Iris_Recognition_System_Using_Histogram_Analysis_-_24_Pages.pdf

IRIS RECOGNITION SYSTEM USING HISTOGRAM ANALYSIS

NORHUDA BINTI OTHMAN

This Report is submitted in partial fulfillment of requirement for the award of Bachelor

of Electronic Engineering (Computer Engineering) With Honours

Faculty of Electronic and Computer Engineering

Universiti Teknikal Malaysia Melaka

June 2012

iii

I declare that this report entitled Iris Recognition using Histogram Analysis is the

result of my own research expect as cited in the differences.

Signature :

Name : NORHUDA BINTI OTHMAN

Date : 15 JUNE 2012

iv

I hereby declare that I have read this report and in my opinion this report is

sufficient in terms of the scope and quality for the award of Bachelor of Electronic

Engineering (Computer Engineering) With Honours.

Signature :

Supervisors Name : MR. ROSMAN BIN ABD RAHIM

Date : 15 JUNE 2012

v

Dedicated, in thankful appreciation for support, encouragement and understandings

to my beloved parent, Mr. Othman Bin Dahali and Mrs. Bainah Binti Sanusi, my

supervisor, Mr. Rosman Bin Abd Rahim, my siblings and all my friends.

vi

ACKNOWLEDGMENT

Alhamdulillah. Thanks to Allah SWT, for giving me the opportunity to

complete this Final Year Project and writing the thesis successfully at Universiti

Teknikal Malaysia Melaka.

First of all, I would like to thank to my supervisor of this Final Year Project,

Mr. Rosman bin Abd Rahim for the valuable guidance and advice. He motivates me

to contribute more idea to my project. I also would like to thank him for propose the

title of my project and showing me some example that related to my project.

Lastly, I would like to thanks and appreciate to my family for their

understandings and supports on me in completing my Final Year Project. Also thanks

to all of my friends and everyone, that has been contributed by supporting my Final

Year Project at Universiti Teknikal Malaysia Melaka.

vii

ABSTRACT

A biometric system provides automatic identification of an individual based

on a unique feature or characteristic possessed by the individual. Iris recognition is

regarded as the most reliable and accurate biometric identification system available.

This report involves the development of iris recognition system using the histogram

analysis method the Local Phase Quantization and Rotation Invariant Local Phase

Quantization to verify the both the uniqueness of human iris as a biometric and

performance. To determine the recognition performance of the iris image system

digitized gray scale has been used and developed. Iris recognition system consists of

the process of image segmentation, feature extraction which includes the RGB image

to grayscale. Then, process transformation image to polar using Polar Transform and

through the conversion process to form a histogram method LPQ and RILPQ. From

this histogram, it will be analyzed to determine or recognize the iris image. To

implement this recognition, the machine learning process has been implementing a

database containing the iris image test database and iris image train database has

been developed. K Nearest Neighbour classifier algorithm will use for recognition in

the Iris Recognition System. An iris recognition system that requires the

comprehension of a complex algorithm was succesfully developed and it is effective

enough when being integrated with a system that requires identity checking. The

overall system that deploy LPQ and RILPQ with histogram analysis was shown to

achieve the initial objectives of this project. It was also proved to attain high

recognition accuracy.

viii

ABSTRAK

Satu sistem biometrik menyediakan pengenalan automatik individu

berdasarkan ciri-ciri yang unik atau ciri-ciri yang dimiliki oleh individu. Iris

pengiktirafan dianggap sebagai sistem pengenalan biometrik yang paling dipercayai

dan tepat. Laporan ini melibatkan pembangunan sistem pengiktirafan iris

menggunakan kaedah analisis histogram Pengkuantuman Fasa Tempatan dan Putaran

yang berbeza Fasa Pengkuantuman Tempatan untuk mengesahkan kedua-dua

keunikan iris manusia sebagai biometrik dan prestasi. Untuk menentukan prestasi

pengiktirafan sistem imej iris didigitalkan skala kelabu telah digunakan dan

dibangunkan. Iris sistem pengiktirafan terdiri daripada proses pengekstrakan

segmentasi, ciri imej yang termasuk imej RGB kepada skala kelabu. Kemudian,

proses transformasi imej kutub menggunakan Kutub Transform dan melalui proses

penukaran untuk membentuk kaedah LPQ histogram dan RILPQ. Dari histogram ini,

ia akan dianalisis untuk menentukan atau mengiktiraf imej iris. Untuk melaksanakan

pengiktirafan ini, proses pembangunan pangkalan data telah dilaksanakan yang

mengandungi imej iris ujian pangkalan data dan pangkalan data bagi melatih imej

iris telah dibangunkan. KNN algoritma digunakan untuk pengiktirafan dalam Sistem

Pengiktirafan Iris. Satu sistem pengiktirafan iris yang memerlukan kefahaman

algoritma yang kompleks telah berjaya dibangunkan dan ia adalah cukup berkesan

apabila disepadukan dengan sistem yang memerlukan semakan identiti. Keseluruhan

sistem yang menghantar LPQ dan RILPQ dengan analisis histogram telah

ditunjukkan untuk mencapai objektif awal projek ini. Ia juga membuktikan mencapai

ketepatan pengiktirafan tinggi.

