面向人脸识别的流形正则化判别特征提取算法研究
|
摘要
人脸识别由于其自然、直观、非接触、安全、快捷等特点，是一项极具发展潜力的生物特征识别技术。人脸识别技术又是一个非常活跃的研究领域，涉及计算机视觉、模式识别、图像处理、机器学习和认知科学等多门学科。对人脸识别技术的研究具有重要的理论意义和应用价值。
如何有效地从人脸图像中提取鉴别特征，是人脸识别需要解决的关键问题。子空间方法是目前众多特征提取技术中的主流方法。一般认为，人脸从某种意义上来说是一种流形结构，人脸数据集是由某些内在变量控制形成的非线性流形。基于流形学习的人脸识别研究引起了人们的广泛关注。本文在人脸识别的研究背景下，在不同的算法发展框架下，针对保局子空间特征提取方法中存在的样本扩展、参数自适应选择、判别扩展、非线性核扩展、非负扩展和小样本尺寸(SSS)等问题，进行了流形正则化判别特征提取的算法研究，并通过标准人脸数据库仿真实验验证了它们的有效性。
本文的主要内容和创新之处可概述如下：
1.针对线性判别分析没有考虑数据内在局部几何结构，难以有效应对对齐裁剪且具有较大照明、表情变化人脸识别问题的缺点，在图嵌入算法发展框架下，提出两种嵌入邻域关系的判别分析方法。其中零空间判别投射(NDPE)算法采用能够同时约简噪音和增强内在差的预降维方法，并有效地利用了类内局部散度矩阵零空间中的判别信息。为避免稠密矩阵特征分解并减少判别信息损失，进一步提出基于谱回归求解方法的正交保局判别映射(SROLPDM)算法。为提高识别性能，分别提出基于选择后的双树复小波特征与人脸图像特征的得分层和复数特征融合人脸识别方法。
2.针对流形学习方法存在样本扩展并且没有充分利用样本类别标签信息，不适用于具有较大照明、表情或姿态变化人脸识别问题的缺点，在片排列框架内提出一种称为判别改进局部切空间排列(DILTSA)的特征提取方法。该方法基于改进型局部切空间排列(ILTSA)，分别保持类内与类间样本的局部几何关系并进行全局排列。所提方法采用散度差形式的目标准则，完全避免了小样本尺寸问题。为进一步提高识别性能，提出增强型Gabor-like小波，在此基础上提出以多分辨率小波特征或融合特征作为DILTSA输入的人脸识别方法。
3.针对保局子空间特征提取的参数自适应选择和非负判别扩展问题，以具有较大遮盖或表情变化的人脸识别问题为应用背景，基于结合改进最大间隔准则和稀疏学习方法，设计判别形式的局部保持目标函数，首先提出称为最大间隔稀疏表示判别映射(MSRDM)的特征提取算法。在此基础上，结合非负矩阵分解(NMF)的部分表示特点，提出一种判别邻域间隔正则化非负矩阵分解(DNMRNMF)算法。该算法能够在最小化逼近误差的同时，使类内样本更加紧凑，类间样本更加远离。
4.针对线性保局子空间特征提取方法，难以有效应对外观变化复杂人脸识别问题的缺点，提出一种正则化核判别局部样条嵌入(RKDLSE)特征提取方法。该方法基于对局部样条嵌入(LSE)的线性逼近和非线性核判别扩展，通过采用不同的方法建模局部类内和类间散度矩阵，采用类内散度矩阵特征值正则化的求解方法，解决了目标函数求解中的矩阵奇异问题。此外，基于新提出的WLTSA非线性降维方法，并以聚类中心作为算法输入，拓展RKDLSE用于图像集人脸识别。
For the merits of being natural, directly perceived, safe and convenient,face recognition is a biometric technology with great developable potential.Face recognition is also one of the most active research areas, which isclosely related to many disciplines such as Computer Vision, PatternRecognition, Image Processing, Machine Learning and Cognitive Psychology,etc. The study of face recognition technology has both tremendous applicationvalues and important theory significance.
A central issue to a successful approach for face recognition is how toextract discriminative features from the facial images. Many featureextraction methods have been proposed and among them the subspaceanalysis has received extensive attention. In a sense, it is commonly acceptedthat human face is a manifold structure. Face dataset is a nonlinear manifoldformed by some inner variables. Face recognition research based on manifoldlearning is attracting more and more attention. Under different algorithmdevelopment frameworks, our research focuses on developing manifoldregularized locality preservation feature extraction algorithms for facerecognition, which can alleviate the out-of-sample, adaptive parameterselection, discriminant extension, nonlinear kernel extension, nonnegativeextension and small sample size (SSS) problems. Each algorithm’seffectiveness has been verified through the simulation experiments onbenchmark face databases.
The main contents and originalities of this dissertation are summarized asfollows:
1. To overcome the drawback of ignoring local geometric structures ofdata set in LDA and therefore not being able to cope with face recognitionproblems that take the aligned and cropped images with differentillumination and expression variations as input, two locality preserving discriminant analysis methods are developed under the graph embeddingframework. To avoid the SSS problem, a preprocessing dimensionalityreduction method which can reduce the noise effect and enhance theintrinsic difference is adopted in Null Space Discriminative ProjectionEmbedding (NDPE) algorithm and NDPE can seek more discriminativeinformation in the null space of locality preservation within-class scattermatrix. To avoid the eigen-decomposition of dense matrices and reducethe loss of discriminative information, spectral regression technologybased orthogonal locality preserving discriminant mapping algorithm ispresented and we name this algorithm SROLPDM. To improverecognition performance, face recognition methods fusing selectedDTCWT features and original face features at score level or usingcomplex number approach are also developed.
2. In manifold learning methods, there exist the out-of-sample problemand the label information of training samples is ignored. Therefore, theoriginal manifold learning methods are not suitable for face recognitionproblems with great illumination, expression or pose variations. Based onthe within-class and between-class local geometry preservation with theirglobal alignment, a novel feature extraction algorithm namedDiscriminant Improved Local Tangent Space Alignment (DILTSA) isproposed in the patch alignment framework. Scatter difference objectivecriterion is adopted in our method and small sample size problem isavoided naturally. To improve the recognition performance, facerecognition methods using augmented Gabor-like wavelet features or thefusion of multi-resolution wavelet features and original face features arealso developed.
3. To the adaptive parameter selection and discriminative nonnegativeextension of locality preserving subspace feature extraction for facerecognition with big occlusion or expression variations, based onintegration of modified maximum margin criterion and sparse learning, bydesigning new locality preserving discriminant objective function, a novel feature extraction method named Maximum Margin SparseRepresentation Discriminant Mapping (MSRDM) is proposed. By takingthe part indication virtue of NMF into account, a low-rank approximationalgorithm named Discriminant Neighborhood Margin Regularized NMF(DNMRNMF) is also proposed. DNMRNMF can minimize theapproximation error while contracting intra-class neighborhoods andexpanding inter-class neighborhoods.
4. To overcome the drawback of not being able to cope with complex facerecognition problems in linear locality preservation subspace featureextraction methods, a novel feature extraction method named RegularizedKernel Discriminant Local Spline Embedding (RKDLSE) is developed.RKDLSE is a linear approximation and discriminant kernel extensionalgorithm of Local Spline Embedding (LSE). Different methods areadopted to model the within-class and between-class scatter matrices.Regularization method is used to resolve the within-class scatter matrixsingularity. Moreover, with the novel WLTSA nonlinear dimensionalityreduction method and using clustering centers as input, RKDLSE isextended to image set face recognition application.
