雷达辐射源个体识别中的分类器设计与子空间学习
|
摘要
雷达辐射源个体识别在整个雷达对抗中占有重要地位。随着雷达技术的迅速发展及新体制雷达的应用,传统的识别方法逐渐失效。因此,从信号处理角度看,需要发展更为有效的个体特征提取方法,而从模式识别角度看,相应的分类器设计及特征的进一步精简与优化也变得至关重要。本文以两种有效的辐射源个体特征为依托,重点研究了雷达辐射源个体识别中的分类器设计及基于子空间学习的降维技术。
在个体特征提取中,分别介绍了基于循环谱零点切片和基于模糊函数切片串联的特征提取方法。在实测雷达辐射源数据上的实验结果充分验证了这两种特征提取方法的有效性,从而为辐射源的个体识别提供了稳定可靠的分类特征。
在分类器设计中,首先介绍了六种能够输出后验概率的分类器,继而通过一定的概率融合函数可靠地实现了多分类器的组合。进一步,考虑到实际系统必须能够拒判库外目标,先由后验概率估计出广义置信度,从而设定一定的门限实现了系统对库外目标的拒判功能,最后给出多种评价系统性能的指标。实验结果验证了分类器组合的优势,以及基于广义置信度的拒识算法的可行性。
在子空间特征降维中,重点研究了线性判别子空间学习和核判别子空间学习两大降维技术。其中,核子空间方法是通过特定的核技巧对线性子空间方法的扩展。我们将目前已有的各种经典算法归纳为单子空间学习和多子空间学习两大类,并研究了各方法在图像识别与雷达辐射源识别中的应用。实验结果表明,无论是线性方法还是核方法,由于考虑到子空间的互补性,多子空间学习具有更鲁棒的识别性能;而由于算法的数据依赖性,从线性方法到核方法的推广不一定意味着识别率的提升,需要根据实际数据的分布和工程需求选择合适的方法。
Radar emitter recognition plays an important role in the ECM. With the rapid development of the radar technology and the utility of new types of radar, exploring more effective feature extraction methods is urgent from the point view of signal processing. From the perspective of pattern recognition, it is also essential to concentrate on the classifier design and feature optimization. Based on the two effective features of radar signals, this thesis deals with classifier design, as well as subspace learning based dimensionality reduction technology, for radar emitter recognition.
For feature extraction, zero-slice of the cyclic spectrum and concatenation of the slices in ambiguity function based methods are studied respectively. The experimental results on the real radar data show the validity of the two methods, which provide reliable features for subsequent classification tasks.
For classifier design, six types of classifiers which can produce posterior probability are introduced. Then classifier combination can be conveniently realized via proper fusion strategy. To further reject the out-of-database targets, generalized confidence is calculated by the posterior probability, and then thresholds are empirically chosen to implement the rejection operation. Finally, several evaluation indices are given. The experimental results demonstrate the advantage of classifier combination, as well as the feasibility of generalized confidence based rejection methods.
For subspace based dimensionality reduction, both linear and kernel discriminant subspace learning are investigated in depth. Therein, kernel-based methods are the extensions of corresponding linear methods via some specific kernel trick. These classical algorithms are summarized into single-subspace learning and multi-subspace learning, which are applied to image recognition and radar emitter recognition. The experimental results indicate that multi-subspace learning algorithms take into account the complementarity of different subspaces and are more robust than single-subspace based methods, no matter linear or kernel cases are concerned. Due to data dependence, kernel extension not always guarantees higher accuracy, so appropriate method needs to be selected according to the actual data distribution and engineering requirements.
