Abstract：Recently proposed marginal fisher analysis (MFA) has obtained better classification results than the traditional linear discriminant analysis (LDA). Based on the separability criterion of MFA, the orthogonal and uncorrelated restrictions are imposed on the base-vectors in this paper. An iterative algorithm for the proposed methods is given and it is proved theoretically that the separability of the proposed methods is better than that of the original MFA. Finally, experimental results on ORL and Yale databases validate the effectiveness of the proposed methods.
[1] Fukunaga K. Introduction to Statistical Pattern Recognition. 2nd Edition. Boston, USA: Academic Press, 1990 [2] Turk M, Pentland A. Eigenfaces for Recognition. Journal of Cognitive Neuroscience, 1991, 3(1): 71-86 [3] Swets D L, Weng J. Using Discriminant Eigenfeatures for Image Retrieval. IEEE Trans on Pattern Analysis and Machine Intelligence, 1996, 18(8): 831-836 [4] Belhumeur P N, Hespanda J, Kiregeman D. Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection. IEEE Trans on Pattern Analysis and Machine Intelligence, 1997, 19(7): 711-720 [5] Loog M, Duin R P W, Haeb-Umbach R. Multiclass Linear Dimension Reduction by Weighted Pairwise Fisher Criteria. IEEE Trans on Pattern Analysis and Machine Intelligence, 2001, 23(7): 762-766 [6] He Xiaofei, Yan Shuicheng, Hu Yuxiao, et al. Face Recognition Using Laplacianfaces. IEEE Trans on Pattern Analysis and Machine Intelligence, 2005, 27(3): 328-340 [7] Yan Shuicheng, Xu Dong, Zhang Benyu, et al. Graph Embedding and Extensions: A General Framework for Dimensionality Reduction. IEEE Trans on Pattern Analysis and Machine Intelligence, 2007, 29(1): 40-51 [8] Yang Jian, Zhang D, Yang Jingyu, et al. Globally Maximizing, Locally Minimizing: Unsupervised Discriminant Projection with Applications to Face and Palm Biometrics. IEEE Trans on Pattern Analysis and Machine Intelligence, 2007, 29(4): 650-664 [9] Duchene J, Leclercq S. An Optimal Transformation for Discriminant and Principal Component Analysis. IEEE Trans on Pattern Analysis and Machine Intelligence, 1988, 10(6): 978-983 [10] Foley D H, Jr Sammon J W. An Optimal Set of Discriminant Vectors. IEEE Trans on Computers, 1975, 24(3): 281-289 [11] Jin Zhong, Yang Jingyu, Hu Zhongshan, et al. Face Recognition Based on the Uncorrelated Discriminant Transformation. Pattern Recognition, 2001, 34(1): 1405-1416 [12] Xu Yong, Yang Jingyu, Jin Zhong. Theory Analysis on FSLDA and ULDA. Pattern Recognition, 2003, 36(12): 3031-3033 [13] Cai Deng, He Xiaofei, Han Jiawei, et al. Orthogonal Laplacianfaces for Face Recognition. IEEE Trans on Image Processing, 2006, 15(11): 3608-3614 [14] Jing Zhong, Yang Jingyu, Lu Jianfeng. Effective Extraction of Optimal Discriminant Features and the Dimension Problem. Chinese Journal of Computers, 2000, 23(1): 108-112 (in Chinese) (金 忠,杨静宇,陆建峰.有效最佳鉴别特征的抽取与维数问题.计算机学报, 2000, 23(1): 108-112) [15] Zhang Xianda. Matrix Analysis and Application. Beijing, China: Tsinghua University Press, 1999 (in Chinese)
2008, Vol. 21 
