计算机科学 ›› 2017, Vol. 44 ›› Issue (3): 296-299.doi: 10.11896/j.issn.1002-137X.2017.03.060

• 图形图像与模式识别 • 上一篇    下一篇

联合Gabor误差字典和低秩表示的人脸识别算法

首照宇,杨晓帆   

  1. 桂林电子科技大学认知无线电与信息处理教育部重点实验室 桂林541004,桂林电子科技大学认知无线电与信息处理教育部重点实验室 桂林541004
  • 出版日期:2018-11-13 发布日期:2018-11-13
  • 基金资助:
    本文受广西自然科学基金(2013GXNSFDA019030,3GXNSFAA019331,4GXNSFDA118035),桂林市科技攻关项目(20130105-6,3-5),桂林电子科技大学研究生科研创新项目(YJCXS201531)资助

Jointing Gabor Error Dictionary and Low Rank Representation for Face Recognition

SHOU Zhao-yu and YANG Xiao-fan   

  • Online:2018-11-13 Published:2018-11-13

摘要: 针对人脸图片的遮挡、伪装、光照及表情变化等问题,根据Gabor特征对遮挡、伪装、光照及表情变化有着更强的鲁棒性的特点,提出了联合Gabor误差字典和低秩表示的人脸识别算法(GDLRR)。首先对训练样本和测试样本分别进行Gabor特征提取,并将这些特征组成待测试的特征字典;然后将一个单位阵进行Gabor特征提取并训练成一个更紧凑的Gabor误差字典;最后联合Gabor误差字典和训练特征字典对测试特征字典进行低秩表示后进行分类识别。各类实验表明,提出的改进算法对人脸识别的各类问题都有着更强的鲁棒性和更高的识别准确率。

关键词: 遮挡,低秩表示,Gabor特征,误差字典,降维

Abstract: Focused on the issues that face images have the problems occlusion,disguise,illumination and facial expression changes in face recognition,an improved face recognition method was proposed.According to the characteristics of the Gabor feature for occlusion,disguise,illumination and facial expression changes has stronger robustness,jointing Gabor error dictionary and low rank representation (GDLRR) for face recognition was proposed.Firstly,the Gabor feature of training samples and testing samples are extracted for making up features dictionaries that is to be tested,respectively.And then,a unit matrix is utilized to extract Gabor feature for training a more compact Gabor error dictionary.Finally,lowest-rank representation of feature dictionary of testing samples is sought for classification by jointing Gabor error dictionary and training feature dictionary.Experiments show that the proposed algorithm has better robustness and recognition results against the different problems in the face recognition.

Key words: Occlusion,Low-rank representation,Gabor feature,Error dictionary,Reduce dimension

[1] TURK M,PENTLAND A.Eigenfaces for recognition [J].Cognitive Neuroscience,1991,3(1):71-86.
[2] BELHUMEUR P N,HESPANHA J P,KRIENGMAN D J.Eigenfaces vs.Fisherfaces:Recognition using class specific linear projection[J].IEEE Trans.Pattern Anal.Machine Intell,1997,19(7):711-720.
[3] YANG J,YANG J Y.Why can LDA be performed in PCAtransformed space?[J].Pattern Recognition,2003,36(2):563-566.
[4] WRIGHT J,YANG Y,GANESH A,et al.Robust Face Recognition via Sparse Representation[J].IEEE Trans.Pattern Anal.Mach.Intell.,2009,31(2):210-227.
[5] YANG M,ZHANG L,ZHANG D.Gabor Feature Based Robust Representation and Classification for Face Recognition with Gabor Occlusion Dictionary[J].Pattern Recognition,2013,6(7):1865-1878.
[6] LIU G,Lin Z,YU Y.Robust subspace segmentation by low-rank representation[C]∥ Proceedings of the 27th International Conference on Machine Learning (ICML-10).2010:663-670.
[7] ZHANG J,DAVIS.Learning structured low-rank representa-tions for image classificition[M]∥ Computer Vision and Pattern Recognition.2013:676-683.
[8] DAUBECHIES I,DEFRIES M,DEMOL C.An iterative thres-holding algorithm for linear inverse problems with a sparsity constraint[J].Commun.Pure,2004,57(11):1413-1457.
[9] BECK A,TEBOULLE M.A fast iterative shrinkage-threshol-ding algorithm for linear inverse problems[J].SIAMJ.Imag.,2009,2009(2):183-202.
[10] BERTSEKAS D P.Computer science and applied mathematics,Constrained Optimization and Lagrange Multiplier Methods[M].Academic Press,Boston,1982.
[11] LIU G,LIN Z,YAN S,et al.Robust recovery of subspace structures by low-rank representation[J].Pattern Analysis and Machine Intelligence,2013,35(1):171-184.
[12] LIU C,WECHSLER H.Gabor feature based classification using the enhanced fisher linear discriminant model for face recognition[J].IEEE Transactions on Image processing,2002,11(4):467-476.
[13] ZHANG L,YANG M,FENG X C.Sparse representation or collaborative representation which helps face recognition?[C]∥Proc.IEEE Int’l Conf.Computer Vision,2011.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!