基于稀疏理论的单样本人脸识别研究
|
摘要
单样本人脸识别问题已发展成为模式识别、人工智能和机器学习领域中的一个热点和难点研究课题。目前大多数人脸识别技术的研究仅集中在怎样提高人脸识别系统的准确率上,并且严重依赖训练样本集的规模和代表性,忽略了由于采集样本的困难或者系统的存储容量有限等所造成的单样本问题。稀疏表征(Sparse Representation, SR)源于对传统信号采样和表示理论的扩展,例如Fourier和Wavelet表示。在过去几年里,SR已被证明在获取、表征和压缩高维信号方面是一个非常强大的工具,并已成功用于解决信号处理、机器学习和模式识别领域中的诸多实际问题。近来,由D. Donoho、E. Candes及T. Tao等人提出的压缩感知(Compressive Sensing,或称Compressed Sensing、Compressed Sampling, CS)理论进一步将稀疏表征思想提升到了一个新的高度。
本文主要是研究基于稀疏理论的单样本人脸识别问题,目标是得到有效的改进算法以提高单样本情况下的识别性能。本文的主要研究工作如下：
(1)对目前人脸识别问题、稀疏理论、单样本人脸识别问题的研究背景、意义、国内外现状及目前面临的挑战等作了综述性分析；
(2)利用Shift、PCA重构、镜像对称变换、下采样等方法结合稀疏表征分类器(Sparse Representation-based Classification, SRC)方法,改善单样本人脸识别：增加了冗余样本,有效地提高了识别率,节省了计算与存储开销,增强了算法的实用性能；
(3)基于稀疏表征理论SR、压缩感知理论CS以及半监督降维技术(Semi-supervised Dimensionality Reduction, SSDR),提出了一个新的基于半监督的算法——半监督的稀疏判别保局投影(Semi-supervised Sparsity Diteriminant Locality Preserving Projections, SSDLPP)。SSDLPP算法在仅仅拥有很少无标号样本的情况下,改善了DLPP (Discriminant Locality Preserving Projections, DLPP)和LPP (Locality Preserving Projections, LPP)的识别性能。
本文在ORL和Yale等人脸数据库上进行了仿真实验,实验结果证明本文所提出的改进的单样本人脸识别算法取得了较好的识别性能。
Single sample face recognition problem has been a hot and difficult research topic in the field of pattern recognition, artificial intelligence and machine learning. Many face recognition techniques focus on how to improve the accuracy of a recognition system, and largely depend on the. size and representative of training set. However, it seems that most of them ignore the potential problem that may stem from the face database at hand, where there may be only one sample image per person, possibly due to the difficulties of collecting samples or the limitations of storage capability of the systems, etc. Sparse representation (SR) is initially proposed as an extension of traditional signal sampling and representation theory such as Fourier representation and wavelet representation. In the past few years, SR has proven to be an extremely powerful tool for acquiring, representing, and compressing high-dimensional signals and has been successfully applied to solve many practical problems in fields of signal processing, machine learning, and pattern recognition. Recently, compressive sensing (CS) was proposed by D. Donoho, E. Candes, T. Tao et al., and makes SR a breakthrough in real applications.
This paper focuses on the single sample problem based sparse representation for face recognition. The goal of this paper is to improve the recognition performance of the single sample face recognition algorithm. The main contributions of this paper are as follows:
(1) This paper briefly introduces the background, significance, status and challenges of current face recognition, sparse theory, and single sample face recognition problem.
(2) This paper generates the multiple images using Shift, PCA Reconstruction, Mirror-symmetry Transformation and Sampling methods combined with Sparse Representation-based Classification (SRC), to further improve the recognition performance. These proposed methods not only increase redundant samples and enhance the recognition performance but also save the computation and storage costs.
(3) Motivated by the recent development of SR, CS and semi-supervised dimensionality reduction (SSDR), a novel SSDR based on l1-graph is presented, namely semi-supervised sparsity discriminant locality preserving projections (SSDLPP). SSDLPP can remarkably improve the recognition performance of discriminant locality preserving projections (DLPP) and locality preserving projections (LPP) even if we have only single training sample and very few extra unlabeled training samples.
Experimental results on both two famous face datasets (ORL and Yale) show that the proposed algorithms substantially improve the recognition performance for the single sample face recognition.
本文主要是研究基于稀疏理论的单样本人脸识别问题,目标是得到有效的改进算法以提高单样本情况下的识别性能。本文的主要研究工作如下：
(1)对目前人脸识别问题、稀疏理论、单样本人脸识别问题的研究背景、意义、国内外现状及目前面临的挑战等作了综述性分析；
(2)利用Shift、PCA重构、镜像对称变换、下采样等方法结合稀疏表征分类器(Sparse Representation-based Classification, SRC)方法,改善单样本人脸识别：增加了冗余样本,有效地提高了识别率,节省了计算与存储开销,增强了算法的实用性能；
(3)基于稀疏表征理论SR、压缩感知理论CS以及半监督降维技术(Semi-supervised Dimensionality Reduction, SSDR),提出了一个新的基于半监督的算法——半监督的稀疏判别保局投影(Semi-supervised Sparsity Diteriminant Locality Preserving Projections, SSDLPP)。SSDLPP算法在仅仅拥有很少无标号样本的情况下,改善了DLPP (Discriminant Locality Preserving Projections, DLPP)和LPP (Locality Preserving Projections, LPP)的识别性能。
本文在ORL和Yale等人脸数据库上进行了仿真实验,实验结果证明本文所提出的改进的单样本人脸识别算法取得了较好的识别性能。
Single sample face recognition problem has been a hot and difficult research topic in the field of pattern recognition, artificial intelligence and machine learning. Many face recognition techniques focus on how to improve the accuracy of a recognition system, and largely depend on the. size and representative of training set. However, it seems that most of them ignore the potential problem that may stem from the face database at hand, where there may be only one sample image per person, possibly due to the difficulties of collecting samples or the limitations of storage capability of the systems, etc. Sparse representation (SR) is initially proposed as an extension of traditional signal sampling and representation theory such as Fourier representation and wavelet representation. In the past few years, SR has proven to be an extremely powerful tool for acquiring, representing, and compressing high-dimensional signals and has been successfully applied to solve many practical problems in fields of signal processing, machine learning, and pattern recognition. Recently, compressive sensing (CS) was proposed by D. Donoho, E. Candes, T. Tao et al., and makes SR a breakthrough in real applications.
