A Novel Null Space-Based Kernel Discriminant Analysis for Face Recognition
The symmetrical decomposition is a powerful method to extract features for image recognition. It reveals the significant discriminative information from the mirror image of symmetrical objects. In this paper, a novel null space kernel discriminant method based on the symmetrical method with a weighted fusion strategy is proposed for face recognition. It can effectively enhance the recognition performance and shares the advantages of Null-space, kernel and symmetrical methods. The experiment results on ORL database and FERET database demonstrate that the proposed method is effective and outperforms some existing subspace methods.
Keywordssymmetrical decomposition symmetrical null-space based kernel LDA weighted fusion strategy face recognition
- 1.Belhumeur, P.N., Hespanha, J.P., Kiregman, D.J.: Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection. IEEE Trans. on PAMI 19(7), 711–720 (1997)Google Scholar
- 2.Zhao, W., Chellappa, R., Philips, P.J.: Subspace Linear Discriminant Analysis for Face Recognition. Tech Report CAR-TR-914 Center for Automation Research, University of Maryland (1999)Google Scholar
- 6.Yang, M.H.: Kernel Eigenfaces vs. Kernel Fisherfaces: Face Recognition Using Kernel Methods. In: Int. Conf. on Automatic Face and Gesture Recognition, pp. 215–220 (2002)Google Scholar
- 7.Liu, W., Wang, Y.H., Li, S.Z., Tan, T.N.: Null Space-based Kernel Fisher Discriminant Analysis for Face Recognition. In: Int. Conf. on Automatic Face and Gesture Recognition, pp. 369–374 (2004)Google Scholar
- 8.Yang, Q., Ding, X.: Symmetrical PCA in Face Recognition. In: IEEE Int. Conf. on Image Processing, pp. 97–100 (2002)Google Scholar