A computationally efficient scheme for feature extraction with kernel discriminant analysis. (February 2016)

• Record Type: Journal Article
• Authors: Min, Hwang-Ki; Hou, Yuxi; Park, Sangwoo; Song, Iickho
• Is Part Of: Pattern recognition. Volume 50(2016:Feb.)
• Page Start: 45
• Page End: 55
• Publication Date: 2016-02
• Subjects: Kernel discriminant analysis -- Computational complexity -- Lagrange method -- Regularization -- Pattern recognition; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2015.08.021
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract: The kernel discriminant analysis (KDA), an extension of the linear discriminant analysis (LDA) and null space-based LDA into the kernel space, generally provides good pattern recognition (PR) performance for both small sample size (SSS) and non-SSS PR problems. Due to the eigen-decomposition technique adopted, however, the original scheme for the feature extraction with the KDA suffers from a high complexity burden. In this paper, we derive a transformation of the KDA into a linear equation problem, and propose a novel scheme for the feature extraction with the KDA. The proposed scheme is shown to provide us with a reduction of complexity without degradation of PR performance. In addition, to enhance the PR performance further, we address the incorporation of regularization into the proposed scheme. Abstract : Highlights: We propose a complexity-reduced scheme for the kernel discriminant analysis. The proposed scheme does not cause degradation of recognition accuracy performance. The core is the transformation of the original problem into a linear equation problem. We further address the incorporation of regularization into the proposed scheme.