空中目标ISAR像特征提取与识别技术研究
|
摘要
逆合成孔径雷达（Inverse Synthetic Aperture Radar，ISAR）作为一种高分辨二维成像设备，能全天候全天时的获取反映目标大小形状结构及姿态等细节信息的二维高分辨雷达图像，可以为目标分类与识别提供丰富的特征信息基于ISAR像的目标识别技术在现代军事应用中发挥着非常重要的作用，受到国内外的热切关注本论文以空中目标识别为背景，针对空中目标ISAR像识别存在的问题，在对空中目标进行几何建模和仿真成像的基础上，对ISAR像预处理技术和空中目标ISAR像的特征提取算法等内容进行了较为深入的研究本文的主要内容概括如下：
第一章阐述了论文的研究背景及其意义，较详尽地介绍了ISAR目标识别技术的发展概况，然后针对空中目标的气动布局，对飞机部件的散射特性和ISAR识别中的关键问题进行了分析，最后归纳和总结了论文的主要工作
第二章研究了空中目标ISAR像的建模仿真和图像预处理技术首先，介绍了ISAR成像的基本原理，探讨了基于Mu1tigen Creator软件的Open Flight模型建模技术，并采用RadBase产生雷达回波，利用Matlab完成目标ISAR成像仿真；然后，分析了目标ISAR像的固有特性以及影响ISAR像的主要因素，并以F14和F18飞机为例讨论了不同姿态下ISAR像的相关性；最后，针对ISAR像的斑点噪声和横条纹干扰特性，提出了相应的干扰抑制算法，并对图像进行归一化处理，提高图像稳定性以利于后续的特征提取与识别
第三章研究了ISAR像方位估计及其仿射不变特征提取算法首先，为了提高模板匹配效率，引入基于目标主轴和基于最小面积外接矩形的两种方位估计思想，在ISAR像强散射中心视角相关性分析基础上，提出了一种基于ISAR像方位估计的相关匹配算法，以方位角估计值为索引缩小模板搜索范围，提高了ISAR像模板匹配的效率；其次，针对ISAR像的平移姿态和尺度不一致性问题，获取了ISAR目标特征点及其仿射不变特征，并存入几何散列表中，然后采用几何散列法实现了目标的匹配识别，该算法能够有效区分不同结构不同视角和尺度目标，并具备较强的局部识别能力
第四章研究了基于非负矩阵分解的ISAR像统计特征提取算法类比于人类视觉认知机制，以非负矩阵分解（NMF）为数学工具获取ISAR像的统计分类特征向量，提出了基于优化判别非负矩阵分解和基于子类判别非负矩阵分解的ISAR目标特征提取算法首先，针对ISAR目标的孤立稀疏强散射中心分布特性，分别采用局部非负矩阵分解（LNMF）和非负稀疏编码（NNSC）获取ISAR目标的局部性和稀疏性特征空间，然后引入线性判别思想，对特征空间特征基进行优化精选，提出了基于优化判别局部非负矩阵分解（ODLNMF）和优化判别非负稀疏编码（ODNNSC）的ISAR目标特征提取算法，从而通过优化特征基的鉴别性和鲁棒性提高ISAR目标分类识别性能；其次，针对ISAR像的多模特性，引入聚类判别分析（CDA）思想，在LNMF和NNSC的目标函数上加入子类判别限制，提出了基于子类判别局部非负矩阵分解（SDLNMF）和子类判别非负稀疏编码（SDNNSC）的ISAR目标特征提取算法，并利用Expectation-Maximization （EM）方法推导了各自的特征基和特征矢量的迭代规则实验证明，聚类判别分析思想的引入提高了目标识别的有效性和稳健性
第五章研究了基于正交邻域子类判别投影（ONSDP）的流形特征提取算法首先，为了提取ISAR高维数据集中的低维流形特征，提出了正交邻域子类判别投影流形学习算法该算法融合了数据局部拓扑保持和子类判别降维思想，通过引入正交约束，使得获取的投影特征保持了原数据的欧式空间结构，有效地解决了ISAR像多模数据的分类问题；另一方面，为了更有效的利用ISAR非线性结构，通过在ONSDP算法加入核函数，提出了核正交邻域子类判别投影（KONSDP）算法，使用径向基核函数代替内积，隐式的将原始的非线性数据通过非线性映射到高维线性可分空间，给出了其推导过程及实验验证；最后，将ISAR像视为二阶张量，通过CDA和ONSDP的张量化发展，提出了张量子类判别分析（TSDA）和张量正交邻域子类判别投影（TONSDP）算法，并给出了推导过程TSDA和TONSDP保留了图像自身行与列之间的结构信息，同时加入了子类判别信息，最终将最优张量子空间求解归结为广义特征向量求解问题论文通过实验验证了所提算法在提高目标识别率方面的优势
第六章总结全文，并指出了下一步需要开展的工作
Inverse Synthetic Aperture Radar (ISAR) can obtain two-dimensional highresolution images, with detailed information, such as scale, shape, structure and gesture,which afford abundant features for classification and recognition of targets, in allweather, day or night. Therefore, the technique of automatic target recognition (ATR)using ISAR images is playing a very important role in modern military applications andis always being paid attention to. In order to resolve the problems of classifying airtargets using ISAR images, several special researches are made on the basis ofgeometric modeling and imaging simulation in this dissertation, including ISAR imagespre-processing and feature extraction algorithms of ISAR images.
In chapter1, the research background and significance is introduced, and thedevelopment of ATR techniques using ISAR images is reviewed. Then aerodynamiclayout of air targets, scattering properties of aircraft components and the key in ATR ofISAR targets are analyzed in detail, followed by the introduction of main content in thisdissertation.
In chapter2, modeling and simulation of ISAR targets, image pre-processingtechnology are inverstigated. At first, the principles of ISAR imaging, Open Flightmodeling with Mu1tigen Creator, simulation of ISAR imaging using Matlab with radarwave generated by RadBase are introduced; then, the inherent characteristics and mainfactors affecting ISAR images are analyzed, followed by correlation stduy betweenISAR images using F14and F18aircrafts as examples; finally, suppression of thespeckle noise and interferential stripes，images normalization are proposed to enhancethe stability of images，which is propitious to the latter feature extraction andrecognition.
In chapter3, algorithms for aspect estimation and affine invariant featureextraction are presented. First of all, in order to improve the efficiency of templatematching, we combine the thoughts of target’s principal axis extraction and minimumenclosed rectangle acquirement, and propose a correlation matching method based onaspect estimation, which narrows the scope of template search by aspect estimation andimprove the efficiency of ISAR template matching; secondly, aimed at resolving theproblems of translation, gesture and scale inconsistencies, we extract key points ofISAR images, constructs affine-invariance features which are then saved in the hashtables, and uses geometric hash as the final classifier. Simulations show the presentedalgorithm not only effectively distinguishes targets with different structures, imagingperspective and scale, but also performs excellently in local recognition.
