Iris localization using Daugman's algorithm

Abstract: Iris recognition system is a reliable and an accurate biometric system. Localization of the iris borders in an eye image can be considered as a vital step in the iris recognition process. There exist many algorithms to segment the iris. One of the segmentation methods, that is used in many commercial iris biometric systems is an algorithm known as a Daugman's algorithm. The aim of this thesis is to implement this algorithm using MATLAB programming environment. The implemented algorithm was tested on the eye images of different quality, such as the images with partly covered iris or low contrast images. The test results demonstrated that the Daugman’s algorithm detects the iris borders in the high quality images with high accuracy. The performance of the algorithm on the lower quality images has been improved by additional preprocessing of these images.