在个体特征提取中,分别介绍了基于循环谱零点切片和基于模糊函数切片串联的特征提取方法。在实测雷达辐射源数据上的实验结果充分验证了这两种特征提取方法的有效性,从而为辐射源的个体识别提供了稳定可靠的分类特征。
在分类器设计中,首先介绍了六种能够输出后验概率的分类器,继而通过一定的概率融合函数可靠地实现了多分类器的组合。进一步,考虑到实际系统必须能够拒判库外目标,先由后验概率估计出广义置信度,从而设定一定的门限实现了系统对库外目标的拒判功能,最后给出多种评价系统性能的指标。实验结果验证了分类器组合的优势,以及基于广义置信度的拒识算法的可行性。
在子空间特征降维中,重点研究了线性判别子空间学习和核判别子空间学习两大降维技术。其中,核子空间方法是通过特定的核技巧对线性子空间方法的扩展。我们将目前已有的各种经典算法归纳为单子空间学习和多子空间学习两大类,并研究了各方法在图像识别与雷达辐射源识别中的应用。实验结果表明,无论是线性方法还是核方法,由于考虑到子空间的互补性,多子空间学习具有更鲁棒的识别性能;而由于算法的数据依赖性,从线性方法到核方法的推广不一定意味着识别率的提升,需要根据实际数据的分布和工程需求选择合适的方法。
Radar emitter recognition plays an important role in the ECM. With the rapid development of the radar technology and the utility of new types of radar, exploring more effective feature extraction methods is urgent from the point view of signal processing. From the perspective of pattern recognition, it is also essential to concentrate on the classifier design and feature optimization. Based on the two effective features of radar signals, this thesis deals with classifier design, as well as subspace learning based dimensionality reduction technology, for radar emitter recognition.
For feature extraction, zero-slice of the cyclic spectrum and concatenation of the slices in ambiguity function based methods are studied respectively. The experimental results on the real radar data show the validity of the two methods, which provide reliable features for subsequent classification tasks.
For classifier design, six types of classifiers which can produce posterior probability are introduced. Then classifier combination can be conveniently realized via proper fusion strategy. To further reject the out-of-database targets, generalized confidence is calculated by the posterior probability, and then thresholds are empirically chosen to implement the rejection operation. Finally, several evaluation indices are given. The experimental results demonstrate the advantage of classifier combination, as well as the feasibility of generalized confidence based rejection methods.
For subspace based dimensionality reduction, both linear and kernel discriminant subspace learning are investigated in depth. Therein, kernel-based methods are the extensions of corresponding linear methods via some specific kernel trick. These classical algorithms are summarized into single-subspace learning and multi-subspace learning, which are applied to image recognition and radar emitter recognition. The experimental results indicate that multi-subspace learning algorithms take into account the complementarity of different subspaces and are more robust than single-subspace based methods, no matter linear or kernel cases are concerned. Due to data dependence, kernel extension not always guarantees higher accuracy, so appropriate method needs to be selected according to the actual data distribution and engineering requirements.
引文
[1]韩俊,何明浩,李胜喜,等.现代雷达辐射源识别技术研究.航天电子对抗. 2008, 24(1): 43-45.
[2]张国柱.雷达辐射源识别技术研究.博士研究生学位论文.国防科学技术大学. 2005.
[3]张葛祥.雷达辐射源信号智能识别方法研究.博士研究生学位论文.西南交通大学. 2005.
[4]张葛详,胡来招,金炜东.雷达辐射源信号脉内特征分析.红外与毫米波学报. 2004, 23(6): 477-480.
[5]陈韬伟.基于脉内特征的雷达辐射源信号分选技术研究.博士研究生学位论文.西南交通大学. 2010.
[6]刘婷.基于循环平稳分析的雷达辐射源特征提取与融合识别.硕士研究生学位论文.西安电子科技大学. 2009.
[7]许丹.辐射源指纹机理及识别方法研究.博士研究生学位论文.国防科学技术大学. 2008.
[8]张政超,李应升,王磊,等.雷达辐射源信号识别研究综述.舰船电子工程. 2009, 29(4): 10-14.
[9]普运伟,金炜东,朱明,等.雷达辐射源信号模糊函数主脊切片特征提取方法.红外与毫米波学报. 2008, 27(2): 133-137.
[10]王磊,姬红兵,李林.基于模糊函数零点切片特征优化的辐射源个体识别.西安电子科技大学学报(自然科学版). 2010, 37(2): 285-289.
