A fast and accurate iris localization method based on gray level statistics and region properties

In this paper, an image gray level statistics and region properties based iris localization method has been proposed. The proposed method starts by finding the coarse pupil region, reducing the effect of the eyelashes connected to pupil region, and tracing the boundary of pupil. Once the pupil is localized, we use a modified Intergrodifferential operator to localize the iris coarsely and finally use a small range Hough transform to get an accurate iris localization result. Experimental results show the superiority of the proposed method in comparison with some of the existing methods.