On improvement for normalizing iris region for a ubiquitous computing

Abstract

Iris patterns are believed to be an important class of biometrics suitable for subject verification and identification applications. An efficient approach for iris recognition through an iris region normalization is presented in this paper. An efficient iris region normalization consists of a doubly polar coordinate and noise region exclude. From this evaluation, we obtain iris code of small size and very high recognition rate. This effort is intended to enable a human authentication in small embedded systems, such as an integrated circuit card (smart cards). © Springer-Verlag Berlin Heidelberg 2005.