A New Localization Method for Iris Recognition Based on Angular Integral Projection Function

Abstract

Iris recognition is one of the most reliable and accurate biometric technologies. Iris localization is crucial for the performance of an iris recognition system, it includes finding the iris boundaries (inner and outer) and the eyelids (lower and upper). In this paper, we propose an efficient iris localization method based on the angular integral projection function (AIPF) to detect the iris boundaries in iris images. The proposed algorithm adopts boundary points detection and curve fitting. First, the approximate pupil center is obtained. Then, two sets of radial boundary points are detected for the iris inner and outer boundaries using AIPF method. Finally, we get the iris boundaries by fitting a circle for each of the above boundary points set. In the recognition stage, we used 2D Gabor filter to extract the iris code for the normalized iris image. At last, the proposed algorithm was tested on CASIA V1.0 iris images database. We evaluate the performance based on the analysis of both False Accept Rate(FAR) and False Reject Rate(FRR) curves. Experimental results show that the proposed iris localization algorithm is efficient and improves iris recognition.