如何有效地从人脸图像中提取鉴别特征，是人脸识别需要解决的关键问题。子空间方法是目前众多特征提取技术中的主流方法。一般认为，人脸从某种意义上来说是一种流形结构，人脸数据集是由某些内在变量控制形成的非线性流形。基于流形学习的人脸识别研究引起了人们的广泛关注。本文在人脸识别的研究背景下，在不同的算法发展框架下，针对保局子空间特征提取方法中存在的样本扩展、参数自适应选择、判别扩展、非线性核扩展、非负扩展和小样本尺寸(SSS)等问题，进行了流形正则化判别特征提取的算法研究，并通过标准人脸数据库仿真实验验证了它们的有效性。
本文的主要内容和创新之处可概述如下：
1.针对线性判别分析没有考虑数据内在局部几何结构，难以有效应对对齐裁剪且具有较大照明、表情变化人脸识别问题的缺点，在图嵌入算法发展框架下，提出两种嵌入邻域关系的判别分析方法。其中零空间判别投射(NDPE)算法采用能够同时约简噪音和增强内在差的预降维方法，并有效地利用了类内局部散度矩阵零空间中的判别信息。为避免稠密矩阵特征分解并减少判别信息损失，进一步提出基于谱回归求解方法的正交保局判别映射(SROLPDM)算法。为提高识别性能，分别提出基于选择后的双树复小波特征与人脸图像特征的得分层和复数特征融合人脸识别方法。
2.针对流形学习方法存在样本扩展并且没有充分利用样本类别标签信息，不适用于具有较大照明、表情或姿态变化人脸识别问题的缺点，在片排列框架内提出一种称为判别改进局部切空间排列(DILTSA)的特征提取方法。该方法基于改进型局部切空间排列(ILTSA)，分别保持类内与类间样本的局部几何关系并进行全局排列。所提方法采用散度差形式的目标准则，完全避免了小样本尺寸问题。为进一步提高识别性能，提出增强型Gabor-like小波，在此基础上提出以多分辨率小波特征或融合特征作为DILTSA输入的人脸识别方法。
3.针对保局子空间特征提取的参数自适应选择和非负判别扩展问题，以具有较大遮盖或表情变化的人脸识别问题为应用背景，基于结合改进最大间隔准则和稀疏学习方法，设计判别形式的局部保持目标函数，首先提出称为最大间隔稀疏表示判别映射(MSRDM)的特征提取算法。在此基础上，结合非负矩阵分解(NMF)的部分表示特点，提出一种判别邻域间隔正则化非负矩阵分解(DNMRNMF)算法。该算法能够在最小化逼近误差的同时，使类内样本更加紧凑，类间样本更加远离。
4.针对线性保局子空间特征提取方法，难以有效应对外观变化复杂人脸识别问题的缺点，提出一种正则化核判别局部样条嵌入(RKDLSE)特征提取方法。该方法基于对局部样条嵌入(LSE)的线性逼近和非线性核判别扩展，通过采用不同的方法建模局部类内和类间散度矩阵，采用类内散度矩阵特征值正则化的求解方法，解决了目标函数求解中的矩阵奇异问题。此外，基于新提出的WLTSA非线性降维方法，并以聚类中心作为算法输入，拓展RKDLSE用于图像集人脸识别。
For the merits of being natural, directly perceived, safe and convenient,face recognition is a biometric technology with great developable potential.Face recognition is also one of the most active research areas, which isclosely related to many disciplines such as Computer Vision, PatternRecognition, Image Processing, Machine Learning and Cognitive Psychology,etc. The study of face recognition technology has both tremendous applicationvalues and important theory significance.
A central issue to a successful approach for face recognition is how toextract discriminative features from the facial images. Many featureextraction methods have been proposed and among them the subspaceanalysis has received extensive attention. In a sense, it is commonly acceptedthat human face is a manifold structure. Face dataset is a nonlinear manifoldformed by some inner variables. Face recognition research based on manifoldlearning is attracting more and more attention. Under different algorithmdevelopment frameworks, our research focuses on developing manifoldregularized locality preservation feature extraction algorithms for facerecognition, which can alleviate the out-of-sample, adaptive parameterselection, discriminant extension, nonlinear kernel extension, nonnegativeextension and small sample size (SSS) problems. Each algorithm’seffectiveness has been verified through the simulation experiments onbenchmark face databases.
The main contents and originalities of this dissertation are summarized asfollows:
1. To overcome the drawback of ignoring local geometric structures ofdata set in LDA and therefore not being able to cope with face recognitionproblems that take the aligned and cropped images with differentillumination and expression variations as input, two locality preserving discriminant analysis methods are developed under the graph embeddingframework. To avoid the SSS problem, a preprocessing dimensionalityreduction method which can reduce the noise effect and enhance theintrinsic difference is adopted in Null Space Discriminative ProjectionEmbedding (NDPE) algorithm and NDPE can seek more discriminativeinformation in the null space of locality preservation within-class scattermatrix. To avoid the eigen-decomposition of dense matrices and reducethe loss of discriminative information, spectral regression technologybased orthogonal locality preserving discriminant mapping algorithm ispresented and we name this algorithm SROLPDM. To improverecognition performance, face recognition methods fusing selectedDTCWT features and original face features at score level or usingcomplex number approach are also developed.
2. In manifold learning methods, there exist the out-of-sample problemand the label information of training samples is ignored. Therefore, theoriginal manifold learning methods are not suitable for face recognitionproblems with great illumination, expression or pose variations. Based onthe within-class and between-class local geometry preservation with theirglobal alignment, a novel feature extraction algorithm namedDiscriminant Improved Local Tangent Space Alignment (DILTSA) isproposed in the patch alignment framework. Scatter difference objectivecriterion is adopted in our method and small sample size problem isavoided naturally. To improve the recognition performance, facerecognition methods using augmented Gabor-like wavelet features or thefusion of multi-resolution wavelet features and original face features arealso developed.
3. To the adaptive parameter selection and discriminative nonnegativeextension of locality preserving subspace feature extraction for facerecognition with big occlusion or expression variations, based onintegration of modified maximum margin criterion and sparse learning, bydesigning new locality preserving discriminant objective function, a novel feature extraction method named Maximum Margin SparseRepresentation Discriminant Mapping (MSRDM) is proposed. By takingthe part indication virtue of NMF into account, a low-rank approximationalgorithm named Discriminant Neighborhood Margin Regularized NMF(DNMRNMF) is also proposed. DNMRNMF can minimize theapproximation error while contracting intra-class neighborhoods andexpanding inter-class neighborhoods.
4. To overcome the drawback of not being able to cope with complex facerecognition problems in linear locality preservation subspace featureextraction methods, a novel feature extraction method named RegularizedKernel Discriminant Local Spline Embedding (RKDLSE) is developed.RKDLSE is a linear approximation and discriminant kernel extensionalgorithm of Local Spline Embedding (LSE). Different methods areadopted to model the within-class and between-class scatter matrices.Regularization method is used to resolve the within-class scatter matrixsingularity. Moreover, with the novel WLTSA nonlinear dimensionalityreduction method and using clustering centers as input, RKDLSE isextended to image set face recognition application.
引文
[1] Zhao W., Chellappa R., Rosenfeld A., Philips P. J. Face recognition: a literature survey[J]. ACMComputing Survey.2003,35:399-458.
[2]边肇祺,张学工.模式识别(第二版).北京:清华大学出版社.2000:32-85.
[3] Seung H., Lee D.The manifold ways of perception[J].Science.2000,290(5500):2268-2269.
[4] Nayar S., Nene S., Murase H. Subspace method for robot vision[J]. IEEE Transactions onRobotics and Automation.1996,12(5):750-758.
[5]杨飞，苏剑波.人脸显性特征的融合构造方法及识别.电子学报.2012,40(3):466-471.
[6]侯书东，孙权森.稀疏保持典型相关分析及在特征融合中的应用.自动化学报.2012,38(4):659-665.
[7]严严，章毓晋.基于视频的人脸识别研究进展.计算机学报.2009,32(5):878-886.
[8]李勇智.图像特征提取方法及其在人脸识别中的应用[博士论文].南京：南京理工大学.2009.
[9] Ding X. Q., Fang C. H. Discussion on some problems in face recognition[J]. Sinobiometrics.2004,3338:47-56.
[10] Kanade T. Computer recognition of human faces. Basel&Stuttgart: Birkhauser Verlag.1977:20-56.
[11] Buhr R. Analyze und klassifikation von gesichtsbildern[J]. ntzArchiv.1986,8(10):245-256.
[12] Yulle A. L., Hallinan P. W., Cohen D. S. Feature extraction from face using deformable templates[J]. International Journal of Computer Vision.1992,8(2):99-111.
[13] Craw I. Recognizing face features and faces[C]. Proc. of IEE Colloquium on Machine Storge andRecognition of Face. London, UK,1992:711-714.
[14] Lee S. Y., Ham Y. K., Park R. H. Recognition of human front faces using knowledge-basedfeature extraction and neuro-fuzzy algorithm[J]. Pattern Recognition.1996,29(11):1863-1876.
[15] Samaria F. Face recognition using hidden markov models:[PhD thesis].Cambridge: University ofCambridge,1994.
[16] Nefian A. Hidden Markov Model-based approach for face detection and recognition:[PhD thesis].Atlanta: Georgia Institute of Technology,1999.
[17]刘小军,王东峰,张丽飞.一种基于奇异值分解和隐马尔可夫模型的人脸识别方法.计算机学报,2003,26(3):340-344.
[18] Hallinan P. A deformable model for the recognition of human faces under arbitrary illumination:
[PhD thesis]. Cambridge: Harvard University.1995.
[19] Lee M. W., Ranganath S.3D deformable face model for pose determination and facesynthesis[C]. Proc. of the10th International Conference on Image Analysis and Processing, Venice,Italy.1999:532-537.
[20] Xiao J., Kanade T. Non-rigid shape and motion recovery: degrenerate deformations[C]. Proc. ofthe2004IEEE Conference on Computer Vision and Pattern Recognition, Washington D C, USA.2004:668-675.
[21] Vetter T., Blanz V. A morphable model for the synthesis of3D faces[C]. Proc. of ACMSIGGRAPH99Conference, Los Angles, USA.1999:187-194.
[22] Kohonen. Self-organization and associative memory. Berlin: Spring-Verlag.1988:37-92.
[23] Cottrell G. W., Fleming M. K. Face recognition using unsupervised feature extraction[C]. Proc.of International Neural Network Conference.1990:322-325.
[24] Lawrence S., Giles C. L., et al. Face recognition: a convolutional neural-network approach[J].IEEE Transactions on Neural Networks.1997,8(1):98-113.
[25] Demers D., Cottrell G. W. Nonlinear dimensionality reduction[C]. Proc. of Advances in NeuralInformation Processing System.1993:580-587.