In chapter4, algorithms for statistical feature extraction of ISAR images based onnon-negative matrix factorization (NMF) are proposed. Introducing visual perceptionconcept to classification, we adopt NMF as a mathematical tool to obtain the statistical feature vector of an ISAR image, and propose approaches of feature extraction based onoptimized discriminant non-negative matrix factorization and subclass discriminantnon-negative matrix factorization. At first, we construct a feature space based on localnon-negative matrix factorization (LNMF) and non-negative sparse coding (NNSC),which is convenient for local and sparse feature extraction, respectively, then optimizethe feature space by screening the feature base to enhance the classification performance,and finally present algorithms for ISAR target recognition based on optimized discri-minant local non-negative matrix factorization (ODLNMF) and optimized discriminantnon-negative sparse coding (ODNNSC); secondly, to deal with the multimodal distribu-tions of ISAR images, constraints inspired by the clustering based discriminant analysis(CDA) are imposed on LNMF and NNSC to deduce algorithms for ISAR target recog-nition, which are based on subclass discriminant local non-negative matrix factorization(SDLNMF) and subclass discriminant non-negative sparse coding (SDNNSC).SDLNMF and SDNNSC extend the LNMF and NNSC decomposition, respectively, byembedding the subclass discriminant constraints, and reformulate the cost function toachieve discriminant projections. Simulations demonstrate the effectiveness and ro-bustness of the proposed method.
In chapter5, algorithms for manifold feature extraction of ISAR images based onorthogonal neighborhood subclass discriminant projections (ONSDP) are presented. Atfirst, ONSDP is proposed to extract low-dimensional feature from high-dimensional andmultimodal ISAR images, which incorporates local topology preserving and subclassdiscriminant dimensionality reduction. Moreover, an orthogonal constraint is introducedto preserve geometric structures of ISAR images; secondly, to explore nonlinear struc-ture of ISAR targets, we add kernel function to ONSDP and propose kernel orthogonalneighborhood subclass discriminant projections (KONSDP), which replaces innerproduct with RBF kernel function and maps implicitly original nonlinear data to a high-er dimensional linearly separable space. Derivation process and verification by simula-tions are given later; thirdly, regarding an ISAR image as the second-order tensor, wepropose algorithm steps as well as derivation process of tensor subclass discriminantanalysis (TSDA) and tensor orthogonal neighborhood subclass discriminant projections(TONSDP), which preserve the structure information between the rows and columns inthe ISAR image and takes advantage of subclass discriminant analysis. Optimization oftensor subspace is finally converted to the problem of generalized eigenvectors Simula-tions verificate the advantage of these methods in target recognition rate.
Charpter6summerizes this dissertation and discusses the future work.
第一章阐述了论文的研究背景及其意义，较详尽地介绍了ISAR目标识别技术的发展概况，然后针对空中目标的气动布局，对飞机部件的散射特性和ISAR识别中的关键问题进行了分析，最后归纳和总结了论文的主要工作
第二章研究了空中目标ISAR像的建模仿真和图像预处理技术首先，介绍了ISAR成像的基本原理，探讨了基于Mu1tigen Creator软件的Open Flight模型建模技术，并采用RadBase产生雷达回波，利用Matlab完成目标ISAR成像仿真；然后，分析了目标ISAR像的固有特性以及影响ISAR像的主要因素，并以F14和F18飞机为例讨论了不同姿态下ISAR像的相关性；最后，针对ISAR像的斑点噪声和横条纹干扰特性，提出了相应的干扰抑制算法，并对图像进行归一化处理，提高图像稳定性以利于后续的特征提取与识别
第三章研究了ISAR像方位估计及其仿射不变特征提取算法首先，为了提高模板匹配效率，引入基于目标主轴和基于最小面积外接矩形的两种方位估计思想，在ISAR像强散射中心视角相关性分析基础上，提出了一种基于ISAR像方位估计的相关匹配算法，以方位角估计值为索引缩小模板搜索范围，提高了ISAR像模板匹配的效率；其次，针对ISAR像的平移姿态和尺度不一致性问题，获取了ISAR目标特征点及其仿射不变特征，并存入几何散列表中，然后采用几何散列法实现了目标的匹配识别，该算法能够有效区分不同结构不同视角和尺度目标，并具备较强的局部识别能力
第四章研究了基于非负矩阵分解的ISAR像统计特征提取算法类比于人类视觉认知机制，以非负矩阵分解（NMF）为数学工具获取ISAR像的统计分类特征向量，提出了基于优化判别非负矩阵分解和基于子类判别非负矩阵分解的ISAR目标特征提取算法首先，针对ISAR目标的孤立稀疏强散射中心分布特性，分别采用局部非负矩阵分解（LNMF）和非负稀疏编码（NNSC）获取ISAR目标的局部性和稀疏性特征空间，然后引入线性判别思想，对特征空间特征基进行优化精选，提出了基于优化判别局部非负矩阵分解（ODLNMF）和优化判别非负稀疏编码（ODNNSC）的ISAR目标特征提取算法，从而通过优化特征基的鉴别性和鲁棒性提高ISAR目标分类识别性能；其次，针对ISAR像的多模特性，引入聚类判别分析（CDA）思想，在LNMF和NNSC的目标函数上加入子类判别限制，提出了基于子类判别局部非负矩阵分解（SDLNMF）和子类判别非负稀疏编码（SDNNSC）的ISAR目标特征提取算法，并利用Expectation-Maximization （EM）方法推导了各自的特征基和特征矢量的迭代规则实验证明，聚类判别分析思想的引入提高了目标识别的有效性和稳健性
第五章研究了基于正交邻域子类判别投影（ONSDP）的流形特征提取算法首先，为了提取ISAR高维数据集中的低维流形特征，提出了正交邻域子类判别投影流形学习算法该算法融合了数据局部拓扑保持和子类判别降维思想，通过引入正交约束，使得获取的投影特征保持了原数据的欧式空间结构，有效地解决了ISAR像多模数据的分类问题；另一方面，为了更有效的利用ISAR非线性结构，通过在ONSDP算法加入核函数，提出了核正交邻域子类判别投影（KONSDP）算法，使用径向基核函数代替内积，隐式的将原始的非线性数据通过非线性映射到高维线性可分空间，给出了其推导过程及实验验证；最后，将ISAR像视为二阶张量，通过CDA和ONSDP的张量化发展，提出了张量子类判别分析（TSDA）和张量正交邻域子类判别投影（TONSDP）算法，并给出了推导过程TSDA和TONSDP保留了图像自身行与列之间的结构信息，同时加入了子类判别信息，最终将最优张量子空间求解归结为广义特征向量求解问题论文通过实验验证了所提算法在提高目标识别率方面的优势
第六章总结全文，并指出了下一步需要开展的工作
Inverse Synthetic Aperture Radar (ISAR) can obtain two-dimensional highresolution images, with detailed information, such as scale, shape, structure and gesture,which afford abundant features for classification and recognition of targets, in allweather, day or night. Therefore, the technique of automatic target recognition (ATR)using ISAR images is playing a very important role in modern military applications andis always being paid attention to. In order to resolve the problems of classifying airtargets using ISAR images, several special researches are made on the basis ofgeometric modeling and imaging simulation in this dissertation, including ISAR imagespre-processing and feature extraction algorithms of ISAR images.
