Iris Localization and Segmentation for Less Constrained and Non Ideal Iris Images

A comprehensive review of iris localization and segmentation techniques for challenging iris images

Abstract

Recognition of iris is one of the fastest identity verification systems. The accuracy of the iris recognition system is also better in comparison of other biometric traits like fingerprint, voice, hand etc. Because of its increased accuracy and uniqueness, it is used in areas which are very security sensitive, medical field, industrial area. The most important process in iris recognition is iris localization because it extracts part of iris which is valid. Iris images are often acquired under less constrained environment which includes near infrared illumination, bad lightning conditions, reflection. The images captured also have noise due to eyelids, eyelashes, eyeglasses. This creates the problem in iris localization and hence decreases accuracy. The goal of this paper is to discuss the literature of recently proposed methods in iris localization and also to compare the performance of these techniques.