In this paper we have presented a fusion of three face recognizers, LFA, LBA and KDDA, -which combines the three Confidence Measure Factors (CMFs) using a RBF neural netwrok. This strategy is used to increase the accuracy of the face recognition system. The recognizer is tested on AT&T-ORL, AR and IITK Face database and the results are found to be more than 95%. © 2004 IEEE.
|Journal||Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics|