Adaptive image processing is one of the most important techniques in visual information processing, especially in early vision such as image restoration, filtering, enhancement, and segmentation. While existing books present some important aspects of the issue, there is not a single book that treats this problem from a viewpoint that is directly linked to human perception - until now. This reference treats adaptive image processing from a computational intelligence viewpoint, systematically and successfully, from theory to applications, using the synergies of neural networks, fuzzy logic, and evolutionary computation. Based on the fundamentals of human perception, this book gives a detailed account of computational intelligence methods and algorithms for adaptive image processing in regularization, edge detection, and early vision.Adaptive Image Processing: A Computational Intelligence Perspective consists of 8 chapters:Chapter 1 - Provides material of an introductory nature to describe the basic concepts and current state-of-the-art in the field of computational intelligence for image restoration and edge detectionChapter 2 - Gives a mathematical description of the restoration problem from the neural network perspective, and describes current algorithms based on this methodChapter 3 - Extends the algorithm presented in chapter 2 to implement adaptive constraint restoration methods for both spatially invariant and spatially variant degradationsChapter 4 - Utilizes a perceptually motivated image error measure to introduce novel restoration algorithmsChapter 5 - Examines how model-based neural networks can be used to solve image restoration problemsChapter 6 - Probes image restoration algorithms, making use of the principles of evolutionary computationChapter 7 - Explores the difficult concept of image restoration when insufficient knowledge of the degrading function is availableChapter 8 - Studies the subject of edge detection and characterization using model-based neural networksThe first to treat adaptive image processing from a computational intelligence perspective, this work provides an excellent reference in R&D practice to researchers and IT technologists, is most suitable for teaching image processing and applied neural network courses, and will be of equal value for technical managers and executives in industries where intelligent visual information processing is required.
Author(s): Ling Guan
Edition: 1
Year: 2001
Language: English
Pages: 288
Adaptive Image Processing, A Computational Intelligence Perspective......Page 1
Preface......Page 5
Acknowledgments......Page 6
Table of Contents......Page 7
Table of Contents......Page 0
1.1 The Importance of Vision......Page 11
1.2 Adaptive Image Processing......Page 12
Edges......Page 13
Textures......Page 14
1.4 Difficulties in Adaptive Image Processing System Design......Page 15
Optimization......Page 17
1.5 Computational Intelligence Techniques......Page 18
Neural Networks......Page 20
Fuzzy Logic......Page 21
Evolutionary Computation......Page 22
1.6 Scope of the Book......Page 23
1.6.1 Image Restoration......Page 24
1.6.2 Edge Characterization and Detection......Page 28
1.7.1 Application of Neural Networks for Image Restoration......Page 29
1.7.3 Application of Fuzzy Set Theory to Adaptive Regularization......Page 30
1.8 Overview of This Book......Page 31
2.1 Image Distortions......Page 33
2.2.1 Degradation Measure......Page 36
2.2.2 Neural Network Restoration......Page 37
2.3 Neural Network Restoration Algorithms in the Literature......Page 39
2.4 An Improved Algorithm......Page 42
2.5 Analysis......Page 44
2.6 Implementation Considerations......Page 46
2.7.2 Efficiency......Page 47
2.7.3 An Application Example......Page 48
2.8 Summary......Page 49
3.1 Introduction......Page 51
3.2 Dealing with Spatially Variant Distortion......Page 53
3.3.1 Motivation......Page 56
3.3.2 The Gradient-Based Method......Page 58
3.3.3 Local Statistics Analysis......Page 67
3.4 Correcting Spatially Variant Distortion Using Adaptive Constraints......Page 76
3.5 Semi-Blind Restoration Using Adaptive Constraints......Page 78