This paper focuses on the single sample problem based sparse representation for face recognition. The goal of this paper is to improve the recognition performance of the single sample face recognition algorithm. The main contributions of this paper are as follows:
(1) This paper briefly introduces the background, significance, status and challenges of current face recognition, sparse theory, and single sample face recognition problem.
(2) This paper generates the multiple images using Shift, PCA Reconstruction, Mirror-symmetry Transformation and Sampling methods combined with Sparse Representation-based Classification (SRC), to further improve the recognition performance. These proposed methods not only increase redundant samples and enhance the recognition performance but also save the computation and storage costs.
(3) Motivated by the recent development of SR, CS and semi-supervised dimensionality reduction (SSDR), a novel SSDR based on l1-graph is presented, namely semi-supervised sparsity discriminant locality preserving projections (SSDLPP). SSDLPP can remarkably improve the recognition performance of discriminant locality preserving projections (DLPP) and locality preserving projections (LPP) even if we have only single training sample and very few extra unlabeled training samples.
Experimental results on both two famous face datasets (ORL and Yale) show that the proposed algorithms substantially improve the recognition performance for the single sample face recognition.
引文
[1]Deng, W., J. Hu, J. Guo, et al. Robust, accurate and efficient face recognition from a single training image:A uniform pursuit approach[J]. Pattern Recognition,2010. 43(5):p.1748-1762.
[2]W.Ahao, R.Chellappa, PJ.Phillips, et al. Face recognition:a literature survey[J]. ACM Computing Surveys (CSUR),2003.35(4):p.399-458.
[3]Phillips, P.J., H. Wechsler, J. Huang, et al. The FERET database and evaluation procedure for face-recognition algorithms[J]. Image and Vision Computing,1998. 16(5):p.295-306.
[4]Rebollo-Neira, L.,Z. Xu. Sparse signal representation by adaptive non-uniform B-spline dictionaries on a compact interval[J]. Signal Processing,2010.90(7):p. 2308-2313.
[5]Baraniuk, R. Compressive sensing[J]. IEEE Signal Processing Magazine,2007. 24(4):p.118-121.
[6]Smith, K. Mind-reading with a brain scan[J]. Nature,2008.
[7]赵松年,姚力,金真,et al.视像整体特征在人类初级视皮层上的稀疏表象：脑功能成像的证据[J].科学通报,2008.53(11):p.1296-1304.
[8]Donoho, D.L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006.52(4):p.1289-1306.
[9]Candes, E. Compressive sampling[C]. in Proceedings of the International Congress of Mathematicians.2006. Madrid, Spain.
[10]Candes, E.J.,T. Tao. Decoding by linear programming[J]. IEEE Transactions on Information Theory,2005.51(12):p.4203-4215.
[11]Candes, E.J.,T. Tao. Reflections on compressed sensing[J]. IEEE Information Theory Society Newslette,2008.58(4):p.20-23.
[12]Wright, J., A.Y. Yang, A. Ganesh, et al. Robust face recognition via sparse representation[J]. IEEE Transactions Pattern Analysis and Machine Intelligence, 2009.31(2):p.210-227.
[13]Hegde, C., M.B. Wakin, R.G. Baraniuk. Random projections for manifold learning[C]. The Twenty-First Annual Conference on Neural Information Processing Systems (NIPS),2007.
[14]Haupt, J.,R. Nowak. Compressive sampling for signal detection[C]. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2007.
[15]Sheikh, M.A.,R.G. Baraniuk. Blind error-free detection of transform-domain watermarks[C]. IEEE International Conference on Image Processing(ICIP),2007.
[16]Ji, S., Y. Xue, L. Carin. Bayesian Compressive Sensing[J]. IEEE Transactions on Signal Processing,2008.56(6):p.2346-2356
[17]Wakin, M.B. The geometry of low-dimensional signal models[J]. PhD Thesis, Rice University, Houston, TX,,2007.
[18]Hubel, D.H.,T.N. Wiesel. Receptive fields of single neurons in the cat's striate cortex[J]. Journal of Physiology,1959.148(3):p.574-591.
[19]Barlow, H.B. The coding of sensory messages[M], in Current Problems in Animal Behaviors, Chapter XIII, Cambridge Universtity Press.1961. p.331-360.
[20]Barlow, H.B. Single units and sensation:a neuron doctrine for perceptual psychology?[J]. Perception,1972.1(4):p.371-394.
[21]Michison, G. The organization of sequential memory: sparse representations and the targeting problem[M]. Organiztion of Neural Networks, VCH Verlagsgesellschaft, Weinheim.1988.347-367.
[22]Olshausen, B.A. Emergence of simple-cell receptive field properties by learning a sparse code for natural images[J]. Nuture,1996.381(6583):p.607-609.
