A Hybridization of Fuzzy Logic and Deterministic Learning Machine for Face Recognition

Abstract

In this chapter, a new method for face recognition (FR) is proposed by integrating fuzzy logic in deterministic learning machine called fuzzy deterministic learning machine (FDLM). The main steps of the proposed approach are the fuzzification and classification step. The fuzzification step is done using π-membership function (MF) to map the grades of association of each input feature to each subject in a fuzzy matrix form and classification is done with the help of fast learning, parameter-free deterministic learning machine. To show the efficacy and superiority of the proposed approach, we have conducted a comparison analysis of the proposed approach with other methods available in the literature on Georgia Tech face database. Experimental results obtained reveal that the proposed approach shows a significant improvement in recognition performance for FR.