ix

TABLE OF CONTENTS

CHAPTER TITLE PAGE

PROJECT TOPIC i

CERTIFICATION STATUS REPORT FORM ii

DECLARATION iii

DEDICATION v

ACKNOWLEDGEMENT vi

ABSTRACT vii

ABSTRAK viii

TABLE OF CONTENTS ix

LIST OF TABLE xiii

LIST OF FIGURES xiv

LIST OF ABBREVIATIONS xvi

LIST OF APPENDICES xvii

I INTRODUCTION 1

1.1 BACKGROUND 1

1.2 PROBLEM STATEMENT 2

1.3 PROJECT OBJECTIVES 2

1.4 SCOPE OF THE PROJECT 2

1.5 PROJECT OUTLINE 3

II LITERATURE REVIEW 4

2.1 IRIS RECOGNITION PROJECT 4

x

2.1.1 Iris Recognition using Harr Wavelet 4

2.1.2 Iris Recognition of Human Iris Patterns 6

2.1.3 Iris Recognition Works 6

2.1.4 Iris Recognition Based on Using 7

Ridgelet and Curvelet Transform

2.1.5 PCA based Iris Recognition using DWT 7

2.1.6 Texture Classification Using Local Phase 8

Quantization (LPQ)

2.1.7 Local phase quantization texture descriptor 9

For protein classification

2.2 BASIC RECOGNITION 10

2.3 IMAGE PROCESSING AND COMPUTER VISION 10

2.3.1 Computer Vision Hierarchy 11

2.3.2 Computer Vision vs. Image Processing 11

2.3.3 Image Formation 12

2.4 HISTOGRAM ANALYSIS 13

2.4.1 Grayscale Images 14

2.4.2 Binary Images 15

2.4.3 Color Images 15

2.4.4 8-bit Color Images 16

2.4.5 Color Quantization 16

2.4.6 Thresholding 17

III METHODOLOGY 20

3.1 PLANNING OF PROJECT 20

3.1.1 Gantt Chart 1 21

3.1.2 Gantt Chart 2 22

3.2 PROJECT METHODOLOGY 23

3.3 PROJECT DESIGN 24

3.4 SEGMENTATION 25

3.5 FEATURE EXTRACTION 26

3.5.1 Images Conversion in Grayscale 26

3.5.2 Polar Transform 27

xi

3.5.3 Local Phase Quantization (LPQ) 29

3.5.4 Rotation Invariant Local Phase Quantization 31

(RILPQ)

3.6 MACHINE LEARNING 33

3.7 RECOGNITION 34

IV RESULT AND DISCUSSION 35

4.1 IRIS RECOGNITION SYSTEM 35

4.2 INTERFACE OF IRIS RECOGNITION SYSTEM 37

(GUI)

