视频图像处理

本书涉及图像处理、模式识别、计算机视觉、机器学习、人工智能等多个交叉领域，介绍了视频图像处理的相关技术和方法，结合作者近年来有关视频图像处理技术的研究与应用实践，详细介绍了从模型、方法到应用系统实现的理论和技术。本书共分十章，主要内容包括：图像复原、目标检测、目标跟踪、图像显著性检测、变化检测、目标识别、场景分类、稀疏表示、深度学习等。本书内容新颖、理论联系实际，可作为计算机科学与技术、信息与通信工程、电子信息工程、软件工程、模式识别与智能系统等相关专业的研究生和高年级本科生、科研人员、工程技术人员参考书。

Preface i
Chapter 1 Introduction 1
1.1 Video Image Restoration 1
1.2 Target Detection in Video Images 1
1.3 Target Tracking in Surveillance Videos 2
1.4 Video Image Saliency Detection 3
1.5 Underwater Video Image Processing 4
1.6 Vision Image Processing for River Surface Velocimetry 5
1.7 Remote Sensing Imagery Processing and Analysis 6
References 7
Chapter 2 Video Image Restoration 10
2.1 Automatic Image De-weathering Using Physical Model and Maximum Entropy 10
2.1.1 Introduction 10
2.1.2 Automatic Contrast Restoration 11
2.1.3 Experimental Results 15
2.1.4 Conclusion 16
2.2 Image Dehazing Using Degradation Model and Group-based Sparse Representation 17
2.2.1 Introduction 17
2.2.2 Preliminaries 18
2.2.3 Presented Approach 18
2.2.4 Experimental Results 20
2.2.5 Conclusion 23
References 23
Chapter 3 Target Detection in Video Images 26
3.1 A Sparse Representation-based Method for Infrared Dim Target Detection under Sea-sky Background 26
3.1.1 Introduction 26
3.1.2 Related Work 27
3.1.3 Proposed Approach 28
3.1.4 Experimental Results 33
3.1.5 Conclusion 39
3.2 Spatiotemporal Saliency Model for Small Moving Object Detection in Infrared Videos 39
3.2.1 Introduction 39
3.2.2 Spatiotemporal Saliency Model for Infrared Videos 40
3.2.3 Experimental Results 45
3.2.4 Conclusion 48
References 49
Chapter 4 Target Tracking in Surveillance Videos 52
4.1Multi-feature Local Sparse Representation for Infrared Pedestrian Tracking 52
4.1.1 Introduction 52
4.1.2 Related Work 53
4.1.3 Proposed Method 54
4.1.4 Experimental Results 61
4.1.5 Conclusion 66
4.2 Spatiotemporal Difference-of-Gaussians Filters for Robust Infrared Small Target Tracking in Various Complex Scenes 67
4.2.1 Introduction 67
4.2.2 Spatiotemporal Gabor Filters 68
4.2.3 Combined Difference-of-Gaussians Filter for Small IR Target Detection 69
4.2.4 Spatiotemporal Difference-of-Gaussians Filters for Small IR Target Tracking 71
4.2.5 Experimental Results 76
4.2.6 Conclusions 86
References 86
Chapter 5 Video Image Saliency Detection 91
5.1 A Novel Visual Saliency Detection Method for Infrared Video Sequences 91
5.1.1 Introduction 91
5.1.2 Proposed Method 93
5.1.3 Experimental Results 102
5.1.4 Conclusion 109
5.2 Visual Saliency Detection Based on In-depth Analysis of Sparse Representation 109
5.2.1 Introduction 109
5.2.2 Related Work 111
5.2.3 Proposed Method 112
5.2.4 Experimental Results 118
5.2.5 Conclusion 125
References 126
Chapter 6 Underwater Video Image Processing 130
6.1 An Integrative Framework for Effective Restoration of Underwater Images 130
6.1.1 Introduction 130
6.1.2 Related Work 131
6.1.3 Proposed Method 133
6.1.4 Experimental Results 138
6.1.5 Conclusion 148
6.2 Combination of Interacting Multiple Models with the Particle Filter for Three-dimensional Target Tracking in Underwater Wireless Sensor Networks 148
6.2.1 Introduction 148
6.2.2 Problem Formulation 149
6.2.3 IMMPF Underwater Target Tracking Algorithm 153
6.2.4 Simulation and Results 155
6.2.5 Conclusion 158
References 158
Chapter 7 Video Image Processing for River Surface Velocimetry 162
7.1 Balloon-borne Spectrum-polarization Imaging for River Surface Velocimetry under Extreme Conditions 162
7.1.1 Introduction 162
7.1.2 Near-infrared Spectrum-polarization Imaging for Flow Tracers Detection 163
7.1.3 Balloon-borne Low-altitude Telemetry System with A Self-stabilized Servo Platform 166
7.1.4 Experimental Results 169
7.1.5 Conclusion 171
7.2 An Information Acquisition Method Based on Dragonfly Vision Mechanism for Observed Target Displacement Measurement 171
7.2.1 Introduction 171
7.2.2 Related Work 172
7.2.3 Optical Characteristics of River Surface 173
7.2.4 Our Approach 174
7.2.5 Experimental Results 176
7.2.6 Conclusion 179
References 179
Chapter 8 Remote Sensing Imagery Processing and Analysis 182
8.1 Multi-class Remote Sensing Objects Recognition Based on Discriminative Sparse Representation 182
8.1.1 Introduction 182
8.1.2 Related Work 184
8.1.3 Proposed Method 185
8.1.4 Experimental Results 193
8.1.5 Conclusion 203
8.2 Integration of Heterogeneous Features for Remote Sensing Scene Classification 203
8.2.1 Introduction 203
8.2.2 Proposed Method 206
8.2.3 Experimental Results 213
8.2.4 Conclusion 221
References 221