Robust and Efficient Approach for Iris Recognition using Wavelet Transform

This paper presents robust and efficient approach for iris recognition using wavelet transform. Iris recognition is accepted as one of the most efficient biometric method. One of the difficult problems in feature based iris recognition is that the matching performance is significantly influenced by many parameters in feature extraction process, which may vary depending on environmental factors of image acquisition. We propose a efficient approach for improved iris recognition system with reduced false acceptance rate (FAR) and false rejection rate (FRR). The technique developed here uses db1, db4 and bior2.4 discrete wavelet transform for feature extraction of normalized iris and encoded the vertical and horizontal (LH3 & HL3) coefficients as iris code which reduces complexity of iris encoding and gives texture information in low and high frequency region. Experimental evaluation using different iris database such as CASIA (versions 1.0, 2.0 and 3.0), UBIRIS (versions 1.0), MMU (versions 1.0 and 2.0) clearly demonstrates that the use of wavelet filters makes it possible to achieve accurate and robust iris recognition and an effective matching performance of the proposed algorithm with lowest EER (0.2% & 0.19%) for MMU and UBIRIS database than other existing approaches.