4.2.1 Layout the GUI system 37

4.2.2 Save the GUI Layout 42

4.2.3 Program the GUI 44

4.3 FUNCTIONALITY OF IRIS 45

RECOGNITION SYSTEM

4.3.1 Show Original Iris Database Button 45

4.3.2 Show Test Iris Database Button 46

4.3.3 Show Train Iris Database Button 47

4.3.4 Select Test Iris Button 48

4.3.5 Feature Extraction Button 49

4.3.6 Load Pre-Trained Database Button 50

4.3.7 Recognise Test Iris Button 51

V CONCLUSION AND RECOMMANDATION 53

4.1 PROJECT CONCLUSION 53

4.2 FUTURE DEVELOPMENT 54

REFERENCES 55

xii

APPENDIX A 57

APPENDIX B 61

APPENDIX C 78

xiii

LIST OF TABLE

TABLE NO TITLE PAGE

3.1 Gantt Chart for semester 1 21

3.2 Gantt Chart for semester 2 22

4.1 Efficiencies of the different process in the system 52

xiv

LIST OF FIGURES

FIGURE NO TITLE PAGE

2.1 Original image 5

2.2 Localized Iris 5

2.3 Iris isolated image 5

2.4 Basic Recognition 10

2.5 Block Diagram Computer vision 10

2.6 Block Diagram image processing 11

2.7 Process of image formation 13

2.8 Histogram Analysis 14

2.9 A 22 pixel area displaying one composite color 17

2.10 Threshold, Density slicing 18

2.11 Classic bi-modal intensity distribution 19

3.1 Project Methodology 23

3.2 Diagram method 24

3.3 Iris Recognition work diagram 25

3.4 Original Iris Image 25

3.5 Iris Image after Segmentation Process 26

3.6 Iris Image in Grayscale 27

3.7 Iris image in Polar Transform 27

3.8 Polar Transform 28

3.9 Produce the Histogram using LPQ method 33

4.1 Create of new GUI file 37

4.2 Dialog of GUIDE Quick Start 38

4.3 Interface to build GUI 38

4.4 Setting the toolbox with name 39

4.5 GUIDE Preferences 39

xv

4.6 New GUI interface with function namely 40

4.7 Complate the GUI Interface 42

4.8 Windows to save the gui file 43

4.9 GUI system 45

4.10 Original Iris Database 46

4.11 Test Iris Database 47

4.12 Train Iris Database 48

4.13 Output when Select Test Iris Button 49

4.14 Output when select Feature Extraction Button 50

4.15 Load Pre-Trained Database Button 51

4.16 Overall result of Iris Recognition System 52

xvi

LIST OF ABBREVIATIONS

LPQ - Local Phase Quantization RILPQ - Rotation Invariance Local Phase Quantization PCA - Principal Component Analysis DWT - Discrete Wavelet Transform RGB - Red, Green, Blue KNN - K-Nearest Neighbour GUI - Graphic User Interface

xvii

LIST OF APPENDICES

NO TITLE PAGE

A Main Source Code of Iris Recognition 57

Using Histogram Anaysis System in MATLAB

B Function Source Code of Iris Recognition 61

Using Histogram Anaysis System in MATLAB

C Source Code of Iris Recognition Using Histogram 78

Anaysis System in Graphic User Interface(GUI)

in MATLAB

CHAPTER I

INTRODUCTION

1.1 Background

`IRIS RECOGNITION` is a biometric technique to identify a person due to

its unique pattern contained in each iris. Advantage of appearing on characteristics of

iris with develops iris recognition system to identify an individual using the image of

their iris. In this proposal, how to identify Iris is based on histogram of the image iris.

The purpose of 'Iris Recognition', biometrical technology for personal identification

and authentication, is to recognize a person from the iris. In fact, the iris pattern is

characterized by a high level of stability and distinctiveness. Every individual has a

unique iris.

Development 'Iris Recognition' uses MATLAB for ease of manipulation of

images and applications. This has the potential to improve the efficiency of the

introduction of the iris, and the system only need to store and process for the

introduction of one-dimensional signal processing is required. The methodology in

the form of this histogram analysis, how it is used in the registration and

identification and performance in terms of positive and negative result will be

presented through this project.

2

1.2 Problem Statement

Problem statement for this project are:

a) This system was developed to help and solve the problem of invasion of

person identity information.

b) Iris recognition system is able to protect the security of any information more

safe and secure.

c) Thus, by having this system it can help to improve the security level

remarkably.

1.3 Project Objectives

The objectives of this project are:

a) To study and comprehend the iris recognition technology by using histogram

analysis.

b) To design and develop an iris based recognition system with histogram

analysis using MATLAB.

c) To analyze the performance of the histogram analysis method to recognize

person.

1.4 Scope of the Project

Firstly, Project focuses on the software development only, not using hardware

for capturing an iris image. The designed system will tested with develop own iris

image database from downloading iris image. Recognition algorithm is implemented

in MATLAB software. MATLAB software provides an excellent rapid application

development with its image processing toolbox, and high level programming

methodology. Other than that, the system using iris image because the iris has many

advantages and benefits for biometric technology. Here are just a few of the benefits

using iris recognition in this system. Iris has high accuracy; one of the main benefits

3

of using eye biometrics is the high accuracy that this technology provides. Everyone

has an iris pattern that is distinct even twins. Both the left and right irises differ from

one another as well. In theory, the accuracy of this technology should be at close to

100%. So, from this advantages will develop of iris recognition system.

1.5 Project Outline

This thesis comprises of five chapters. The first chapter briefly discusses the

overviews about the project such as introduction, objectives, problem statements and

scope of this project.

Chapter II describes about the research and information about the project.

Every facts and information, which found through by any references had been

selected. This literature review has been explained about the experienced project of

the Iris Recognition and other project uses the same method for recognition using

histogram.

Chapter III will discuss about the methodology that have been used in this

project. The project must be understand first and make the research about previous

project. The method that have been used are develop the software, troubleshooting

the program and lastly the project have been presented. This chapter also consists

about the project design software using more that one method.

