caa a22 4500 60554123X CHVBK 20210128100915.0 cr unu---uuuuu 210128e20150501xx s 000 0 eng 10.1007/s00371-014-0991-9 doi (NATIONALLICENCE)springer-10.1007/s00371-014-0991-9 Efficient kernel discriminative common vectors for classification [Elektronische Daten] [Jianwei Zheng, Qiongfang Huang, Shengyong Chen, Wanliang Wang] Kernel discriminant analysis (KDA) which operates in the reproducing kernel Hilbert space (RKHS) is a very popular approach to dimensionality reduction. Kernel discriminative common vectors (KDCV) shares the same modified Fisher linear discriminant criterion with KDA and guarantees a 100% recognition rate for the training set samples as well as favorable generalization performance. However, KDCV has the disadvantage of high computational complexity in both the training and the testing stage. This paper attempts to improve the computation efficiency of KDCV by two strategies. First, the Cholesky decomposition is introduced to obtain the projection matrix instead of eigen-decomposition. Second, we replace the matrix operation with vector operation in the testing process which reduces the computational complexity. Extensive experiments on COIL images dataset, ORL faces dataset, PIE faces dataset, and USPS handwritten digits dataset demonstrate that the proposed algorithm is more efficient than the traditional KDCV algorithm without loss of accuracy. Springer-Verlag Berlin Heidelberg, 2014 Feature extraction nationallicence Kernel discriminative common vector nationallicence Cholesky decomposition nationallicence Kernel trick nationallicence Zheng Jianwei College of Computer Science and Technology, Zhejiang University of Technology, 288 Liuhe Road, 310023, Hangzhou, China aut Huang Qiongfang College of Computer Science and Technology, Zhejiang University of Technology, 288 Liuhe Road, 310023, Hangzhou, China aut Chen Shengyong College of Computer Science and Technology, Zhejiang University of Technology, 288 Liuhe Road, 310023, Hangzhou, China aut Wang Wanliang College of Computer Science and Technology, Zhejiang University of Technology, 288 Liuhe Road, 310023, Hangzhou, China aut The Visual Computer Springer Berlin Heidelberg 31/5(2015-05-01), 643-655 0178-2789 31:5<643 2015 31 371 https://doi.org/10.1007/s00371-014-0991-9 text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s00371-014-0991-9 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Zheng Jianwei College of Computer Science and Technology, Zhejiang University of Technology, 288 Liuhe Road, 310023, Hangzhou, China aut NATIONALLICENCE 700 1- Huang Qiongfang College of Computer Science and Technology, Zhejiang University of Technology, 288 Liuhe Road, 310023, Hangzhou, China aut NATIONALLICENCE 700 1- Chen Shengyong College of Computer Science and Technology, Zhejiang University of Technology, 288 Liuhe Road, 310023, Hangzhou, China aut NATIONALLICENCE 700 1- Wang Wanliang College of Computer Science and Technology, Zhejiang University of Technology, 288 Liuhe Road, 310023, Hangzhou, China aut NATIONALLICENCE 773 0- The Visual Computer Springer Berlin Heidelberg 31/5(2015-05-01), 643-655 0178-2789 31:5<643 2015 31 371