[11] Lei Wang, Hongbing Ji. Optimizing Zero-Slice Feature of Ambiguity Function for Radar Emitter Identification. In International Conference on Information, Communications and Signal Processing (ICICS). 2009: 1-4.
[12]王磊,姬红兵,史亚.基于模糊函数代表性切片的运动雷达辐射源识别.系统工程与电子技术. 2010, 32(8): 1630-1634.
[13] Lei Wang, Hongbing Ji, Ya Shi. Feature extraction and optimization of representative-slice in ambiguity function for moving radar emitter recognition. In IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP). 2010: 2246-2249.
[14] Lin Li, Hongbing Ji, Lei Wang. Specific Radar Emitters Recognition Based on Wavelet Packet Transform and Probabilistic SVM. In IEEE International Conference on Information and Automation (ICIA). 2009: 1308-1313.
[15]张国柱,黄可生,姜文利,等.基于信号包络的辐射源细微特征提取方法.系统工程与电子技术. 2006, 28(6): 795-798.
[16]朱明,金炜东,胡来招.一种基于Spectrum原子的雷达辐射源信号识别方法.电子与信息学报. 2009, 31(1): 188-191.
[17]刘海军,柳征,姜文利,等.一种基于云模型的辐射源识别方法.电子与信息学报. 2009, 31(9): 2079-2083.
[18]唐健仁,朱元清,徐庆,等.用神经网络进行雷达辐射源识别研究.空军雷达学院学报. 2007, 21(1): 25-27.
[19]郭小宾,王壮,胡卫东.基于贝叶斯网络分类器的雷达辐射源识别方法.火力与指挥控制. 2006, 31(2): 36-39.
[20] Zhang Z, Guan X, and He Y. Study on Radar Emitter Recognition Signal Based on Rough Sets and RBF Neural Network. In International Conference on Machine Learning and Cybernetics. 2009: 1225-1230.
[21] B.W. Gillespie, and L.E. Atlas. Optimizing time-frequency kernels for classification. IEEE Trans. Signal Processing. 2001, 49(3): 485-496.
[22] L. Cohen. Time-frequency distributions-a review. Proceedings of IEEE. 1989, 77(7): 941-981.
[23]刘成林,谭铁牛.模式识别研究进展.中国计算机学会通讯. 2007, 3(12): 45-52.
[24] R.O. Duda, P.E. Hart, and D.G. Stork. Pattern Classification (2nd Edition). John Wiley & Sons. 2001.
[25] A.K. Jain, R.P.W. Duin, J. Mao. Statistical pattern recognition: a review. IEEE Trans. PAMI. 2000, 22(1): 4-37.
[26] Ludmila I. Kuncheva. Combining Pattern Classifiers: Methods and Algorithms. John Wiley & Sons. 2004.
[27] J. Kittler, M. Hatef, R.P.W. Duin, et al.. On combining classifiers. IEEE Trans. PAMI. 1998, 20(3): 226-239.
[28] I. Rish, An empirical study of the naive Bayes classifier. In IJCAI Workshop on Empirical Methods in Artificial Intelligence. 2001: 41-46.
[29] J.C.Platt.Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods.In Advances in Large Margin Classifiers. 1999: 61-74.
[30] T.F.Wu, C.J.Lin, and R.C.Weng.Probability estimates for multi-class classification by pairwise coupling.Journal of Machine Learning Research. 2004, 5: 975-1005
[31]柴晶.雷达高分辨距离像目标识别的拒判算法和特征提取技术研究.博士研究生学位论文.西安电子科技大学. 2010.
[32]陈渤.基于核方法的雷达高分辨距离像目标识别技术研究.博士研究生学位论文.西安电子科技大学. 2008.
[33]林晓帆,丁晓青,吴佑寿.字符识别的置信度分析.清华大学学报(自然科学版). 1998, 38(9): 47-50.
[34]林晓帆,丁晓青,吴佑寿.最近邻分类器置信度估计的理论分析.科学通报. 1998, 43(3): 322-325.
