Discriminative analysis of direct linear discriminant analysis for face recognition

被引：0
|
作者：
Zhao W.-F. [1 ,2 ]
Shen H.-B. [1 ]
Yan X.-L. [1 ]
机构：
[1] Institute of VLSI Design, Zhejiang University
[2] Department of Information Science and Electronic Engineering, Zhejiang University
来源：
Zhejiang Daxue Xuebao (Gongxue Ban)/Journal of Zhejiang University (Engineering Science) | 2010年 / 44卷 / 08期
关键词：
Direct linear discriminant analysis (DLDA); Face recognition; Linear discriminant analysis (LDA); Principal components analysis (PCA); Small sample size;
D O I：
10.3785/j.issn.1008-973X.2010.08.008
中图分类号：
学科分类号：
摘要：
Direct linear discriminant analysis (DLDA) is claimed that it can take advantage of all the information, both within and outside of the within-class scatter matrix's null space. In order to analyze this claim's flaw in theory, the discriminative characteristics of DLDA for face recognition was studied. Since the optimal solution of DLDA is inside the range space of the between-class scatter matrix, a theoretical analysis was unfold via the following three aspects: the relationship between the within-class and between-class scatter matrix's range space, the relationship between the within-class scatter matrix's null space and the between-class scatter matrix's range space, the characteristics of DLDA under keeping all the discriminative vectors. The results show that: in undersampled cases DLDA nearly can not make use of the information inside the null space of the within-class scatter matrix, thus some discriminative information may be lost; DLDA is degenerated as PCA of the between-class scatter matrix with all nonzero principal components if it keeps the complete discriminative vectors found. The comparative results on the face database, ORL and YALE, indicate that DLDA is inferior to the other extensions of linear discriminant analysis in terms of recognition accuracy. Which is consistent with the theoretical analysis.
引用
收藏
页码：1479 / 1483
页数：4
相关论文
共 12 条
  • [1] Krzanowski W.J., Jonathan P., Mccarthy W.V., Et al., Discriminant analysis with singular covariance matrices: Methods and applications to spectroscopic data, Applied Statistics, 44, 11, pp. 101-115, (1995)
  • [2] Yu H., Yang J., A direct LDA algorithm for high-dimensional data-with application to face recognition, Pattern Recognition, 34, 10, pp. 2067-2070, (2001)
  • [3] Lu J., Plataniotis K.N., Venetsanopoulos A.N., Regularization studies of linear discriminant analysis in small sample size scenarios with application to face recognition, Pattern Recognition Letters, 26, 2, pp. 181-191, (2005)
  • [4] Zheng Y.-J., Guo Z.-B., Yang J., Et al., DLDA/QR: A robust direct LDA algorithm for face recognition and its theoretical foundation, Lecture Notes in Computer Science, 4426, pp. 379-387, (2007)
  • [5] Li X.-R., Zhao G.-Z., Zhao L.-Y., Improved kernel direct Fisher discriminant analysis and face recognition, Journal of Zhejiang University: Engineering Science, 42, 4, pp. 583-589, (2008)
  • [6] Huang R., Liu Q., Lu H., Et al., Solving the small sample size problem of LDA, Proceedings of International Conference on Pattern Recognition, 3, pp. 29-32, (2002)
  • [7] Cevikalp H., Neamtu M., Wilkes M., Et al., Discriminative common vector method with kernels, IEEE Transactions on neural networks, 17, 6, pp. 1550-1565, (2006)
  • [8] Park H., Park C.H., A comparison of generalized linear discriminant analysis algorithms, Pattern Recognition, 41, 3, pp. 1083-1097, (2008)
  • [9] Gao H., Davis J.W., Why direct LDA is not equivalent to LDA, Pattern Recognition, 39, 5, pp. 1002-1006, (2006)
  • [10] Ye J.-P., Xiong T., Computational and theoretical analysis of null space and orthogonal linear discriminant analysis, Journal of Machine Learning Research, 7, pp. 1183-1204, (2006)
12