A new fuzzy linear mapping technique for facial feature extraction and recognition

Fast feature matching process and high recognition rates are two important issues in face recognition systems. In this paper, a fuzzy neural network based face recognition system which addresses on these two subjects is proposed. This face recognition system uses a fuzzy linear mapping for feature extraction and a fuzzy neural network for fuzzy feature matching. Experimental results show that the proposed face recognition system has two advantages: 1) higher recognition rates are achieved in comparison with statistical feature extraction methods; and 2) feature matching process using fuzzy neural network is faster than minimum distance classifier.