The Biometric Computing: Recognition and Registration

Cover
Half Title
Title Page
Copyright Page
Contents
Preface
Editors
Contributors
Part I: The Biometric Computing – Fundamentals & Definitions
	1. Acquisition and Computation for Data in Biometric System
		1.1 Introduction
		1.2 Elements of Biometric System
			1.2.1 Image Acquisition
			1.2.2 Feature Extraction
			1.2.3 Recognition
		1.3 Performance Measures
		1.4 Applications of Face Recognition
		1.5 Image Acquisition Methods
			1.5.1 Fingerprint Recognition Methods
			1.5.2 Face Recognition
			1.5.3 Iris Recognition
			1.5.4 Multi-Biometrics
			1.5.5 Hand Geometry
		1.6 Conclusion
		References
	2. Advances in Unconstrained Handprint Biometrics
		2.1 Background
			2.1.1 Palm Print and Knuckle Print Biometrics
				2.1.1.1 Palm Print
				2.1.1.2 Inner Knuckle Print
		2.2 Related Works
		2.3 Palm Print and Finger Knuckle Print Traditional Sensors
		2.4 Proposed Single Sensor Single Shot Multi-Biometric System
			2.4.1 Image Acquisition
			2.4.2 Preprocessing Techniques
			2.4.3 ROI Extraction
				2.4.3.1 Palm Print ROI Extraction
				2.4.3.2 Inner Knuckle Print ROI Extraction
			2.4.4 Image Enhancement and Transformation
			2.4.5 Feature Extraction and Matching
			2.4.6 Two-Stage Fusion
		2.5 Experimental Analysis and Evaluation Parameters
		2.6 Conclusions
		References
	3. Voiceprint-Based Biometric Template Identifications
		3.1 Introduction
		3.2 Related Work
			3.2.1 Unsupervised Learning
			3.2.2 Pronunciation Modeling
			3.2.3 Phonetic Distance Measurements
		3.3 Phoneme Substitution Cost Matrix
			3.3.1 Classification of Phonemes Based on Articulatory Features
			3.3.2 Phonetic Distance
			3.3.3 Phonetic Distance Computations
		3.4 Dynamic Phone Warping (DPW)
			3.4.1 DPW Algorithm
			3.4.2 Experimentation Details, Results and Analysis
		3.5 Critical Distance Criteria
			3.5.1 Definitions
			3.5.2 Critical Distance Estimation (CDE) Algorithm
			3.5.3 Experimentation
			3.5.4 Results and Analysis
				3.5.4.1 Estimation of Parameter γ
				3.5.4.2 Estimation of Parameter δ
		3.6 Biometric Template for Speaker Identification
			3.6.1 Typical Speaker Recognition System
			3.6.2 Speaker Recognition Models
				3.6.2.1 Minimum-Distance Classifier Model
				3.6.2.2 Acoustic Model
				3.6.2.3 GMM Model
				3.6.2.4 Hybrid HMM/VQ-Based Model
			3.6.3 Pronunciation Voiceprint-Based Speaker Recognition Model
				3.6.3.1 Preparation of Speaker Models
				3.6.3.2 Speaker Identification
			3.6.4 Speaker Recognition Algorithm
			3.6.5 Experimentation Details
				3.6.5.1 Data Sets Source and Sample Size
				3.6.5.2 Experimental Setup
				3.6.5.3 Experimentation Methodology
			3.6.6 Results and Discussion
				3.6.6.1 Speaker Models
				3.6.6.2 Speaker Identification
		3.7 Conclusions
		References
	4. Behavioral Biometrics: A Prognostic Measure for Activity Recognition
		4.1 Introduction to Behavioral Biometrics
		4.2 Keystroke Dynamics
			4.2.1 User Enrollment
			4.2.2 Feature Extraction
			4.2.3 Classification of Keystroke Dynamics
			4.2.4 Public Database for Keystroke Dynamics
			4.2.5 Commercial Applications
			4.2.6 Limitations of Keystroke Biometrics
		4.3 Speaker Recognition
			4.3.1 User Enrollment
			4.3.2 Voice Feature Extraction
			4.3.3 Speaker Recognition Public Databases
		4.4 Handwriting and Signature Recognition
			4.4.1 Handwritten Text and Signature Public Databases
			4.4.2 Feature Extraction
		4.5 Computer-/Mobile Application-Based Behavioral Biometrics
		4.6 Biomedical Signals as Biometrics
		4.7 Conclusions
		References
	5. Finger Biometric Recognition with Feature Selection