Chapter IV, describe about the result and discussion. The result is presented

more to Interface of Iris Recognition System, function of the system and

performance of system and movement meanwhile discussion presented more to the

problem that occur along this project session and that problem.

Finally, Chapter V tells about conclusion and recommendation. The

conclusion describes about the task that have been done during this project. The

recommendation is added to give an opinion and also an improvement on how the

future works should have done.

4

CHAPTER II

LITERATURE REVIEW

This chapter will explain about the literature review which is related to this

Iris Recognition or method will be uses in project. Information about the Iris

Recognition and histogram analysis in LPQ and RILPQ method has been studied

from different resources to perform this project

2.1 Iris Recognition Project

This section will discuss about the previous Iris Recognition project that have

been developed by the previous researcher.

2.1.1 Iris Recognition using Harr Wavelet

According to C. H. Daouk, L. A. El-Esber, F. D. Kammoun and M. A. Al

Alaoui, Iris Recognition, Iris Recognition using Matlab for its ease in image

manipulation and wavelet applications. The first step of project consists of images

acquisition. Then, the pictures size and type are manipulated in order to be able

subsequently to process them. Once the preprocessing step is achieved, it is

necessary to localize the iris and unwrap it. This stage, can extract the texture of the

5

iris using Haar Wavelets. Finally, compare the coded image with the already coded

iris in order to find a match or detect an imposter.

Figure 2.1: Original image

a) Image acquisition

b) Image manipulation

c) Iris localization

Figure 2.2: Localized Iris

Figure 2.3: Iris isolated image

d) Feature Extraction

e) Mapping

f) Haar Wavelets

6

Most previous implementations have made use of Gabor wavelets to extract

the iris patterns. But, since we are very keen on keeping our total computation time

as low as possible, we decided that building a neural network especially for this task

would be too time consuming and selecting another wavelet would be more

appropriate.

2.1.2 Iris Recognition of Human Iris Patterns

Accoding to Libor Masek, Recognition of Human Iris Patterns for Biometric

Identification, and the first stage of iris recognition is to isolate the actual iris region

in a digital eye image. The iris region, shown in Figure 1.1, can be approximated by

two circles, one for the iris or sclera boundary and another, interior to the first, for

the iris/pupil boundary. The eyelids and eyelashes normally occlude the upper and

lower parts of the iris region. Also, specular reflections can occur within the iris

region corrupting the iris pattern. A technique is required to isolate and exclude these

artefacts as well as locating the circular iris region.

The success of segmentation depends on the imaging quality of eye images.

Images in the CASIA iris database do not contain specular reflections due to the use

of near infra-red light for illumination. However, the images in the LEI database

contain these specular reflections, which are caused by imaging under natural light.

Also, persons with darkly pigmented irises will present very low contrast between

the pupil and iris region if imaged under natural light, making segmentation more

difficult. The segmentation stage is critical to the success of an iris recognition

system, since data that is falsely represented as iris pattern data will corrupt the

biometric templates generated, resulting in poor recognition rates.

2.1.3 Iris Recognition Works

At the outset of this project, some key tasks were identied that needed to be

carried out to full the aim of creating a working prototype of an iris recognition

system. The task was to read in an image of an eye and display this on screen. From

7

this stage the iris then needs to be isolated from the rest of the image; to do this

accurately the pupil, iris, and eyelids all need to be identied. This isolation was

originally specifed to be carried out manually by the user by clicking points on the

image. The next task for the program was to calculate features of the iris pattern

using the algorithms described by Professor John Daugman. The iris bitcodes that are

obtained from this can then be used to compare against a database of other such

bitcodes to and the identity of the individual.

2.1.4 Iris Recognition Based on Using Ridgelet and Curvelet Transform

There are different methods for personal identification with using biometric

characteristics. In general, biometric is an individual identification ability based on

physiological characteristics such as fingerprint, handwriting, retina, iris and face.

There are many advantages of employing biometric system for identification but

there are also some disadvantages. We can mention to high recognition accuracy,

uniqueness, and no needs to memorize a code as advantages and low public

acceptance, and complex or expensive equipments as disadvantages. Any way the

advantages of using the biometric systems are more than its drawback, so using is

increasing daily. Although using iris patterns for personal identification have been

begun in the last 19th century, it takes more attention nowadays. In addition iris

recognition system is a noninvasive method.

2.1.5 PCA based Iris Recognition using DWT

There are many iris recognition systems and the first automatic system was

developed by Daugman using efficient integrodifferential operator, which is still

popular in todays most of the iris recognition systems. In the zero crossings

representation method the image is decomposed using wavelet transform and the

required features are extracted from the image.