In chapter1, the research background and significance is introduced, and thedevelopment of ATR techniques using ISAR images is reviewed. Then aerodynamiclayout of air targets, scattering properties of aircraft components and the key in ATR ofISAR targets are analyzed in detail, followed by the introduction of main content in thisdissertation.
In chapter2, modeling and simulation of ISAR targets, image pre-processingtechnology are inverstigated. At first, the principles of ISAR imaging, Open Flightmodeling with Mu1tigen Creator, simulation of ISAR imaging using Matlab with radarwave generated by RadBase are introduced; then, the inherent characteristics and mainfactors affecting ISAR images are analyzed, followed by correlation stduy betweenISAR images using F14and F18aircrafts as examples; finally, suppression of thespeckle noise and interferential stripes，images normalization are proposed to enhancethe stability of images，which is propitious to the latter feature extraction andrecognition.
In chapter3, algorithms for aspect estimation and affine invariant featureextraction are presented. First of all, in order to improve the efficiency of templatematching, we combine the thoughts of target’s principal axis extraction and minimumenclosed rectangle acquirement, and propose a correlation matching method based onaspect estimation, which narrows the scope of template search by aspect estimation andimprove the efficiency of ISAR template matching; secondly, aimed at resolving theproblems of translation, gesture and scale inconsistencies, we extract key points ofISAR images, constructs affine-invariance features which are then saved in the hashtables, and uses geometric hash as the final classifier. Simulations show the presentedalgorithm not only effectively distinguishes targets with different structures, imagingperspective and scale, but also performs excellently in local recognition.
In chapter4, algorithms for statistical feature extraction of ISAR images based onnon-negative matrix factorization (NMF) are proposed. Introducing visual perceptionconcept to classification, we adopt NMF as a mathematical tool to obtain the statistical feature vector of an ISAR image, and propose approaches of feature extraction based onoptimized discriminant non-negative matrix factorization and subclass discriminantnon-negative matrix factorization. At first, we construct a feature space based on localnon-negative matrix factorization (LNMF) and non-negative sparse coding (NNSC),which is convenient for local and sparse feature extraction, respectively, then optimizethe feature space by screening the feature base to enhance the classification performance,and finally present algorithms for ISAR target recognition based on optimized discri-minant local non-negative matrix factorization (ODLNMF) and optimized discriminantnon-negative sparse coding (ODNNSC); secondly, to deal with the multimodal distribu-tions of ISAR images, constraints inspired by the clustering based discriminant analysis(CDA) are imposed on LNMF and NNSC to deduce algorithms for ISAR target recog-nition, which are based on subclass discriminant local non-negative matrix factorization(SDLNMF) and subclass discriminant non-negative sparse coding (SDNNSC).SDLNMF and SDNNSC extend the LNMF and NNSC decomposition, respectively, byembedding the subclass discriminant constraints, and reformulate the cost function toachieve discriminant projections. Simulations demonstrate the effectiveness and ro-bustness of the proposed method.
In chapter5, algorithms for manifold feature extraction of ISAR images based onorthogonal neighborhood subclass discriminant projections (ONSDP) are presented. Atfirst, ONSDP is proposed to extract low-dimensional feature from high-dimensional andmultimodal ISAR images, which incorporates local topology preserving and subclassdiscriminant dimensionality reduction. Moreover, an orthogonal constraint is introducedto preserve geometric structures of ISAR images; secondly, to explore nonlinear struc-ture of ISAR targets, we add kernel function to ONSDP and propose kernel orthogonalneighborhood subclass discriminant projections (KONSDP), which replaces innerproduct with RBF kernel function and maps implicitly original nonlinear data to a high-er dimensional linearly separable space. Derivation process and verification by simula-tions are given later; thirdly, regarding an ISAR image as the second-order tensor, wepropose algorithm steps as well as derivation process of tensor subclass discriminantanalysis (TSDA) and tensor orthogonal neighborhood subclass discriminant projections(TONSDP), which preserve the structure information between the rows and columns inthe ISAR image and takes advantage of subclass discriminant analysis. Optimization oftensor subspace is finally converted to the problem of generalized eigenvectors Simula-tions verificate the advantage of these methods in target recognition rate.
Charpter6summerizes this dissertation and discusses the future work.
引文
[1] Ausherman D A, Kozma A, Walker J L, et al. Development in radar imaging[J].IEEE Trans. On AES,1984,20(4):363-399.
[2]孙文峰.雷达目标识别技术述评[J].雷达与对抗,2001(3):1-8.
[3] Bhanu B. Automatic target recognition：State of the art survey[J]. IEEE Trans．onAES,1986,22(4):364-379.
[4]刘宏伟,杜兰,袁莉等.雷达高分辨距离像目标识别研究进展[J].电子与信息学报,2005,27(8):1328-1334.
[5]郭桂蓉,庄钊文,陈曾平.电磁特征抽取与目标识别[M].长沙:国防科技大学出版社,1996.
[6]黄培康,殷红成,许小剑.雷达目标特性[M].北京:电子工业出版社,2005.
[7]保铮,邢孟道,王彤.雷达成像技术[M].北京:电子工业出版社,2005.
[8]刘永坦.雷达成像技术[M].哈尔滨:哈尔滨工业大学出版社,1999.
[9] Hovanessian S A. Introduction to synthetic array and imaging radars[M]. ArtechHouse,1980.
[10] Brown W M. Synthetic aperture radar[J]. IEEE Trans.AES,1967,3(2):217-299.
[11] Wiley C A. Synthetic aperture radars[J]. IEEE Trans. On AES,1985,21(3):440-443.
[12] Curlander J C, Mcdonough R N. Synthetic aperture radar systems and signal pro-cessing[M]. John Wiley&Sons,1991.
[13]张直中.合成孔径逆合成孔径和成像雷达[J].现代雷达,1985,7(56).
[14] Chen C C, Andrews H C. Target-motion-induced Radar Imaging[J]. IEEETrans.AES,1967,3(2):217-299.
[15] Prickett M J, Chen C C. Principle of inverse synthetic aperture radar[J]. EASCONrecord,1980:340-345.
[16]张直中.微波成像技术[M].北京:科学出版社,1990.
