文献类型：期刊文献

英文题名：Local sparse representation projections for face recognition

作者：Lai, Zhihui[1,2];Li, Yajing[3];Wan, Minghua[2,4];Jin, Zhong[2]

机构：[1]Harbin Inst Technol, Biocomp Res Ctr, Shenzhen Grad Sch, Shenzhen 518055, Peoples R China;[2]Nanjing Univ Sci & Technol, Sch Comp Sci, Nanjing 210094, Jiangsu, Peoples R China;[3]E China Univ Sci & Technol, Sch Informat Sci & Engn, Shanghai 200237, Peoples R China;[4]Nanchang Hangkong Univ, Sch Informat Engn, Nanchang 330063, Jianxi, Peoples R China

年份：2013

卷号：23

期号：7-8

起止页码：2231

外文期刊名：NEURAL COMPUTING & APPLICATIONS

收录：;EI(收录号：20134716999529);WOS:【SCI-EXPANDED(收录号:WOS:000326889800043)】；

基金：This work is partially supported by the Natural Science Foundation of China under grant No. 61203376, 61203243, 61005005, 61005008, 61105054, Hi-Tech Research and Development Program of China under grant No. 2006AA01Z119 and China Post-doctoral Science Foundation funded project 2012M511479.

语种：英文

外文关键词：Sparse representation; Manifold learning; Dimensionality reduction; Feature extraction

摘要：How to define the sparse affinity weight matrices is still an open problem in existing manifold learning algorithm. In this paper, we propose a novel supervised learning method called local sparse representation projections (LSRP) for linear dimensionality reduction. Differing from sparsity preserving projections (SPP) and the recent manifold learning methods such as locality preserving projections (LPP), LSRP introduces the local sparse representation information into the objective function. Although there are no labels used in the local sparse representation, it still can provide better measure coefficients and significant discriminant abilities. By combining the local interclass neighborhood relationships and sparse representation information, LSRP aims to preserve the local sparse reconstructive relationships of the data and simultaneously maximize the interclass separability. Comprehensive comparison and extensive experiments show that LSRP achieves higher recognition rates than principle component analysis, linear discriminant analysis and the state-of-the-art techniques such as LPP, SPP and maximum variance projections.