Cuevas / Zaldívar / Perez-Cisneros Applications of Evolutionary Computation in Image Processing and Pattern Recognition

1. Auflage 2016
ISBN: 978-3-319-26462-2
Verlag: Springer Nature Switzerland
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band 100, 274 Seiten

Reihe: Intelligent Systems Reference Library

This book presents the use of efficient Evolutionary Computation (EC) algorithms for solving diverse real-world image processing and pattern recognition problems. It provides an overview of the different aspects of evolutionary methods in order to enable the reader in reaching a global understanding of the field and, in conducting studies on specific evolutionary techniques that are related to applications in image processing and pattern recognition. It explains the basic ideas of the proposed applications in a way that can also be understood by readers outside of the field. Image processing and pattern recognition practitioners who are not evolutionary computation researchers will appreciate the discussed techniques beyond simple theoretical tools since they have been adapted to solve significant problems that commonly arise on such areas. On the other hand, members of the evolutionary computation community can learn the way in which image processing and pattern recognition problems can be translated into an optimization task. The book has been structured so that each chapter can be read independently from the others. It can serve as reference book for students and researchers with basic knowledge in image processing and EC methods.

Autoren/Hrsg.

Cuevas, Erik

Zaldívar, Daniel

Perez-Cisneros, Marco

Inhaltsverzeichnis

1;Foreword;6
2;Preface;8
3;Contents;12
4;1 Introduction;17
4.1;Abstract;17
4.2;1.1 Definition of an Optimization Problem;17
4.3;1.2 Classical Optimization;18
4.4;1.3 Evolutionary Computation Methods;21
4.4.1;1.3.1 Structure of an Evolutionary Computation Algorithm;22
4.5;References;24
5;2 Image Segmentation Based on Differential Evolution Optimization;25
5.1;Abstract;25
5.2;2.1 Introduction;25
5.3;2.2 Gaussian Approximation;26
5.4;2.3 Differential Evolution Algorithms;27
5.5;2.4 Determination of Thresholding Values;29
5.6;2.5 Experimental Results;30
5.7;2.6 Conclusions;36
5.8;References;37
6;3 Motion Estimation Based on Artificial Bee Colony (ABC);39
6.1;Abstract;39
6.2;3.1 Introduction;39
6.3;3.2 Artificial Bee Colony (ABC) Algorithm;43
6.3.1;3.2.1 Biological Bee Profile;43
6.3.2;3.2.2 Description of the ABC Algorithm;43
6.3.3;3.2.3 Initializing the Population;44
6.3.4;3.2.4 Send Employed Bees;44
6.3.5;3.2.5 Select the Food Sources by the Onlooker Bees;45
6.3.6;3.2.6 Determine the Scout Bees;45
6.4;3.3 Fitness Approximation Method;45
6.4.1;3.3.1 Updating the Individual Database;46
6.4.2;3.3.2 Fitness Calculation Strategy;46
6.4.3;3.3.3 Presented ABC Optimization Method;48
6.5;3.4 Motion Estimation and Block Matching;50
6.6;3.5 BM Algorithm Based on ABC with the Estimation Strategy;51
6.6.1;3.5.1 Initial Population;52
6.6.2;3.5.2 The ABC-BM Algorithm;54
6.7;3.6 Experimental Results;55
6.7.1;3.6.1 ABC-BM Results;55
6.7.1.1;3.6.1.1 Distortion Performance;57
6.7.1.2;3.6.1.2 Search Efficiency;59
6.7.2;3.6.2 Results on H.264;59
6.7.3;3.6.3 Experiments with High Definition Sequences;62
6.8;3.7 Conclusions;64
6.9;References;65
7;4 Ellipse Detection on Images Inspired by the Collective Animal Behavior;68
7.1;Abstract;68
7.2;4.1 Introduction;68
7.3;4.2 Collective Animal Behavior Algorithm (CAB);71
7.3.1;4.2.1 Description of the CAB Algorithm;71
7.3.1.1;4.2.1.1 Initializing the Population;71