[17]张澄波.综合孔径雷达[M].北京:科学出版社,1989.
[18] Brown W M. Range-Doppler imaging with motion through resolution cells[J].IEEE Trans.AES,1969,5(1):98-102.
[19] Chen C C, Andrews H C. Target-motion-induced radar imaging[J]. IEEE Trans.AES,1980,16(1):2-14.
[20]曹志道,刘永坦,宿富林.实验ISAR外场实验及数据处理[C].信息获取与处理技术主题十周年汇报,1996:56-64.
[21]保铮,邓文彪,杨军.散射重心跟踪的ISAR运动补偿方法[J].西安电子科技大学学报,1990,17(3):1-8.
[22]朱兆达,邱晓辉,佘志舜.用改进的多普勒中心跟踪法进行ISAR运动补偿[J].电子学报,1997,25(3):65-69.
[23]邱晓辉.运动目标雷达成像研究[D].南京航空航天大学,1995.
[24] Kennaugh E M, Moffatt D L. Transient and Impulse Response Approximations[J].Proc. IEEE,1965,53(8):893-901.
[25] Baum C E. On the Singularity Expansion Method for the Solution of Electromag-netic Interaction Problems[J]. AFWL Interaction Notes,1971.
[26] Wicke P E Z, Kiss I. A new implementations of Mellin Transform and its applica-tion to radar classification of ships[J]. IEEE trans on pattern analysis and machineintelligence,1983,5(2):598-606.
[27] Zyweck A, Bogner R E. Radar Target Classification of Commercial Aircraft[J]. IEEE Trans. AES,1996,32(2):598-606.
[28] Gabel R A. machine intelligence and radar image understanding[C]. In proceed-ings of the twenty-first asilomar conference on signals, systems, and computers,pacific grove，CA,1987:775-879.
[29] Jaenicke R A. structure from limited motion of complex rigid objects[C]. In pro-ceedings of IEEE workshop on motion, Irvine, CA,1989:256-263.
[30] Paul F H. radar image understanding for complex space objects[C]. Proc.SPIE1381, Boston MA,1991:55-65.
[31] Hu Y K, Yang Y M. Automatic Target Recognition of ISAR Images Based onHausdorff Distance[C].1st Asian and Pacific Conference on Synthetic ApertureRadar, Huangshan,2007:477-479.
[32] Saidi M N, Daoudi K, Khenchaf A. Automatic target recognition of aircraft mod-els based on ISAR images[C]. Geoscience and Remote Sensing Symposium, CapeTown,2009:685-688.
[33] Saidi M N, Toumi A, Khenchaf A, et al. Pose estimation for ISAR image classifi-cation[C]. In Proceedings of IGARSS, Brest, France,2010:4620-4623.
[34] Saidi M N, Hoeltzener B, Toumi A, et al. Automatic recognition of ISAR Images:Target shapes features extraction[C]. ICTTA3rd International Conference, Da-mascus,2008:1-6.
[35] Maki A, Fukui K, Onoguchi K, et al. ISAR Image Analysis by Subspace Method:Automatic Extraction and Identification of Ship Profile[C].11th internationalconference on image analysis and processing, Palermo,2001:523-528.
[36]张兴敢.逆合成孔径雷达成像及目标识别[D].南京:南京航空航天大学,2001.
[37]高兴斌,刘永坦. ISAR目标像的特征提取和特征选择[J].哈尔滨工业大学学报,1994,26(5):77281.
[38]张善文,甄蜀春,郑全弟等.雷达目标特征提取的一种方法[J].电子与信息学报,2000,24(12):1994-1997.
[39]范春彦,李晓曼,付红卫等. ISAR成像的处理与分类方法[J].电光与控制,2003,10(2):70-72.
[40]马君国,赵宏钟,李保国.基于二维小波变换的空间目标识别算法[J].国防科技大学学报,2006,28(1):57-61.
[41]马君国.空间雷达目标特征提取与识别方法研究[D].长沙:国防科学技术大学,2006.
[42]许人灿,刘朝军,黄小红等.基于超分辨ISAR成像的空中目标自动识别[J].系统工程与电子技术,2006,28(1):46-48.
[43] Botha E C. Classification of aerospace targets using superresolution ISAR imag-es[C]. Communications and Singnal Processing Proceeding, Stellenbosch,1994:138-145.
[44]季战领,冯晓毅,阎昆.一种ISAR图像中飞机目标识别的新特征[J].计算机工程,2010,36(23):192-193,196.
[45]李大芳.基于ISAR图像目标识别算法的研究[D].哈尔滨:哈尔滨工业大学,2004.
[46]陈善丽.逆合成孔径雷达图像识别研究[D].杭州:浙江大学,2008.
[47] Rosenbach K, Schiller J. Identification of aircraft on the basis of2-D radar imag-es[C]. IEEE international Radar Conference, Alexandria, VA,1995:405-409.
[48] Kim K, Seo D, Kim H. Efficient Classification of ISAR Images[J]. IEEE Trans.Antennas and Propagation,2005,53(5):1611-1621.
[49] Musman S, Kerr D, Bachmann C. Automatic recognition of ISAR ship images[J].IEEE Trans. Aerosp. Electron. Syst.,1996,32(4):1392-1404.
[50] Pastina D, Spina C. Multi-feature based automatic recognition of ship targets inISAR[J]. IET Radar Sonar Navig.,2009,3(4):406-423.
[51] Zeljkovi V, Li Q, Vincelette R, et al. Automatic algorithm for inverse syntheticaperture radar images recognition and classification [J]. IET Radar Sonar Navig.,2010,1(4):96-109.
[52] Zeljkovi V, Li Q, Vincelette R. Noise resistant algorithm for radar imagesrecognition and classification[C]. Signal and Data Processing of Small Targets,SPIE, San Diego,2009:744507.
[53] Zeljkovi V, Li Q, Vincelette R. Aircraft Identification by Unions of ISAR Imag-es[C]. Proceedings of the5th International ICST Mobile Multimedia Communica-tions Conference, Brussels,2009.
[54]崔延鹏,胡建伟,杨绍全等.利用尺度不变量特征的ISAR二维像自动识别技术[J].西安电子科技大学学报,2009,36(4):725-728.
[55]刘烽,许家栋.基于ISAR图像序列的目标识别方法研究[C].第九届全国雷达学术年会,烟台,2004:580-583.
[56] Chia T, Tien S, Lu Y. recognize aircraft in ISAR images[J]. Journal Of Infor-mation Science And Engineering,2007(23):299-313.
[57] Maskall G T, Webb A R. Nonlinear feature extraction applied to ISAR images oftargets for classification[C]. the international society for optical engineering pro-cessings of SPIE,2001:255-265.
[58] Maskall G T. An application of nonlinear feature extraction to the classification ofISAR images[C]. RADAR2002, UK:405-408.