[23]BA, O.,F. DJ. Sparse coding with an overcomplete basis set: a strategy employed by V1[J]. Vision Research,1997.37(23):p.3311-25.
[24]Olshausen, B.A.,D.J. Field. Sparse coding of sensory inputs[J]. Current Opinion in Neurobiology,2004.14(4):p.481-487.
[25]Reddy, L.,N. Kanwisher. Coding of visual objects in the ventral stream[J]. Current Opinion in Neurobiology,2006.16(4):p.408-414.
[26]Riesenhuber, M.,T. Poggio. Are cortical models really bound by the "binding problem"?[J]. Neuron,1999.24(1):p.87-93.
[27]Bell, A.J.,T.J. Sejnowski. The "independent components" of natural scenes are edge filters[J]. Vision Research,1997.37(23):p.3327-3338.
[28]Tenenbaum, J.,W. Freeman. Separating style and content with bilinear models. 2000; Available from:http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.33. 4444.
[29]Hyvarinen, A.,P.O. Hoyer. A two-layer sparse coding model learns simple and complex cell receptive fields and topography from natural images[J]. Vision Research,2001.41(18):p.2413-2423.
[30]张海,王尧,常象宇,et al. L1/2正则化[J].中国科学信息科学,2010.40(3):p.412-422.
[31]Li, Y.,S.-i. Amari. Two conditions for equivalence of 0-norm solution and 1-norm solution in sparse representation[J]. IEEE Transactions on Neural Networks,2010. 21(7):p.1189-1196.
[32]Zhou, G., Z. Yang, S. Xie, et al. Mixing matrix estimation from sparse mixtures with unknown number of sources [J]. IEEE Transactions on Neural Networks,2011. 22(2):p.211-221.
[33]Shi, G, J. Lin, X. Chen, et al. UWB echo signal detection with ultra-low rate sampling based on compressed sensing[J]., IEEE Transactions on Circuits and Systems Ⅱ:Express Briefs,2008.55(4):p.379-383.
[34]Zhang, L.,W.-D. Zhou. Sparse ensembles using weighted combination methods based on linear programming[J]. Pattern Recognition,2010.44(1):p.97-106.
[35]Li, Y.-R., D.-Q. Dai, L. Shen. Multiframe super-resolution reconstruction using sparse directional regularization[J]. Circuits and Systems for Video, IEEE Transactions on,2010.20(7):p.945-956.
[36]Ji, S., L. Yuan, Y.-X. Li, et al. Drosophila gene expression pattern annotation using sparse features and term-term interactions[C]. Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining,2009: p.407-416.
[37]Qiao, L., S. Chen, X. Tan. Sparsity preserving projections with applications to face recognition[J]. Pattern Recognition,2010.43(1):p.331-341.
[38]Lai, Z., M. Wan, Z. Jin, et al. Sparse two-dimensional local discriminant projections for feature extraction[J]. Neurocomputing,2011.74(4):p.629-637.
[39]Cai, D.,X. He. Sparse concept coding for visual analysis[C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2011.
[40]Wu, F., Y. Yuan, Y. Zhuang. Heterogeneous feature selection by group lasso with logistic regression[C], in Proceedings of the international conference on Multimedia.2010. Firenze, Italy:ACM.
[41]Zhang, Z., G. Dai, M.I. Jordan. Bayesian generalized kernel mixed models[J]. J. Mach. Learn. Res.,2011.12:p.111-139.
[42]Liu, B., S. Chen, M. Qian, et al. Sparse norm-regularized reconstructive coefficients learning[C].2009 Ninth IEEE International Conference on Data Mining,2009:p.854-859.
[43]Zhao, R., X. Liu, C.-C. Li, et al. Wavelet denoising via sparse representation[J]. Science in China Series F:Information Sciences,2009.52(8):p.1371-1377.
[44]Ma, B.,T. Wang. Head pose estimation using sparse representation[C]. Proceedings of the 2010 Second International Conference on Computer Engineering and Applications - Volume 02,2010:p.389-392.
[45]Sun, S.,J. Shawe-Taylor. Sparse semi-supervised learning using conjugate functions[J]. J. Mach. Learn. Res.,2010.9999:p.2423-2455.
[46]Xu, R., B. Zhang, Q. Ye, et al. Human detection in images via L1-norm minimization learning. [C]. IEEE International Conference on Acoustics, Speech and Signal Processing,2010:p.3566-3569.
[47]Li, Z., Z. Shi, X. Liu, et al. A novel sparse coding model based on structural similarity[C]. IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP) 2010:p.4170-4173
[48]Liu, W., S. Zheng, S. Jia, et al. Sparse nonnegative matrix factorization with the elastic net[C]. IEEE International Conference on Bioinformatics and Biomedicine (BIBM),2010:p.265-268.
[49]Tibshirani, R. Regression shrinkage and selection via the lasso[J]. J. Roy. Statist. Soc. Ser. B,1996.58(1):p.267-288.
[50]Efron, B., T. Hastie, I. Johnstone, et al. Least angle regression[J]. The Annals of Statistics,2004.32(2):p.407-499.
[51]Zou, H.,T. Hastie. Regularization and variable selection via the Elastic Net[J]. Journal of the Royal Statistical Society B,2005.67:p.301-320.
[52]Mackey, L. Deflation methods for sparse PCA[C]. Advances in Neural Information Processing Systems 21,2009:p.1017-1024.
[53]Zass, R.,A. Shashua. Nonnegative sparse PCA[C]. In Neural Information Processing Systems,2007.
[54]Zou, H., T. Hastie, R. Tibshirani. Sparse principal component analysis[J]. Journal of Computational and Graphical Statistics,2006.15(2):p.265-286.
