Journal of Mathematical Sciences: Advances and ApplicationsAuthor's: Jianqiang Gao and Lizhong Xu
Pages: [75] - [97]
Received Date: May 31, 2014
Submitted by:
In this paper, a kernel fuzzy discriminant analysis minimum distance-based approach for the classification of face images is proposed to deal with face classification problem (we call this method mdkfda/qr as an abbreviation). A superiority of the mdkfda/qr is its computational efficiency and can avoid the singularity. In the proposed method, the membership degree is incorporated into the definition of between-class and within-class scatter matrices to get fuzzy between-class and within-class scatter matrices. The mdkfda/qr approach was compared with kernel discriminant analysis (KDA) and fuzzy discriminant analysis (FDA) two algorithms in terms of classification accuracy. Experiments on ORL and FERET, two real face datasets are performed to test and evaluate the effectiveness of the proposed algorithm on classification accuracy. The results show that the effect of mdkfda/qr method can achieve higher classification accuracy than KDA and FDA methods.
kernel fuzzy discriminant analysis, fuzzy membership, QR decomposition, classification accuracy, mdkfda/qr.
