Automatic iris segmentation and eyelids and eyelashes detection
Xu, GZ, ZF Zhang, YD Ma, Yadong Shao, Xu, GZ (reprint author), Lanzhou Univ, Sch Informat Sci & Engn, Lanzhou 730000, Peoples R China. · Lanzhou University Institutional Repository · 2006
Reliable segmentation of iris is a necessary precondition for any iris-based biometric identification system. It is also important to isolate eyelids and eyelashes from a segmented iris image because they degrade the performance of an iris recognition system. In this paper a novel and robust method for automatic iris segmentation and eyelids and eyelashes detection is described. Such method is composed of three main parts. (a) Find the local rectangle region with the minimum intensity mean and extend it to locate pupil in an ingenious way. (b) Select two small sectors crossing over the iris outer boundary from the left and right part of iris image to locate outer iris respectively. (c) Divide the segmented iris into 26 blocks and choose the blocks with larruping intensity mean as eyelids and eyelashes. The method was implemented and tested in two iris database sets, i.e. CASIA and MMU Iris databases, with different contrast quality. The experimental results show that the proposed method is very effective for iris recognition.