·462· 智能系统学报 第14卷 染，有效减少了样本中的污染对实验结果造成的 [J].Neurocomputing,2015,168:70-80. 负面影响。因此SWLRR算法样本在有无遮挡的 [12]CANDES E J,LI Xiaodong,MA Yi,et al.Robust prin- 情况下都有着较为突出的性能。在AR、Extended cipal component analysis[J].Journal of the ACM,2011, Yale B、CMU PIE数据库上的实验验证了SWLRR 58(311. 算法在不同情况下的有效性和对噪声的鲁棒性。 [13]LIU Guangcan,LIN Zhouchen,YAN Shuicheng,et al. 下一步将尝试将本文思想与其他解决样本中存在 Robust recovery of subspace structures by low-rank rep- 污染的算法结合，进一步提升算法性能。 resentation[J].IEEE transactions on pattern analysis and machine intelligence,2013,35(1):171-184. 参考文献： [14]MA Long,WANG Chunheng,XIAO Baihua,et al.Sparse representation for face recognition based on discriminat- [1]TURK M,PENTLAND A.Eigenfaces for recognition[J]. ive low-rank dictionary learning[C]//Proceedings of 2012 Journal of cognitive neuroscience,1991,3(1):71-86. IEEE Conference on Computer Vision and Pattern Recog- [2]BELHUMEUR P N.HESPANHA J P.KRIEGMAN D J. nition.Providence,USA,2012:2586-2593 Eigenfaces vs.fisherfaces:recognition using class specific [15]ZHANG Yangmuzi,JIANG Zhuolin,DAVIS L S.Learn- linear projection[J].IEEE transactions on pattern analysis ing structured low-rank representations for image classi- and machine intelligence,997,19(7):711-720. fication[C]//Proceeding of 2013 IEEE Conference on [3]LIU Chengjun,WECHSLER H.Gabor feature based clas- Computer Vision and Pattern Recognition.Portland, sification using the enhanced fisher linear discriminant USA,2013:676-683. model for face recognition[J].IEEE transactions on image [16]NGUYEN H,YANG Wankou,SHENG Biyun,et al.Dis- processing,.2002,11(4):467-476. criminative low-rank dictionary learning for face recogni- [4]HE Xiaofei,YAN Shuicheng,HU Yuxiao,et al.Face re- tion[J].Neurocomputing,2016,173(3):541-551 cognition using Laplacianfaces[J].IEEE transactions on [17]CHANG Heyou,ZHENG Hao.Weighted discriminative pattern analysis and machine intelligence,2005,27(3): dictionary learning based on lowrank representation 328-340. [J].Journal of physics:conference series,2017,90: [5]WRIGHT J,YANG A Y,GANESH A,et al.Robust face 012009. recognition via sparse representation[J].IEEE transactions [18]ZHANG Zheng,XU Yong,SHAO Ling,et al.Discrimin- on pattern analysis and machine intelligence,2009,31(2): 210-227 ative block-diagonal representation learning for image re- [6]MIN R,DUGELAY J L.Improved combination of LBP cognition[J].IEEE transactions on neural networks and and sparse representation based classification (SRC)for 1 earning systems,,2018,29(7):3111-3125. face recognition[C]//Proceedings of 2011 IEEE ICME. [19]CHEN Jie,YI Zhang.Sparse representation for face re- Barcelona,Spain,2011:1-6. cognition by discriminative low-rank matrix recovery[J]. [7]CHEN C F,WEI C P.WANG Y C F.Low-rank matrix re- Journal of visual communication and image representa- covery with structural incoherence for robust face recogni- tion,2014,25(5):763-773. tion[C]//Proceedings of 2012 IEEE Conference on Com- [20]COSTEIRA J P,KANADE T.A multibody factorization puter Vision and Pattern Recognition.Providence,USA. method for independently moving objects[J].Internation- 2012:2618-2625 al journal of computer vision,1998,29(3):159-179. [8]YANG Meng,ZHANG Lei,YANG Jian,et al.Regular- [21]ZHUANG Liansheng,GAO Haoyuan,LIN Zhouchen,et ized robust coding for face recognition[J].IEEE transac- al.Non-negative low rank and sparse graph for semi-su- tions on image processing,2013,22(5):1753-1766 pervised learning[C]//Proceedings of 2012 IEEE Confer- [9]ZHANG Lei,YANG Meng,FENG Xiangchu.Sparse rep- ence on Computer Vision and Pattern Recognition. resentation or collaborative representation:which helps Providence,USA,2012:2328-2335. face recognition?[C]//Proceedings of 2011 ICCV.Bar- [22]LIN Zhouchen,LIU Risheng,SU Zhixun.Linearized al- celona,Spain,2011:471-478. ternating direction method with adaptive penalty for low- [10]DENG Weihong,HU Jiani,GUO Jun.Extended SRC:un- rank representation[J].Advance in neural information dersampled face recognition via intraclass variant diction- processing systems,2011:612-620. ary].IEEE transactions on pattern analysis and machine [23]LIN Zhouchen,CHEN Minming,MA Yi.The augmented intelligence,2012.34(9y:1864-1870. lagrange multiplier method for exact recovery of corrup- [11]YANG Meng,ZHU Pengfei,LIU Feng,et al.Joint repres- ted low-rank matrices[R].Urbana-Champaign:Uni- entation and pattern learning for robust face recognition versity of Illinois at Urbana-Champaign,2009.染，有效减少了样本中的污染对实验结果造成的 负面影响。因此 SWLRR 算法样本在有无遮挡的 情况下都有着较为突出的性能。