A Fast and Robust Method for Iris Localization

摘要

A novel method for localization of iris images is proposed in this paper. Especially the iris localization plays a vital role in an iris recognition system such that it increases both speed and performances of the recognition system to a great extent. The process of iris localization is used to detect an annular region between the pupil (black portion) and the sclera (white portion) of an eye. Here, for iris localization, we use a modified version of Circular Hough transform for both pupil and iris segmentation. Extensive experimental results have been performed to show that proposed algorithm has satisfying performance as compared to existing methods for iris localization. The performance of the proposed system is tested with IITD and MMU1 iris databases.