[35] J.F. Pitrelli, J. Subrahmonia, M.P. Perrone, Confidence modeling for handwriting recognition: algorithms and applications. International Journal of Document Analysis. 2006, 8(1): 35-46.
[36] A.M. Martinez, and A.C. Kak. PCA versus LDA. IEEE Trans. PAMI. 2001, 23(2): 228-233.
[37] N.V. Boulgouris, K.N. Plataniotis, E.M. Tzanakou . Biometrics: Theory, Methods and Applications. John Wiley & Sons. 2009.
[38] P.N. Belhumeur, J.P. Hespanha, and D.J. Kriegman. Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection. IEEE Trans. PAMI. 1997, 19(7): 711-720.
[39] K. Fukunnaga. Introduction to Statistical Pattern Recognition. Academic Press. 1990.
[40] C.J. Liu, and H. Wechsler. Enhanced Fisher Linear Discriminant Models for Face Recognition. ICPR. 1998, 2: 1368-1372.
[41] J.H. Friedman, Regularized discriminant analysis. Journal of the American Statistical Association. 1989, 84(405):165-175.
[42] T. Hastie, A. Buja, and R. Tibshirani. Penalized discriminant analysis. The Annals of Statistics. 1995, 23(1): 73-102.
[43] L. Chen, H. Liao, M. Ko, et al.. A new LDA-based face recognition system which can solve the small sample size. Pattern Recognition. 2000, 33(10): 1713-1726.
[44] Rui Huang, Qingshan Liu, Hanqing Lu, et al.. Solving the Small Sample Size Problem of LDA. ICPR. 2002, 3: 29-32.
[45] H. Yu, and J. Yang. A direct LDA algorithm for high-dimensional data with application to face recognition. Pattern Recognition. 2001, 34(10): 2067-2070.
[46] J. Ye. Characterization of a family of algorithms for generalized discriminant analysis on undersampled problems. Journal of Machine Learning Research. 2005, 6:483-502.
[47] P. Howland, and H. Park. Generalizing discriminant analysis using the generalizedsingular value decomposition. IEEE Trans. PAMI. 2004, 26(8): 995-1006.
[48] Xiaogang Wang, and Xiaoou Tang. Dual-Space Linear Discriminant Analysis for Face Recognition. In IEEE international Conference on CVPR. 2004, 2:564-569.
[49] Hui Gao, and J.W. Davis. Why direct LDA is not equivalent to LDA. Pattern Recognition. 2006, 39: 1002-1006.
[50] Xudong Jiang, B. Mandal, and A. Kot. Eigenfeature Regularization and Extraction in Face Recognition. IEEE Trans. PAMI. 2008, 30(3): 383-394.
[51] Xiaogang Wang, and Xiaoou Tang. Random sampling LDA for face recognition. In IEEE international conference on CVPR. 2004, 2: 586-591.
[52]张丽梅,乔立山,陈松灿.基于张量模式的特征提取及分类器设计综述.山东大学学报. 2009, 39(1): 6-14, 40.
[53] J. Yang, David Zhang, A.F. Frangi, et al.. Two dimensional PCA: a new approach to appearance-based face representation and recognition. IEEE Trans. PAMI. 2004, 26(1): 131-137.
[54] J. Ye, R. Janardan, and Q. Li. Two-Dimensional Linear DiscriminantAnalysis. Neural Information Processing Systems (NIPS). 2005: 1569-1576.
[55] Shuicheng Yan, Dong Xu, Qiang Yang, et al.. Discriminant Analysis with Tensor Representation. In IEEE international conference on CVPR. 2005, 1: 526-532.
[56] Haiping Lu, K.N. Plataniotis, and A.N. Venetsanopoulos. MPCA: Multilinear Principal Component Analysis of Tensor Objects. IEEE Trans. Neural Networks. 2008, 19(1): 18-39.
[57] M. Loog, and R.P.W. Duin. Linear dimensionality reduction via a heteroscedastic extension of LDA: the Chernoff criterion. IEEE Trans. PAMI. 2004, 26(6): 732-739.
