计算机工程与应用2019，Vol.55Issue(6)：126-132,7.DOI:10.3778/j.issn.1002-8331.1712-0148
基于结构化低秩恢复的鲁棒人脸识别算法
Robust Face Recognition Algorithm Based on Structured Low-Rank Recovery
摘要
Abstract
Due to the self-expressive property of data, the samples from same class have similar representations over a given dictionary, which means the representations of all samples possess block-diagonal structure. But in view of the exis-tence of various corruptions in face images, the sample’s subspace structure may be destroyed. In order to handle this problem, many low-rank representation based methods are proposed to recovery clear components of data, but only low-rank constraint can not transform the original training samples to ideal low-rank subspace perfectly. A Robust Structured Low-Rank Recovery algorithm(RSLRR)is proposed. RSLRR utilizes an ideal label matrix to promotethe low-rank repre-sentation tend to a block-diagonal structure which can excavate more potential structural information. Meanwhile, in order to alleviate the structural information loss caused by above strict label pursuing, another relax regularization term are applied to weaken the negative influence of off-block-diagonal coefficients. Afterwards, a discriminative and structured dictionary can be obtained by RSLRR algorithm, and then a low-rank projection matrix is computed to project all test sam-ples into its corresponding low-rank subspaces efficiently. Experimental results on AR and CMU PIE databases demon-strate the effectiveness and robustness of the proposed RSLRR algorithm.关键词
人脸识别/块对角结构/低秩恢复/子空间投影/特征提取Key words
face recognition/block-diagonal structure/low-rank recovery/subspace projection/feature extraction分类
信息技术与安全科学引用本文复制引用
陈哲,吴小俊..基于结构化低秩恢复的鲁棒人脸识别算法[J].计算机工程与应用,2019,55(6):126-132,7.基金项目
国家自然科学基金(No.71301104,No.71271138) （No.71301104,No.71271138）
高等学校博士学科点专项科研基金资助课题(No.20133120120002) （No.20133120120002）
上海市教育委员会科研创新项目(No.14YZ088) （No.14YZ088）
上海市一流学科项目(No.S1201YLXK) （No.S1201YLXK）
沪江基金(No.A14006). （No.A14006）
