We consider the problem of fusing colour information to enhance the performance of a face authentication system. The discriminatory information potential of a vast range of colour spaces is investigated. The verification process is based on the normalised correlation in an LDA feature space. A sequential search approach which is in principle similar to the "plus L and take away R" algorithm is applied in order to find an optimum subset of the colour spaces. The colour based classifiers are combined using the SVM classifier. We show that by fusing colour information using the proposed method, the resulting decision making scheme considerably outperforms the intensity based verification system.
展开▼