[58] K. Fukunaga, and J. Mantock. Nonparametric discriminant analysis. IEEE Trans. PAMI. 1983, 5: 671-678.
[59] T. Hastie, and R. Tibshirani. Discriminant analysis by Gaussian mixtures. Journal of the Royal Statistical Society, Series B. 1996, 58(1): 155-176.
[60] H.S. Seung, and D.D. Lee. The Manifold Ways of Perception. Science. 2000, 290: 2268-2269.
[61] S.T. Roweis, and L.K. Saul. Nonlinear Dimensionality Reduction by Locally Linear Embedding. Science. 2000, 290: 2323-2326.
[62] X.F. He, and P. Niyogi. Locality preserving projections. Neural Information Processing Systems (NIPS). 2003.
[63]陶晓燕.基于支持向量机和流形学习的分类方法研究.博士研究生学位论文.西安电子科技大学. 2008.
[64]杨静宇,金忠,杨健.模式特征抽取研究进展. 2009中国自动化大会暨两化融合高峰会议. 2009.
[65] B. Sch?lkopf, A.J. Sm?la, and K.R. Müller. Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation. 1998, 10(5): 1299-1319.
[66] S. Mika, G. R?tsch, J. Weston, et.al.. Fisher discriminant analysis with kernels. In IEEE international Workshop on Neural Networks for Signal Processing. IX, 1999: 41-48.
[67] G. Baudat, and F. Anouar. Generalized discriminant analysis using a kernel approach. Neural Computation. 2000, 12(10): 2385-2404.
[68] Qingshan Liu, and Rui Huang. Face Recognition Using Kernel Based Fisher Discriminant Analysis. In IEEE international conference on Autumatic Face and Gesture Recognition. 2002: 187-191.
[69] J. Yang, A. F. Frangi, J.-y. Yang, et al.. KPCA plus LDA: a Complete Kernel Fisher Discriminant Framework for Feature Extraction and Recognition, IEEE Trans. PAMI. 2005, 27(2), 230-244.
[70] K.R. Müller, S. Mika, G. R?tsch, et al.. An Introduction to Kernel-Based Learning Algorithms. IEEE Trans. Neural Networks. 2001, 12(2): 181-201.
[71] Wei Liu, Yunhong Wang, Stan Z. Li, et.al. Null space-based kernel fisher discriminant analysis for face recognition. In IEEE international conference on Autumatic Face and Gesture Recognition. 2004: 369-374.
[72] J. Lu, K.N. Plataniotis, and A.N. Venetsanopoulos. Face recognition using kernel direct discriminant analysis algorithms, IEEE Trans. Neural Networks. 2003, 14(1):117-126.
[73] Xudong Jiang, B. Mandal, and A. Kot. Complete discriminant evaluation and feature extraction in kernel space for face recognition. Machine Vision and Applications. 2009, 20: 35-46.
[74]周伟达.核机器学习方法研究.博士研究生学位论文.西安电子科技大学. 2003.
[75] ORL数据库. http://www.cs.uiuc.edu/homes/dengcai2.
[76] UMIST数据库. http://www.sheffield.ac.uk/eee/research/iel/research/face.html.
[77] USPS数据库. http://www.kernel-machines.org/.
[78]程云鹏,张凯院,徐仲.矩阵论(第3版).西北工业大学出版社. 2006.
[2]张国柱.雷达辐射源识别技术研究.博士研究生学位论文.国防科学技术大学. 2005.
[3]张葛祥.雷达辐射源信号智能识别方法研究.博士研究生学位论文.西南交通大学. 2005.
[4]张葛详,胡来招,金炜东.雷达辐射源信号脉内特征分析.红外与毫米波学报. 2004, 23(6): 477-480.
[5]陈韬伟.基于脉内特征的雷达辐射源信号分选技术研究.博士研究生学位论文.西南交通大学. 2010.
[6]刘婷.基于循环平稳分析的雷达辐射源特征提取与融合识别.硕士研究生学位论文.西安电子科技大学. 2009.
