High Speed Method for Eyelid Segmentation in Iris Image
Wang Xiu-xin · Journal of Engineering Graphics · 2009
Some part of iris is usually occluded by the eyelid.In order to ensure high recognition rate and remove the disturbing factor in iris image,an algorithm of segment detection is proposed to detect eyelid and remove it.Firstly,the circular arcs approximation and the dynamic clustering of iris circular arcs are proceeded;secondly,the circular arcs are approximated again after dynamic clustering.Experiments show that the algorithm is faster and more precise than Daugman’s Algorithm for Circular Detection and the Hough Transform under same conditions.The recognition experimental results show that the approach is helpful to improve pattern classification effect and recognition reliability.