Weighted PCA space and its application in face recognition

In this paper, we propose a new PCA based subspace approach for pattern recognition. The conventional PCA feature space is first converted to a WPCA feature space with unit variance by weighting the features and then face recognition is performed in the new space. Detailed theoretical derivation and analysis are presented and simulation results on AR and ORL face databases are given. The simulation results indicate that the proposed approach is superior to conventional PCA approach in recognition accuracy under the same computation complexity.