Table of Contents
Preface vii
Acknowledgment xv
Chapter I Overview 1
Introduction 1
Biometric Recognition Technologies 5
Main Problems in Biometric Recognition 8
Book Perspective 12
References 16
Section I Biometric Discriminant Analysis
Chapter II Discriminant Analysis for Biometric Recognition 25
Linear Discriminant Analysis 25
LDA for Solving the Small Sample Size Problems 27
Organization of Section I 28
References 28
Chapter III Discriminant Criteria for Pattern Classification 30
Introduction 30
Large Margin Linear Projection Classifier 34
Minimum Norm Minimum Squared-Error Classifier 38
Maximum Scatter Difference Classifier 43
Summary 55
References 55
Chapter IV Orthogonal Discriminant Analysis Methods 58
Introduction 58
Orthogonalized Fisher Discriminant 63
Fisher Discriminant with Schur Decomposition 65
Comparison of Orthogonal Discriminant Analysis Methods 70
Summary 75
References 76
Chapter V Parameterized Discriminant Analysis Methods 78
Parameterized Direct Linear Discriminant Analysis 78
Weighted LDA in the Range of Within-Class Scatter Matrix 94
Summary 103
References 103
Chapter VI Two Novel Facial Feature Extraction Methods 106
Multiple Maximum Scatter Difference 106
Feature Extraction Based on Coefficients of Variances 119
Summary 131
References 132
Section II Tensor Technology
Chapter VII Tensor Space 135
Background 135
Basic Notations 136
Tensor Decomposition 142
Tensor Rank 145
References 148
Chapter VIII Tensor Principal Component Analysis 150
Introduction 150
Basic Algorithms 152
Applications to Biometric Verification 161
Summary and Discussion 168
References 170
Chapter IX TensorLinear Discriminant Analysis 172
Introduction 172
Basic Algorithms 173
Applications to Biometric Verification 191
Summary and Discussion 198
References 199
Chapter X Tensor Independent Component Analysis and Tensor Non-Negative Factorization 202
Introduction 202
Tensor Independent Component Analysis 205
Tensor Non-Negative Factorization (NF) 213
Applications to Biometric Verification 220
Summary 223
References 223
Chapter XI Other Tensor Analysis and Further Direction 226
Introduction 226
Tensor-Based Classifiers 228
Other Tensor Subspace Analysis 239
Summary 249
References 250
Section III Biometric Fusion
Chapter XII From Single Biometrics to Multi-Biometrics 254
Introduction 254
Biometric and Multi-Biometric Fusion: Definition and Notation 257
Performance Evaluation of Biometric Techniques 263
Research and Development of Multi-Biometrics 265
References 268
Chapter XIII Feature Level Fusion 273
Introduction 273
Schemes to Fuse Features at the Feature Level 275
Face and Palm Print Fusion at the Feature Level 276
Fusion of Multiple Feature Presentations 294
Comments 301
References 301
Chapter XIV Matching Score Level Fusion 305
Introduction 305
Matching Score Fusion Rules 308
Normalization Procedures of Matching Scores 314
Exemplification: Information Fusion of Face and Palmprint 318
Comments 323
References 323
Chapter XV Decision Level Fusion 328
Introduction 328
Rules and Methods of Decision Level Fusion 330
Selecting Classifiers Based on Correlations Between Classifiers 337
A Case Study of Group Decision-Based Face Recognition 338
Comments on Biometric Fusion at the Three Levels 343
References 345
Chapter XVI Book Summary 349
Content Summary 349
Method Applicability 351
Comments on Multi-Biometrics Development 352
References 357
Glossary 359
About the Authors 364
Index 366