[7]许丹.辐射源指纹机理及识别方法研究.博士研究生学位论文.国防科学技术大学. 2008.
[8]张政超,李应升,王磊,等.雷达辐射源信号识别研究综述.舰船电子工程. 2009, 29(4): 10-14.
[9]普运伟,金炜东,朱明,等.雷达辐射源信号模糊函数主脊切片特征提取方法.红外与毫米波学报. 2008, 27(2): 133-137.
[10]王磊,姬红兵,李林.基于模糊函数零点切片特征优化的辐射源个体识别.西安电子科技大学学报(自然科学版). 2010, 37(2): 285-289.
[11] Lei Wang, Hongbing Ji. Optimizing Zero-Slice Feature of Ambiguity Function for Radar Emitter Identification. In International Conference on Information, Communications and Signal Processing (ICICS). 2009: 1-4.
[12]王磊,姬红兵,史亚.基于模糊函数代表性切片的运动雷达辐射源识别.系统工程与电子技术. 2010, 32(8): 1630-1634.
[13] Lei Wang, Hongbing Ji, Ya Shi. Feature extraction and optimization of representative-slice in ambiguity function for moving radar emitter recognition. In IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP). 2010: 2246-2249.
[14] Lin Li, Hongbing Ji, Lei Wang. Specific Radar Emitters Recognition Based on Wavelet Packet Transform and Probabilistic SVM. In IEEE International Conference on Information and Automation (ICIA). 2009: 1308-1313.
[15]张国柱,黄可生,姜文利,等.基于信号包络的辐射源细微特征提取方法.系统工程与电子技术. 2006, 28(6): 795-798.
[16]朱明,金炜东,胡来招.一种基于Spectrum原子的雷达辐射源信号识别方法.电子与信息学报. 2009, 31(1): 188-191.
[17]刘海军,柳征,姜文利,等.一种基于云模型的辐射源识别方法.电子与信息学报. 2009, 31(9): 2079-2083.
[18]唐健仁,朱元清,徐庆,等.用神经网络进行雷达辐射源识别研究.空军雷达学院学报. 2007, 21(1): 25-27.
[19]郭小宾,王壮,胡卫东.基于贝叶斯网络分类器的雷达辐射源识别方法.火力与指挥控制. 2006, 31(2): 36-39.
[20] Zhang Z, Guan X, and He Y. Study on Radar Emitter Recognition Signal Based on Rough Sets and RBF Neural Network. In International Conference on Machine Learning and Cybernetics. 2009: 1225-1230.
[21] B.W. Gillespie, and L.E. Atlas. Optimizing time-frequency kernels for classification. IEEE Trans. Signal Processing. 2001, 49(3): 485-496.
[22] L. Cohen. Time-frequency distributions-a review. Proceedings of IEEE. 1989, 77(7): 941-981.
[23]刘成林,谭铁牛.模式识别研究进展.中国计算机学会通讯. 2007, 3(12): 45-52.
[24] R.O. Duda, P.E. Hart, and D.G. Stork. Pattern Classification (2nd Edition). John Wiley & Sons. 2001.
[25] A.K. Jain, R.P.W. Duin, J. Mao. Statistical pattern recognition: a review. IEEE Trans. PAMI. 2000, 22(1): 4-37.
[26] Ludmila I. Kuncheva. Combining Pattern Classifiers: Methods and Algorithms. John Wiley & Sons. 2004.
[27] J. Kittler, M. Hatef, R.P.W. Duin, et al.. On combining classifiers. IEEE Trans. PAMI. 1998, 20(3): 226-239.
[28] I. Rish, An empirical study of the naive Bayes classifier. In IJCAI Workshop on Empirical Methods in Artificial Intelligence. 2001: 41-46.
[29] J.C.Platt.Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods.In Advances in Large Margin Classifiers. 1999: 61-74.
