Integration of wavelet packet decomposition and random secret in iris verification

This paper presents a novel method of integrating the user-specific secret pseudo-random number and iris feature extracted from wavelet packet transform as a feature vector for iris verification. The proposed method helps to improve the traditional iris verification system which depends solely on iris feature. The security risks such as iris fabrication can be avoided through the token replacement. Experimental results show that the proposed method has an encouraging performance. The optimum result is obtained for mean scheme wavelet packet transform with secret pseudo-random number using 2 levels of decomposition, in which an excellent verification rate is produced.