Advances in Biometrics (eBook)

Preface 6
Acknowledgment 8
Contents 9
Editor's Biography 11
1 Introduction to Biometrics and Special Emphasis on Myanmar Sign Language Recognition 13
1.1 Introduction and Background 13
1.2 Classification 16
1.3 Societal and Ethical Issues 17
1.3.1 Ethical Issues 18
1.4 Soft Biometrics 18
1.5 Biometric Standards, Protocols, and Databases 19
1.5.1 Standards 20
1.5.2 Protocols 21
1.5.3 Databases 21
1.6 Myanmar Sign Language Recognition 22
1.6.1 Sign Language Recognition 23
1.6.2 Myanmar Sign Language (MSL) Recognition and MIIT Database 23
1.6.3 MSL Implementation and Results 25
1.7 Conclusions 32
References 32
2 Handling the Hypervisor Hijacking Attacks on Virtual Cloud Environment 36
2.1 Introduction 36
2.2 Related Works 38
2.3 Background Theory of Proposed System 39
2.3.1 Virtualization Concept 39
2.3.2 Detection Method Based on Behavior Approach 40
2.3.2.1 Translation Lookaside Buffer TLB-Based Approach 41
2.3.2.2 CPU-Based Detection Approach 41
2.3.3 Hypervisor Model 41
2.3.3.1 OS Virtualization 42
2.3.3.2 Hardware Emulation 42
2.3.3.3 Para-Virtualization 43
2.3.4 Hypervisor Hijacking Thread Types 45
2.4 Serious Vulnerabilities in Virtualization 46
2.4.1 VM Sprawl 47
2.4.2 Hyper-jacking Attack 47
2.4.3 VM Escape Method 48
2.4.4 Denial-of-Service Attack 48
2.4.5 Incorrect VM Isolation 48
2.4.6 Unsecured VM Migration or VMotion 48
2.4.7 Host and Guest Vulnerabilities 48
2.5 Hacker Lifestyle 49
2.5.1 White Hat Hacker 49
2.5.2 Black Hat Hacker 49
2.5.3 Gray Hat Hacker 50
2.6 Cyber-Attack Lifecycle 50
2.6.1 Phase 1: Reconnaissance 50
2.6.2 Phase 2: Initial Compromise 50
2.6.3 Phase 3: Establish Foothold 51
2.6.4 Phase 4: Lateral Movement 51
2.6.5 Phase 5: Target Attainment 51
2.6.6 Phase 6: Ex-filtration, Corruption, and Disruption 51
2.6.7 Phase 7: Malicious Activities (Fig. 2.10) 51
2.7 Implementation Framework for Protecting Mechanism 52
2.7.1 VM Creation of Virtual Network Configuration 52
2.7.2 Tested Methodology 54
2.8 Behavior-Based Analysis for Hypervisor Detection 56
2.9 Protecting and Mitigation Technique for System Hardening 57
2.10 Future Studies 59
2.11 Conclusion 60
References 60
3 Proposed Effective Feature Extraction and Selection for Malicious Software Classification 62
3.1 Introduction 62
3.2 Related Work 64
3.3 Malicious Software Family Classification System 66
3.3.1 Analyzing Malware Samples and Generating Reports 67
3.3.2 Labeling Malicious Samples 68
3.3.3 Extracting Malicious Features 69
3.3.4 Applying N-gram 70
3.3.5 Representing and Selecting Malicious Features 71
3.3.6 Classifying Malware vs Benign Using Machine Learning 72
3.4 Results and Discussion 73
3.5 Conclusion 78
References 81
4 Feature-Based Blood Vessel Structure Rapid Matching and Support Vector Machine-Based Sclera Recognition with Effective Sclera Segmentation 83
4.1 Introduction 83
4.2 Proposed Design Methodologies 86
4.2.1 Pre-processing Process 86
4.2.1.1 Iris Segmentation 87
4.2.1.2 Sclera Segmentation 88
4.2.1.3 Sclera Blood Vein Enhancement 89
4.2.2 Features Extraction 90
4.2.3 Features Training and Classification 91
4.2.3.1 By K-d Tree-Based Matching Identifier 91
4.2.3.2 By SVM Classifier 93
4.3 Experimental Results and Performance Comparisons 94
4.4 Conclusions 99
References 99
5 Different Parameter Analysis of Class-1 Generation-2 (C1G2) RFID System Using GNU Radio 101
5.1 Introduction 101
5.2 RFID EPC C1G2 Protocol 102
5.2.1 Representation of RFID EPC Protocol in GNU Radio 102
5.2.2 Representation of BER in GNU for C1G2 Protocol 103
5.3 Introduction to RFID Authentication Factor 104
5.3.1 Single-Factor Authentication 105
5.3.2 Multi-factor Authentication 105
5.3.3 RFID Factor Authentication Application 106
5.3.4 Biometric Hash Functions 107
5.4 Digital Modulation Scheme for RFID System 107
5.4.1 Binary Amplitude Shift Keying (ASK) 108
5.4.2 Binary Frequency Shift Keying (BFSK) 108
5.4.3 Phase Shift Key (PSK) 108
5.4.3.1 Binary Phase Shift Key (BPSK) 109
5.4.3.2 Quadrature Phase Shift Key (QPSK) 109
5.4.3.3 Quadrature Amplitude Modulation (QAM) 110