[26] Lin S. H., Kung S. Y., Lin L. J., et al. Face recognition/detection by probalilistic decision-basedneural network[J]. IEEE Transactions on Neural Network.1997,8(1):114-132.
[27] Gutta S., Wechsler H. Face recognition using hybrid classifiers[J]. Pattern Recogntion.1997,30(4):539-553.
[28] Luh G, C., Lin C. Y. PCA based immune networks for human face recognition[J]. Applied SoftComputing.2011,11:1743-1752.
[29] Wong Y. W., Seng K. P., Ang L. M. Radial basis function neural network with incrementallearning for face recognition[J]. IEEE Transactions on Systems, Man, and Cybernetics-Part B:Cybernetics.2011,41(4):940-949.
[30] Sudha N., Mohan A. R., Meher P. K. A self-configurable systolic architecture for face recognitionsystem based on principal componet neural network[J]. IEEE Transactions on Circuits and Systemsfor Video Technology.2011,21(8):1071-1084.
[31] Vapnik V. The nature of statistical learning theory. New York: Springer.1995.
[32] Philips P. J. Support vector machine applied to face recognition[C]. In: Advanced in NeuralInformation Processing System II. MIT Press.1998:803-809.
[33] Guo G., Li S. Z., Chan K. L. Support vector machine for face recognition[J]. Image and VisionComputing.2001,19(9-10):631-638.
[34] Ko J., Byun H. N-division output coding method applied to face recognition[J]. PatternRecognition Letters.2003,24(16):3115-3123.
[35] Heisele B., Ho P., Wu J., Poggio T. Face recognition: component-based versus globalapproaches[J]. Computer Vision and Image Understanding.2003,91(1-2):6-21.
[36] Ivanov Y., Heisele B., Serre T. Using component features for face recognition[C]. Proceedings ofthe Sixth IEEE International Conference on Automatic Face and Gesture Recognition (FGR’04).2004:421-426.
[37] Casasent D., Yuan C. Face recognition with pose and illumination variations using new SVRDMsupport-vector machine [J]. Optical Engineering.2004,43(8):1804-1813.
[38] Ko J., Byun H. Combining SVM classifiers for multiclass problem: its application to facerecognition[C]. AVBPA2003, LNCS2688.2003:531-539.
[39] Li Z., Tang X. Bayesian face recognition using support vector machine and face clustering[C].Proceedings of the2004IEEE Computer Society Conference on Computer Vision and PatternRecognition (CVPR’04).2004:374-380.
[40] Lu J., Plataniotis K. N., Venetsanopoulos A.N. Face recognition using feature optimization andv-support vector learning[C]. Proceedings of the IEEE2001International Workshop on NeuralNetworks for Signal Processing, Falmouth, MA, USA. September2001:373-382.
[41] Deniz O., Castrillon M., Hernandez M. Face recognition using independent component analysisand support vector machines[J]. Pattern Recognition Letters.2003,24(13):2153-2157.
[42] Gumus E., Killic N., Sertbas A., Ucan O. N. Evaluation of face recognition techniques usingPCA, wavelet and SVM[J]. Expert Systems with Applications.2010,37(9):6404-6408.
[43] Kanan H. R., Faez K. GA-based optimal selection of PZMI features for face recognition[J].Applied Mathematics and Computation.2008,205(2):706-715.
[44] Smach F., Cedric L., Gauthier J. P., et al. Generalized Fourier descriptors with applications toobjects recognition in SVM context[J]. J. Math. Imaging Vis.2008,30:43-71.
[45] Zhang B.,Chen X.,Shan S.,Gao W. Nonlinear face recognition based on maximum average margincriterion[C]. Proc. of the2005IEEE Computer Society Conference on Computer Vision and PatternRecognition(CVPR’05).2005:554-559.
[46] Song F., Yang J., Liu S. Large margin linear projection and face recongnition[J]. PatternRecognition.2004,37(9):1953-1955.
[47] Fortuna J., David C. Improved support vector classification using PCA and ICA feature spacemodification[J]. Pattern Recognition.2004,37(6):1117-1129.
[48] Kim S. K., Park Y. J., Toh K. A., Lee S. SVM-based feature extraction for face recognition[J].Pattern Recognition.2010,43:2871-2881.
[49] Lades M., Vorbruggen J. C., Buhmann J., et al. Distortion invariant object recognition in thedynamic link architecture[J]. IEEE Transactions on computers,1993,42(3):300-311.
[50] Wiskott L., Fwllous J. M., Kruger N., et al. Face recognition by elstic bunch graph matching[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.1997,19(7):775-779.
[51]李云峰.基于Gabor小波变换的人脸识别[博士论文].大连:大连理工大学.2006.
[52] Liu C. J.,Wechsler H. Gabor feature based classification using the enhanced Fisher lineardiscriminant model for face recognition[J]. IEEE Transactions on Image Processing.2002,11(4):467-476.
[53]刘青山,卢汉清,马颂德.综述人脸识别中的子空间方法.自动化学报.2003,29(6):900-911.
[54] Turk M. A., Pentland A. P. Face recognition using Eigenfaces[C]. Proceedings of IEEEConference on Computer Vision and Pattern Recognition, Maui, Hawaii, USA.1991:586-591.
[55] Belhumeur P. N., Hespanha J. P., Kriegman D. J. Eigenfaces vs. Fisherfaces: recognition usingclass-specific linear projection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.1997,19(7):711-720.
[56] He X. F., Yan S. C., Hu Y. X., et al. Face recognition using Laplacianfaces[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence.2005,27(3):328-340.
[57] Chen S., Zhu Y. Subpattern-based principle component analysis[J]. Pattern Recognition.2004,37:1081-1083.
[58] Tan K., Chen S. Adaptively weighted sub-pattern PCA for face recognition[J]. Neurocomputing.2005,64:505-511.
[59] Kumar K. V., Negi A. SubXPCA and a generalized feature partitioning approach to principalcomponent analysis[J]. Pattern Recognition.2008,41:1398-1409.
[60] Tan X., Liu J., Chen S. Sub-intrapersonal space analysis for face recognition[J]. Neurocomputing.2006,69:1796-1801.
[61] Zhu Y., Liu J., Chen S. Semi-random subspace method for face recognition[J]. Image and VisionComputing.2009,27:1358-1370.
[62] Wang S., Zhou C., Chen Y., et al. A novel face recognition method based on sub-pattern andtensor[J]. Neurocomputing.2011,74:3553-3564.
[63] Yang J., Zhang D., Frangi A. F., Yang J. Two-dimensional PCA: a new approach toappearance-based face representation and recognition[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence.2004,26(1):131-137.
[64] Zuo W., Zhang D., Wang K. An assembled matrix distance metric for2DPCA-based imagerecognition[J]. Pattern Recognition Letters.2006,27:210-216.
[65] Meng J., Zhang W. Volume measure in2DPCA-based face recognition[J]. Pattern RecognitionLetters.2007,28:1203-1208.
[66] Xu Y., Zhang D., Yang J., Yang J. Y. An approach for directly extracting features from matrixdata and its application in face recognition[J]. Neurocomputing.2008,71:1857-1865.
[67] Zhang D., Zhou Z.(2D)2PCA: two-directional two-dimensional PCA for efficient facerepresentation and recognition[J]. Neurocomputing.2005,69:224-231.
[68] Ren C. X., Dai D. Q. Incremental learning of bidirectional principal components for facerecognition[J]. Pattern Recognition.2010,43:318-330.
[69] Sun Y., Chen S., Yin B. Color face recognition based on quaternion matrix representation[J].Pattern Recognition Letters.2011,32:597-605.
[70] Ye J. Generalized low rank approximation of matrices[C]. The Proceedings of Twenty-firstInternational Conference on Machine Learning.2004:887-894.
[71] Liang Z., Zhang D., Shi P. The theoretical analysis of GLRAM and its applications[J]. PatternRecognition.2007,40:1032-1041.
[72] Lu C., Liu W., An S. A simplified GLRAM algorithm for face recognition[J]. Neurocomputing.2008,72:212-217.
[73] Scholkopf B.,Smola A.,Muller K.Nonlinear component analysis as a kernel eigenvalueproblem[J].Neural Computation.1998,10(5):1299-1319.
[74] Yang M., Ahuja N.,Kriegman D.Face recognition using kernel eigenfaces[C]. Proc. IEEE Int’lConf. Image Processing, Sept.2000:37-40.
[75] Kim K., Jung K.,Kim H. Face recognition using kernel principal component analysis[J]. IEEESignal Processing Letters,2002,9(2):40-42.
[76] Yang J., Yang J., Frangi A.Combined fisherfaces framework [J]. Image and Vision Computing,2003,21(12):1037-1044.
[77] Liu C. Gabor-based kernel PCA with fractional power polynomial models for face recognition[J].IEEE Trans. Pattern Analysis and Machine Intelligence,2004,26(5):572-581.
[78] Xie X., Lam K. Kernel PCA with doubly nonlinear mapping for face recognition[C]. Proceedingsof2005international symposium on intelligent signal processing and communication Systems,2005:73-76.