[30] T.F.Wu, C.J.Lin, and R.C.Weng.Probability estimates for multi-class classification by pairwise coupling.Journal of Machine Learning Research. 2004, 5: 975-1005
[31]柴晶.雷达高分辨距离像目标识别的拒判算法和特征提取技术研究.博士研究生学位论文.西安电子科技大学. 2010.
[32]陈渤.基于核方法的雷达高分辨距离像目标识别技术研究.博士研究生学位论文.西安电子科技大学. 2008.
[33]林晓帆,丁晓青,吴佑寿.字符识别的置信度分析.清华大学学报(自然科学版). 1998, 38(9): 47-50.
[34]林晓帆,丁晓青,吴佑寿.最近邻分类器置信度估计的理论分析.科学通报. 1998, 43(3): 322-325.
[35] J.F. Pitrelli, J. Subrahmonia, M.P. Perrone, Confidence modeling for handwriting recognition: algorithms and applications. International Journal of Document Analysis. 2006, 8(1): 35-46.
[36] A.M. Martinez, and A.C. Kak. PCA versus LDA. IEEE Trans. PAMI. 2001, 23(2): 228-233.
[37] N.V. Boulgouris, K.N. Plataniotis, E.M. Tzanakou . Biometrics: Theory, Methods and Applications. John Wiley & Sons. 2009.
[38] P.N. Belhumeur, J.P. Hespanha, and D.J. Kriegman. Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection. IEEE Trans. PAMI. 1997, 19(7): 711-720.
[39] K. Fukunnaga. Introduction to Statistical Pattern Recognition. Academic Press. 1990.
[40] C.J. Liu, and H. Wechsler. Enhanced Fisher Linear Discriminant Models for Face Recognition. ICPR. 1998, 2: 1368-1372.
[41] J.H. Friedman, Regularized discriminant analysis. Journal of the American Statistical Association. 1989, 84(405):165-175.
[42] T. Hastie, A. Buja, and R. Tibshirani. Penalized discriminant analysis. The Annals of Statistics. 1995, 23(1): 73-102.
[43] L. Chen, H. Liao, M. Ko, et al.. A new LDA-based face recognition system which can solve the small sample size. Pattern Recognition. 2000, 33(10): 1713-1726.
[44] Rui Huang, Qingshan Liu, Hanqing Lu, et al.. Solving the Small Sample Size Problem of LDA. ICPR. 2002, 3: 29-32.
[45] H. Yu, and J. Yang. A direct LDA algorithm for high-dimensional data with application to face recognition. Pattern Recognition. 2001, 34(10): 2067-2070.
[46] J. Ye. Characterization of a family of algorithms for generalized discriminant analysis on undersampled problems. Journal of Machine Learning Research. 2005, 6:483-502.
[47] P. Howland, and H. Park. Generalizing discriminant analysis using the generalizedsingular value decomposition. IEEE Trans. PAMI. 2004, 26(8): 995-1006.
[48] Xiaogang Wang, and Xiaoou Tang. Dual-Space Linear Discriminant Analysis for Face Recognition. In IEEE international Conference on CVPR. 2004, 2:564-569.
[49] Hui Gao, and J.W. Davis. Why direct LDA is not equivalent to LDA. Pattern Recognition. 2006, 39: 1002-1006.
[50] Xudong Jiang, B. Mandal, and A. Kot. Eigenfeature Regularization and Extraction in Face Recognition. IEEE Trans. PAMI. 2008, 30(3): 383-394.
[51] Xiaogang Wang, and Xiaoou Tang. Random sampling LDA for face recognition. In IEEE international conference on CVPR. 2004, 2: 586-591.
[52]张丽梅,乔立山,陈松灿.基于张量模式的特征提取及分类器设计综述.山东大学学报. 2009, 39(1): 6-14, 40.
[53] J. Yang, David Zhang, A.F. Frangi, et al.. Two dimensional PCA: a new approach to appearance-based face representation and recognition. IEEE Trans. PAMI. 2004, 26(1): 131-137.
[54] J. Ye, R. Janardan, and Q. Li. Two-Dimensional Linear DiscriminantAnalysis. Neural Information Processing Systems (NIPS). 2005: 1569-1576.
