A COMPARATIVE STUDY OF DISTANCE METRICS USED IN FACE RECOGNITION

Face recognition(FR) has become a specialized application area within the field of computer, among which the appearance-based approaches are popular. The appearance-based FR can be induced into the subspace projection step and a nearest neighbor classifier. Because face image data have the property of high dimension, the subspace projection methods should be employed for dimensionality reduction. They include Principal component analysis (PCA), Linear discriminant analysis (LDA), Independent component analysis (ICA) and Non-negative matrix factorization (NMF). In this paper, all the subspace projection algorithms are explored and face recognition system based on the four techniques presented. The simulation experiments are implemented with ORL face dataset, the results presented and analyzed. Also the four techniques are compared in recognition performance for FR.