[79] Kong H.,Li X., et. al. Generalized2D principal component analysis[C].Proceedings ofInternational Joint Conference on Neural Networks, Montreal, Canada, July31-August4,2005:108-113.
[80] Wilks S. S. Mathematical Statistics. New York: Wiley.1962:577-578.
[81] Foley D. H., Sammon J. W. An optimal set of discriminant vectors[J]. IEEE Transactions onComputer.1975,24(3):281-289.
[82] Jin Z., Yang J. Y., Hu Z. S., et al. Face recognition based on uncorrelated discriminanttransformation[J]. Pattern Recognition.2001,34(7):1405-1416.
[83] Yu H., Yang J. A direct LDA algorithm for high-dimensional data with application to facerecognition[J]. Pattern Recognition.2001,34:2067-2070.
[84] Chen L. F., Liao H.Y. M., Ko M. T. A new LDA-based face recognition system which can solvethe small sample size problem[J]. Pattern Recognition.2000,33:1713-1726.
[85] Dai D. Q., Yuen P. C. Regularized discriminant analysis and its application to face recognition[J].Pattern Recognition.2003,36:845-847.
[86] Yang J., Frangi A. F., Yang J., Zhang D., Jin Z. KPCA plus LDA: a complete kernel fisherdiscriminant framework for feature extraction and recognition[J]. IEEE Tranactions on PatternAnalysis and Machine Intelligence.2005,27(2):230-244.
[87] Li H. F., Jiang T., Zhang K. Efficient and robust feature extraction by maximum margincriterion[J]. IEEE Transactions on Neural Networks.2006,17(1):157-165.
[88] Li M., Yuan B. Z.2D-LDA: a statistical linear discriminant analysis for image matrix[J]. PatternRecognition.2005,26:527-532.
[89] Ye J. P., Janardan R., Li Q. Two-dimensional linear discriminant analysis[C]. Proceedings of theConference on Adavances in Neural Information Processing Systems.2005:1569-1576.
[90] Yu W., Wang Z., Chen W. A new framework to combine vertical and horizontal information forface recognition[J]. Neurocomputing.2009,72:1084-1091.
[91] Yang W., Yan X., Zhang L., Sun C. Feature extraction based on fuzzy2DLDA[J].Neurocomputing.2010,73:1556-1561.
[92] Wang J.G., Sung E., Yau W. Y. Incremental two-dimensional linear discriminant analysis withapplications to face recognition[J]. Journal of Network and Computer Applications.2010,33:314-322.
[93] Luo D., Ding C., Huang H. Symmetric two dimensional linear discriminant analysis (2DLDA)[C].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.2009:2820-2827.
[94] Zheng W., Zou C., Zhao L. Weighted maximum margin discriminant analysis with kernels[J].Neurocomputing.2005,67:357-362.
[95] Zheng Y., Yang J., Yang J., Wu X. A reformative kernel Fisher discriminant algorithm and itsapplication to face recognition[J]. Neurocomputing.2006,69:1806-1810.
[96] Liang Z., Zhang D.,Shi P. Robust kernel discriminant analysis and its application to featureextraction and recognition[J]. Neurocomputing.2006,69:928-933.
[97] Chen W. S., Yuen P. C., Huang J., Dai D. Q. Kernel machine-based one-parameter regularizedFisher discriminant method for face recognition[J]. IEEE Transactions on Systems, Man, andCybernetics-Part B: Cybernetics.2005,35(4):659-669.
[98] Yu X., Wang X., Liu B. A direct kernel uncorrelated discriminant analysis algorithm[J]. IEEESignal Processing Letters.2007,14(10):742-745.
[99] Zhang L., Zhou W. D., Chang P. C. Generalized nonlinear discriminant analysis and its smallsample size problems[J]. Neurocomputing.2011,74:568-574.
[100] Zeng W. J., Li X. L., Zhang X. D., Cheng E. Kernel-based nonlinear discriminant analysis usingminimum squared errors criterion for multiclass and undersampled problems[J]. Signal Processing.2010,90:2333-2343.
[101] Chen W. S., Yuen P. C., Xie X. Kernel machine-based rank-lifting regularized discriminantanalysis method for face recognition[J]. Neurocomputing.2011,74:2953-2960.
[102] Tenenbaum J. B., Silva V., Langford J. C. A global geometric framework for nonlineardimensionality reduction[J]. Science.2000,290(5500):2319-2323.
[103] Roweis S. T., Saul L. K. Nonlinear dimensionality reduction by locally linear embedding[J].Science.2000,290(5500):2323-2326.
[104] Balasubramanian M., Schwartz E. L., Tenenbaum J. B. The Isomap algorithm and topologicalstability[J]. Science.2002,295(5552):7a.
[105] Belkin M., Niyogi P. Laplacian Eigenmap for dimensionality reduction and datarepresentation[J]. Neural Computation.2003,15(6):1373-1396.
[106] Donoho D. L., Grimes C. Hessian eigenmaps: new locally linear embedding techniques forhigh-dimensional data[C]. Proc. of the National Academy of Science.2003,100(10):5591-5596.
[107] Zhang Z., Zha H. Principal manifolds and nonlinear dimensionality reduction via tangent spacealignment[J]. SIAM J. ScI. Comput.,2004,26:313-338.
[108] He X. F., Niyogi P. Locality preserving projections[C]. Proc. of Advances in Neural InformationProcessing System16, Vancouver, Canada,2003.
[109] He X. F., Cai D., Yan S. C., et al. Neighborhood preserving embedding[C]. Proc. of10th IEEEInternational Conference on Computer Vision, Beijing, China.2005:1208-1213.
[110] Zhang T., Yang J., Zhao D., Ge X. Linear local tangent space alignment and applicaiton to facerecognition[J]. Neurocomputing.2007(70):1547-1553.
[111] Yan S., Xu D., Zhang B.,et al. Graph embedding and extensions: a general framework fordimensionality reduction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2007,29(1):40-51.
[112] Hastie T. Principal curves and surfaces. Laboratory for computational Statistics StanfordUniversity, Department of Statistics Tecnnical Report11,1984.
[113] Hastie T., Stuetzle W. Principal curves[J]. Journal of the American Statistical Association.1989,84(406):502-516.
[114] Weinberger K., Sha F., Lawrence K. Learning a kernel matrix for nonlinear dimensionalityreduction[C].Proceedings of the Twenty-first International Conference on Machine Learning.2004:839-846.
[115] Brun A.,Westin C. Herberthson M.,et al. Fast manifold learning based on Riemannian normalcoordinates[C].Proc. of the14th Scandinavian Conference on Image Analysis,2005:920-929.
[116] Hinton G., Roweis S. Stochastic Neighbor Embedding[C]. Advances in Neural InformationProcessing Systems.2002.
[117] Brand M. Charting a manifold[C]. Advances in Neural Information Processing Systems.2003:961-968.
[118] Coifman R. et al. Geometric diffusions as a tool for harmonic analysis and structure definitionof data: diffusion maps[C]. Proc. of the National Academy of Sciences.2005,102:7426-7431.
[119] Li Y. Alignment of overlapping locally scaled patches for multidimensional scaling anddimensionality reduction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2008,30(3):438-450.
[120] Lin T., Zha H. Riemannian Manifold Learning[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence.2008,30(5):796-809.
[121] Bengio Y., Paiement J.,Vincent P. Out-of-sample extensions for LLE,Isomap, MDS, Eigenmaps,and Spectral clustering[C]. Advances in Neural Information Processing Systems,2003:177-184.
[122] Brand M., Kun H. A unifying theorem for spectral embedding and clustering[C]. Proc. of the9th International Conference on Artificial Intelligence and Statistics,2003.
[123] Ridder D., Kouropteva O., Okun O., et al. Supervised locally linear embedding[C]. Proc. of theICANN2003of Lecture Notes in Computer Science.2003,2714:333-341.
[124] Zeng X., Luo S. A supervised subspace learning algorithm: supervised neighborhood preservingembedding[J]. Lecture Notes in Computer Science.2007,4632:81-88.
[125] Pang Y. H., Andrew B. Maximum neighborhood margin criterion in face recognition[J]. OpticalEngineering.2009,48(4):047205.
[126] Pang Y., Liu Z., Yu N. A new nonlinear feature extraction method for face recognition[J].Neurocomputing.2006,69:949-953.
[127] Li B., Zheng C., Huang D. Locally linear embedding: an efficient method for facerecognition[J]. Pattern Recognition.2008,41(12):3813-3821.
[128] Hu H. Orthogonal neighborhood preserving discriminant analysis for face recognition[J].Pattern Recognition.2008,41(6):2045-2054.
[129] Sugiyama M. Dimensionality reduction of multimodal labeled data by local Fisher discriminantanalysis[J]. Journal of Machine Learning Research.2007,8:1027-1061.
[130] Cai D., He X.,Zhou K.,Han J., Bao H. Locality sensitive discriminant analysis[C]. Proceedingsof2007International Joint Conference on Artificial Intelligence, Hyderabad,India,2007:708-713.
[131] Yan Y., Zhang Y. J. Discriminant projection embedding for face and palmprint recognition[J].Neurocomputing.2008,71(16-18):3534-3543.