[55]Jenatton, R., G. Obozinski, F. Bach. Structured sparse principal component analysis[J].2009.
[56]Lee, D., W. Lee, Y. Lee, et al. Super-sparse principal component analyses for high-throughput genomic data[J]. BMC bioinformatics,2010.11(1):p.296.
[57]Clemmensen, L., T. Hastie, B. Ersb(?)ll. Sparse discriminant analysis[J]. Technical Report,2008.
[58]Hastie, T., A. Buja, R. Tibshirani. Penalized discriminant analysis[J]. Annals of Statistics,1995.23(1):p.73-102.
[59]Qiao, L., S. Chen, X. Tan. Sparsity preserving discriminant analysis for single training image face recognition[J]. Pattern Recognition Letters,2010.31(5):p. 422-429.
[60]Wang, J., K.N. Plataniotis, J. Lu, et al. On solving the face recognition problem with one training sample per subject[J]. Pattern Recognition,2006.39(9):p. 1746-1762.
[61]Tan, X., S. Chen, Z.-H. Zhou, et al. Face recognition from a single image per person:A survey [J]. Pattern Recognition,2006.39(9):p.1725-1745.
[62]Wu, J.,Z. Zhou. Face recognition with one training image per person[J]. Pattern Recognition Letters,2002.23(14):p.1711-1719.
[63]Chen, S., D. Zhang, Z.-H. Zhou. Enhanced (PC)2 A for face recognition with one training image per person[J]. Pattern Recognition Letters,2004.25(14):p. 1173-1181.
[64]Zhang, D., S. Chen, Z. Zhou. A new face recognition method based on SVD perturbation for single example image per person[J]. Applied mathematics and computation,2005.163:p.895-907.
[65]Yang, J., D. Zhang, A. Frangi, et al. Two-dimensional PCA:a new approach to appearance-based face representation and recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2004.26(1):p.131-137.
[66]Jung, H.-C., B.-W. Hwang, S.-W. Lee. Authenticating corrupted face image based on noise mode[C]. in Proceedings of the Sixth IEEE International Conference on Automatic Face and Gesture Recognition.2004.
[67]Wang, J., K.N. Plataniotis, A.N. Venetsanopoulos. Selecting discriminant eigenfaces for face recognition [J]. Pattern Recognition Letters,2005.26(10):p. 1470-1482.
[68]Torre, F.D.l., R. Gross, S. Baker, et al. Representational oriented component analysis (ROCA) for face recognition with one sample image per training class[C]. in Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) 2005:IEEE Computer Society.
[69]Martinez, A.M. Recognizing imprecisely localized, partially occluded and expression variant face from a single sample per class [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002.25(6):p.748-763.
[70]Martinez, A.M. Recognizing expression variant faces from a single sample image per class[C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR),2003.1.
[71]Yang, Q.,X. Ding. Symmetrical PAC in face recognition[C]. IEEE 2002 International Conference on Image Processing(ICIP),2002.2:p.97-100.
[72]Gutta, S.,H. Wechsler. Face recognition using asymmetric faces[M], in Biometric Authentication, D. Zhang and A.K. Jain, Editors.2004, Springer Berlin/ Heidelberg. p.1-24.
[73]Poggio, T.,T. Vetter. Recognition and structure from one 2D model view: observations on prototypes, object classes and symmetries[R]. Technical Report, Cambridge, MA, USA.:Massachusetts Institute of Technology.1992.
[74]Goldstein, A.J., L.D. Harmon, A.B. Lesk. Identification of human faces[J]. Proceedings of the IEEE 1971.59(5):p.748-760
[75]Brunelli, R.,T. Poggio. Face recognition: features versus templates[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1993.15(10):p. 1042-1052.
[76]Wu, C., C. Liu, H.-Y. Shum, et al. Automatic eyeglasses removal from face images[J]. IEEE Transactions on Pattern Analysis And Machine Intelligence,2004. 26(3):p.322-336.
[77]Manjunath, B.S., R. Chellappa, C. von der Malsburg. A feature based approach to face recognition[C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR),1992:p.373-378.
[78]Lades, M., J.C. Vorbruggen, J. Buhmann, et al. Distortion Invariant Object Recognition in the Dynamic Link Architecture[J]. IEEE Transactions on Computers, 1993.42(3):p.300-311.
[79]Wiskott, L., J.-M. Fellous, N. Kriiger, et al. Face recognition by elastic bunch graph matching[M], in Computer Analysis of Images and Patterns, G. Sommer, K. Daniilidis, and J. Pauli, Editors.1997, Springer Berlin/Heidelberg, p.456-463.
[80]Gao, Y.,Y. Qi. Robust visual similarity retrieval in single model face databases[J]. Pattern Recognition,2005.38(7):p.1009-1020.
[81]Tan, X., S. Chen, Z.-H. Zhou, et al. Recognizing partially occluded, expression variant faces from fingle training image per person with SOM and soft kNN ensemble[J]. IEEE Transactions on Neural Networks,2005.16(4):p.875-886.
[82]Chen, S., J. Liu, Z.-H. Zhou. Making FLDA applicable to face recognition with one sample per person[J]. Pattern Recognition,2004.37(7):p.1553-1555.
[83]Le, H.-S.,H. Li. Recognizing frontal face images using hidden Markov models with one training image per person[C].17th International Conference on Pattern Recognition (ICPR'04) 2004.1:p.318-321.
[84]Le, H.-S.,H. Li. Recognizing frontal face images using hidden Markov models with one training image per person[C].17th International Conference on Pattern Recognition (ICPR) 2004.1:p.318-321.
