Iris localization by means of adaptive thresholding and Circular Hough Transform

In this paper, a new iris localization method is presented for mobile devices. The proposed system uses both the intensity and saturation thresholdings on the captured eye images to determine the iris boundary and sclera area, respectively. The estimated iris boundary pixels placed outside the sclera are removed. The remaining pixels are mainly the iris boundary inside the sclera. Then a circular Hough transform is applied to such iris boundary pixels in order to localize the iris. The experiments are carried out on 60 iris images taken by an HTC mobile device from 10 different persons with both the left and right eye images available per person. We also evaluate the proposed algorithm on the MICHE datasets for iphone5, Samsung Galaxy S4, and Samsung Galaxy Tab2. The evaluation results obtained show that the developed system can successfully localize iris on the tested images.