[132] Cai D., He X. F., Han J. W. Spectral regression for efficient regularized subspace learning[C].Proc. of the11th IEEE International Conference on Computer Vision, Brazil,2007:1-8.
[133] Shen L., Li B. A review on Gabor wavelets for face recognition[J]. Pattern Anal. Applic.,2006,9:273-292.
[134] Angel S., Isaac M. Cristina C., et al. Recent advances in face biometrics with Gabor wavelets: areview[J]. Pattern Recognition Letters.2010,31:372-381.
[135] Kingsbury N. G. The dual-tree complex wavelet transform: a new efficient tool for imagerestoration and enhancement[C]. Proc. European Signal Processing Conf.,1998:319-322.
[136] http://cvc.yale.edu/projects/yalefaces/yalefaces.html
[137] Martinez A., Benavente R. The AR database. CVC Technical Report No.24,1998.
[138] Sim T., Baker S., Bsat M. The CMU pose, illumination, and expression database[J]. IEEETransactions on Pattern Analysis and Machine Intelligence.2003,25(12):1615-1618.
[139] Bressan M., Vitria J. Nonparametric discriminant analysis and nearest neighbor classification[J].Pattern Recognition Letters.2003,24(15):2743-2749.
[140] Zhang T., Tao D., Li X., Yang J. Patch Alignment for Dimensionality Reduction[J].IEEETransactions on Knowledge and Data Engineering.2009,21(9):1299-1313.
[141] Zhang P., Qiao H., Zhang B. An improved local tangent space alignment method for manifoldlearning[J]. Pattern Recognition Letters.2011,32:181-189.
[142] Chaudhury K. N., Unser M. Construction of Hilbert transform pairs of wavelet bases andGabor-like transforms [J]. IEEE Transactions on Signal Processing.2009,57:3411-3425.
[143] http://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html
[144] http://www.sheffield.ac.uk/eee/research/iel/research/face
[145] 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.
[146] Cai D., He X., Han J. Spectral regression: a unified approach for sparse subspace learning[C].Proceedings of the7th IEEE International Conference on Data Mining, ICDM2007,2007:73-82.
[147] He R., Hu B., Zhen W., Kong X. Robust principal component analysis based on maximumcorrentropy criterion [J]. IEEE Transactions on Image Processing.2011,20(6):1485-1494.
[148] He R., Zhen W., Hu B. Maximum correntropy criterion for robust face recognition[J]. IEEETransactions on Pattern Analysis and Machine Intelligence.2011,33(8):1561-1576.
[149] Cheng B., Yang J., Yan S., et al. Learning with L(1)-graph for image analysis[J]. IEEETransactions on Image Processing.2010,19(4):858-866.
[150] Qiao L., Chen S., Tan X. Sparsity preserving projections with applications to face recognition[J].Pattern Recognition.2010,43(1):331-341.
[151] Lee D. D., Seung H. S. Learning the parts of objects by non-negative matrix factorization[J].Nature.1999,401:788-791.
[152] Li S. Z., Hou X. W., Zhang H. J., et al. Learning spatially localized parts-basedrepresentation[C]. Proc. of the IEEE International Conference on Computer Vision and PatternRecognition(CVPR), Kaual, Hawaii,2001:207-212.
[153] Hoyer P. O. Non-negative matrix factorization with sparseness constraints[J]. Journal ofMachine Learning Research.2004,5:1457-1469.
[154] Wang Y., Jia Y., Hu C., et al. Fisher non-negative matrix factorization for learning localfeatures[C]. Proc. of the Asian Conference on Computer Vision (ACCV), Jeju Island, Korea,2004.
[155] Zafeiriou S., Tefas A., Buciu I., et al. Exploiting discriminant information in nonnegative matrixfactorization with application to frontal face verification[J]. IEEE Transactions on Neural Networks.2006,17(3):683-695.
[156] Cai D., He X., Wu X., et al. Non-negative matrix factorization on manifold[C]. Proc. of theIEEE International Conference on Data Mining (ICDM), Pisa, Italy.2008:63-72.
[157] Gu Q., Zhou J. Local learning regularized nonnegative matrix factorization[C]. Proc. of the Int.Joint Conf. Artif. Intell.,2009:1046-1051.
[158] Lee D. D., Seung H. S. Algorithms for non-negative matrix factorization[C]. In NIPS,2000:556-562.
[159] Ding C., Li T., Jordan M. I. Convex and semi-nonnegative matrix factorizations[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2010,32(1):45-55.
[160] Xiang S. M., Nie F. P., Zhang C. S. Nonlinear dimensionality reduction with local splineembedding[J]. IEEE Trans. Knowledge Data Eng.2009,21(9):1285-1298.
[161] Lei Y. K., Ding Z. G., Hu R. X., et al. Orthogonal local spline discriminant projection withapplication to face recognition[J]. Pattern Recognition Letters.2011,32:615-625.
[162] Lu J., Plataniotis K. N., Venetsanopoulos A. N. Regularization studies of linear discriminantanalysis in small sample size scenarios with application to face recognition[J]. Pattern RecognitionLetters.2005,26:181-191.
[163] Jiang X., Mandal B., Kot A. Eigenfeature regularization and extraction in face recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.2008,30(3):383-394.
[164] Fan W., Yeung D. Face recognition with image sets using hierarchically extracted exemplarsfrom appearance manifolds[C]. Proc. of the7th International Conference on Automatic Face andGesture Recognitio.2006:177-182.
[165] Kim T. K.,Kitter J.,Cipolla R. Discriminative learning and recognition of image set classesusing canonical correlations[J]. IEEE Trans. Pattern Anal. Mach. Intell.2007,29(6):1-14.
[166] Hamm J., Lee D. D. Grassmann discriminant analysis: a unifying view on subspace-basedlearning[C]. Proc. of the25th International Conference on Machine Learning.2008:376-383.
[167] Wang R.,Shan S., Chen X., Gao W. Manifold-manifold distance with application to facerecognition based on image set[C]. Proc. of IEEE Computer Society Conference on Computer Visionand Pattern Recogntion.2008:1-8.
[168] Wang T., Shi P. Kernel Grassmannian distances and discriminant analysis for face recognitionfrom image sets[J]. Pattern Recognition Letters.2009,30(13):1161-1165.
[169] Mehrtash T. H., Conrad S., Sareh S., et al. Graph embedding discriminant analysis onGrassmannian manifolds for improved image set matching[C]. Proc. of Computer Vision and PatternRecognition (CVPR).2011:2705-2712.
[170] Yamaguchi O., Fukui K., Maeda K. Face recognition using temporal image sequence[C]. Proc.Of IEEE International Conference on Automatic Face and Gesture Recognition.1998:318-323.
[171] Kim T. K.,Arandjelovic O., Cipolla R. Boosted manifold principal angels for image set basedrecognition[J]. Pattern Recognition,2007,40(9):2475-2484.
[172] Chang J. M., Kirby M., Peterson C. Set-to-set face recognition under variations of pose andillumination[C]. Proc. of Biometric Symposium,2007:1-6.
[173] Hadid A.,Pietikainen M. From still image to video-based face recognition: an experimentalanalysis[C]. Proc. of the Sixth IEEE International Conference on Automatic Face and GestureRecognition,2004:17-19.
[174] http://www.itl.nist.gov/iad/humanid/feret/feret_master.html
[175] Lee K. C., Ho J., Yang M. H., et al. Visual tracking and recognition using probabilisticappearance manifolds[J]. Computer Vision and Image Understanding,2005,99(3):303-331.
[2]边肇祺,张学工.模式识别(第二版).北京:清华大学出版社.2000:32-85.
[3] Seung H., Lee D.The manifold ways of perception[J].Science.2000,290(5500):2268-2269.
[4] Nayar S., Nene S., Murase H. Subspace method for robot vision[J]. IEEE Transactions onRobotics and Automation.1996,12(5):750-758.
[5]杨飞，苏剑波.人脸显性特征的融合构造方法及识别.电子学报.2012,40(3):466-471.
[6]侯书东，孙权森.稀疏保持典型相关分析及在特征融合中的应用.自动化学报.2012,38(4):659-665.
[7]严严，章毓晋.基于视频的人脸识别研究进展.计算机学报.2009,32(5):878-886.
[8]李勇智.图像特征提取方法及其在人脸识别中的应用[博士论文].南京：南京理工大学.2009.
[9] Ding X. Q., Fang C. H. Discussion on some problems in face recognition[J]. Sinobiometrics.2004,3338:47-56.
[10] Kanade T. Computer recognition of human faces. Basel&Stuttgart: Birkhauser Verlag.1977:20-56.
[11] Buhr R. Analyze und klassifikation von gesichtsbildern[J]. ntzArchiv.1986,8(10):245-256.
[12] Yulle A. L., Hallinan P. W., Cohen D. S. Feature extraction from face using deformable templates[J]. International Journal of Computer Vision.1992,8(2):99-111.
[13] Craw I. Recognizing face features and faces[C]. Proc. of IEE Colloquium on Machine Storge andRecognition of Face. London, UK,1992:711-714.
[14] Lee S. Y., Ham Y. K., Park R. H. Recognition of human front faces using knowledge-basedfeature extraction and neuro-fuzzy algorithm[J]. Pattern Recognition.1996,29(11):1863-1876.
