Iris Recognition Using Robust Algorithm for Eyelid, Eyelash and Shadow Avoiding
Zyad Thalji, Mutasem K. Alsmadi · 2013
Abstract: Eyelids, eyelashes and shadows are three major challenges for effective iris segmentation, which have not been adequately addressed in the current literature. In this paper, we present a novel method for Iris recognition (IR) based on color images. Firstly; extracting of the color values from Iris images using RGB color space, which is represented as vectors in the form of 3D-RGB color space. Secondly; determine the center pupil and external contour of iris circle. Thirdly; taking a sub Iris image to extracted the iris print, then merge the three sub image in one array to make the aggregation. Then, a smart prediction model is established to determine an appropriate threshold for eyelash and (eyelid or shadow) detection. Finally; Fourier Fast Transform (FFT) is used to recognize the Iris print. Our proposed method has been applied on 100 Iris images which obtained from CASIA iris images database. Experimental results on the challenging CASIA-iris image database demonstrate that the proposed method outperforms state-of-the-art methods in both accuracy and speed, where the overall accuracy results of the recognition rate and error rate are equal to 98.82%, 1.18 % respectively.