Biometrics and Biometric Fingerprint Recognition Essay examples

Submitted By Sindhu-Vasamsetti
Words: 2901
Pages: 12

Optimization of biometric Fingerprint Recognition parameters using Genetic Algorithms

Report submitted for

CPSC - 6126
Fall 2014
Term Paper
By
Krishna Sindhuri Nagavolu
13th December, 2014

Optimization of biometric fingerprint recognition parameters using Genetic Algorithms

Abstract
This research paper discusses about parameter optimization for biometric fingerprint recognition with the use of genetic algorithms. An accurate access control system is very important in domains like identification checks at airports, government organizations, FBI’s, scientific working groups like
NASA, US defense department and driver’s license office. The main reason these organizations prefer Biometrics is because it measures physiological characteristics like fingerprints, iris patterns, facial recognition, retina recognition, ear recognition and DNA analysis. Based on the features of the biometric sensors used, the system can detect whether a person is an authorized user or not. Though there are many other methods for identification, biometric sensors are considered to be very reliable and accurate. The main objective of this paper is to build a fingerprint recognizer that is fully automatic and can minimize the errors caused while matching the fingerprints.
Keywords: Biometrics, genetic algorithms, Parameter optimization, fingerprints 2|Page

Optimization of biometric fingerprint recognition parameters using Genetic Algorithms

Table of Contents

Page Number

Abstract ……………………………………………………………………………..2

1. Introduction…………………………………………………………………………4

2. Genetic Algorithms……………………………………………………………..5

3. Design of Genetic Optimizer……………………………………………….7

4. Experimental Setup……………………………………………………………9

5. Proposed Solution……………………………………………………………11

6. Conclusion…………………………………………………………………………12

7. Acknowledgments and References……………………………………13

3|Page

Optimization of biometric fingerprint recognition parameters using Genetic Algorithms 1.

Introduction

Traditional identification methods have several disadvantages like identity theft or hacking and biometric methods are considered to be one of the solutions to overcome these disadvantages. Biometrics is a science of establishing the identity of an individual based on his or her physical or behavioral patterns. The most common and well-known biometric identification system used is fingerprint recognition. Although there are many other biometric identification systems attracting the modern world, fingerprint recognition is considered as a standard model. They are widely used because of their individuality and immutability.
The main disadvantages of biometrics are fraud rate, insult rate and equal error rate (EER). Fraud rate is the rate at which a biometric system mistakenly authenticates a person as a user and insult rate is the rate at which the system mistakenly does not authenticate the correct user… When the fraud rate and the error rate are same, then it is said to be equal error rate (EER). In other words EER can be determined as a point where false reject rates are equal to false accept rates. This article provides a method where the biometric system uses genetic algorithms for fingerprint recognition that has very low EER and is more robust. For more information on Biometrics refer [3].
Optimization is a process of making a system as effective as possible.