Face Recognition Based on WKPCA

Principal Component Analysis is acknowledged one of the most powerful techniques for feature extraction. As a nonlinear form of Principal Component Analysis, Kernel Principal Component Analysis has been applied abroad in face recognition in recent years. However, the principal component analysis and kernel principal component analysis treat each dimension feature of face images as equals in feature extraction. But in fact, different features play different roles in face recognition. In this paper, we adopt a Gaussian distribution function as a weight function, which can give prominence to the key features in face recognition. The proposed method is combined with the kernel principal component analysis to calculate the weighted subspace. Experimental results on the normal ORL face database show the proposed method is effective.