[59] Maskall G T, Webb A R. Nonlinear feature extraction for MMW image classifica-tion: a supervised approach[C]. Proc.SPIE,2002:353-363.
[60]何强,蔡洪,韩壮志等.基于非线性流形学习的ISAR目标识别研究[J].电子学报,2010,38(3):585-590.
[61]蔡洪,何强,韩壮志等.空间平滑局部保持投影用于ISAR目标识别[J].光电工程,2009,36(7):8-13.
[62]蔡洪,何强,韩壮志等.基于局部保持判别子空间的ISAR目标识别[J].中国电子科学研究院学报,2009,4(6):656-660.
[63]朱兆达,殷军,邬小青.基于信号时间-频率分析的多运动目标ISAR成像[J].现代雷达,1993,34(5):33-37.
[64] Soumekh M, Nugroho S. ISAR imaging of an airborne DC-9[C]. IEEE interna-tional conference on acoustic,speech and signal processing,1993:465-468.
[65] Walker J L. Range Doppler imaging of rotating objects[J]. Ieee Transactions OnAerospace And Electronic Systems,1980,16(1):23-52.
[66] Chen C C. Multi-frequency imaging of radar turntable data[J]. IEEE Transactionson Aerospace and Electronic Systems,1980,16(1):15-22.
[67]朱自强,陈迎春,王晓璐等.现代飞机的空气动力设计[M].北京:国防工业出版社,2011.
[68]飞机气动布局简介. http://wenku.baidu.com/view/6c4e2ed3360cba1aa811da43.html.
[69]黄凤岗,宋克欧.模式识别[M].哈尔滨:哈尔滨工程大学出版社,1998.
[70]边肇祺,张学工.模式识别（第二版）[M].北京:清华大学出版社,2000.
[71] Suvorova S, Schroeder J. Automated target recognition using the Karhunen–Loéve transform with invariance[J]. Digital Signal Processing,2002,12(2-3):295-306.
[72]王世晞,贺志国.基于PCA特征的快速SAR图像目标识别方法[J].国防科技大学学报,2007,30(3):136-140.
[73] Mishra A K. Validation of PCA and LDA for SAR ATR[C]. IEEE Region10Conference, Hyderabad,2008:1-6.
[74]韩萍,吴仁彪,王兆华.基于KPCA准则的SAR图像目标特征提取与识别[J].电子与信息学报,2003,25(10):1297-1301.
[75]成功,赵巍,毛士艺.基于快速提升KLDA准则的MSTAR SAR目标特征提取与识别研究[J].航空学报,2007,28(3):667-672.
[76]汪玲.逆合成孔径雷达成像关键技术研究[D].南京:南京航空航天大学,2006.
[77]庄钊文,袁乃昌.雷达散射截面测量一紧凑场理论与技术[M].长沙:国防科技大学出版社,2000.
[78]王胜廷.高分辨率SAR仿真图像生成技术及其在目标识别中的应用[D].长沙:国防科学技术大学,2011.
[79]王乘,周均清,李利军. creator可视化仿真建模技术[M].武汉:华中科技大学出版社,2000.
[80] Blasband C, Jafolla J. RadBaseTM: A Physics-Based Radar Database GenerationToolkit. http://www.surfaceoptics.com/data/pdf/RadBase2_tech_paper.PDF.2001.
[81]孙即祥.数字图像处理[M].石家庄:河北教育出版社,1993.
[82]王润生.图像理解[M].长沙:国防科技大学出版社,1995.
[83]吴京.信号与系统分析[M].长沙:国防科技大学出版社,1999.
[84] Zhang X G, Bai Y C. ISAR Target2D Images Pre-processing in ATR[C].2009World Congress on Computer Science and Information Engineering, Los Angeles,2009:542-544.
[85]计科峰. SAR图像目标特征提取与分类方法研究[D].长沙:国防科学技术大学,2003.
[86] Wehner D R. High Resolution Radar,2nd ed[M]. Norwood, Massachusetts: Ar-tech House,1995.
[87]高贵,何娟,匡纲要等. SAR图像目标方位角估计方法综述[J].信号处理,2008,24(3):438-443.
[88]辛宁,王国宏,张静.改进的SAR图像车辆目标姿态角估计方法[J].火力与指挥控制,2008,33(5):57-60.
[89]程鹏飞,闫浩文,韩振辉.一个求解多边形最小面积外接矩形的算法[J].工程图学学报,2008,29(1):122-126.
[90]孙广家.计算机图形学(第三版)[M].北京:清华大学出版社,1998.
[91] Chiang H C. Feature based classification with Application to Synthetic ApertureRadar[D]. Columbus: The Ohio State University,1999.
[92]孙季丰,方少元.几何型哈什法和图像的快速合成[J].华南理工大学学报,1998,26(2):75-80.
[93] Wolfson H J, Rigoutsos I. Geometric hashing: an overview[J]. IEEE Computa-tional Science and Engineering,1997,4(4):10-21.
[94] Lamdan Y, Wolfson H J. Geometric hashing:A general and efficient model-basedcognition scheme[C].2nd IEEE International Conference on Comput Vision,Tamp FL,1988:238-249.
[95] Lamdan Y, Schualtz J T, Wolfson H J. Affine invariant model-based objectrecognition[J]. IEEE Trans on Robotics and Automation,1990,6(5):576-587.
[96] Lamdan Y, Schualtz J T, Wolfson H J. Object recognition by Affine invariantmatching[C]. Proc IEEE Conf Comput Vision Pattern, Recog., Ann Anbor, MI,1988:335-344.
[97] Chum O, Perdoch M, Matas J. Geometric min-Hashing Finding a (thick) needle ina haystack[C]. Proc.IEEE Conf.on Computer Vision and Pattern Recognition,Miami, FL, USA,2009:10-21.
[98] Zhe J, Teoh A B J, Thian S O, et al. Secure minutiae-based fingerprint templatesusing random triangle hashing[C]. Proceedings of the1st International Visual In-formatics Conference on Visual Informatics: Bridging Research and Practice,Kuala Lumpur, Malaysia,2009:521-531.
[99] He Z, Lu J, Kuang G. A fast SAR target recognition approach using PCA fea-tures[C]. Proceeding of the Fourth International Conference on Image andGraphics, Sichuan,2007:580-585.
[100]Huang C W, Lee K C. Application of ica technique to PCA based radar targetrecognition[J]. Progress In Electromagnetics Research,2010,105:157-170.
[101]Terry C. Visual perception, theory and practice[M]. New York: Pergamon,1981.
[102]Marr D. Vision: A Computational Investigation into the Human Representationand Processing of Visual Information[M]. New York: Freeman,1982.
[103]Lee D D, Seung H S. Learning the parts of objects by non-negative matrix factor-ization[J]. Nature,1999,401(21):788-792.
