An improved iris recognition system based on the fusion of the curvelet and DTCWT

Abstract

In recent years, iris recognition has been emerged as one of the most popular biometric techniques because it guarantees high universality, distinctiveness, permanence, collectability, performance, acceptability, circumvention. In the paper we propose an improved system for iris recognition with high accuracy by fusing curvelet and dual tree complex wavelet transform. In our system, the main features are extracted from pre-processed/normalized iris images using both curvelet and Dual Tree Complex Wavelet Transform (DTCWT) tranforms. After performing different classifiers independently, all the results are fused to get final classification in the decision level to increase the accuracy of system. Finally, the random forest classifier and CATIA dataset are used to measure the performance of the proposed method. The experimental results show that the technique of the paper based on fusion of the curvelet and DTCWT is promising when compared with other existing similar techniques.