Design and Implementation of Iris Pattern Recognition Based on Wireless Network Systems

Table of Contents

1. Chapter One : Introduction

1.1 General Introduction

1.2 Literature Survey

1.3 Aim of the Work

1.4 Thesis Layout

2. Chapter Two : Theoretical Background

2.1 Introduction

2.2 Biometrics and Human Iris

2.3 Iris Recognition System

2.3.1 Capturing Sample

2.3.2 Iris Segmentation

2.3.3 Iris Normalization

2.3.4 Feature Encoding

2.3.4.1 1D Log-Gabor Wavelets

2.3.4.2 Genetic Algorithm

2.3.5 Pattern Matching

2.3.6 Iris Recognition Preformance Measurements

2.4 Image Compression

2.4.1 Image Compression Using PCA

2.4.2 Image Quality Measurements

2.5 Wireless Network System

2.5.1 Wireless Network Architectures

2.5.2 Network Communication

2.5.3 Network Monitoring

3. Chapter Three : Hardware and Software Design

3.1 General Introduction

3.2 General System Architecture

3.3 Input Data

3.4 Main Components of the Iris Recognition System

3.4.1 Iris Segmentaion

3.4.2 Iris Normalization

3.4.3 Feature Extraction

3.4.3.1 1D Log-Gabor Wavelets

3.4.3.2 Genetic Algorithm Optimization Method

3.4.4 Template Matching

3.5 Image Compression/Decompression

3.5.1 One-Stage Image Compression

3.5.2 The Proposed Multi-Stage Image Compression

3.6 Wireless Network

3.6.1 Proposed Network’s Hardware Design

3.6.2 Network’s Software Design

3.7 Decision Making

4. Chapter Four :System Implementation Results and Discussion

4.1 Introduction

4.2 System Requirement

4.3 Iris Pattern Recognition Methods Comparison

4.4 Image Compression/Decompression

4.4.1 One-Stage Image Compression/Decompression

4.4.2 Multi-Stage Image Compression/Decompression

4.5 Wireless Network

5. Chapter Five : Conclusions and Suggestions for Future Work

5.1 Conclusions

5.2 Suggestions for Future Work

Objectives and Research Themes

The primary objective of this thesis is to design and implement a high-speed, accurate iris pattern recognition system integrated into a wireless network environment. The research addresses the challenges of transmitting biometric data over bandwidth-constrained wireless channels by developing efficient image compression techniques, while simultaneously improving recognition performance through an optimized genetic algorithm compared against the traditional Libor Masek approach.

• Development of GUI-based iris recognition systems using both Libor Masek and Genetic algorithms in MATLAB.
• Implementation of image compression and multi-stage decompression to optimize data transmission over wireless links.
• Design and configuration of a wireless network architecture utilizing TCP/IP client-server sockets for image transfer.
• Comparative analysis of recognition accuracy, execution time, and network transmission performance.

Excerpt from the Book

Summary of Chapters

Chapter One : Introduction: Provides an overview of biometrics and iris recognition, highlighting the motivation for integrating wireless network technology and the need for data compression in distributed systems.

Chapter Two : Theoretical Background: Covers the fundamental concepts of biometric systems, the human iris, iris recognition stages, image compression using PCA, and wireless network communication protocols.

Chapter Three : Hardware and Software Design: Details the architecture of the proposed system, including the implementation of the iris recognition algorithms, the design of the wireless network using MATLAB, and the multi-stage compression technique.

Chapter Four :System Implementation Results and Discussion: Presents the experimental outcomes, comparing the performance of Libor Masek and genetic algorithms, evaluating compression metrics, and assessing the network transmission efficiency.

Chapter Five : Conclusions and Suggestions for Future Work: Summarizes the key findings of the research, confirming the superiority of the genetic algorithm in speed and accuracy, and proposes future enhancements such as mesh networking and hybrid recognition methods.

Keywords

Iris Recognition, Genetic Algorithm, Libor Masek, Principal Component Analysis (PCA), Wireless Network, TCP/IP, Image Compression, Biometrics, Network Monitoring, MATLAB, Hamming Distance, Multi-stage Compression, Network Architecture, Performance Measurement, Data Transmission.

Frequently Asked Questions

What is the core focus of this thesis?

This thesis focuses on designing and implementing an efficient iris pattern recognition system that operates over a wireless network, incorporating data compression to ensure fast and secure transmission.

What are the primary thematic areas covered in this work?

The research spans three major domains: iris pattern recognition algorithms, digital image compression techniques (specifically PCA), and wireless network system design using TCP/IP protocols.

What is the main goal or research question of this study?

The goal is to enhance iris recognition performance while solving the data transmission overhead in distributed biometric systems by combining advanced recognition algorithms with multi-stage image compression.

Which scientific methods are utilized for recognition?

The research utilizes two distinct methods: the enhanced Libor Masek algorithm and a custom-developed, optimized genetic algorithm, both implemented and compared within a MATLAB environment.

What topics are discussed in the main body of the work?

The main body covers the theoretical basis of iris recognition, system hardware and software design, the implementation of PCA-based compression, and the detailed results of the comparison between the two recognition methods in terms of accuracy and execution time.

Which specific keywords define the scope of this work?

Key terms include Iris Recognition, Genetic Algorithm, PCA, Wireless Network, TCP/IP, Image Compression, and biometric performance metrics such as FAR and FRR.

How does the proposed multi-stage image compression enhance network security?

The multi-stage compression makes it computationally difficult for unauthorized parties (hackers) to determine the original state or the compression levels applied to the iris image, thereby protecting data integrity during network transit.

Why was the genetic algorithm found to be superior in this study?

Experimental results indicated that the genetic algorithm achieved faster recognition times and maintained lower error rates compared to the traditional Libor Masek method when testing against the CASIA database.