[15] Samaria F. Face recognition using hidden markov models:[PhD thesis].Cambridge: University ofCambridge,1994.
[16] Nefian A. Hidden Markov Model-based approach for face detection and recognition:[PhD thesis].Atlanta: Georgia Institute of Technology,1999.
[17]刘小军,王东峰,张丽飞.一种基于奇异值分解和隐马尔可夫模型的人脸识别方法.计算机学报,2003,26(3):340-344.
[18] Hallinan P. A deformable model for the recognition of human faces under arbitrary illumination:
[PhD thesis]. Cambridge: Harvard University.1995.
[19] Lee M. W., Ranganath S.3D deformable face model for pose determination and facesynthesis[C]. Proc. of the10th International Conference on Image Analysis and Processing, Venice,Italy.1999:532-537.
[20] Xiao J., Kanade T. Non-rigid shape and motion recovery: degrenerate deformations[C]. Proc. ofthe2004IEEE Conference on Computer Vision and Pattern Recognition, Washington D C, USA.2004:668-675.
[21] Vetter T., Blanz V. A morphable model for the synthesis of3D faces[C]. Proc. of ACMSIGGRAPH99Conference, Los Angles, USA.1999:187-194.
[22] Kohonen. Self-organization and associative memory. Berlin: Spring-Verlag.1988:37-92.
[23] Cottrell G. W., Fleming M. K. Face recognition using unsupervised feature extraction[C]. Proc.of International Neural Network Conference.1990:322-325.
[24] Lawrence S., Giles C. L., et al. Face recognition: a convolutional neural-network approach[J].IEEE Transactions on Neural Networks.1997,8(1):98-113.
[25] Demers D., Cottrell G. W. Nonlinear dimensionality reduction[C]. Proc. of Advances in NeuralInformation Processing System.1993:580-587.
[26] Lin S. H., Kung S. Y., Lin L. J., et al. Face recognition/detection by probalilistic decision-basedneural network[J]. IEEE Transactions on Neural Network.1997,8(1):114-132.
[27] Gutta S., Wechsler H. Face recognition using hybrid classifiers[J]. Pattern Recogntion.1997,30(4):539-553.
[28] Luh G, C., Lin C. Y. PCA based immune networks for human face recognition[J]. Applied SoftComputing.2011,11:1743-1752.
[29] Wong Y. W., Seng K. P., Ang L. M. Radial basis function neural network with incrementallearning for face recognition[J]. IEEE Transactions on Systems, Man, and Cybernetics-Part B:Cybernetics.2011,41(4):940-949.
[30] Sudha N., Mohan A. R., Meher P. K. A self-configurable systolic architecture for face recognitionsystem based on principal componet neural network[J]. IEEE Transactions on Circuits and Systemsfor Video Technology.2011,21(8):1071-1084.
[31] Vapnik V. The nature of statistical learning theory. New York: Springer.1995.
[32] Philips P. J. Support vector machine applied to face recognition[C]. In: Advanced in NeuralInformation Processing System II. MIT Press.1998:803-809.
[33] Guo G., Li S. Z., Chan K. L. Support vector machine for face recognition[J]. Image and VisionComputing.2001,19(9-10):631-638.
[34] Ko J., Byun H. N-division output coding method applied to face recognition[J]. PatternRecognition Letters.2003,24(16):3115-3123.
[35] Heisele B., Ho P., Wu J., Poggio T. Face recognition: component-based versus globalapproaches[J]. Computer Vision and Image Understanding.2003,91(1-2):6-21.
[36] Ivanov Y., Heisele B., Serre T. Using component features for face recognition[C]. Proceedings ofthe Sixth IEEE International Conference on Automatic Face and Gesture Recognition (FGR’04).2004:421-426.
[37] Casasent D., Yuan C. Face recognition with pose and illumination variations using new SVRDMsupport-vector machine [J]. Optical Engineering.2004,43(8):1804-1813.
[38] Ko J., Byun H. Combining SVM classifiers for multiclass problem: its application to facerecognition[C]. AVBPA2003, LNCS2688.2003:531-539.
[39] Li Z., Tang X. Bayesian face recognition using support vector machine and face clustering[C].Proceedings of the2004IEEE Computer Society Conference on Computer Vision and PatternRecognition (CVPR’04).2004:374-380.
[40] Lu J., Plataniotis K. N., Venetsanopoulos A.N. Face recognition using feature optimization andv-support vector learning[C]. Proceedings of the IEEE2001International Workshop on NeuralNetworks for Signal Processing, Falmouth, MA, USA. September2001:373-382.
[41] Deniz O., Castrillon M., Hernandez M. Face recognition using independent component analysisand support vector machines[J]. Pattern Recognition Letters.2003,24(13):2153-2157.
[42] Gumus E., Killic N., Sertbas A., Ucan O. N. Evaluation of face recognition techniques usingPCA, wavelet and SVM[J]. Expert Systems with Applications.2010,37(9):6404-6408.
[43] Kanan H. R., Faez K. GA-based optimal selection of PZMI features for face recognition[J].Applied Mathematics and Computation.2008,205(2):706-715.
[44] Smach F., Cedric L., Gauthier J. P., et al. Generalized Fourier descriptors with applications toobjects recognition in SVM context[J]. J. Math. Imaging Vis.2008,30:43-71.
[45] Zhang B.,Chen X.,Shan S.,Gao W. Nonlinear face recognition based on maximum average margincriterion[C]. Proc. of the2005IEEE Computer Society Conference on Computer Vision and PatternRecognition(CVPR’05).2005:554-559.
[46] Song F., Yang J., Liu S. Large margin linear projection and face recongnition[J]. PatternRecognition.2004,37(9):1953-1955.
[47] Fortuna J., David C. Improved support vector classification using PCA and ICA feature spacemodification[J]. Pattern Recognition.2004,37(6):1117-1129.
[48] Kim S. K., Park Y. J., Toh K. A., Lee S. SVM-based feature extraction for face recognition[J].Pattern Recognition.2010,43:2871-2881.
[49] Lades M., Vorbruggen J. C., Buhmann J., et al. Distortion invariant object recognition in thedynamic link architecture[J]. IEEE Transactions on computers,1993,42(3):300-311.
[50] Wiskott L., Fwllous J. M., Kruger N., et al. Face recognition by elstic bunch graph matching[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.1997,19(7):775-779.
[51]李云峰.基于Gabor小波变换的人脸识别[博士论文].大连:大连理工大学.2006.
[52] Liu C. J.,Wechsler H. Gabor feature based classification using the enhanced Fisher lineardiscriminant model for face recognition[J]. IEEE Transactions on Image Processing.2002,11(4):467-476.
[53]刘青山,卢汉清,马颂德.综述人脸识别中的子空间方法.自动化学报.2003,29(6):900-911.
[54] Turk M. A., Pentland A. P. Face recognition using Eigenfaces[C]. Proceedings of IEEEConference on Computer Vision and Pattern Recognition, Maui, Hawaii, USA.1991:586-591.
[55] Belhumeur P. N., Hespanha J. P., Kriegman D. J. Eigenfaces vs. Fisherfaces: recognition usingclass-specific linear projection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.1997,19(7):711-720.
[56] He X. F., Yan S. C., Hu Y. X., et al. Face recognition using Laplacianfaces[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence.2005,27(3):328-340.
[57] Chen S., Zhu Y. Subpattern-based principle component analysis[J]. Pattern Recognition.2004,37:1081-1083.
[58] Tan K., Chen S. Adaptively weighted sub-pattern PCA for face recognition[J]. Neurocomputing.2005,64:505-511.
[59] Kumar K. V., Negi A. SubXPCA and a generalized feature partitioning approach to principalcomponent analysis[J]. Pattern Recognition.2008,41:1398-1409.
[60] Tan X., Liu J., Chen S. Sub-intrapersonal space analysis for face recognition[J]. Neurocomputing.2006,69:1796-1801.
[61] Zhu Y., Liu J., Chen S. Semi-random subspace method for face recognition[J]. Image and VisionComputing.2009,27:1358-1370.
[62] Wang S., Zhou C., Chen Y., et al. A novel face recognition method based on sub-pattern andtensor[J]. Neurocomputing.2011,74:3553-3564.
[63] Yang J., Zhang D., Frangi A. F., Yang J. Two-dimensional PCA: a new approach toappearance-based face representation and recognition[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence.2004,26(1):131-137.
[64] Zuo W., Zhang D., Wang K. An assembled matrix distance metric for2DPCA-based imagerecognition[J]. Pattern Recognition Letters.2006,27:210-216.
[65] Meng J., Zhang W. Volume measure in2DPCA-based face recognition[J]. Pattern RecognitionLetters.2007,28:1203-1208.
[66] Xu Y., Zhang D., Yang J., Yang J. Y. An approach for directly extracting features from matrixdata and its application in face recognition[J]. Neurocomputing.2008,71:1857-1865.
[67] Zhang D., Zhou Z.(2D)2PCA: two-directional two-dimensional PCA for efficient facerepresentation and recognition[J]. Neurocomputing.2005,69:224-231.
[68] Ren C. X., Dai D. Q. Incremental learning of bidirectional principal components for facerecognition[J]. Pattern Recognition.2010,43:318-330.