[104]Guillametd D, Vitria J. Non-negative matrix factorization for face recognition[C].Proceeding of the5th Catalonian Conference on AI, Berlin,2002:336-344.
[105]Karvonen J. Non-negative matrix factorization with sparseness constraints for seaice SAR feature extraction[C].7th European Conference on Synthetic ApertureRadar, Friedrichshafen,2008:1-4.
[106]龙泓琳,皮亦鸣,曹宗杰.基于非负矩阵分解的SAR图像目标识别[J].电子学报,2010,38(6):1-5.
[107]Li S Z, Hou X, Zhang H, et al. Learning spatially localized, part-based representa-tion[J]. Computer Vision and Pattern Recognition,2001,1(8):207-212.
[108]Hoyer P O. Non-negative Sparse Coding[J]. Neural Networks for Signal Pro-cessing,2002:557-565.
[109]Shastri B J, Levine M D. Face recognition using localized features based onnon-negative sparse coding[J]. Machine Vision And Applications,2007,18:107-122.
[110]陈霖.知觉过程是从哪里开始的?[C].中国科学院第十二次院士大会学术报告,2004.
[111]刘维湘,郑南宁,游屈波.非负矩阵分解及其在模式识别中的应用[J].科学通报,2006,51(3):241-250.
[112]郑南宁.认知与计算机视觉[C].国家杰出青年科学基金实施十周年学术报告,2004.
[113]Buciu I, Pitas I. NMF, LNMF, and DNMF modeling of neural receptive fields in-volved in human facial expression perception[J]. J. Vis. Commun. Image Rep.,2006,17(5):958-969.
[114]Hoyer P O. Modeling receptive fields with non-negative sparse coding[J]. Com-putational Neuroscience,2003,52/54(0):547-552.
[115]Knight B W, Omurtag A, Sirovich L. The approach of a neuron population firingrate to a new equilibrium: an exact theoretical result[J]. Neural Computation,2000,12(5):1045-1055.
[116]Guillamet D, Vitria J Chiele B. Introducing a weighted non-negative matrix fac-torization for image classification[J]. Pattern Recognition Letters,2003,24(14):2447-2454.
[117]罗四维.视觉信息认知计算理论[M].北京:科学出版社,2010.
[118]Paatero P, Tapper U. Positive matrix factorization: A non-negative factor modelwith optimal utilization of error estimates of data values[J]. Environmetrics,1994,5:111-126.
[119]Lee D D, Seung H S. Algorithms for non-negative matrix factorization[J]. Ad-vances Neural Information Processing Systems,2001,13:556-562.
[120]Sauer T, Schroth A. Robust range alignment algorithm via hough transform in anISAR imaging system[J]. Ieee Transactions On Aerospace And Electronic Sys-tems,1995,31:1173-1177.
[121]Olshausen B A, Field D J. Emergency of simple-cell receptive-field properties bylearning a sparse code for natural images[J]. Nature,1996,381:607-609.
[122]Olshausen B A, Field D J. Sparse coding with an overcomplete basis set：a strat-egy employed by V1[J]. Vision Research,1997,37(23):3311-3325.
[123]Yang J C, Yu K, Gong Y H, et al. Linear Spatial Pyramid Matching Using SparseCoding for Image Classification[J]. Computer Vision and Pattern Recognition,2009:1794-1801.
[124]Sivaram G S V S, Nemala S K, Elhilali M, et al. Sparse Coding for SpeechRecognition[J]. Acoustics Speech and Signal Processing,2010:4346-4349.
[125]Ghanem B, Ahuja N. Sparse Coding of Linear Dynamical Systems with an Appli-cation to Dynamic Texture Recognition[C].20th International Conference on Pat-tern Recognition,2010:987-990.
[126]Hoyer P O. Non-negative Matrix Factorization with Sparseness Constraints[J].Journal Of Machine Learning Research,2004,5:1457-1469.
[127]Chen B, Liu H, Bao Z. A kernel optimization method based on the localized ker-nel fisher criterion[J]. Pattern Recognition,2008,41(3):1098-1109.
[128]Kim T K, Kittler J. Locally linear discriminant analysis for multimodally distrib-uted classes for face recognition with a single model image[J]. IEEE Trans. Pat-tern Analysis and Machine Intelligence,2005,27(3):318-327.
[129]Fukunaga K. Introduction to statistical pattern recognition[M]. Boston: AcademicPress Inc.,1990.
[130]Chen X, Huang T. Facial Expression Recognition: a Clustering-Based Ap-proach[J]. Pattern Recognition Letters,2003,24(9-10):1295-1302.
[131]Dempster A P, Laird N M, Rubin D B. Maximum Likelihood from IncompleteData via the EM Algorithm[J]. Royal Stat. Soc.,2001,39:1-38.
[132]Likas A, Vlassis N, Verbeek J J. The global k-means clustering algorithm[J]. Pat-tern Recognition,2003,36:451-461.
[133]Thomas M, Jason C, Robert H, et al. An analysis of fully polarimetric W-BandISAR imagery on seven scale model main battle tanks for use in target recogni-tion[C].2002:17-26.
[134]Lu H, Fainman Y, Robert H. Image manifolds[C]. Proc.SPIE,1998:52-63.
[135]Seung H S, Lee D D. The manifold ways of perception[J]. Science,2000,290(5500):2268-2269.
[136]Belkin M, Niyogi P. Laplacian eigenmaps and spectral techniques for embeddingand clustering[J]. Adv Neural Inform Proc Syst,2002,14:585-591.
[137]Roweis S T, Saul L K. Nonlinear dimensionality reduction by locally linear em-bedding[J]. Science,2000,290(5500):2323-2326.
[138]Tenenbaum J B, Silva V D, Langford J C. A global geometric framework for non-linear dimensionality reduction[J]. Science,2000,290(5500):2319-2323.
[139]He X F, Niyogi P. Locality preserving projections[J]. Advan Neural Inform ProcSyst.,2004,16:153-160.
[140]He X F, Cai D, Yan S. neighborhood preserving embedding[C]. Proceedings ofthe10th IEEE international conference on computer vision, Beijing,China,2005:1208-1213.
[141]Sugiyama M. Dimensionality reduction of multimodal labeled data by local fisherdiscriminant analysis[J]. Journal Of Machine Learning Research,2007,8:1027-1061.
[142]Yan S, Xu D, Zhang B. Graph embedding and extensions：A general frameworkfor dimensionality reduction[J]. IEEE transaction on pattern analysis and machineintelligence,2007,29(1):40-51.
[143]罗磊.基于流形学习的毫米波探测器目标识别方法研究[D].南京理工大学,2010.
[144]Belkin M, Niyogi P. Laplacian eigenmaps for dimensionality reduction and datarepresentation[J]. Neural Computation,2003,15(6):1373-1396.