7.3.1.2;4.2.1.2 Keep the Position of the Best Individuals;72
7.3.1.3;4.2.1.3 Move from or to Nearby Neighbors;72
7.3.1.4;4.2.1.4 Move Randomly;73
7.3.1.5;4.2.1.5 Compete for the Space Within of a Determined Distance (Update the Memory);73
7.3.1.6;4.2.1.6 Computational Procedure;74
7.4;4.3 Ellipse Detection Using CAB;75
7.4.1;4.3.1 Data Preprocessing;75
7.4.2;4.3.2 Individual Representation;75
7.4.3;4.3.3 Objective Function;77
7.4.4;4.3.4 Implementation of CAB for Ellipse Detection;79
7.5;4.4 The Multiple Ellipse Detection Procedure;80
7.6;4.5 Experimental Results;81
7.6.1;4.5.1 Ellipse Localization;82
7.6.1.1;4.5.1.1 Synthetic Images;82
7.6.1.2;4.5.1.2 Natural Images;82
7.6.2;4.5.2 Shape Discrimination Tests;83
7.6.3;4.5.3 Ellipse Approximation: Occluded Ellipse and Ellipsoidal Detection;84
7.6.4;4.5.4 Performance Comparison;84
7.7;4.6 Conclusions;90
7.8;References;91
8;5 Template Matching by Using the States of Matter Algorithm;93
8.1;Abstract;93
8.2;5.1 Introduction;93
8.3;5.2 States of Matter;95
8.4;5.3 States of Matter Search (SMS);97
8.4.1;5.3.1 Definition of Operators;97
8.4.1.1;5.3.1.1 Direction Vector;97
8.4.1.2;5.3.1.2 Collision;98
8.4.1.3;5.3.1.3 Random Positions;99
8.4.1.4;5.3.1.4 Best Element Updating;99
8.4.2;5.3.2 SMS Algorithm;100
8.4.2.1;5.3.2.1 General Procedure;100
8.4.2.2;5.3.2.2 The Complete Algorithm;100
8.4.2.3;5.3.2.3 Initialization;102
8.4.2.4;5.3.2.4 Gas State;103
8.4.2.5;5.3.2.5 Liquid State;103
8.4.2.6;5.3.2.6 Solid State;104
8.5;5.4 Fitness Approximation Method;104
8.5.1;5.4.1 Updating Individual Database;105
8.5.2;5.4.2 Fitness Calculation Strategy;105
8.5.3;5.4.3 Presented Optimization SMS Method;108
8.6;5.5 Template Matching Process;109
8.7;5.6 TM Algorithm Based on SMS with the Estimation Strategy;110
8.7.1;5.6.1 The SMS-TM Algorithm;111
8.8;5.7 Experimental Results;113
8.9;5.8 Conclusions;117
8.10;References;118
9;6 Estimation of Multiple View Relations Considering Evolutionary Approaches;120
9.1;Abstract;120
9.2;6.1 Introduction;120
9.3;6.2 View Relations from Point Correspondences;123
9.4;6.3 Random Sampling Consensus (RANSAC) Algorithm;126
9.5;6.4 Clonal Selection Algorithm (CSA);127
9.5.1;6.4.1 Definitions;127
9.5.2;6.4.2 CSA Operators;128
9.5.3;6.4.3 Clonal Proliferation Operator (T_{\rm{P}}^{\rm{C}} );128
9.5.4;6.4.4 Affinity Maturation Operator (T_{M}^{\rm{A}} );129
9.5.5;6.4.5 Clonal Selection Operator (T_{\rm{S}}^{\rm{C}} );130
9.6;6.5 Method for Geometric Estimation Using CSA;130
9.6.1;6.5.1 Computational Procedure;132
9.7;6.6 Experimental Results;135
9.7.1;6.6.1 Fundamental Matrix Estimation with Synthetic Data;136
9.7.2;6.6.2 Fundamental Matrix Estimation with Real Images;139
9.7.3;6.6.3 Homography Estimation with Synthetic Data;142
9.7.4;6.6.4 Homography Estimation with Real Images;145
9.8;6.7 Conclusions;147
9.9;References;149
10;7 Circle Detection on Images Based on an Evolutionary Algorithm that Reduces the Number of Function Evaluations;152
10.1;Abstract;152
10.2;7.1 Introduction;153
10.3;7.2 The Adaptive Population with Reduced Evaluations (APRE) Algorithm;155
10.3.1;7.2.1 Initialization;156
10.3.2;7.2.2 Selecting the Population to Be Evolved;157
10.3.3;7.2.3 The Number of Elements N_{e}^{k} to Be Selected;157
10.3.4;7.2.4 Selection Strategy for Building {{\bf P}}^{k};159
10.3.5;7.2.5 Exploration Operation;160
10.3.6;7.2.6 DE Mutation Operator;161
10.3.7;7.2.7 Trigonometric Mutation Operator;161
10.3.7.1;7.2.7.1 Computational Procedure;161
10.3.8;7.2.8 Fitness Estimation Strategy;163
10.3.9;7.2.9 Memory Updating;166
10.3.10;7.2.10 Exploitation Operation;166
10.3.11;7.2.11 Computational Procedure;168