在 AR、Extended Yale B、CMU PIE 数据库上的实验验证了 SWLRR 算法在不同情况下的有效性和对噪声的鲁棒性。 下一步将尝试将本文思想与其他解决样本中存在 污染的算法结合，进一步提升算法性能。 参考文献： TURK M, PENTLAND A. Eigenfaces for recognition[J]. Journal of cognitive neuroscience, 1991, 3(1): 71–86. [1] BELHUMEUR P N, HESPANHA J P, KRIEGMAN D J. Eigenfaces vs. fisherfaces: recognition using class specific linear projection[J]. IEEE transactions on pattern analysis and machine intelligence, 997, 19(7): 711–720. [2] LIU Chengjun, WECHSLER H. Gabor feature based clas￾sification using the enhanced fisher linear discriminant model for face recognition[J]. IEEE transactions on image processing, 2002, 11(4): 467–476. [3] HE Xiaofei, YAN Shuicheng, HU Yuxiao, et al. Face re￾cognition using Laplacianfaces[J]. IEEE transactions on pattern analysis and machine intelligence, 2005, 27(3): 328–340. [4] WRIGHT J, YANG A Y, GANESH A, et al. Robust face recognition via sparse representation[J]. IEEE transactions on pattern analysis and machine intelligence, 2009, 31(2): 210–227. [5] MIN R, DUGELAY J L. Improved combination of LBP and sparse representation based classification (SRC) for face recognition[C]//Proceedings of 2011 IEEE ICME. Barcelona, Spain, 2011: 1–6. [6] CHEN C F, WEI C P, WANG Y C F. Low-rank matrix re￾covery with structural incoherence for robust face recogni￾tion[C]//Proceedings of 2012 IEEE Conference on Com￾puter Vision and Pattern Recognition. Providence, USA, 2012: 2618–2625. [7] YANG Meng, ZHANG Lei, YANG Jian, et al. Regular￾ized robust coding for face recognition[J]. IEEE transac￾tions on image processing, 2013, 22(5): 1753–1766. [8] ZHANG Lei, YANG Meng, FENG Xiangchu. Sparse rep￾resentation or collaborative representation: which helps face recognition?[C]//Proceedings of 2011 ICCV. Bar￾celona, Spain, 2011: 471–478. [9] DENG Weihong, HU Jiani, GUO Jun. Extended SRC: un￾dersampled face recognition via intraclass variant diction￾ary[J]. IEEE transactions on pattern analysis and machine intelligence, 2012, 34(9): 1864–1870. [10] YANG Meng, ZHU Pengfei, LIU Feng, et al. Joint repres￾entation and pattern learning for robust face recognition [11] [J]. Neurocomputing, 2015, 168: 70–80. CANDÈS E J, LI Xiaodong, MA Yi, et al. Robust prin￾cipal component analysis[J]. Journal of the ACM, 2011, 58(3): 11. [12] LIU Guangcan, LIN Zhouchen, YAN Shuicheng, et al. Robust recovery of subspace structures by low-rank rep￾resentation[J]. IEEE transactions on pattern analysis and machine intelligence, 2013, 35(1): 171–184. [13] MA Long, WANG Chunheng, XIAO Baihua, et al. Sparse representation for face recognition based on discriminat￾ive low-rank dictionary learning[C]//Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recog￾nition. Providence, USA, 2012: 2586–2593. [14] ZHANG Yangmuzi, JIANG Zhuolin, DAVIS L S. Learn￾ing structured low-rank representations for image classi￾fication[C]//Proceeding of 2013 IEEE Conference on Computer Vision and Pattern Recognition. Portland, USA, 2013: 676–683. [15] NGUYEN H, YANG Wankou, SHENG Biyun, et al. Dis￾criminative low-rank dictionary learning for face recogni￾tion[J]. Neurocomputing, 2016, 173(3): 541–551. [16] CHANG Heyou, ZHENG Hao. Weighted discriminative dictionary learning based on lowrank representation [J]. Journal of physics: conference series, 2017, 90: 012009. [17] ZHANG Zheng, XU Yong, SHAO Ling, et al. Discrimin￾ative block-diagonal representation learning for image re￾cognition[J]. IEEE transactions on neural networks and learning systems, 2018, 29(7): 3111–3125. [18] CHEN Jie, YI Zhang. Sparse representation for face re￾cognition by discriminative low-rank matrix recovery[J]. Journal of visual communication and image representa￾tion, 2014, 25(5): 763–773. [19] COSTEIRA J P, KANADE T. A multibody factorization method for independently moving objects[J]. Internation￾al journal of computer vision, 1998, 29(3): 159–179. [20] ZHUANG Liansheng, GAO Haoyuan, LIN Zhouchen, et al. Non-negative low rank and sparse graph for semi-su￾pervised learning[C]//Proceedings of 2012 IEEE Confer￾ence on Computer Vision and Pattern Recognition. Providence, USA, 2012: 2328–2335. [21] LIN Zhouchen, LIU Risheng, SU Zhixun. Linearized al￾ternating direction method with adaptive penalty for low￾rank representation[J]. Advance in neural information processing systems, 2011: 612-620. [22] LIN Zhouchen, CHEN Minming, MA Yi. The augmented lagrange multiplier method for exact recovery of corrup￾ted low-rank matrices[R]. Urbana-Champaign: Uni￾versity of Illinois at Urbana-Champaign, 2009. [23] ·462· 智 能 系 统 学 报 第 14 卷