Colour feature selection for an unconstrained iris recognition system

Abstract

We consider the problem of fusing colour information to enhance the performance of an iris authentication system. The discriminatory information potential of a vast range of colour channels is investigated. The verification process is based on open-source implementation made by Libor Masek. A sequential search approach was used which is similar to the 'plus L and take away R' algorithm that is applied in order to find an optimum subset of the colour spaces. The colour based classifiers are combined using the SVM classifier to find optimum colour features. We show that by fusing colour information using the proposed method, the resulting decision making scheme considerably outperforms the intensity based verification system.