Affine invariant Iris identification using angular and radial partitioning

This paper presents a novel human identification system based on features obtained from iris images using angular and radial partitionings. The identification task in the proposed system is composed of two general stages including feature extraction and decision making. In the feature extraction stage, first all of the images are normalized in a preprocessing step. Then, the sketch (pattern) of the iris is extracted from iris images. Two feature vectors based on the angular and radial partitionings are extracted from the sketch image. In the next stage, the extracted feature vectors are analyzed using 1D discrete Fourier transform and the Manhattan metric is used to measure the closeness of the feature vectors to have a compression on them. Experimental results on a database, including 960 iris images obtained from 64 subjects, demonstrated an average true identification accuracy rate more than 98 percent for the proposed system. The identification task in the proposed system is rotation and scale invariant and robust against translation.