您的位置：山东大学 -> 科技期刊社 -> 《山东大学学报（工学版）》

山东大学学报(工学版) ›› 2018, Vol. 48 ›› Issue (1): 15-20.doi: 10.6040/j.issn.1672-3961.0.2017.005

• • 上一篇    下一篇

基于低秩表示投影的无监督人脸特征提取

张振月,李斐,江铭炎 *   

  1. 山东大学信息科学与工程学院, 山东 济南 250100
  • 收稿日期:2017-01-03 出版日期:2018-02-20 发布日期:2017-01-03
  • 通讯作者: 江铭炎(1964— ),男,江苏苏州人,教授,博士,主要研究方向为计算机视觉,机器学习,智能优化.E-mail:jiangmingyan@sdu.edu.cn E-mail:zhangzhenyue0@163.com
  • 作者简介:张振月(1989— ),男,山东菏泽人,硕士研究生,主要研究方向为低秩表示,人脸识别,机器学习.E-mail:zhangzhenyue0@163.com
  • 基金资助:
    国家自然科学基金资助项目(61201370);山东省自然科学基金资助项目(ZR2014FM039)

Unsupervised face image feature extraction based on low-rank representation projection

ZHANG Zhenyue, LI Fei, JIANG Mingyan*   

  1. School of Information Science and Engineering, Shandong University, Jinan 250100, Shandong, China
  • Received:2017-01-03 Online:2018-02-20 Published:2017-01-03

PDF (PC)

587

摘要： 为了构造数据之间的自适应邻接图,同时克服稀疏表示系数和协同表示系数互相独立、提取全局信息弱的缺陷,提出采用低秩表示(low-rank representation, LRR)系数构造权重矩阵的流形学习算法,即低秩表示投影(low-rank representation projections, LRRP)和判别低秩表示投影(discriminative low-rank representation projections, DLRRP)。在新算法中,将低秩表示系数表征的样本之间的邻接关系保留在特征空间;同时利用低秩系数的聚类性质,在优化目标中加入类内散度最小化项,计算出具有判别性的投影矩阵。试验结果表明,在真实人脸图像库上与其他几种流形学习算法相比,LRRP和DLRRP能够取得更好的识别率。提出的新算法是有效的特征提取算法,能够丰富流形学习框架。

关键词: 人脸识别, 低秩表示, 流形学习, 邻接图, 特征提取

Abstract: In order to construct the adaptive adjacency graph between data points, and also to overcome the disadvantage that the coefficients of sparse representation and collaborative representation were independent, the low-rank representation projections(LRRP)and discriminative low-rank representation projections(DLRRP)were proposed. In these two manifold learning methods, the weighted matrix was constructed by low-rank representation(LRR). The adjacencies defined by the coefficients were preserved in the feature space. By virtue of the clustering property of the coefficients, an within-class scatter minimum term was added in the optimization objective, which leaded to a discriminative projection. The experimental results showed that compared with other manifold learning algorithms, LRRP and DLRRP could obtain the better recognition accuracies. The proposed methods were effective feature extraction algorithms and enriched the manifold learning framework.

Key words: manifold learning, low-rank representation, face recognition, feature extraction, adjacency graph

中图分类号: 