[145]Hu H F. Orthogonal neighborhood preserving discriminant analysis for facerecognition[J]. Pattern Recognition,2008,41(6):2045-2054.
[146]Zhu M, Martinez A M. Subclass discriminant analysis[J]. IEEE transaction onpattern analysis and machine intelligence,2006,28(8):1274-1286.
[147]谢钧,刘剑.一种新的局部判别投影方法[J].计算机学报,2011,34(11):2243-2250.
[148]He X, Yan S, Hu Y. Face recognition using Laplacianfaces[J]. IEEE transactionon pattern analysis and machine intelligence,2005,27(3):328-340.
[149]Fukunaga K. introduction to statistical pattern recognition[M]. boston: academicpress,1990.
[150]Yang M H. kernel eigenfaces vs. kernel fisherfaces：face recognition using kernelmethods[C]. Proceedings of the fifth IEEE international conference on automaticface and gesture recognition, Washington, DC, USA,2002:215-220.
[151]祝磊,朱善安. KSLPP:新的人脸识别算法[J].浙江大学学报(工学版),2007,41(7):1066-1069.
[152]庞彦伟,俞能海,沈道义.基于核邻域保持投影的人脸识别[J].电子学报,2006,34(8):1542-1544.
[153]Chen B, Liu H, Bao Z. Speeding up SVM in Test Phase：Application to RadarHRRP ATR[C]. The Proc．of ICONIP06, Hongkong, China,2006.
[154]John S T, Nello C. Kernel methods for pattern analysis[M]. Cambridge UniversityPress,2004.
[155]Vapnik V N. the nature of statistical learning theory[M]. new york: springerverlag,1995.
[156]陈渤.基于核方法的雷达高分辨距离像目标识别技术研究[D].西安:西安电子科技大学,2008.
[157]Liu S, Ruan Q. orthogonal tensor neighborhood preserving embedding for facialexpression recognition[J]. Pattern Recognition,44(7):1497-1513.
[158]Zheng D, Du X, Cui L. tensor locality preserving projections for face recogni-tion[C]. IEEE international conference on systems man and cybernetics(SMC),Istanbul,2010:2347-2350.
[159]He X, Cai D, Niyogi P. Tensor subspace analysis[C]. Proceeding of the Confer-ence on Advances in the Neural Information Processing Systems, Vancouver,Canada,2005:249-254.
[160]Ye J P, Janardan R, Li Q. Two-dimensional linear discriminant analysis[C]. Proc.of Advances in Neural Information Processing Systems, Vancouver, Canada,2004:1-7.
[161]王兵,黄钰林,杨建宇等.基于最大异类距离特征提取的SAR目标识别方法[J].中国科学,2011,41(10):1388-1392.
[2]孙文峰.雷达目标识别技术述评[J].雷达与对抗,2001(3):1-8.
[3] Bhanu B. Automatic target recognition：State of the art survey[J]. IEEE Trans．onAES,1986,22(4):364-379.
[4]刘宏伟,杜兰,袁莉等.雷达高分辨距离像目标识别研究进展[J].电子与信息学报,2005,27(8):1328-1334.
[5]郭桂蓉,庄钊文,陈曾平.电磁特征抽取与目标识别[M].长沙:国防科技大学出版社,1996.
[6]黄培康,殷红成,许小剑.雷达目标特性[M].北京:电子工业出版社,2005.
[7]保铮,邢孟道,王彤.雷达成像技术[M].北京:电子工业出版社,2005.
[8]刘永坦.雷达成像技术[M].哈尔滨:哈尔滨工业大学出版社,1999.
[9] Hovanessian S A. Introduction to synthetic array and imaging radars[M]. ArtechHouse,1980.
[10] Brown W M. Synthetic aperture radar[J]. IEEE Trans.AES,1967,3(2):217-299.
[11] Wiley C A. Synthetic aperture radars[J]. IEEE Trans. On AES,1985,21(3):440-443.
[12] Curlander J C, Mcdonough R N. Synthetic aperture radar systems and signal pro-cessing[M]. John Wiley&Sons,1991.
[13]张直中.合成孔径逆合成孔径和成像雷达[J].现代雷达,1985,7(56).
[14] Chen C C, Andrews H C. Target-motion-induced Radar Imaging[J]. IEEETrans.AES,1967,3(2):217-299.
[15] Prickett M J, Chen C C. Principle of inverse synthetic aperture radar[J]. EASCONrecord,1980:340-345.
[16]张直中.微波成像技术[M].北京:科学出版社,1990.
[17]张澄波.综合孔径雷达[M].北京:科学出版社,1989.
[18] Brown W M. Range-Doppler imaging with motion through resolution cells[J].IEEE Trans.AES,1969,5(1):98-102.
[19] Chen C C, Andrews H C. Target-motion-induced radar imaging[J]. IEEE Trans.AES,1980,16(1):2-14.
[20]曹志道,刘永坦,宿富林.实验ISAR外场实验及数据处理[C].信息获取与处理技术主题十周年汇报,1996:56-64.
[21]保铮,邓文彪,杨军.散射重心跟踪的ISAR运动补偿方法[J].西安电子科技大学学报,1990,17(3):1-8.
[22]朱兆达,邱晓辉,佘志舜.用改进的多普勒中心跟踪法进行ISAR运动补偿[J].电子学报,1997,25(3):65-69.
[23]邱晓辉.运动目标雷达成像研究[D].南京航空航天大学,1995.
[24] Kennaugh E M, Moffatt D L. Transient and Impulse Response Approximations[J].Proc. IEEE,1965,53(8):893-901.
[25] Baum C E. On the Singularity Expansion Method for the Solution of Electromag-netic Interaction Problems[J]. AFWL Interaction Notes,1971.
[26] Wicke P E Z, Kiss I. A new implementations of Mellin Transform and its applica-tion to radar classification of ships[J]. IEEE trans on pattern analysis and machineintelligence,1983,5(2):598-606.
[27] Zyweck A, Bogner R E. Radar Target Classification of Commercial Aircraft[J]. IEEE Trans. AES,1996,32(2):598-606.
[28] Gabel R A. machine intelligence and radar image understanding[C]. In proceed-ings of the twenty-first asilomar conference on signals, systems, and computers,pacific grove，CA,1987:775-879.
[29] Jaenicke R A. structure from limited motion of complex rigid objects[C]. In pro-ceedings of IEEE workshop on motion, Irvine, CA,1989:256-263.
[30] Paul F H. radar image understanding for complex space objects[C]. Proc.SPIE1381, Boston MA,1991:55-65.
[31] Hu Y K, Yang Y M. Automatic Target Recognition of ISAR Images Based onHausdorff Distance[C].1st Asian and Pacific Conference on Synthetic ApertureRadar, Huangshan,2007:477-479.
