A New Approach in Face Recognition: Duplicating Facial Images Based on Correlation Study

Abstract

In this paper, we describe a new approach in face recognition by which the recognition accuracy can be increased substantially. In our approach a detail correlation study has been made on standard database such as Yale face database. Based on correlation coefficient, face images of individuals are duplicated in train set or test set or in both category. Then face databases are tested with standard face recognition algorithms such as Eigen face, Fischer face, KPCA, ICA and 2DPCA. In all these methods, arrangement of faces based on our approach gives better result. The software used to test all these algorithms is open source Scilab.