5.4.3.4 Analysis of Digital Modulation Schemes over AWGN Channel 110
5.4.4 PSK over AWGN Channel 110
5.4.4.1 QPSK over AWGN Channel 111
5.4.4.2 QAM over AWGN Channel 111
5.4.4.3 ASK over AWGN Channel 111
5.4.4.4 FSK over AWGN Channel 111
5.4.4.5 BPSK over AWGN Channel 114
5.4.4.6 QPSK over AWGN Channel 114
5.4.4.7 QAM over AWGN Channel 114
5.5 Proposed Methodology 114
5.6 Results and Discussion 118
5.6.1 Performance of Detection Methods 118
5.6.2 Bit Error Rate for Digital Modulation Techniques 124
5.7 Conclusion 125
References 125
6 Design of Classifiers 127
6.1 Introduction 127
6.2 Cognitive Principles 128
6.3 Classification Problem 129
6.4 Classifier Models 130
6.5 Self-Regulatory Resource Allocation Network (SRAN) 130
6.6 Metacognitive Neural Network (McNN) 131
6.6.1 Learning Strategies 132
6.6.2 Knowledge Measures 133
6.7 Metacognitive Fuzzy Inference System (McFIS) 133
6.8 Projection-Based Learning with McNN (PBL McNN) 134
6.9 Metacognitive Extreme Learning Machine (McELM) 136
6.10 Summary 136
Bibliography 136
7 Social Impact of Biometric Technology: Myth and Implications of Biometrics: Issues and Challenges 139
7.1 Introduction 139
7.2 Biometric Myths and Misrepresentation 140
7.3 Vulnerability or Susceptibility Points of a Biometric System 145
7.4 Matter of Concerns of Biometrics 147
7.4.1 Biometric Framework Configuration Concerns 147
7.4.2 Confirmation 147
7.4.3 Liveness Detection 148
7.4.4 Collapse Rates 148
7.4.5 Circumvention and Repudiation 149
7.4.6 Handicapped Non-registrable Users 150
7.4.7 Adaptability 150
7.5 Impediments of Biometrics 151
7.6 Challenges, Difficulties, and Issues of Biometric System 153
7.6.1 Needs of Multi-Model Biometrics 153
7.7 Conclusion 163
References 164
8 Segmentation and Classification of Retina Images Using Wavelet Transform and Distance Measures 166
8.1 Introduction 166
8.2 Structure of the Eye 168
8.2.1 Lexicon of Terms for the Eye 169
8.2.2 Facts of Diabetic Eye Disease 171
8.3 Symptoms and Detection 171
8.3.1 Retinal Analysis 173
8.3.2 Causes of an Occlusion 173
8.3.3 Risk Factors of Retinal Vessel Occlusion 174
8.3.4 Blood Vessels 175
8.3.5 Exudates 177
8.3.6 Microaneurysms 177
8.3.7 Retinal Hemorrhage 178
8.4 Literature Review 179
8.4.1 Our Contribution 183
8.5 Proposed Methodology 183
8.5.1 Disease Classification 186
8.5.2 Datasets 186
8.5.3 Distance Measures 187
8.6 Results and Discussion 188
8.7 Conclusion 190
References 190
9 Language-Based Classification of Document Images Using Hybrid Texture Features 192
9.1 Introduction 192
9.2 Literature Review 197
9.3 Proposed Methodology 198
9.3.1 Preprocessing 198
9.3.2 Proposed Hybrid Texture Features 201
9.3.2.1 Stationary Wavelet Transform (SWT) 201
9.3.2.2 Histogram of Oriented Gradients (HOG) 206
9.3.3 SVM Classifier 209
9.4 Experimental Results 210
9.5 Conclusion 215
References 217
10 Research Trends and Systematic Review of Plant Phenotyping 219
10.1 Introduction 219
10.2 Related Work 223
10.3 Experimental Setup 225
10.3.1 Dataset Gathering 225
10.3.2 Preprocessing Module 226
10.3.3 Height Calculation Module 226
10.3.4 Region of Interest Calculation 227
10.4 Results and Discussion 227
10.5 Conclusion and Future Work 231
References 232
11 Case Studies on Biometric Application for Quality-of-Experience Evaluation in Communication 234
11.1 Introduction 234
11.2 Experimental Setting 236
11.2.1 Game for Experiment 236
11.2.2 Experiment Preparation 236
11.2.3 Subjective Evaluation Experiment Procedure 240
11.3 EEG Measurement 240
11.3.1 EEG Measurement Apparatus and Measurement Points 240
11.3.2 EEG Frequency Bands and Significance 240
11.3.3 EEG Power Spectrum 243
11.4 Experimental Results 243
11.4.1 EEG Power Spectrum Comparison 243
11.4.2 Average Increase Rate of Each Waveform 247
11.4.3 Average Increase Rate of Each Electrode Positions 248
11.4.4 Comprehensive Comparison 248
11.4.5 Player Level Comparison 249
11.4.6 Chinese and Japanese Comparison 251
11.5 Conclusions 253
References 253
12 Nearest Neighbor Classification Approach for Bilingual Speaker and Gender Recognition 255
