An Algorithm of Iris Location Based on Gray Projection and Improved Hough Transform

Abstract

For the purpose of localizing iris precisely and quickly, a fast iris location algorithm using gray projection and Hough transform is presented based on the gray distribution features of eye image. First, threshold segmentation is applied to locate the pupil from original image of eyes. Then gray projection is used to locate the inner edge of iris and the edge information is extracted by enhancing the iris edge image. At last, the prior knowledge of pupil circle centre and radii combined with the improved Hough transform is utilized to locate the iris edge accurately. Experimental results show that this algorithm improves the locating speed with better locating results.