[69] Sun Y., Chen S., Yin B. Color face recognition based on quaternion matrix representation[J].Pattern Recognition Letters.2011,32:597-605.
[70] Ye J. Generalized low rank approximation of matrices[C]. The Proceedings of Twenty-firstInternational Conference on Machine Learning.2004:887-894.
[71] Liang Z., Zhang D., Shi P. The theoretical analysis of GLRAM and its applications[J]. PatternRecognition.2007,40:1032-1041.
[72] Lu C., Liu W., An S. A simplified GLRAM algorithm for face recognition[J]. Neurocomputing.2008,72:212-217.
[73] Scholkopf B.,Smola A.,Muller K.Nonlinear component analysis as a kernel eigenvalueproblem[J].Neural Computation.1998,10(5):1299-1319.
[74] Yang M., Ahuja N.,Kriegman D.Face recognition using kernel eigenfaces[C]. Proc. IEEE Int’lConf. Image Processing, Sept.2000:37-40.
[75] Kim K., Jung K.,Kim H. Face recognition using kernel principal component analysis[J]. IEEESignal Processing Letters,2002,9(2):40-42.
[76] Yang J., Yang J., Frangi A.Combined fisherfaces framework [J]. Image and Vision Computing,2003,21(12):1037-1044.
[77] Liu C. Gabor-based kernel PCA with fractional power polynomial models for face recognition[J].IEEE Trans. Pattern Analysis and Machine Intelligence,2004,26(5):572-581.
[78] Xie X., Lam K. Kernel PCA with doubly nonlinear mapping for face recognition[C]. Proceedingsof2005international symposium on intelligent signal processing and communication Systems,2005:73-76.
[79] Kong H.,Li X., et. al. Generalized2D principal component analysis[C].Proceedings ofInternational Joint Conference on Neural Networks, Montreal, Canada, July31-August4,2005:108-113.
[80] Wilks S. S. Mathematical Statistics. New York: Wiley.1962:577-578.
[81] Foley D. H., Sammon J. W. An optimal set of discriminant vectors[J]. IEEE Transactions onComputer.1975,24(3):281-289.
[82] Jin Z., Yang J. Y., Hu Z. S., et al. Face recognition based on uncorrelated discriminanttransformation[J]. Pattern Recognition.2001,34(7):1405-1416.
[83] Yu H., Yang J. A direct LDA algorithm for high-dimensional data with application to facerecognition[J]. Pattern Recognition.2001,34:2067-2070.
[84] Chen L. F., Liao H.Y. M., Ko M. T. A new LDA-based face recognition system which can solvethe small sample size problem[J]. Pattern Recognition.2000,33:1713-1726.
[85] Dai D. Q., Yuen P. C. Regularized discriminant analysis and its application to face recognition[J].Pattern Recognition.2003,36:845-847.
[86] Yang J., Frangi A. F., Yang J., Zhang D., Jin Z. KPCA plus LDA: a complete kernel fisherdiscriminant framework for feature extraction and recognition[J]. IEEE Tranactions on PatternAnalysis and Machine Intelligence.2005,27(2):230-244.
[87] Li H. F., Jiang T., Zhang K. Efficient and robust feature extraction by maximum margincriterion[J]. IEEE Transactions on Neural Networks.2006,17(1):157-165.
[88] Li M., Yuan B. Z.2D-LDA: a statistical linear discriminant analysis for image matrix[J]. PatternRecognition.2005,26:527-532.
[89] Ye J. P., Janardan R., Li Q. Two-dimensional linear discriminant analysis[C]. Proceedings of theConference on Adavances in Neural Information Processing Systems.2005:1569-1576.
[90] Yu W., Wang Z., Chen W. A new framework to combine vertical and horizontal information forface recognition[J]. Neurocomputing.2009,72:1084-1091.
[91] Yang W., Yan X., Zhang L., Sun C. Feature extraction based on fuzzy2DLDA[J].Neurocomputing.2010,73:1556-1561.
[92] Wang J.G., Sung E., Yau W. Y. Incremental two-dimensional linear discriminant analysis withapplications to face recognition[J]. Journal of Network and Computer Applications.2010,33:314-322.
[93] Luo D., Ding C., Huang H. Symmetric two dimensional linear discriminant analysis (2DLDA)[C].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.2009:2820-2827.
[94] Zheng W., Zou C., Zhao L. Weighted maximum margin discriminant analysis with kernels[J].Neurocomputing.2005,67:357-362.
[95] Zheng Y., Yang J., Yang J., Wu X. A reformative kernel Fisher discriminant algorithm and itsapplication to face recognition[J]. Neurocomputing.2006,69:1806-1810.
[96] Liang Z., Zhang D.,Shi P. Robust kernel discriminant analysis and its application to featureextraction and recognition[J]. Neurocomputing.2006,69:928-933.
[97] Chen W. S., Yuen P. C., Huang J., Dai D. Q. Kernel machine-based one-parameter regularizedFisher discriminant method for face recognition[J]. IEEE Transactions on Systems, Man, andCybernetics-Part B: Cybernetics.2005,35(4):659-669.
[98] Yu X., Wang X., Liu B. A direct kernel uncorrelated discriminant analysis algorithm[J]. IEEESignal Processing Letters.2007,14(10):742-745.
[99] Zhang L., Zhou W. D., Chang P. C. Generalized nonlinear discriminant analysis and its smallsample size problems[J]. Neurocomputing.2011,74:568-574.
[100] Zeng W. J., Li X. L., Zhang X. D., Cheng E. Kernel-based nonlinear discriminant analysis usingminimum squared errors criterion for multiclass and undersampled problems[J]. Signal Processing.2010,90:2333-2343.
[101] Chen W. S., Yuen P. C., Xie X. Kernel machine-based rank-lifting regularized discriminantanalysis method for face recognition[J]. Neurocomputing.2011,74:2953-2960.
[102] Tenenbaum J. B., Silva V., Langford J. C. A global geometric framework for nonlineardimensionality reduction[J]. Science.2000,290(5500):2319-2323.
[103] Roweis S. T., Saul L. K. Nonlinear dimensionality reduction by locally linear embedding[J].Science.2000,290(5500):2323-2326.
[104] Balasubramanian M., Schwartz E. L., Tenenbaum J. B. The Isomap algorithm and topologicalstability[J]. Science.2002,295(5552):7a.
[105] Belkin M., Niyogi P. Laplacian Eigenmap for dimensionality reduction and datarepresentation[J]. Neural Computation.2003,15(6):1373-1396.
[106] Donoho D. L., Grimes C. Hessian eigenmaps: new locally linear embedding techniques forhigh-dimensional data[C]. Proc. of the National Academy of Science.2003,100(10):5591-5596.
[107] Zhang Z., Zha H. Principal manifolds and nonlinear dimensionality reduction via tangent spacealignment[J]. SIAM J. ScI. Comput.,2004,26:313-338.
[108] He X. F., Niyogi P. Locality preserving projections[C]. Proc. of Advances in Neural InformationProcessing System16, Vancouver, Canada,2003.
[109] He X. F., Cai D., Yan S. C., et al. Neighborhood preserving embedding[C]. Proc. of10th IEEEInternational Conference on Computer Vision, Beijing, China.2005:1208-1213.
[110] Zhang T., Yang J., Zhao D., Ge X. Linear local tangent space alignment and applicaiton to facerecognition[J]. Neurocomputing.2007(70):1547-1553.
[111] Yan S., Xu D., Zhang B.,et al. Graph embedding and extensions: a general framework fordimensionality reduction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2007,29(1):40-51.
[112] Hastie T. Principal curves and surfaces. Laboratory for computational Statistics StanfordUniversity, Department of Statistics Tecnnical Report11,1984.
[113] Hastie T., Stuetzle W. Principal curves[J]. Journal of the American Statistical Association.1989,84(406):502-516.
[114] Weinberger K., Sha F., Lawrence K. Learning a kernel matrix for nonlinear dimensionalityreduction[C].Proceedings of the Twenty-first International Conference on Machine Learning.2004:839-846.
[115] Brun A.,Westin C. Herberthson M.,et al. Fast manifold learning based on Riemannian normalcoordinates[C].Proc. of the14th Scandinavian Conference on Image Analysis,2005:920-929.
[116] Hinton G., Roweis S. Stochastic Neighbor Embedding[C]. Advances in Neural InformationProcessing Systems.2002.
[117] Brand M. Charting a manifold[C]. Advances in Neural Information Processing Systems.2003:961-968.
[118] Coifman R. et al. Geometric diffusions as a tool for harmonic analysis and structure definitionof data: diffusion maps[C]. Proc. of the National Academy of Sciences.2005,102:7426-7431.
[119] Li Y. Alignment of overlapping locally scaled patches for multidimensional scaling anddimensionality reduction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2008,30(3):438-450.
[120] Lin T., Zha H. Riemannian Manifold Learning[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence.2008,30(5):796-809.
[121] Bengio Y., Paiement J.,Vincent P. Out-of-sample extensions for LLE,Isomap, MDS, Eigenmaps,and Spectral clustering[C]. Advances in Neural Information Processing Systems,2003:177-184.
