DWT-based Feature Extraction and Radon Transform based Contrast Enhancement for Improved Iris Recognition

Iris based authentication system is essentially a pattern recognition technique that makes use of iris patterns, which are statistically unique, for the purpose of personal identification. In this paper, a novel method for recognition of iris patterns is proposed by using a combination of pre-processing using Radon transform and Top hat filtering, feature extraction using Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT), and feature selection using Binary Particle Swarm Optimization. Individual stages are examined and an attempt is made to improve each stage. Radon transform is used for detecting lines present in the iris textures and Top hat filtering is used for image enhancement. A combination of DWT and DCT is used to extract the salient iris features. A BPSO-based feature selection algorithm is used to search the feature vector space for the optimal feature subset. Experimental results on two benchmark iris databases, namely Phoenix and IITD, depict promising performance of the proposed techniques for iris recognition.