[85]Michael W. Marcellin, Michael J. Gormish, Ali Bilgin, et al. An overview of JPEG-2000[C]. in Proceedings of the Data Compression Conference.2000.
[86]Michael Elad,M. Aharon. Image denoising via sparse and redundant representations over learned dictionaries[J]. IEEE transactions on image processing a publication of the IEEE Signal Processing Society,2006.15(12):p.3736-3745.
[87]Yang, J., J. Wright, Y. Ma, et al. Image super-resolution as sparse representation of raw image patches[C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2008:p.1-8.
[88]Mark A. Davenport, R Marco F. Duarte, R Michael B. Wakin, et al. The smashed filter for compressive classification and target recognition[C]. in Proceedings of IS&T/SPIE Symposium on Electronic Imaging:Computational Imaging 2007.
[89]Huang, K.,S. Aviyente. Sparse representation for signal classification[C]. Advances in Neural Information Processing Systems(NIPS),2006.
[90]Baraniuk, R. A lecture on compressive sensing[J]. IEEE Signal Processing Magazine,2007.
[91]Boufounos, P.,R. Baraniuk. Quantization of sparse representations[C]. in Proceedings of the 2007 Data Compression Conference.2007:IEEE Computer Society.
[92]Wright, J., Y. Ma, J. Mairal, et al. Sparse representation for computer vision and pattern recognition[J]. Proceedings of IEEE,2009.
[93]Georghiades, A.S., P.N. Belhumeur, D.J. Kriegman. 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):p. 643-660.
[94]Basri, R.,D.W. Jacobs. Lambertian reflectance and linear subspaces[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003.25(2):p. 218-233.
[95]Donoho, D.L.,M. Elad. Optimally sparse representation in general (nonorthogonal) dictionaries via l1 minimization[J]. Proceedings of the National Academy of Sciences of the United States of America,2003.100(5):p.2197-2202.
[96]Donoho, D.L. For most large underdetermined systems of linear equations the minimal l1-norm solution is also the sparsest solution[J]. Communications on Pure and Applied Mathematics,2006.59(6):p.797-829.
[97]Wright, J., A.Y. Yang, A. Ganesh, et al. Robust face recognition via sparse representation——APPENDIX[J]. IEEE Transactions Pattern Analysis and Machine Intelligence,2009.31(2):p.210-227.
[98]Lee, K.-C., J. Ho, D.J. Kriegman. Acquiring Linear Subspaces for Face Recognition under Variable Lighting[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,2005.27(5):p.684-698.
[99]Stan Z. Li ,J. Lu. Face Recognition Using the Nearest Feature Line Method[J]. IEEE Transactions on Neural Networks,1999.10(2):p.439-443.
[100]Liu, J., S. Chen, z. Zhou, et al. Single image subspace for face recognition[J]. Analysis and Modeling of Faces and Gestures,2007.4778:p.205-219.
[101]杨琼,丁晓青.对称主分量分析及其在人脸识别中的应用[J].计算机学报,2003.26(9).
[102]Yin, H., P. Fu, S. Meng. Sampled FLDA for face recognition with single training image per person[J]. Neurocomputing,2006.69(16-18):p.2443-2445.
[103]Cai, D., X. He, J. Han. Semi-supervised discriminant analysis[C]. IEEE 11th International Conference on Computer Vision,2007:p.1-7.
[104]Song, Y, F. Nie, C. Zhang, et al. A unified framework for semi-supervised dimensionality reduction[J]. Pattern Recognition,2008.41(9):p.2789-2799.
[105]Jianhui, c., Y. Jieping, L. Qi. Integrating global and local structures:a least squares framework for dimensionality reduction[C]. IEEE Conference on Computer Vision and Pattern Recognition,2007.
[106]Yu, W., X. Teng, C. Liu. Face recognition using discriminant locality preserving projections[J]. Image and Vision Computing,2006.24(3):p.239-248.
[2]W.Ahao, R.Chellappa, PJ.Phillips, et al. Face recognition:a literature survey[J]. ACM Computing Surveys (CSUR),2003.35(4):p.399-458.
[3]Phillips, P.J., H. Wechsler, J. Huang, et al. The FERET database and evaluation procedure for face-recognition algorithms[J]. Image and Vision Computing,1998. 16(5):p.295-306.
[4]Rebollo-Neira, L.,Z. Xu. Sparse signal representation by adaptive non-uniform B-spline dictionaries on a compact interval[J]. Signal Processing,2010.90(7):p. 2308-2313.
[5]Baraniuk, R. Compressive sensing[J]. IEEE Signal Processing Magazine,2007. 24(4):p.118-121.
[6]Smith, K. Mind-reading with a brain scan[J]. Nature,2008.
[7]赵松年,姚力,金真,et al.视像整体特征在人类初级视皮层上的稀疏表象：脑功能成像的证据[J].科学通报,2008.53(11):p.1296-1304.
[8]Donoho, D.L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006.52(4):p.1289-1306.
[9]Candes, E. Compressive sampling[C]. in Proceedings of the International Congress of Mathematicians.2006. Madrid, Spain.
[10]Candes, E.J.,T. Tao. Decoding by linear programming[J]. IEEE Transactions on Information Theory,2005.51(12):p.4203-4215.
[11]Candes, E.J.,T. Tao. Reflections on compressed sensing[J]. IEEE Information Theory Society Newslette,2008.58(4):p.20-23.
[12]Wright, J., A.Y. Yang, A. Ganesh, et al. Robust face recognition via sparse representation[J]. IEEE Transactions Pattern Analysis and Machine Intelligence, 2009.31(2):p.210-227.
