Robust iris recognition algorithm for non-cooperative environment

Non-cooperative iris recognition system results in blurred and noisy images that can degrade the performance of the system. This paper proposes a novel feature extraction algorithm for non-cooperative environment. The proposed iris subdivision method reduces the effect of noise due to non-cooperative behaviour in an authentication system. Optimal projection analysis is proposed to obtain the rotation invariant features from the directional subband coefficients of the normalised image block. A fusion algorithm combines the matching scores from individual subimages to reduce the false rejection rate. The performance of the proposed method is compared with other algorithms on UBIRIS iris image database.