Supervised sparsity preserving projections for face feature extraction

CORC > 兰州理工大学 > 兰州理工大学

	Supervised sparsity preserving projections for face feature extraction
	Ding, Zhengjian; Dai, Zhiyang
刊名	Journal of Computational Information Systems
	2012-10-01
卷号	8 期号:19 页码:7947-7954
关键词	Discriminant analysis Feature extraction Dimensionality reduction method Discriminating power Discriminative information Face database Face feature extraction Objective functions Sparse representation Sparsity preserving projections Supervised Supervised algorithm Weight matrices
ISSN号	15539105
英文摘要	Sparsity preserving projections (SPP) is a recently presented unsupervised dimensionality reduction method. In this paper, we propose two supervised algorithms for extending SPP named sparsity discriminant analysis (SDA) and sparsity preserving projections with discriminative information (DISPP), respectively. Both methods firstly extract within-class sparse representation matrix and between-class sparse representation matrix from the sparse reconstructive weight matrix (SRWM). Then, SDA seeks a transformed space to minimize the within-class reconstructive errors and simultaneously maximize the between-class reconstructive errors. DISPP constructs a new weight matrix, which fuses more withinclass representation information and less between-class representation information, to replace the SRWM in its objective function, thus boosting the discriminating power. Experiments based on Yale and AR face databases demonstrate the effectiveness of the proposed methods. © 2012 Binary Information Press.
语种	英语
出版者	Binary Information Press, P.O. Box 162, Bethel, CT 06801-0162, United States
内容类型	期刊论文
源URL	[http://ir.lut.edu.cn/handle/2XXMBERH/111117]
专题	兰州理工大学
作者单位	School of Computer and Communication, Lanzhou University of Technology, Lanzhou 730050, China
推荐引用方式 GB/T 7714	Ding, Zhengjian,Dai, Zhiyang. Supervised sparsity preserving projections for face feature extraction[J]. Journal of Computational Information Systems,2012,8(19):7947-7954.
APA	Ding, Zhengjian,&Dai, Zhiyang.(2012).Supervised sparsity preserving projections for face feature extraction.Journal of Computational Information Systems,8(19),7947-7954.
MLA	Ding, Zhengjian,et al."Supervised sparsity preserving projections for face feature extraction".Journal of Computational Information Systems 8.19(2012):7947-7954.