Base article 1
Chapter1 Introduction 1
1.1 Overview 1
1.1.1 Concept about the Visual Perception 1
1.1.2 The Development of Visual Perception Technology 2
1.1.3 Classification of Visual Perception System 4
1.2 A Visual Perception Hardware-base 6
1.2.1 Image Sensing 6
1.2.2 Image Acquisition 22
1.2.3 PC Hardware Requirements for VPS 27
Exercises 31
Chapter2 Foundations of Image Processing 32
2.1 Basic Processing Methods for Gray Image 32
2.1.1 Spatial Domain Enhancement Algorithm 32
2.1.2 Frequency Domain Enhancement Algorithm 43
2.2 Edge Detection of Gray Image 50
2.2.1 Threshold Edge Detection 51
2.2.2 Gradient-based Edge Detection 53
2.2.3 Laplacian Operator 56
2.2.4 Canny Edge Operator 58
2.2.5 Mathematical Morphological Method 63
2.2.6 Brief Description of Other Algorithms 66
2.3 Binarization Processing and Segmentation of Image 67
2.3.1 General Description 67
2.3.2 Histogram-based Valley-point Threshold Image Binarization 68
2.3.3 OTSU Algorithm 68
2.3.4 Minimum Error Method of Image Segmentation 70
2.4 Color Image Enhancement 71
2.4.1 Color Space and Its Transformation 71
2.4.2 Histogram Equalization of Color Levels in Color Image 74
2.5 Color Image Edge Detection 76
2.5.1 Color Image Edge Detection Based on Gradient Extreme Value 76
2.5.2 Practical Method for Color Image Edge Detection 79
Exercises 80
Chapter3 Mathematical Model of the Camera 83
3.1 Geometric Transformations of Image Space 83
3.1.1 Homogeneous Coordinates 84
3.1.2 Orthogonal Transformation and Rigid Body Transformation 84
3.1.3 Similarity Transformation and Affine Transformation 85
3.1.4 Perspective Transformation 86
3.2 Image Coordinate System and Its Transformation 88
3.2.1 Image Coordinate System 88
3.2.2 Image Coordinate Transformation 90
3.3 Common Method of Calibration Camera Parameters 94
3.3.1 Step Calibration Method 95
3.3.2 Calibration Algorithm Based on More than One Free Plane 97
3.3.3 Non-linear Distortion Parameter Calibration Method 99
Exercises 101
Chapter4 Visual Perception Identification Algorithms 104
4.1 Image Feature Extraction and Identification Algorithm 105
4.1.1 Decision Theory Approach 105
4.1.2 Statistical Classification Method 112
4.1.3 Feature Classification Discretion Similarity about the Image Recognition Process 114
4.2 Principal Component Analysis 116
4.2.1 Principal Component Analysis Principle 116
4.2.2 Kernel Principal Component Analysis 118
4.2.3 PCA-based Image Recognition 122
4.3 Support Vector Machines 125
4.3.1 Main Contents of Statistical Learning Theory 126
4.3.2 Classification-Support Vector Machine 130
4.3.3 Solution to the Nonlinear Regression Problem 136
4.3.4 Algorithm of Support Vector Machine 139
4.3.5 Image Characteristics Identification Based on SVM 145
4.4 Moment Invariants and Normalized Moments of Inertia 146
4.4.1 Moment Theory 147
4.4.2 Normalized Moment of Inertia 149
4.5 Template Matching and Similarity 157
4.5.1 Spatial Domain Description of Template Matching 157
4.5.2 Frequency Domain Description of Template Matching 162
4.6 Object Recognition Based on Color Feature 171
4.6.1 Image Colorimetric Processing 171
4.6.2 Construction of Color-Pool 173
4.6.3 Object Recognition Based on Color 175
