PartⅠ：Pattern Recognition and Machine Intelligence 3
1 A Review of Applications of Evolutionary Algorithms in Pattern Recognition 3
1.1 Introduction 3
1.2 Basic Notions of Evolutionary Algorithms 4
1.3 A Review of EAs in Pattern Recognition 15
1.4 Future Research Directions 21
1.5 Conclusions 23
References 24
2 Pattern Discovery and Recognition in Sequences 29
2.1 Introduction 29
2.2 Sequence Patterns and Pattern Discovery-A Brief Review 31
2.3 Our Pattern Discovery Framework 42
2.4 Conclusion 58
References 58
3 A Hybrid Method of Tone Assessment for Mandarin CALL System 61
3.1 Introduction 61
3.2 Related Work 65
3.3 Proposed Approach 67
3.4 Experimental Procedure and Analysis 73
3.5 Conclusions 77
References 78
4 Fusion with Infrared Images for an Improved Performance and Perception 81
4.1 Introduction 81
4.2 The Principle of Infrared Imaging 82
4.3 Fusion with Infrared Images 83
4.4 Applications 102
4.5 Summary 104
References 105
5 Feature Selection and Ranking for Pattern Classification in Wireless Sensor Networks 109
5.1 Introduction 109
5.2 General Approach 112
5.3 Sensor Ranking 115
5.4 Experiments 118
5.5 Summary,Discussion and Conclusions 134
References 136
6 Principles and Applications of RIDED-2D-A Robust Edge Detection Method in Range Images 139
6.1 Introduction 140
6.2 Definitions and Analysis 143
6.3 Principles of Instantaneous Denoising and Edge Detection 146
6.4 Experiments and Evaluations 151
6.5 Discussions and Applications 159
6.6 Conclusions and Prospects 163
References 165
PartⅡ：Computer Vision and Image Processing 171
7 Lens Shading Correction for Dirt Detection 171
7.1 Introduction 171
7.2 Background 174
7.3 Our Proposed Method 180
7.4 Experimental Results 184
7.5 Conclusions 194
References 194
8 Using Prototype-Based Classification for Automatic Knowledge Acquisition 197
8.1 Introduction 197
8.2 Prototype-Based Classification 198
8.3 Methodology 204
8.4 Application 206
8.5 Results 208
8.6 Conclusion 211
References 211
9 Tracking Deformable Objects with Evolving Templates for Real-Time Machine Vision 213
9.1 Introduction 213
9.2 Problem Formulation 216
9.3 Search Framework for Computing Template Position 218
9.4 Updating Framework for Computing Template Changes 224
9.5 Multiple Object Tracking and Intensity Information 227
9.6 Experiments and Results 229
9.7 Conclusions and Future Work 233
References 234
10 Human Extremity Detection for Action Recognition 237
10.1 Introduction 237
10.2 Relevant Works 239
10.3 Extremities as Points on a Contour 241
10.4 Extremities as Image Patches 250
10.5 Experimental Results 254
10.6 Conclusion 258
References 259
11 Ensemble Learning for Object Recognition and Tracking 261
11.1 Introduction 261
11.2 Random Subspace Method 264
11.3 Boosting Method 269
References 276
12 Depth Image Based Rendering 279
12.1 Introduction 279
12.2 Depth Image Based Rendering 283
12.3 Disocclusions 289
12.4 Other Challenges 304
12.5 Conclusion 307
References 308
PartⅢ：Face Recognition and Forensics 313
13 Gender and Race Identification by Man and Machine 313
13.1 Introduction 313
13.2 Background 314
13.3 Silhouetted Profile Faces 315
13.4 Frontal Faces 319
13.5 Fusing the Frontal View and Silhouetted Profile View Classifiers 323
13.6 Human Experiments 324
13.7 Observations and Discussion 329
13.8 Concluding Remarks 330
References 331
14 Common Vector Based Face Recognition Algorithm 335
14.1 Introduction 335
14.2 Algorithm Description 339
14.3 Two Methods Based on Common Vector 347
14.4 Experiments and Results 350
14.5 Conclusion and Future Research 358
References 358
15 A Look at Eye Detection for Unconstrained Environments 361
15.1 Introduction 361
15.2 Related Work 363
15.3 Machine Learning Approach 364
15.4 Correlation Filter Approach 368
15.5 Experiments 370
15.6 Conclusions 385
References 386
16 Kernel Methods for Facial Image Preprocessing 389
16.1 Introduction 389
16.2 Kernel PCA 391
16.3 Kernel Methods for Nonlinear Image Preprocessing 392
16.4 Face Image Preprocessing Using KPCA 397
16.5 Summary 407
References 408
17 Fingerprint Identification-Ideas,Influences,and Trends of New Age 411
17.1 Introduction 411
17.2 System Architecture and Applications of Fingerprint Matching 415
17.3 The Early Years 419
17.4 Recent Feature Extraction Techniques-Addressing Core Problem 423
17.5 Conclusion and Future Directions 438
References 439
18 Subspaces Versus Submanifolds-A Comparative Study of Face Recognition 447
18.1 Introduction 447
18.2 Notation and Definitions 449
18.3 Brief Review of Subspace-Based Face Recognition Algorithms 451
18.4 Submanifold-Based Algorithms for Face Recognition 453
18.5 Experiments Results and Analysis 472
18.6 Conclusion 480
References 481
19 Linear and Nonlinear Feature Extraction Approaches for Face Recognition 485
19.1 Introduction 485
19.2 Linear Feature Extraction Methods 488
