Linear Dimension Reduction Techniques
Linear dimension reduction technique reduces the dimension of biometric data using a linear transform. The linear transform is always learned by optimization of a criterion. Biometric data are then projected onto the range space of this transform. Subsequent processing will then be performed in that lower-dimensional space.
In biometrics, data are always represented in vectors and the dimensionality is always very high. It would be computationally expensive to process them directly by many algorithms. Moreover, it is sometimes desirable to exact robust, informative or discriminative information from the data. For these reasons, a lower-dimensional subspace is always found such that most important information of data is retained for linear representation. Among the techniques for learning a subspace, linear dimension reduction methods are always popular.
KeywordsLinear Discriminant Analysis Independent Component Analysis Dimension Reduction Independent Component Analysis Nonnegative Matrix Factorization
- 5.D.D. Lee, H.S. Seung, Algorithms for non-negative matrix factorization, in Advances in Neural Information Processing Systems, Denver, 2000, pp. 556–562Google Scholar
- 8.X. He, P. Niyogi, Locality preserving projections, in Advances in Neural Information Processing Systems, Vancouver, 2003, pp. 153–160Google Scholar
- 9.R.A., Fisher, The use of multiple measures in taxonomic problems. Ann. Eugen. 7, 179–188 (1936)Google Scholar
- 10.P.N. Belhumeour, J.P. Hespanha, D.J. Kriegman, Eigenfaces vs. fisherfaces: recognition using class specific linear projection. IEEE Trans. Pattern Anal. Mach. Intell. 19 (7), 711–720 (1997)Google Scholar
- 15.D. Lin, X. Tang, Inter-modality face recognition, in European Conference on Computer Vision, Graz, 2006Google Scholar
- 16.D. Cai, X. He, J. Han, Semi-supervised discriminant analysis, in IEEE International Conference on Computer Vision, Rio de Janeiro, 2007Google Scholar
- 17.J. Yang, D. Zhang, A.F. Frangi, J.y. Yang, Two-dimensional PCA: a new approach to appearance-based face representation and recognition. IEEE Trans. Pattern Anal. Mach. Intell. 26 (1), 131–137 (2004)Google Scholar
- 19.H. Kong, X. Li, L. Wang, E.K. Teoh, J.-G. Wang, R. Venkateswarlu, Generalized 2D principal component analysis, in IEEE International Joint Conference on Neural Networks, Oxford, vol. 1, 2005, pp. 108–113Google Scholar
- 20.J.P. Ye, R. Janardan, Q. Li, Two-dimensional linear discriminant analysis, in Advances in Neural Information Processing Systems, Vancouver, 2004, pp. 1569–1576Google Scholar