4.7 Image Fuzzy Recognition Method 176
4.7.1 Fuzzy Content Feature and Fuzzy Similarity Degree 176
4.7.2 Extraction of Fuzzy Structure 178
4.7.3 Fuzzy Synthesis Decision-making of Image Matching 183
Exercises 188
Chapter5 Detection Principle of Visual Perception 191
5.1 Single View Geometry and Detection Principle of Monocular Visual Perception 191
5.1.1 Single Vision Coordinate System 191
5.1.2 Basic Algorithm for Single Vision Detection 192
5.1.3 Engineering Technology Based on Single View Geometry 192
5.2 Detection Principle of Binocular Visual Perception 195
5.2.1 Two-view Geometry and Detection of Binocular Perception 196
5.2.2 Epipolar Geometry Principle 200
5.2.3 Determination Method of Spatial Coordinates 204
5.2.4 Camera Calibration in Binocular Visual Perception System 207
5.3 Theoretical Basis for Multiple Visual Perception Detection 217
5.3.1 Tensor Geometry Principle 218
5.3.2 Geometric Properties of Three Visual Tensor 221
5.3.3 Operation of Three-visual Tensor 226
5.3.4 Constraint Matching Feature Points of Three-visual Tensor 228
5.3.5 Three-visual Tensor Restrict the Three Visual Restraint Feature Line’s Matching 231
Exercises 236
Application article 238
Chapter6 Practical Technology of Intelligent Visual Perception 238
6.1 Automatic Monitoring System and Method of Load Limitation of The Bridge 238
6.1.1 The Basic Composition of The System 239
6.1.2 System Algorithm 241
6.2 Intelligent Identification System for Billet Number 244
6.2.1 System Control Program 245
6.2.2 Recognition Algorithm 245
6.3 Verification of Banknotes-Sorting Based on Image Information 251
6.3.1 Preprocessing of the Banknotes Image 252
6.3.2 Distinction Between Old and New Banknotes 252
6.3.3 Distinction of the Denomination and Direction of the Banknotes 253
6.3.4 Banknotes Fineness Detection 255
6.4 Intelligent Collision Avoidance Technology of Vehicle 258
6.4.1 Basic Hardware Configuration 258
6.4.2 Road Obstacle Recognition Algorithm 259
6.4.3 Smart Algorithm of Anti-collision to Pedestrians 262
6.5 Intelligent Visual Perception Control of Traffic Lights 267
6.5.1 Overview 267
6.5.2 The Core Algorithm of Intelligent Visual Perception Control of Traffic Lights 267
Exercises 272
Appendix 275
Ⅰ Least Square and Common Algorithms in Visual Perception Detection 275
Ⅰ.1 Basic Idea of the Algorithm 275
Ⅰ.2 Common Least Square Algorithms in Visual Perception Detection 276
Ⅰ.2.1 Least Square of Linear System of Equations 276
Ⅰ.2.2 Least Square Solution of Nonlinear Homogeneous System of Equations 278
Ⅱ Theory and Method of BAYES Decision 281
Ⅱ.1 Introduction 281
Ⅱ.2 BAYES Classification Decision Mode 281
Ⅱ.2.1 BAYES Classification of Minimum Error Rate 281
Ⅱ.2.2 BAYES Classification Decision of Minimum Risk 283
Ⅲ Statistical Learning and VC-dimension Theorem 285
Ⅲ.1 Bounding Theory and VC-dimension Principle 285
Ⅲ.2 Generalized Capability Bounding 286
Ⅲ.3 Structural Risk Minimization Principle of Induction 287
Ⅳ Optimality Conditions on Constrained Nonlinear Programming Problem 288
Ⅳ.1 Kuhn-Tucker Condition 288
Ⅳ.1.1 Gordon Lemma 288
Ⅳ.1.2 Fritz John Theorem 288
Ⅳ.1.3 Proof of the Kuhn-Tucker Condition 289
Ⅳ.2 Karush-Kuhn-Tucker Condition 291
Subject Index 293
References 300