KERNEL GENERALIZED NONLINEAR DISCRIMINANT ANALYSIS ALGORITHM FOR PATTERN RECOGNITION (WA-P3)

Author(s) : Guang Dai (College of Computer Science, Zhejiang University, China) Yuntao Qian (College of Computer Science, Zhejiang University, China)

Abstract : Linear discriminant analysis (LDA) is a very effective tool used for dimensionality reduction and feature extraction in pattern recognition. However, the LDA is inadequate to describe the complex and nonlinear pattern. To solve this problem, kernel nonlinear discriminant analysis (K-NDA) has been proposed. Although successful in many cases, classic K-NDA also suffers from the small sample size problem (SSSP) and loses some discriminatory information as same as classic LDA. In this paper, a novel K-NDA, i.e., the kernel generalized nonlinear discriminant analysis (KG-NDA) algorithm is introduced to effectively overcome these problems, and it also views the optimal discriminant analysis vectors as a global transform in the feature space at a certain extent. It not only can deal with the nonlinear problem, but also can effectively solve the SSSP. The KG-NDA is the applied to the experiments on face recognition, and the results tested on two popular databases demonstrate that this method is very effective.

Menu