10.4;7.3 Implementation of APRE-Based Circle Detector;169
10.4.1;7.3.1 Individual Representation;169
10.4.2;7.3.2 Objective Function;171
10.4.3;7.3.3 The Multiple Circle Detection Procedure;171
10.5;7.4 Results on Multi-circle Detection;172
10.6;7.5 Conclusions;178
10.7;References;179
11;8 Otsu and Kapur Segmentation Based on Harmony Search Optimization;181
11.1;Abstract;181
11.2;8.1 Introduction;181
11.3;8.2 Harmony Search Algorithm;183
11.3.1;8.2.1 The Harmony Search Algorithm;183
11.3.1.1;8.2.1.1 Initializing the Problem and the Algorithm Parameters;184
11.3.1.2;8.2.1.2 Harmony Memory Initialization;184
11.3.1.3;8.2.1.3 Improvisation of New Harmony Vectors;185
11.3.1.4;8.2.1.4 Updating the Harmony Memory;185
11.3.1.5;8.2.1.5 Computational Procedure;186
11.4;8.3 Image Multilevel Thresholding (MT);186
11.4.1;8.3.1 Between---Class Variance (Otsu's Method);187
11.4.2;8.3.2 Entropy Criterion Method (Kapur's Method);189
11.5;8.4 Multilevel Thresholding Using Harmony Search Algorithm (HSMA);191
11.5.1;8.4.1 Harmony Representation;191
11.5.2;8.4.2 HMA Implementation;191
11.5.3;8.4.3 Parameter Setting;193
11.6;8.5 Experimental Results;194
11.6.1;8.5.1 Otsu's Results;196
11.6.2;8.5.2 Kapur's Results;196
11.6.3;8.5.3 Comparisons;202
11.6.3.1;8.5.3.1 Comparison Between Otsu and Kapur HSMA;205
11.6.3.2;8.5.3.2 Comparison Among HSMA and Other MT Approaches;207
11.7;8.6 Conclusions;210
11.8;References;213
12;9 Leukocyte Detection by Using Electromagnetism-like Optimization;215
12.1;Abstract;215
12.2;9.1 Introduction;216
12.3;9.2 Electromagnetism-like Optimization Algorithm (EMO);218
12.4;9.3 Circle Detection Using EMO;221
12.4.1;9.3.1 Data Preprocessing;221
12.4.2;9.3.2 Particle Representation;221
12.4.3;9.3.3 Objective Function;222
12.4.4;9.3.4 EMO Implementation;223
12.5;9.4 The White Blood Cell Detector;225
12.5.1;9.4.1 Image Preprocessing;226
12.5.2;9.4.2 The Modified EMO-Based Circle Detector;227
12.5.3;9.4.3 Numerical Example;229
12.6;9.5 Experimental Results;231
12.7;9.6 Comparisons to Other Methods;232
12.7.1;9.6.1 Detection Comparison;232
12.7.2;9.6.2 Robustness Comparison;233
12.7.3;9.6.3 Stability Comparison;234
12.8;9.7 Conclusions;237
12.9;References;238
13;10 Automatic Segmentation by Using an Algorithm Based on the Behavior of Locust Swarms;240
13.1;Abstract;240
13.2;10.1 Introduction;240
13.3;10.2 Biological Fundamentals and Mathematical Models;243
13.3.1;10.2.1 Solitary Phase;244
13.3.2;10.2.2 Social Phase;246
13.4;10.3 The Locust Search (LS) Algorithm;246
13.4.1;10.3.1 Solitary Operation (A);248
13.4.2;10.3.2 Social Operation (B);250
13.4.3;10.3.3 Complete LS Algorithm;252
13.4.4;10.3.4 Discussion About the LS Algorithm;253
13.5;10.4 Numerical Experiments over Benchmark Functions;254
13.5.1;10.4.1 Uni-modal Test Functions;255
13.5.2;10.4.2 Multimodal Test Functions;257
13.6;10.5 Gaussian Mixture Modelling;258
13.7;10.6 Segmentation Algorithm Based on LS;259
13.7.1;10.6.1 New Objective Function Jnew;260
13.7.2;10.6.2 Complete Segmentation Algorithm;262
13.8;10.7 Segmentation Results;264
13.8.1;10.7.1 Performance of LS Algorithm in Image Segmentation;264
13.8.1.1;10.7.1.1 First Image;264
13.8.1.2;10.7.1.2 Second Image;265
13.8.2;10.7.2 Histogram Approximation Comparisons;267
13.8.2.1;10.7.2.1 Convergence;269
13.8.2.2;10.7.2.2 Accuracy;270
13.8.2.3;10.7.2.3 Computational Cost;272
13.8.3;10.7.3 Performance Evaluation of the Segmentation Results;274
13.8.3.1;10.7.3.1 Evaluation Criteria;275
13.8.3.2;10.7.3.2 Experimental Protocol;275
13.9;10.8 Conclusions;277
13.10;References;278
14;Appendix A: RANSAC Algorithm;281
15;Appendix B: List of Benchmark Functions;283