  • TP391
[1] ROWEIS S T, SAUL L K. Nonlinear dimensionality reduction by locally linear embedding[J]. Science, 2000, 290:2323-2326.
[2] BELKIN M, NIYOGI P. Laplacian eigenmaps and spectral techniques for embedding and clustering[C] //Proceedings of International Conference on Neural Information Processing Systems: Natural and Synthetic. Vancouver, Canada: MIT Press, 2001: 585-591.
[3] HE Xiaofei, NIYOGI P. Locality preserving projections[C] //Proceedings of the Seventeenth Annual Conference on Neural Information Processing Systems. Massachusetts, USA:MIT Press, 2003.
[4] 黄璞,唐振民.无参数局部保持投影及人脸识别[J].模式识别与人工智能,2013, 26(9):865-871. HUANG Pu, TANG Zhenmin. Parameter-free locality preserving projections and face recognition[J]. Pattern Recognition and Artificial Intelligence, 2013, 26(9):865-871.
[5] QIAO Lishan, CHEN Songcan, TAN Xiaoyang. Sparsity preserving projections with applications to face recognition[J]. Pattern Recognition, 2010, 43(1):331-341.
[6] ZHANG Lei, YANG Meng, FENG Xiangchu. Sparse representation or collaborative representation: which helps face recognition?[C] //Proceedings of 2011 International Conference on Computer Vision. Barcelona, Spain: IEEE, 2011:471-478.
[7] YANG Wankou, WANG Zhenyu, SUN Changyin. A collaborative representation based projections method for feature extraction[J]. Pattern Recognition, 2015, 48:20-27.
[8] 杨国亮,谢乃俊,罗璐,等.基于空间约束低秩图的人脸识别[J].计算机科学, 2014, 41(8):297-300. YANG Guoliang, XIE Naijun, LUO Lu, et al. Low-rank graph with spatial constraint for face recognition[J]. Computer Science, 2014, 41(8):297-300.
[9] LIU Guangcan, LIN Zhouchen, YU Yong. Robust subspace segmentation by low-rank representation[C] //Proceedings of International Conference on Machine Learning. Haifa, Israel: Omnipress, 2010:663-670.
[10] WRIGHT J, WRIGHT J, GANESH A, et al. Robust principal component analysis: exact recovery of corrupted low-rank matrices by convex optimization[C] // Proceedings of International Conference on Neural Information Processing Systems. Vancouver, Canada: Curran Associates Inc., 2009:2080-2088.
[11] HE Xiaofei, MA Weiying, ZHANG Hongjiang. Learning an image manifold for retrieval[C] //Proceedings of ACM International Conference on Multimedia. New York, USA:ACM, 2004:17-23.
[12] LIU Guangcan, LIN Zhouchen, YAN Shuicheng, et al. Robust recovery of subspace structures by low-rank representation[J]. IEEE Transaction on Pattern and Machine Recognition, 2013, 35(1): 171-184.
[13] GAN Guojun, NG K P. Subspace clustering using affinity propagation[J]. Pattern Recognition, 2015, 48(4):1455-1464.
[14] BOYD S, VANDERBERGHE L. Convex optimization[M]. New York, USA: Cambridge University Press, 2007:75-78.
[15] MARTINEZ A, BENAVENTE R. The AR face database[R]. USA, Purdue University West Lafayette:Computer Vision Center: Technical Report, 1998.
[16] GEORGHIADES A S, BELHUMEUR P N, KRIEGMAN D J. From few to many: illumination cone models for face recognition under variable lighting and pose[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(6): 643-660.
[17] PHILLIPSP J, MOON H, RIZVI A, et al. The FERET valuation methodology for face recognition algorithms[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(10): 1090-1104.
[18] SIM T, BAKER S, BSAT M. The CMU pose, illumination, and expression database[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(12): 1615-1618.
[19] JOLLIFFE I T. Principal component analysis[M]. New York, USA: Springer-Verlag, 2002:98-99.
[20] ZHANG Limei, CHEN Songcan, QIAO Lishan. Graph optimization for dimensionality reduction with sparsity constraints[J]. Pattern Recognition, 2012, 45(3): 1205-1210.
[1] 王国新,陈凤东,刘国栋. 基于彩色伪随机编码结构光特征提取方法[J]. 山东大学学报(工学版), 2018, 48(5): 55-60.
[2] 叶子云,杨金锋. 一种基于加权图模型的手指静脉识别方法[J]. 山东大学学报(工学版), 2018, 48(3): 103-109.
[3] 包塔拉,马剑,甘祖旺. 基于几何特征与流形距离的锂电池健康评估[J]. 山东大学学报(工学版), 2017, 47(5): 157-165.
[4] 于青民,李晓磊,翟勇. 基于改进EMD和数据分箱的轴承内圈故障特征提取方法[J]. 山东大学学报(工学版), 2017, 47(3): 89-95.
[5] 郭超,杨燕,江永全,宋祎. 基于多视图分类集成的高铁工况识别[J]. 山东大学学报(工学版), 2017, 47(1): 7-14.
[6] 周凯,元昌安,覃晓,郑彦,冯文铎. 基于核贝叶斯压缩感知的人脸识别[J]. 山东大学学报(工学版), 2016, 46(3): 74-78.
[7] 钟智彦,文志强, 张潇云,叶德刚. 基于半色调图像的邻域相似性描述子方法[J]. 山东大学学报(工学版), 2016, 46(3): 58-64.
[8] 卢丹, 周以齐. 基于EEMD和CWT的挖掘机座椅振动分析[J]. 山东大学学报(工学版), 2015, 45(3): 58-64.
[9] 任捷怡, 吴小俊. 一种改进的协方差鉴别学习方法[J]. 山东大学学报(工学版), 2015, 45(1): 9-12.
[10] 于海晶1,2, 李桂菊1*. 基于差分盒维数的彩色烟雾图像识别[J]. 山东大学学报(工学版), 2014, 44(1): 35-40.
[11] 谢志华. 一种新的血流建模方法及其在红外人脸识别中的应用[J]. 山东大学学报(工学版), 2013, 43(5): 1-5.
[12] 郭慧玲,王士同*,闫晓波. 基于广义旋转不变性核函数的人脸识别[J]. 山东大学学报(工学版), 2012, 42(5): 71-79.
[13] 李慧1,2,胡云1,3,李存华1. 基于粗糙集理论的瓦斯灾害信息特征提取技术[J]. 山东大学学报(工学版), 2012, 42(5): 91-95.
[14] 王洪元,封磊,冯燕,程起才. 流形学习算法在中文文本分类中的应用[J]. 山东大学学报(工学版), 2012, 42(4): 8-12.
[15] 曹红根1,袁宝华1,朱辉生2. 结合对比度信息与LBP的分块人脸识别[J]. 山东大学学报(工学版), 2012, 42(4): 29-34.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2018 《山东大学学报(工学版)》编辑部　
地址: 济南市山大南路27号（250100）　 电话: 0531-88366735　 E-mail: xbgxb@sdu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发