19.3 Non-Linear Feature Extraction Methods 489
19.4 Conclusions 511
References 512
20 Facial Occlusion Reconstruction Using Direct Combined Model 515
20.1 Introduction 515
20.2 Direct Combined Model Algorithm 518
20.3 Reconstruction System 523
20.4 Experimental Results 528
20.5 Conclusions 530
References 530
21 Generative Models and Probability Evaluation for Forensic Evidence 533
21.1 Introduction 534
21.2 Generative Models of Individuality 535
21.3 Application to Birthdays 538
21.4 Application to Human Heights 540
21.5 Application to Fingerprints 544
21.6 Summary 557
References 559
22 Feature Mining and Pattern Recognition in Multimedia Forensics-Detection of JPEG Image Based Steganography,Double-Compression,Interpolations and WAV Audio Based Steganography 561
22.1 Introduction 562
22.2 Related Works 566
22.3 Statistical Characteristics and Modification 568
22.4 Feature Mining for JPEG Image Forensics 573
22.5 Derivative Based Audio Steganalysis 576
22.6 Pattern Recognition Techniques 581
22.7 Experiments 584
22.8 Conclusions 600
References 601
PartⅣ：Biometric Authentication 607
23 Biometric Authentication 607
23.1 Introduction 608
23.2 Basic Operations of a Biometric System 616
23.3 Biometrics Standardization 621
23.4 Certification of Biometric System 622
23.5 Cloud Service—Web Service Authentication 624
23.6 Challenges of Large Scale Deployment of Biometric Systems 625
23.7 Conclusion 628
References 630
24 Radical-Based Hybrid Statistical-Structural Approach for Online Handwritten Chinese Character Recognition 633
24.1 Introduction 633
24.2 Overview of Radical-Based Approach 635
24.3 Formation of Radical Models 637
24.4 Radical-Based Recognition Framework 643
24.5 Experiments 650
24.6 Concluding Remarks 653
References 654
25 Current Trends in Multimodal Biometric System—Rank Level Fusion 657
25.1 Introduction 657
25.2 Multimodal Biometric System 660
25.3 Fusion in Multimodal Biometric System 664
25.4 Rank Level Fusion 667
25.5 Conclusion 672
References 672
26 Off-line Signature Verification by Matching with a 3D Reference Knowledge Image—From Research to Actual Application 675
26.1 Introduction 675
26.2 Used Signature Data 676
26.3 Image Types Used for Feature Extraction and Evaluation 678
26.4 Skills of Forgery Creation of Used Forgeries 680
26.5 Previous Work and Motivation for 3D RKI 682
26.6 3D Reference Knowledge of Signature 689
26.7 Ammar Matching Technique 691
26.8 Feature Extraction 694
26.9 Distance Measure and Verification 695
26.10 Experimental Results and Discussion 696
26.11 Limited Results are Shown and Discussed 702
26.12 AMT Features and Signature Recognition 703
26.13 AMT and Closely Related Works 703
26.14 nansition from Research to Prototyping then Pilot Project and Actual Use 704
26.15 Conclusions 706
References 707
27 Unified Entropy Theory and Maximum Discrimination on Pattern Recognition 709
27.1 Introduction 709
27.2 Unified Entropy Theory in Pattern Recognition 710
27.3 Mutual-Information—Discriminate Entropy in Pattern Recognition 715
27.4 Mutual Information Discrimination Analysis in Pattern Recognition 717
27.5 Maximum MI principle 719
27.6 Maximum MI Discriminate SubSpace Recognition in Handwritten Chinese Character Recognition 721
27.7 Conclusion 725
References 731
28 Fundamentals of Biometrics—Hand Written Signature and Iris 733
28.1 Prologue 733
28.2 Fundamentals of Handwritten Signature 735
28.3 Acquisition 750
28.4 Databases 751
28.5 Signature Analysers 752
28.6 Off-line Methods 754
28.7 On-line Methods 756
28.8 Fundamentals of Iris 760
28.9 Feature Extraction 764
28.10 Preprocessing 771
28.11 Iris Image Databases 776
28.12 Iris Analyzers 776
28.13 Conclusion 780
References 781
29 Recent Trends in Iris Recognition 785
29.1 Introduction 785
29.2 Basic Modules of Iris Recognition 786
29.3 Performance Measures 789
29.4 Limitations of Current Techniques 791
29.5 Future Scope 792
References 793
30 Using Multisets of Features and Interactive Feature Selection to Get Best Qualitative Performance for Automatic Signature Verification 797
30.1 Introduction 797
30.2 Signature Data 799
30.3 ASV Systems Using Threshold-Based Decision 801
30.4 MSF and Its Performance 807
30.5 IFS and QP 813
30.6 Conclusion 819
References 819
31 Fourier Transform in Numeral Recognition and Signature Verification 823
31.1 Concepts of Digital Transforms 823
31.2 Orthonormal System of Trigonometric Functions 824
31.3 Introduction to Discrete Fourier Transform 828
31.4 Properties of DFT 830
31.5 DFT Calculation Problem 833
31.6 Description of a Numeral Through Fourier Coefficents 837
31.7 Numeral Recognition Through Fourier Transform 842
31.8 Signature Verification Systems Trough Fourier Analysis 847
31.9 On-line Signature Verification System Based on Fourier Analysis of Strokes 851
References 854
Index 859