		5.1 Introduction
			5.1.1 Objectives
		5.2 Related Works
		5.3 Preprocessing Methodology of Finger Geometry
			5.3.1 Elementary Preprocessing of Hand Image
			5.3.2 Finger Segmentation
			5.3.3 Geometric Feature Set Computation
		5.4 Feature Selection Algorithms
		5.5 Experimentation
			5.5.1 Database Description
			5.5.2 Classifier Description
			5.5.3 Experimental Description
				5.5.3.1 Identification
				5.5.3.2 Feature Overfittting
				5.5.3.3 Verification
		5.6 Conclusion
		Acknowledgment
		References
Part II: The Biometric Computing – Algorithms & Methodologies
	6. Iris Recognition Systems in a Non-Cooperative Environment
		6.1 Introduction
		6.2 Iris Recognition Systems from Images and Video
			6.2.1 Segmentation Iris Texture Region
				6.2.1.1 Viterbi-Based Segmentation Algorithm
				6.2.1.2 Contrast-Adjusted Hough Transform Segmentation Algorithm
				6.2.1.3 Weighted Adaptive Hough and Ellipsopolar Transform
				6.2.1.4 Modified Hough Transform Segmentation Algorithm
			6.2.2 Normalization Process
			6.2.3 Iris Image Fusion
			6.2.4 Features Iris Texture
				6.2.4.1 Taxonomy of Iris Feature Extraction Methods
				6.2.4.2 Statistical Methods
				6.2.4.3 Signal Processing Methods
				6.2.4.4 Combined Methods
				6.2.4.5 Feature Learning Methods
		6.3 Databases in a Non-Cooperative Environments
			6.3.1 CASIA-V3-Interval
			6.3.2 CASIA-V4-Thousand
			6.3.3 UBIRIS-V1
			6.3.4 Multiple Biometrics Grand Challenge MBGC
		6.4 Experimental Results in Iris Recognition Systems in a Non-Cooperative Environments
			6.4.1 Fusion Segmentation and Quality Evaluation as Part of Iris Recognition System in Non-Cooperative Environments
			6.4.2 Perspectives of the Use of Deep Neural Networks in the Improvement of the Accuracy of Iris Biometric Systems
		6.5 Conclusion
		Acknowledgments
		References
	7. Slap Fingerprint Authentication and Its Limitations
		7.1 Introduction
		7.2 State of the Art
			7.2.1 Working of a Biometric System
			7.2.2 Performance Evaluation
				7.2.2.1 Performance Evaluation in Verification
			7.2.3 Performance Evaluation in Identification
		7.3 Slap Image Authentication
			7.3.1 Slap Segmentation
				7.3.1.1 Framework of Slap Segmentation
				7.3.1.2 Existing Approaches
				7.3.1.3 Challenges in Component Detection
				7.3.1.4 Challenges in Fingerprint Component Detection
				7.3.1.5 Challenges in Hand Detection
			7.3.2 Single Fingerprint Matching
				7.3.2.1 Existing Approaches
				7.3.2.2 Challenges
			7.3.3 Fusion
		7.4 Future Research Directions
		7.5 Conclusions
		References
	8. The Reality of People Re-Identification Task, Where We Are and Where We Are Going: A Review
		8.1 Introduction
			8.1.1 Person Re-Identification Formalization
		8.2 Evaluation Datasets
		8.3 Current Approaches
			8.3.1 Feature Representation
				8.3.1.1 Appearance-Based Models
				8.3.1.2 Motion-Based Models
				8.3.1.3 Biometrics-Based Models
				8.3.1.4 Spatiotemporal Features
			8.3.2 Distance Metrics or Learning Metrics
		8.4 Systems Evaluation and Discussion
			8.4.1 VIPeR Dataset Results
			8.4.2 ETHZ Dataset Results
			8.4.3 iLIDS Dataset Results
			8.4.4 CAVIAR4REID Dataset Results
			8.4.5 GRID Dataset Results
			8.4.6 CUHK01 Dataset Results
			8.4.7 Market-1501 Dataset Results
			8.4.8 PRID2011 Dataset Results
			8.4.9 PRID450S Dataset Results
		8.5 Discussion and Conclusions
		References
	9. Optimization of SVM-Based Hand Gesture Recognition System Using Particle Swarm Optimization and Plant Growth Simulation Algorithm
		9.1 Introduction
			9.1.1 Background
			9.1.2 Significance
