Retina Identification Based on the Blood Vessels Sketch Using Zernike Moments

A new human identification system based on features obtained from retina images using Zernike moments and wavelet transform is presented in this paper. The proposed algorithm is composed of two major steps including feature extraction and decision making. In the feature extraction stage, first each image is normalized in a preprocessing step. Then 1 level of discrete wavelet transform is applied on the preprocessed images to produce wavelet image. The blood vessels‘ pattern is then extracted from the lowest frequency subband of the wavelet image. After vessels‘ pattern extraction, 2 orders of Zernike moments are extracted from vessels‘ pattern as image feature vectors. The extracted feature vectors are rotation and scale invariant and robust against translation. In the next stage, a fuzzy decision making approach with Manhattan distances of two feature vectors as input and similarity measure as output has is used to make the final decision. Experimental results on a database, including 360 retina images obtained from 40 subjects, demonstrated an average true identification accuracy rate equal to 96 percent for the proposed system.

Keywords:

Biometrics , retina identification , Zernike moments , wavelet transform , fuzzy logic