[13]Hegde, C., M.B. Wakin, R.G. Baraniuk. Random projections for manifold learning[C]. The Twenty-First Annual Conference on Neural Information Processing Systems (NIPS),2007.
[14]Haupt, J.,R. Nowak. Compressive sampling for signal detection[C]. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2007.
[15]Sheikh, M.A.,R.G. Baraniuk. Blind error-free detection of transform-domain watermarks[C]. IEEE International Conference on Image Processing(ICIP),2007.
[16]Ji, S., Y. Xue, L. Carin. Bayesian Compressive Sensing[J]. IEEE Transactions on Signal Processing,2008.56(6):p.2346-2356
[17]Wakin, M.B. The geometry of low-dimensional signal models[J]. PhD Thesis, Rice University, Houston, TX,,2007.
[18]Hubel, D.H.,T.N. Wiesel. Receptive fields of single neurons in the cat's striate cortex[J]. Journal of Physiology,1959.148(3):p.574-591.
[19]Barlow, H.B. The coding of sensory messages[M], in Current Problems in Animal Behaviors, Chapter XIII, Cambridge Universtity Press.1961. p.331-360.
[20]Barlow, H.B. Single units and sensation:a neuron doctrine for perceptual psychology?[J]. Perception,1972.1(4):p.371-394.
[21]Michison, G. The organization of sequential memory: sparse representations and the targeting problem[M]. Organiztion of Neural Networks, VCH Verlagsgesellschaft, Weinheim.1988.347-367.
[22]Olshausen, B.A. Emergence of simple-cell receptive field properties by learning a sparse code for natural images[J]. Nuture,1996.381(6583):p.607-609.
[23]BA, O.,F. DJ. Sparse coding with an overcomplete basis set: a strategy employed by V1[J]. Vision Research,1997.37(23):p.3311-25.
[24]Olshausen, B.A.,D.J. Field. Sparse coding of sensory inputs[J]. Current Opinion in Neurobiology,2004.14(4):p.481-487.
[25]Reddy, L.,N. Kanwisher. Coding of visual objects in the ventral stream[J]. Current Opinion in Neurobiology,2006.16(4):p.408-414.
[26]Riesenhuber, M.,T. Poggio. Are cortical models really bound by the "binding problem"?[J]. Neuron,1999.24(1):p.87-93.
[27]Bell, A.J.,T.J. Sejnowski. The "independent components" of natural scenes are edge filters[J]. Vision Research,1997.37(23):p.3327-3338.
[28]Tenenbaum, J.,W. Freeman. Separating style and content with bilinear models. 2000; Available from:http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.33. 4444.
[29]Hyvarinen, A.,P.O. Hoyer. A two-layer sparse coding model learns simple and complex cell receptive fields and topography from natural images[J]. Vision Research,2001.41(18):p.2413-2423.
[30]张海,王尧,常象宇,et al. L1/2正则化[J].中国科学信息科学,2010.40(3):p.412-422.
[31]Li, Y.,S.-i. Amari. Two conditions for equivalence of 0-norm solution and 1-norm solution in sparse representation[J]. IEEE Transactions on Neural Networks,2010. 21(7):p.1189-1196.
[32]Zhou, G., Z. Yang, S. Xie, et al. Mixing matrix estimation from sparse mixtures with unknown number of sources [J]. IEEE Transactions on Neural Networks,2011. 22(2):p.211-221.
[33]Shi, G, J. Lin, X. Chen, et al. UWB echo signal detection with ultra-low rate sampling based on compressed sensing[J]., IEEE Transactions on Circuits and Systems Ⅱ:Express Briefs,2008.55(4):p.379-383.
[34]Zhang, L.,W.-D. Zhou. Sparse ensembles using weighted combination methods based on linear programming[J]. Pattern Recognition,2010.44(1):p.97-106.
[35]Li, Y.-R., D.-Q. Dai, L. Shen. Multiframe super-resolution reconstruction using sparse directional regularization[J]. Circuits and Systems for Video, IEEE Transactions on,2010.20(7):p.945-956.
[36]Ji, S., L. Yuan, Y.-X. Li, et al. Drosophila gene expression pattern annotation using sparse features and term-term interactions[C]. Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining,2009: p.407-416.
[37]Qiao, L., S. Chen, X. Tan. Sparsity preserving projections with applications to face recognition[J]. Pattern Recognition,2010.43(1):p.331-341.
[38]Lai, Z., M. Wan, Z. Jin, et al. Sparse two-dimensional local discriminant projections for feature extraction[J]. Neurocomputing,2011.74(4):p.629-637.
[39]Cai, D.,X. He. Sparse concept coding for visual analysis[C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2011.
[40]Wu, F., Y. Yuan, Y. Zhuang. Heterogeneous feature selection by group lasso with logistic regression[C], in Proceedings of the international conference on Multimedia.2010. Firenze, Italy:ACM.
[41]Zhang, Z., G. Dai, M.I. Jordan. Bayesian generalized kernel mixed models[J]. J. Mach. Learn. Res.,2011.12:p.111-139.
[42]Liu, B., S. Chen, M. Qian, et al. Sparse norm-regularized reconstructive coefficients learning[C].2009 Ninth IEEE International Conference on Data Mining,2009:p.854-859.
[43]Zhao, R., X. Liu, C.-C. Li, et al. Wavelet denoising via sparse representation[J]. Science in China Series F:Information Sciences,2009.52(8):p.1371-1377.
[44]Ma, B.,T. Wang. Head pose estimation using sparse representation[C]. Proceedings of the 2010 Second International Conference on Computer Engineering and Applications - Volume 02,2010:p.389-392.