12.1 Introduction 255
12.1.1 Variants of Speaker Recognition 257
12.2 Applications of Speech Processing and Speaker Recognition 258
12.3 Limitations of Speaker Recognition 260
12.4 Issues and Challenges 260
12.5 Related Work 262
12.6 Proposed Method 263
12.6.1 Dataset 263
12.6.2 Flow Chart 263
12.6.3 Methodology 264
12.7 Results 268
12.8 Conclusions 270
References 270
13 Effective Security and Access Control Framework for Multilevel Organizations 273
13.1 Introduction 273
13.1.1 Introduction 273
13.1.2 Objectives of the Proposed System 274
13.2 Related Work 274
13.3 Background Theory 275
13.3.1 What Are External and Internal Threats? 275
13.3.2 Security Control Model 276
13.3.3 Bell-LaPadula Model 276
13.3.4 Integrity 277
13.3.5 Levels of Integrity 277
13.3.6 Strict Integrity Policy 277
13.3.7 Strict Integrity Access Control Model 278
13.3.8 Conditions of Integrity 278
13.3.9 Levels of Entity to Control Security 278
13.4 Access Control 280
13.4.1 Identification 280
13.4.2 Account Authentication 280
13.4.3 Authorization 280
13.4.4 Accountability 280
13.4.5 Types of Access Control 281
13.5 Biba Security Model 281
13.5.1 Access Modes of Biba Model 282
13.5.2 Policies of Biba Model 282
13.5.3 Advantages and Disadvantages of Biba Model 282
13.6 Security Controls 283
13.6.1 Account Authentication Control 283
13.6.2 Handling User Access Control 283
13.6.3 Using User Input Control 284
13.6.4 Handling Communication and Data Transfer Controls 284
13.6.5 How to Handle Employees' Daily Operations Controls? 284
13.7 What Is Multilevel Organization? 284
13.8 Secure Framework for Multilevel Organizations 285
13.8.1 Secure Framework for Multilevel Organizations (SFMO) 285
13.8.2 Features of Secure Framework for Multilevel Organizations (FSFMO) 286
13.8.3 Design of Proposed System 288
13.8.4 System Flow of Proposed System 289
13.8.5 Case Study for Multilevel Organizations 289
13.9 Privacy Policies at Workplace 292
13.10 Conclusion 293
References 294
14 Dimensionality Reduction and Feature Matching in Functional MRI Imaging Data 295
14.1 Introduction 295
14.2 Existing Works 298
14.3 Methods 301
14.3.1 Sliced Inversed Regression 301
14.3.2 Principal Component-Sliced Inverse Regression 302
14.3.3 F-magnetic Resonance Imaging Pain Prediction 302
14.3.3.1 Experimental Design 302
14.3.3.2 PC-SIR Analyses 303
14.3.3.3 Pain Prediction 303
14.3.4 Manifold Learning 303
14.3.5 Manifold Embedding and Regularization 304
14.3.5.1 M-Magnetic Resonance ImagingR1 (Affine Combination) 305
14.3.5.2 M-Magnetic Resonance ImagingR2 305
14.3.5.3 Manifold Regularization and Reconstruction 306
14.3.6 An Initial Image Processing 307
14.3.7 Image Segmentation 307
14.3.8 Classification Methods 309
14.4 Conclusion 309
References 310
15 Classification of Biometrics and Implementation Strategies 312
15.1 Introduction 312
15.2 Traits 313
15.2.1 Physiological Biometric Traits 313
15.2.2 Behavior Biometric Traits 315
15.3 Classification of Biometric System 317
15.3.1 Unimodal Biometric System 317
15.3.2 Multimodal Biometric System 318
15.4 Implementation Strategies 318
15.5 Set of Features 320
15.6 Information Representation Through Features 321
15.6.1 Training and Testing Involved in Biometric System 322
15.6.2 Principal Component Analysis for Face Recognition 322
15.6.3 Eigenimage for Ear Recognition 324
15.6.4 Hamming Distance-Based Iris Recognition 324
15.6.5 Haar Wavelet Transform for Foot Recognition 326
15.7 Template Representation 326
15.7.1 Face Recognition 327
15.7.2 Ear Recognition 331
15.7.3 Iris Recognition 333
15.7.4 Footprint Recognition 335
Bibliography 336
16 Advances in 3D Biometric Systems 338
16.1 Introduction 338
16.2 Developments in 3D Biometric Systems 339
16.2.1 Face Recognition 339
16.2.2 Fingerprint Recognition 341
16.2.3 Iris Recognition 342
16.3 Anti-spoofing 343
16.3.1 Face Anti-spoofing 344
16.3.2 Fingerprint Anti-spoofing 347
16.3.3 Iris Anti-spoofing 347
16.4 Open-Source Softwares 348
16.5 Conclusions 349
References 349
Index 352