3.6 Implementation Considerations......Page 82
3.7.1 Efficiency......Page 83
3.7.2 An Application Example......Page 84
3.8.1 Problem Formulation......Page 86
3.8.2 Problem Solution......Page 87
3.8.3 Conditions for KKT Theory to Hold......Page 90
3.9 Summary......Page 92
4.1 Introduction......Page 94
4.2 Motivation......Page 95
4.3 A LVMSE-Based Cost Function......Page 96
4.3.1 The Extended Algorithm for the LVMSE-Modified Cost Function......Page 97
4.3.2 Analysis......Page 101
4.4 A Log LVMSE-Based Cost Function......Page 105
4.4.1 The Extended Algorithm for the Log LVR-Modified Cost Function......Page 106
4.4.2 Analysis......Page 108
4.6.1 Color Image Restoration......Page 111
4.6.4 Robustness Evaluation......Page 114
4.7 Summary......Page 118
5.1 Model-Based Neural Network......Page 119
5.1.1 Weight-Parameterized Model-Based Neuron......Page 120
5.2 Hierarchical Neural Network Architecture......Page 121
5.4 HMBNN for Adaptive Image Processing......Page 123
5.6 Adaptive Regularization: An Alternative Formulation......Page 124
5.6.1 Correspondence with the General HMBNN Architecture......Page 126
5.7 Regional Training Set Definition......Page 130
5.8 Determination of the Image Partition......Page 132
5.9 The Edge-Texture Characterization......Page 134
5.10 The ETC Fuzzy HMBNN for Adaptive Regularization......Page 138
5.11 Theory of Fuzzy Sets......Page 139
5.12 Edge-Texture Fuzzy Model Based on ETC Measure......Page 141
5.13 Architecture of the Fuzzy HMBNN......Page 143
5.14 Estimation of the Desired Network Output......Page 145
5.15 Fuzzy Prediction of Desired Gray Level Value......Page 146
5.15.1 Definition of the Fuzzy Estimator Membership Function......Page 147
5.15.2 Fuzzy Inference Procedure for Predicted Gray Level Value......Page 148
5.15.3 Defuzzification of the Fuzzy Set G......Page 149
5.15.4 Regularization Parameter Update......Page 150
5.15.5 Update of the Estimator Fuzzy Set Width Parameters......Page 152
5.16 Experimental Results......Page 153
5.17 Summary......Page 162
6.1 Introduction......Page 163
6.2.1 Genetic Algorithm......Page 164
6.2.2 Evolutionary Strategy......Page 165
6.2.3 Evolutionary Programming......Page 166
6.3 The ETC-pdf Image Model......Page 168
6.4 Adaptive Regularization Using Evolutionary Programming......Page 172
6.4.1 Competition under Approximate Fitness Criterion......Page 176
6.4.2 Choice of Optimal Regularization Strategy......Page 177
6.5 Experimental Results......Page 180
6.6.1 Hierarchical Cluster Model......Page 187
6.7 Summary......Page 188
7.1 Introduction......Page 190
7.1.2 Blur Identification by Recursive Soft Decision......Page 192
7.2.1 Formulation of Blind Image Deconvolution as an Evolutionary Strategy......Page 193
7.2.2 Knowledge-Based Reinforced Mutation......Page 200
7.2.3 Perception-Based Image Restoration......Page 204
7.2.4 Recombination Based on Niche-Space Residency......Page 206
7.2.5 Performance Evaluation and Selection......Page 208
7.3.1 Recursive Subspace Optimization......Page 210
7.3.2 Hierarchical Neural Network for Image Restoration......Page 211
7.3.3 Soft Parametric Blur Estimator......Page 217
7.3.4 Blur Identification by Conjugate Gradient Optimization......Page 218
7.3.5 Blur Compensation......Page 221
7.4 Simulation Examples......Page 223
7.4.1 Identification of 2D Gaussian Blur......Page 224
7.4.2 Identification of 2D Gaussian Blur from Degraded Image with Additive Noise......Page 226
7.4.3 Identification of 2D Uniform Blur by CRL......Page 227
7.5 Conclusions......Page 232
8.1 Introduction......Page 235
8.2.1 Input-Parameterized Model-Based Neuron......Page 236
8.2.3 Edge Characterization and Detection......Page 238
8.3 Network Architecture......Page 240
8.3.3 Neuron in Sub-Network......Page 242
8.3.4 Dynamic Tracking Neuron......Page 243
8.3.5 Binary Edge Configuration......Page 244
8.3.6 Correspondence with the General HMBNN Architecture......Page 245
8.4.1 Determination of for Sub-Network......Page 246
8.5 Recognition Stage......Page 247
8.5.2 Identification of Secondary Edge Points......Page 248
8.6 Experimental Results......Page 249
8.7 Summary......Page 253
References......Page 259