[45]Sun, S.,J. Shawe-Taylor. Sparse semi-supervised learning using conjugate functions[J]. J. Mach. Learn. Res.,2010.9999:p.2423-2455.
[46]Xu, R., B. Zhang, Q. Ye, et al. Human detection in images via L1-norm minimization learning. [C]. IEEE International Conference on Acoustics, Speech and Signal Processing,2010:p.3566-3569.
[47]Li, Z., Z. Shi, X. Liu, et al. A novel sparse coding model based on structural similarity[C]. IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP) 2010:p.4170-4173
[48]Liu, W., S. Zheng, S. Jia, et al. Sparse nonnegative matrix factorization with the elastic net[C]. IEEE International Conference on Bioinformatics and Biomedicine (BIBM),2010:p.265-268.
[49]Tibshirani, R. Regression shrinkage and selection via the lasso[J]. J. Roy. Statist. Soc. Ser. B,1996.58(1):p.267-288.
[50]Efron, B., T. Hastie, I. Johnstone, et al. Least angle regression[J]. The Annals of Statistics,2004.32(2):p.407-499.
[51]Zou, H.,T. Hastie. Regularization and variable selection via the Elastic Net[J]. Journal of the Royal Statistical Society B,2005.67:p.301-320.
[52]Mackey, L. Deflation methods for sparse PCA[C]. Advances in Neural Information Processing Systems 21,2009:p.1017-1024.
[53]Zass, R.,A. Shashua. Nonnegative sparse PCA[C]. In Neural Information Processing Systems,2007.
[54]Zou, H., T. Hastie, R. Tibshirani. Sparse principal component analysis[J]. Journal of Computational and Graphical Statistics,2006.15(2):p.265-286.
[55]Jenatton, R., G. Obozinski, F. Bach. Structured sparse principal component analysis[J].2009.
[56]Lee, D., W. Lee, Y. Lee, et al. Super-sparse principal component analyses for high-throughput genomic data[J]. BMC bioinformatics,2010.11(1):p.296.
[57]Clemmensen, L., T. Hastie, B. Ersb(?)ll. Sparse discriminant analysis[J]. Technical Report,2008.
[58]Hastie, T., A. Buja, R. Tibshirani. Penalized discriminant analysis[J]. Annals of Statistics,1995.23(1):p.73-102.
[59]Qiao, L., S. Chen, X. Tan. Sparsity preserving discriminant analysis for single training image face recognition[J]. Pattern Recognition Letters,2010.31(5):p. 422-429.
[60]Wang, J., K.N. Plataniotis, J. Lu, et al. On solving the face recognition problem with one training sample per subject[J]. Pattern Recognition,2006.39(9):p. 1746-1762.
[61]Tan, X., S. Chen, Z.-H. Zhou, et al. Face recognition from a single image per person:A survey [J]. Pattern Recognition,2006.39(9):p.1725-1745.
[62]Wu, J.,Z. Zhou. Face recognition with one training image per person[J]. Pattern Recognition Letters,2002.23(14):p.1711-1719.
[63]Chen, S., D. Zhang, Z.-H. Zhou. Enhanced (PC)2 A for face recognition with one training image per person[J]. Pattern Recognition Letters,2004.25(14):p. 1173-1181.
[64]Zhang, D., S. Chen, Z. Zhou. A new face recognition method based on SVD perturbation for single example image per person[J]. Applied mathematics and computation,2005.163:p.895-907.
[65]Yang, J., D. Zhang, A. Frangi, et al. Two-dimensional PCA:a new approach to appearance-based face representation and recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2004.26(1):p.131-137.
[66]Jung, H.-C., B.-W. Hwang, S.-W. Lee. Authenticating corrupted face image based on noise mode[C]. in Proceedings of the Sixth IEEE International Conference on Automatic Face and Gesture Recognition.2004.
[67]Wang, J., K.N. Plataniotis, A.N. Venetsanopoulos. Selecting discriminant eigenfaces for face recognition [J]. Pattern Recognition Letters,2005.26(10):p. 1470-1482.
[68]Torre, F.D.l., R. Gross, S. Baker, et al. Representational oriented component analysis (ROCA) for face recognition with one sample image per training class[C]. in Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) 2005:IEEE Computer Society.
[69]Martinez, A.M. Recognizing imprecisely localized, partially occluded and expression variant face from a single sample per class [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002.25(6):p.748-763.
[70]Martinez, A.M. Recognizing expression variant faces from a single sample image per class[C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR),2003.1.
[71]Yang, Q.,X. Ding. Symmetrical PAC in face recognition[C]. IEEE 2002 International Conference on Image Processing(ICIP),2002.2:p.97-100.
[72]Gutta, S.,H. Wechsler. Face recognition using asymmetric faces[M], in Biometric Authentication, D. Zhang and A.K. Jain, Editors.2004, Springer Berlin/ Heidelberg. p.1-24.
[73]Poggio, T.,T. Vetter. Recognition and structure from one 2D model view: observations on prototypes, object classes and symmetries[R]. Technical Report, Cambridge, MA, USA.:Massachusetts Institute of Technology.1992.
[74]Goldstein, A.J., L.D. Harmon, A.B. Lesk. Identification of human faces[J]. Proceedings of the IEEE 1971.59(5):p.748-760
[75]Brunelli, R.,T. Poggio. Face recognition: features versus templates[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1993.15(10):p. 1042-1052.
[76]Wu, C., C. Liu, H.-Y. Shum, et al. Automatic eyeglasses removal from face images[J]. IEEE Transactions on Pattern Analysis And Machine Intelligence,2004. 26(3):p.322-336.
