Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
531865 | Pattern Recognition | 2007 | 8 Pages |
Abstract
Iris segmentation is an important step for automatic iris recognition. This paper presents a new iris segmentation method for hand-held capture device. We use a geometrical method for pupil detection. The bottom point of pupil is used as the reference point for pupil localization because it is insensitive to pupil dilation and not affected by the top eyelid or eyelashes. To decrease computational cost, the outer (or limbus) boundary of iris is localized based on shrunk image using Hough transform and modified Canny edge detector. The lower part of iris pattern is used for recognition in order to reduce the occlusion by eyelashes and eyelids. Experimental results demonstrate that the proposed method has an encouraging performance.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Vision and Pattern Recognition
Authors
Xiaofu He, Pengfei Shi,