[122] Brand M., Kun H. A unifying theorem for spectral embedding and clustering[C]. Proc. of the9th International Conference on Artificial Intelligence and Statistics,2003.
[123] Ridder D., Kouropteva O., Okun O., et al. Supervised locally linear embedding[C]. Proc. of theICANN2003of Lecture Notes in Computer Science.2003,2714:333-341.
[124] Zeng X., Luo S. A supervised subspace learning algorithm: supervised neighborhood preservingembedding[J]. Lecture Notes in Computer Science.2007,4632:81-88.
[125] Pang Y. H., Andrew B. Maximum neighborhood margin criterion in face recognition[J]. OpticalEngineering.2009,48(4):047205.
[126] Pang Y., Liu Z., Yu N. A new nonlinear feature extraction method for face recognition[J].Neurocomputing.2006,69:949-953.
[127] Li B., Zheng C., Huang D. Locally linear embedding: an efficient method for facerecognition[J]. Pattern Recognition.2008,41(12):3813-3821.
[128] Hu H. Orthogonal neighborhood preserving discriminant analysis for face recognition[J].Pattern Recognition.2008,41(6):2045-2054.
[129] Sugiyama M. Dimensionality reduction of multimodal labeled data by local Fisher discriminantanalysis[J]. Journal of Machine Learning Research.2007,8:1027-1061.
[130] Cai D., He X.,Zhou K.,Han J., Bao H. Locality sensitive discriminant analysis[C]. Proceedingsof2007International Joint Conference on Artificial Intelligence, Hyderabad,India,2007:708-713.
[131] Yan Y., Zhang Y. J. Discriminant projection embedding for face and palmprint recognition[J].Neurocomputing.2008,71(16-18):3534-3543.
[132] Cai D., He X. F., Han J. W. Spectral regression for efficient regularized subspace learning[C].Proc. of the11th IEEE International Conference on Computer Vision, Brazil,2007:1-8.
[133] Shen L., Li B. A review on Gabor wavelets for face recognition[J]. Pattern Anal. Applic.,2006,9:273-292.
[134] Angel S., Isaac M. Cristina C., et al. Recent advances in face biometrics with Gabor wavelets: areview[J]. Pattern Recognition Letters.2010,31:372-381.
[135] Kingsbury N. G. The dual-tree complex wavelet transform: a new efficient tool for imagerestoration and enhancement[C]. Proc. European Signal Processing Conf.,1998:319-322.
[136] http://cvc.yale.edu/projects/yalefaces/yalefaces.html
[137] Martinez A., Benavente R. The AR database. CVC Technical Report No.24,1998.
[138] Sim T., Baker S., Bsat M. The CMU pose, illumination, and expression database[J]. IEEETransactions on Pattern Analysis and Machine Intelligence.2003,25(12):1615-1618.
[139] Bressan M., Vitria J. Nonparametric discriminant analysis and nearest neighbor classification[J].Pattern Recognition Letters.2003,24(15):2743-2749.
[140] Zhang T., Tao D., Li X., Yang J. Patch Alignment for Dimensionality Reduction[J].IEEETransactions on Knowledge and Data Engineering.2009,21(9):1299-1313.
[141] Zhang P., Qiao H., Zhang B. An improved local tangent space alignment method for manifoldlearning[J]. Pattern Recognition Letters.2011,32:181-189.
[142] Chaudhury K. N., Unser M. Construction of Hilbert transform pairs of wavelet bases andGabor-like transforms [J]. IEEE Transactions on Signal Processing.2009,57:3411-3425.
[143] http://www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html
[144] http://www.sheffield.ac.uk/eee/research/iel/research/face
[145] 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.
[146] Cai D., He X., Han J. Spectral regression: a unified approach for sparse subspace learning[C].Proceedings of the7th IEEE International Conference on Data Mining, ICDM2007,2007:73-82.
[147] He R., Hu B., Zhen W., Kong X. Robust principal component analysis based on maximumcorrentropy criterion [J]. IEEE Transactions on Image Processing.2011,20(6):1485-1494.
[148] He R., Zhen W., Hu B. Maximum correntropy criterion for robust face recognition[J]. IEEETransactions on Pattern Analysis and Machine Intelligence.2011,33(8):1561-1576.
[149] Cheng B., Yang J., Yan S., et al. Learning with L(1)-graph for image analysis[J]. IEEETransactions on Image Processing.2010,19(4):858-866.
[150] Qiao L., Chen S., Tan X. Sparsity preserving projections with applications to face recognition[J].Pattern Recognition.2010,43(1):331-341.
[151] Lee D. D., Seung H. S. Learning the parts of objects by non-negative matrix factorization[J].Nature.1999,401:788-791.
[152] Li S. Z., Hou X. W., Zhang H. J., et al. Learning spatially localized parts-basedrepresentation[C]. Proc. of the IEEE International Conference on Computer Vision and PatternRecognition(CVPR), Kaual, Hawaii,2001:207-212.
[153] Hoyer P. O. Non-negative matrix factorization with sparseness constraints[J]. Journal ofMachine Learning Research.2004,5:1457-1469.
[154] Wang Y., Jia Y., Hu C., et al. Fisher non-negative matrix factorization for learning localfeatures[C]. Proc. of the Asian Conference on Computer Vision (ACCV), Jeju Island, Korea,2004.
[155] Zafeiriou S., Tefas A., Buciu I., et al. Exploiting discriminant information in nonnegative matrixfactorization with application to frontal face verification[J]. IEEE Transactions on Neural Networks.2006,17(3):683-695.
[156] Cai D., He X., Wu X., et al. Non-negative matrix factorization on manifold[C]. Proc. of theIEEE International Conference on Data Mining (ICDM), Pisa, Italy.2008:63-72.
[157] Gu Q., Zhou J. Local learning regularized nonnegative matrix factorization[C]. Proc. of the Int.Joint Conf. Artif. Intell.,2009:1046-1051.
[158] Lee D. D., Seung H. S. Algorithms for non-negative matrix factorization[C]. In NIPS,2000:556-562.
[159] Ding C., Li T., Jordan M. I. Convex and semi-nonnegative matrix factorizations[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2010,32(1):45-55.
[160] Xiang S. M., Nie F. P., Zhang C. S. Nonlinear dimensionality reduction with local splineembedding[J]. IEEE Trans. Knowledge Data Eng.2009,21(9):1285-1298.
[161] Lei Y. K., Ding Z. G., Hu R. X., et al. Orthogonal local spline discriminant projection withapplication to face recognition[J]. Pattern Recognition Letters.2011,32:615-625.
[162] Lu J., Plataniotis K. N., Venetsanopoulos A. N. Regularization studies of linear discriminantanalysis in small sample size scenarios with application to face recognition[J]. Pattern RecognitionLetters.2005,26:181-191.
[163] Jiang X., Mandal B., Kot A. Eigenfeature regularization and extraction in face recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.2008,30(3):383-394.
[164] Fan W., Yeung D. Face recognition with image sets using hierarchically extracted exemplarsfrom appearance manifolds[C]. Proc. of the7th International Conference on Automatic Face andGesture Recognitio.2006:177-182.
[165] Kim T. K.,Kitter J.,Cipolla R. Discriminative learning and recognition of image set classesusing canonical correlations[J]. IEEE Trans. Pattern Anal. Mach. Intell.2007,29(6):1-14.
[166] Hamm J., Lee D. D. Grassmann discriminant analysis: a unifying view on subspace-basedlearning[C]. Proc. of the25th International Conference on Machine Learning.2008:376-383.
[167] Wang R.,Shan S., Chen X., Gao W. Manifold-manifold distance with application to facerecognition based on image set[C]. Proc. of IEEE Computer Society Conference on Computer Visionand Pattern Recogntion.2008:1-8.
[168] Wang T., Shi P. Kernel Grassmannian distances and discriminant analysis for face recognitionfrom image sets[J]. Pattern Recognition Letters.2009,30(13):1161-1165.
[169] Mehrtash T. H., Conrad S., Sareh S., et al. Graph embedding discriminant analysis onGrassmannian manifolds for improved image set matching[C]. Proc. of Computer Vision and PatternRecognition (CVPR).2011:2705-2712.
[170] Yamaguchi O., Fukui K., Maeda K. Face recognition using temporal image sequence[C]. Proc.Of IEEE International Conference on Automatic Face and Gesture Recognition.1998:318-323.
[171] Kim T. K.,Arandjelovic O., Cipolla R. Boosted manifold principal angels for image set basedrecognition[J]. Pattern Recognition,2007,40(9):2475-2484.
[172] Chang J. M., Kirby M., Peterson C. Set-to-set face recognition under variations of pose andillumination[C]. Proc. of Biometric Symposium,2007:1-6.
[173] Hadid A.,Pietikainen M. From still image to video-based face recognition: an experimentalanalysis[C]. Proc. of the Sixth IEEE International Conference on Automatic Face and GestureRecognition,2004:17-19.
[174] http://www.itl.nist.gov/iad/humanid/feret/feret_master.html
[175] Lee K. C., Ho J., Yang M. H., et al. Visual tracking and recognition using probabilisticappearance manifolds[J]. Computer Vision and Image Understanding,2005,99(3):303-331.