[77]Manjunath, B.S., R. Chellappa, C. von der Malsburg. A feature based approach to face recognition[C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR),1992:p.373-378.
[78]Lades, M., J.C. Vorbruggen, J. Buhmann, et al. Distortion Invariant Object Recognition in the Dynamic Link Architecture[J]. IEEE Transactions on Computers, 1993.42(3):p.300-311.
[79]Wiskott, L., J.-M. Fellous, N. Kriiger, et al. Face recognition by elastic bunch graph matching[M], in Computer Analysis of Images and Patterns, G. Sommer, K. Daniilidis, and J. Pauli, Editors.1997, Springer Berlin/Heidelberg, p.456-463.
[80]Gao, Y.,Y. Qi. Robust visual similarity retrieval in single model face databases[J]. Pattern Recognition,2005.38(7):p.1009-1020.
[81]Tan, X., S. Chen, Z.-H. Zhou, et al. Recognizing partially occluded, expression variant faces from fingle training image per person with SOM and soft kNN ensemble[J]. IEEE Transactions on Neural Networks,2005.16(4):p.875-886.
[82]Chen, S., J. Liu, Z.-H. Zhou. Making FLDA applicable to face recognition with one sample per person[J]. Pattern Recognition,2004.37(7):p.1553-1555.
[83]Le, H.-S.,H. Li. Recognizing frontal face images using hidden Markov models with one training image per person[C].17th International Conference on Pattern Recognition (ICPR'04) 2004.1:p.318-321.
[84]Le, H.-S.,H. Li. Recognizing frontal face images using hidden Markov models with one training image per person[C].17th International Conference on Pattern Recognition (ICPR) 2004.1:p.318-321.
[85]Michael W. Marcellin, Michael J. Gormish, Ali Bilgin, et al. An overview of JPEG-2000[C]. in Proceedings of the Data Compression Conference.2000.
[86]Michael Elad,M. Aharon. Image denoising via sparse and redundant representations over learned dictionaries[J]. IEEE transactions on image processing a publication of the IEEE Signal Processing Society,2006.15(12):p.3736-3745.
[87]Yang, J., J. Wright, Y. Ma, et al. Image super-resolution as sparse representation of raw image patches[C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2008:p.1-8.
[88]Mark A. Davenport, R Marco F. Duarte, R Michael B. Wakin, et al. The smashed filter for compressive classification and target recognition[C]. in Proceedings of IS&T/SPIE Symposium on Electronic Imaging:Computational Imaging 2007.
[89]Huang, K.,S. Aviyente. Sparse representation for signal classification[C]. Advances in Neural Information Processing Systems(NIPS),2006.
[90]Baraniuk, R. A lecture on compressive sensing[J]. IEEE Signal Processing Magazine,2007.
[91]Boufounos, P.,R. Baraniuk. Quantization of sparse representations[C]. in Proceedings of the 2007 Data Compression Conference.2007:IEEE Computer Society.
[92]Wright, J., Y. Ma, J. Mairal, et al. Sparse representation for computer vision and pattern recognition[J]. Proceedings of IEEE,2009.
[93]Georghiades, A.S., P.N. Belhumeur, D.J. Kriegman. 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):p. 643-660.
[94]Basri, R.,D.W. Jacobs. Lambertian reflectance and linear subspaces[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003.25(2):p. 218-233.
[95]Donoho, D.L.,M. Elad. Optimally sparse representation in general (nonorthogonal) dictionaries via l1 minimization[J]. Proceedings of the National Academy of Sciences of the United States of America,2003.100(5):p.2197-2202.
[96]Donoho, D.L. For most large underdetermined systems of linear equations the minimal l1-norm solution is also the sparsest solution[J]. Communications on Pure and Applied Mathematics,2006.59(6):p.797-829.
[97]Wright, J., A.Y. Yang, A. Ganesh, et al. Robust face recognition via sparse representation——APPENDIX[J]. IEEE Transactions Pattern Analysis and Machine Intelligence,2009.31(2):p.210-227.
[98]Lee, K.-C., J. Ho, D.J. Kriegman. Acquiring Linear Subspaces for Face Recognition under Variable Lighting[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,2005.27(5):p.684-698.
[99]Stan Z. Li ,J. Lu. Face Recognition Using the Nearest Feature Line Method[J]. IEEE Transactions on Neural Networks,1999.10(2):p.439-443.
[100]Liu, J., S. Chen, z. Zhou, et al. Single image subspace for face recognition[J]. Analysis and Modeling of Faces and Gestures,2007.4778:p.205-219.
[101]杨琼,丁晓青.对称主分量分析及其在人脸识别中的应用[J].计算机学报,2003.26(9).
[102]Yin, H., P. Fu, S. Meng. Sampled FLDA for face recognition with single training image per person[J]. Neurocomputing,2006.69(16-18):p.2443-2445.
[103]Cai, D., X. He, J. Han. Semi-supervised discriminant analysis[C]. IEEE 11th International Conference on Computer Vision,2007:p.1-7.
[104]Song, Y, F. Nie, C. Zhang, et al. A unified framework for semi-supervised dimensionality reduction[J]. Pattern Recognition,2008.41(9):p.2789-2799.
[105]Jianhui, c., Y. Jieping, L. Qi. Integrating global and local structures:a least squares framework for dimensionality reduction[C]. IEEE Conference on Computer Vision and Pattern Recognition,2007.
[106]Yu, W., X. Teng, C. Liu. Face recognition using discriminant locality preserving projections[J]. Image and Vision Computing,2006.24(3):p.239-248.