		9.2 Literature Survey
			9.2.1 Survey on Database
			9.2.2 Survey on Feature Extraction
			9.2.3 Survey on Classification
			9.2.4 Survey on Optimization-Based Hand Gesture Recognition
		9.3 Methodology
			9.3.1 Image Database
			9.3.2 Image Preprocessing
			9.3.3 Feature Extraction
			9.3.4 Classification
			9.3.5 Optimization
				9.3.5.1 Particle Swarm Optimization (PSO)
				9.3.5.2 Plant Growth Simulation Algorithm (PGSA)
		9.4 Result and Discussion
		9.5 Conclusions
		References
	10. Internet of Biometric Things: Standardization Activities and Frameworks
		10.1 Introduction
			10.1.1 Biometrics Technology Evolution
			10.1.2 Biometrics Signals Processing
		10.2 Biometrics Technology and Internet of Things
			10.2.1 Influence of IoT on the Development of Biometrics Technology
			10.2.2 Role of Cloud Computing
			10.2.3 Biometric Sensors
		10.3 Internet of Biometric Things
			10.3.1 IoBT Devices
			10.3.2 Examples of Cloud-Centric Benefits
			10.3.3 Behavioral Analytics Based on Multimedia Content
		10.4 Internet of Multimedia Things
			10.4.1 Multimedia Communication in IoMT
			10.4.2 IoMT Standardization Process
		10.5 Conclusions
		References
Part III: The Biometric Computing – Futuristic Research & Case Studies
	11. Deep Neural Networks for Biometric Identification Based on Non-Intrusive ECG Acquisitions
		11.1 Introduction
		11.2 State-of-the-Art Overview
			11.2.1 ECG-Based Biometrics
			11.2.2 Deep Learning for Signals and Biometrics
		11.3 A CNN for ECG Biometrics
			11.3.1 General Structure Overview
			11.3.2 Convolutional and Pooling Layers
			11.3.3 Fully Connected Layers
			11.3.4 Optimization and Regularization
				11.3.4.1 Optimizer and Loss
				11.3.4.2 Dropout
				11.3.4.3 Data Augmentation
		11.4 Baseline Algorithm
		11.5 Results and Benchmarking
		11.6 Conclusion
		Acknowledgments
		References
	12. Recent Advances in Biometric Recognition for Newborns
		12.1 Introduction
		12.2 State-of-the-Art
		12.3 Current Technological Response and Its Limitations
		12.4 Newborn Recognition Techniques
			12.4.1 RFID Bracelets
			12.4.2 Medical Techniques
			12.4.3 Footprint
			12.4.4 Face Identification
			12.4.5 Fingerprints
			12.4.6 Palmprint
			12.4.7 Ear
			12.4.8 Iris Recognition
			12.4.9 Soft Biometrics
		12.5 Multimodal Biometrics for Newborn Recognition
		12.6 Proposed Framework and Future Work
		12.7 Conclusions
		References
	13. Paradigms of Artificial Intelligence in Biometric Computing
		13.1 Introduction
		13.2 Types of Biometric Technologies
			13.2.1 Fingerprint Recognition
			13.2.2 Face Recognition
			13.2.3 Voice Recognition
			13.2.4 Iris Recognition
			13.2.5 Handwriting and Signature Recognition
			13.2.6 Behavioral Recognition
		13.3 Artificial Intelligence in Biometric Applications
			13.3.1 Artificial Neural Network
			13.3.2 Support Vector Machine
		13.4 Performance Metrics for Testing the Biometric System
		13.5 Conclusion
		References
	14. Face Recognition in Low-Resolution Space
		14.1 Introduction
		14.2 Low-Resolution Face Recognition Problem
		14.3 Solutions of Low-Resolution Face Recognition Problem
		14.4 Literature Review
			14.4.1 SR-Based Methods
			14.4.2 UFS-Based Methods
		14.5 Low-Resolution Face Recognition System via SR Technique
			14.5.1 Image Super-Resolution
		14.6 Feature Extraction
		14.7 Face Recognition
		14.8 Results and Discussions
			14.8.1 Experimental Results on the LFW Database
			14.8.2 Experimental Results on the ORL Database
			14.8.3 Experimental Results on the AR Database
			14.8.4 Experimental Results on the EYB Database
		14.9 Conclusion
		References
Index