[55] Shuicheng Yan, Dong Xu, Qiang Yang, et al.. Discriminant Analysis with Tensor Representation. In IEEE international conference on CVPR. 2005, 1: 526-532.
[56] Haiping Lu, K.N. Plataniotis, and A.N. Venetsanopoulos. MPCA: Multilinear Principal Component Analysis of Tensor Objects. IEEE Trans. Neural Networks. 2008, 19(1): 18-39.
[57] M. Loog, and R.P.W. Duin. Linear dimensionality reduction via a heteroscedastic extension of LDA: the Chernoff criterion. IEEE Trans. PAMI. 2004, 26(6): 732-739.
[58] K. Fukunaga, and J. Mantock. Nonparametric discriminant analysis. IEEE Trans. PAMI. 1983, 5: 671-678.
[59] T. Hastie, and R. Tibshirani. Discriminant analysis by Gaussian mixtures. Journal of the Royal Statistical Society, Series B. 1996, 58(1): 155-176.
[60] H.S. Seung, and D.D. Lee. The Manifold Ways of Perception. Science. 2000, 290: 2268-2269.
[61] S.T. Roweis, and L.K. Saul. Nonlinear Dimensionality Reduction by Locally Linear Embedding. Science. 2000, 290: 2323-2326.
[62] X.F. He, and P. Niyogi. Locality preserving projections. Neural Information Processing Systems (NIPS). 2003.
[63]陶晓燕.基于支持向量机和流形学习的分类方法研究.博士研究生学位论文.西安电子科技大学. 2008.
[64]杨静宇,金忠,杨健.模式特征抽取研究进展. 2009中国自动化大会暨两化融合高峰会议. 2009.
[65] B. Sch?lkopf, A.J. Sm?la, and K.R. Müller. Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation. 1998, 10(5): 1299-1319.
[66] S. Mika, G. R?tsch, J. Weston, et.al.. Fisher discriminant analysis with kernels. In IEEE international Workshop on Neural Networks for Signal Processing. IX, 1999: 41-48.
[67] G. Baudat, and F. Anouar. Generalized discriminant analysis using a kernel approach. Neural Computation. 2000, 12(10): 2385-2404.
[68] Qingshan Liu, and Rui Huang. Face Recognition Using Kernel Based Fisher Discriminant Analysis. In IEEE international conference on Autumatic Face and Gesture Recognition. 2002: 187-191.
[69] J. Yang, A. F. Frangi, J.-y. Yang, et al.. KPCA plus LDA: a Complete Kernel Fisher Discriminant Framework for Feature Extraction and Recognition, IEEE Trans. PAMI. 2005, 27(2), 230-244.
[70] K.R. Müller, S. Mika, G. R?tsch, et al.. An Introduction to Kernel-Based Learning Algorithms. IEEE Trans. Neural Networks. 2001, 12(2): 181-201.
[71] Wei Liu, Yunhong Wang, Stan Z. Li, et.al. Null space-based kernel fisher discriminant analysis for face recognition. In IEEE international conference on Autumatic Face and Gesture Recognition. 2004: 369-374.
[72] J. Lu, K.N. Plataniotis, and A.N. Venetsanopoulos. Face recognition using kernel direct discriminant analysis algorithms, IEEE Trans. Neural Networks. 2003, 14(1):117-126.
[73] Xudong Jiang, B. Mandal, and A. Kot. Complete discriminant evaluation and feature extraction in kernel space for face recognition. Machine Vision and Applications. 2009, 20: 35-46.
[74]周伟达.核机器学习方法研究.博士研究生学位论文.西安电子科技大学. 2003.
[75] ORL数据库. http://www.cs.uiuc.edu/homes/dengcai2.
[76] UMIST数据库. http://www.sheffield.ac.uk/eee/research/iel/research/face.html.
[77] USPS数据库. http://www.kernel-machines.org/.
[78]程云鹏,张凯院,徐仲.矩阵论(第3版).西北工业大学出版社. 2006.