Study on Algorithm of Eyelash Occlusions Detection Based on Endpoint Identification

In iris recognition, eyelash occlusions will reduce the accuracy of iris recognition, an algorithm of eyelash occlusions detection based on endpoint identification is proposed in this paper. By analyzing the structural characteristics of eyelashes, this algorithm identifies the endpoints of eyelashes projection in the outer circularity of iris, and finally uses a sector region to represent the eyelash occlusions region. Removing the sector and the pupil, the rest of the outer circularity area is the effective region of iris. Experimental results demonstrate that the region of eyelash occlusions detected by the method in this paper can contain all eyelash pixels in the outer circularity region and the effective region of iris is "pure".