ABSTRACT
Palm vein recognition is a biometric identification method that integrates automated scientific applications to provide a way or means to identify a person by scanning a person’s palm vein in order to gain access. Palm vein recognition technology is secure because the authentication data exists inside the body, it is very difficult to forge and it is highly accurate.
The palm vein feature extraction task is another challenging problem in the infrared hand palm recognition tasks. In this research two set of extraction techniques have been suggested to represent the veins grid attributes: (i) the Linear Discriminant Analysis and (ii) the Principal Component Analysis. Each extracted feature list is assembled as a feature vector used to distinguish between palms belonging to different persons.
The training and testing were performed on static printed images of palm vein for the database with the palm vein samples taken with different angular prints and that of normal print. The experiments were performed for the recognition system at cropped resolutions of 20x20, 30x30, 40x40, 50x50, and 60x60 pixels. The system developed had average recognition performance accuracy of 96.86% and 90.45% for the LDA and PCA respectively. LDA performs better in terms of accuracy.
The best recognition performance was obtained when the LDA algorithm was used and this however shows why it is commonly used in authentication and verification process. The implemented system serves as an extendable foundation for future research.
TABLE OF CONTENTS
Page
Title page i
Certification ii
Dedication iii
Acknowledgements iv
Table of Content v
List of Figures ix
List of Tables x
Abstract xi
CHAPTER ONE
INTRODUCTION 1
1.1 Preamble 1
1.2 Statement of problem 5
1.3 Aim and Objectives 6
1.4 Method of study 6
CHAPTER TWO
LITERATURE REVIEW 7
2.1 Biometrics7
2.1.1 Enrollment Module 10
2.1.2 Identification Module 10
2.2 Biometrics Performance 11
2.3 Historical Background of Feature Extraction 11
2.4 Overview of Palm Vein Feature Extraction 13
2.4.1 The Description of PalmVein 13
2.4.2 The Principal Line Detection and Datum Point Location 16
2.5 Steps Taken in Palm Vein Feature Extraction 18
2.5.1 Preprocessing 18
2.5.1.1 Binarizing the Palm Image 18
2.5.1.2 Boundary Tracking 18
2.5.1.3 Feature Detection 19
2.5.1.4 Establishing a Coordination System 19
2.5.1.5 Extracting the Central Part 20
2.5.2 ROI Extraction 20
2.5.3 Feature Extraction and Matching 20
2.5.3.1 Line Based Approach 20
2.5.3.2 Subspace Based Approach 22
2.5.3.3 Statistical Approach 22
2.5.3.4 Coding Approaches 22
2.5.4 Accept and Reject 23
2.6 Palm vein Features Extraction Process 23
2.6.1 Image Acquisition 24
2.6.2 Feature Extraction 24
2.6.2.1 Line- Based 25
2.6.2.2 Appearance-Based 26
2.6.2.3 Texture-Based 26
2.6.3 Features Matching 28
2.6.3.1 Geometry Based Matching 28
2.6.3.2 Features Based Matching 29
2.7 Performance Measures 30
2.8 Feature Extraction based on Linear Discriminant Analysis 31
2.8.1 Algorithm for Linear Discriminant Analysis 31
2.9 Feature Extraction based on Principal Component Analysis 32
2.9.1 Algorithm for Principal Component Analysis 33
CHAPTER THREE
RESEARCH METHODOLOGY 34
3.1 Background Theory 34
3.2 Stages in the Development of the Proposed System 36
3.2.1 Palmvein Image Acquisition 36
3.2.2 Image Preprocessing 36
3.2.2.1 Image Enhancement 36
3.2.2.2 Segmentation 37
3.2.3 Histogram Equalization37
3.2.4 Gray Scale Conversion 38
3.2.5 Cropping 38
3.2.6 Noise Filtering 38
3.3 Recognizing images using Eigenspace 39
3.4 Training and Classification 40
3.5 Recognition and Testing 40
CHAPTER FOUR
RESULTS AND DISCUSSION 42
4.1 Analysis of Result 42
4.2 Discussion42
4.2.1 Computational Time 42
4.2.2 Accuracy 43
4.2.3 False Rejection Rate 43
4.2.4 False Acceptance Rate 43
CHAPTER FIVE
CONCLUSION AND RECOMMENDATIONS 60
5.1 Conclusion60
5.2 Recommendation 60
REFERENCES 61
APPENDIX A 65
LIST OF FIGURES
Figure Page
2.1 The Line Pattern of Palm Vein 15
2.2 The Distribution of Palm Vein Features 17
2.3 ROI Extraction 21
3.1 Palm Vein Recognition System 35
4.1 Graph Showing the Result of Computational Time 45
4.2 Graph Showing Accuracy Performance of PCA and LDA Feature
Extraction Comparison of PCA and LDA. 49
4.3 GUI Window Developed 51
4.4: Information Window 52
4.5 Equalization Operation 53
4.6 Raw Histogram of a Palm Vein Image 54
4.7 Equilised Palm Vein 56
4.8 Coded-PCA Features 57
4.9 Coded-LDA Features 58
LIST OF TABLES
Table Page
4.1.1 Computation time 44
4.1.2 Accuracy 46
4.1.3 False Rejection rate 47
4.1.4 False Acceptance rate 48