11;12.7 Soft Computing Method;366
15.12;12.8 Summary;367
15.13;References;368
16;Index;369