[32] Saidi M N, Daoudi K, Khenchaf A. Automatic target recognition of aircraft mod-els based on ISAR images[C]. Geoscience and Remote Sensing Symposium, CapeTown,2009:685-688.
[33] Saidi M N, Toumi A, Khenchaf A, et al. Pose estimation for ISAR image classifi-cation[C]. In Proceedings of IGARSS, Brest, France,2010:4620-4623.
[34] Saidi M N, Hoeltzener B, Toumi A, et al. Automatic recognition of ISAR Images:Target shapes features extraction[C]. ICTTA3rd International Conference, Da-mascus,2008:1-6.
[35] Maki A, Fukui K, Onoguchi K, et al. ISAR Image Analysis by Subspace Method:Automatic Extraction and Identification of Ship Profile[C].11th internationalconference on image analysis and processing, Palermo,2001:523-528.
[36]张兴敢.逆合成孔径雷达成像及目标识别[D].南京:南京航空航天大学,2001.
[37]高兴斌,刘永坦. ISAR目标像的特征提取和特征选择[J].哈尔滨工业大学学报,1994,26(5):77281.
[38]张善文,甄蜀春,郑全弟等.雷达目标特征提取的一种方法[J].电子与信息学报,2000,24(12):1994-1997.
[39]范春彦,李晓曼,付红卫等. ISAR成像的处理与分类方法[J].电光与控制,2003,10(2):70-72.
[40]马君国,赵宏钟,李保国.基于二维小波变换的空间目标识别算法[J].国防科技大学学报,2006,28(1):57-61.
[41]马君国.空间雷达目标特征提取与识别方法研究[D].长沙:国防科学技术大学,2006.
[42]许人灿,刘朝军,黄小红等.基于超分辨ISAR成像的空中目标自动识别[J].系统工程与电子技术,2006,28(1):46-48.
[43] Botha E C. Classification of aerospace targets using superresolution ISAR imag-es[C]. Communications and Singnal Processing Proceeding, Stellenbosch,1994:138-145.
[44]季战领,冯晓毅,阎昆.一种ISAR图像中飞机目标识别的新特征[J].计算机工程,2010,36(23):192-193,196.
[45]李大芳.基于ISAR图像目标识别算法的研究[D].哈尔滨:哈尔滨工业大学,2004.
[46]陈善丽.逆合成孔径雷达图像识别研究[D].杭州:浙江大学,2008.
[47] Rosenbach K, Schiller J. Identification of aircraft on the basis of2-D radar imag-es[C]. IEEE international Radar Conference, Alexandria, VA,1995:405-409.
[48] Kim K, Seo D, Kim H. Efficient Classification of ISAR Images[J]. IEEE Trans.Antennas and Propagation,2005,53(5):1611-1621.
[49] Musman S, Kerr D, Bachmann C. Automatic recognition of ISAR ship images[J].IEEE Trans. Aerosp. Electron. Syst.,1996,32(4):1392-1404.
[50] Pastina D, Spina C. Multi-feature based automatic recognition of ship targets inISAR[J]. IET Radar Sonar Navig.,2009,3(4):406-423.
[51] Zeljkovi V, Li Q, Vincelette R, et al. Automatic algorithm for inverse syntheticaperture radar images recognition and classification [J]. IET Radar Sonar Navig.,2010,1(4):96-109.
[52] Zeljkovi V, Li Q, Vincelette R. Noise resistant algorithm for radar imagesrecognition and classification[C]. Signal and Data Processing of Small Targets,SPIE, San Diego,2009:744507.
[53] Zeljkovi V, Li Q, Vincelette R. Aircraft Identification by Unions of ISAR Imag-es[C]. Proceedings of the5th International ICST Mobile Multimedia Communica-tions Conference, Brussels,2009.
[54]崔延鹏,胡建伟,杨绍全等.利用尺度不变量特征的ISAR二维像自动识别技术[J].西安电子科技大学学报,2009,36(4):725-728.
[55]刘烽,许家栋.基于ISAR图像序列的目标识别方法研究[C].第九届全国雷达学术年会,烟台,2004:580-583.
[56] Chia T, Tien S, Lu Y. recognize aircraft in ISAR images[J]. Journal Of Infor-mation Science And Engineering,2007(23):299-313.
[57] Maskall G T, Webb A R. Nonlinear feature extraction applied to ISAR images oftargets for classification[C]. the international society for optical engineering pro-cessings of SPIE,2001:255-265.
[58] Maskall G T. An application of nonlinear feature extraction to the classification ofISAR images[C]. RADAR2002, UK:405-408.
[59] Maskall G T, Webb A R. Nonlinear feature extraction for MMW image classifica-tion: a supervised approach[C]. Proc.SPIE,2002:353-363.
[60]何强,蔡洪,韩壮志等.基于非线性流形学习的ISAR目标识别研究[J].电子学报,2010,38(3):585-590.
[61]蔡洪,何强,韩壮志等.空间平滑局部保持投影用于ISAR目标识别[J].光电工程,2009,36(7):8-13.
[62]蔡洪,何强,韩壮志等.基于局部保持判别子空间的ISAR目标识别[J].中国电子科学研究院学报,2009,4(6):656-660.
[63]朱兆达,殷军,邬小青.基于信号时间-频率分析的多运动目标ISAR成像[J].现代雷达,1993,34(5):33-37.
[64] Soumekh M, Nugroho S. ISAR imaging of an airborne DC-9[C]. IEEE interna-tional conference on acoustic,speech and signal processing,1993:465-468.
[65] Walker J L. Range Doppler imaging of rotating objects[J]. Ieee Transactions OnAerospace And Electronic Systems,1980,16(1):23-52.
[66] Chen C C. Multi-frequency imaging of radar turntable data[J]. IEEE Transactionson Aerospace and Electronic Systems,1980,16(1):15-22.
[67]朱自强,陈迎春,王晓璐等.现代飞机的空气动力设计[M].北京:国防工业出版社,2011.
[68]飞机气动布局简介. http://wenku.baidu.com/view/6c4e2ed3360cba1aa811da43.html.
[69]黄凤岗,宋克欧.模式识别[M].哈尔滨:哈尔滨工程大学出版社,1998.
[70]边肇祺,张学工.模式识别（第二版）[M].北京:清华大学出版社,2000.
[71] Suvorova S, Schroeder J. Automated target recognition using the Karhunen–Loéve transform with invariance[J]. Digital Signal Processing,2002,12(2-3):295-306.
[72]王世晞,贺志国.基于PCA特征的快速SAR图像目标识别方法[J].国防科技大学学报,2007,30(3):136-140.
[73] Mishra A K. Validation of PCA and LDA for SAR ATR[C]. IEEE Region10Conference, Hyderabad,2008:1-6.
[74]韩萍,吴仁彪,王兆华.基于KPCA准则的SAR图像目标特征提取与识别[J].电子与信息学报,2003,25(10):1297-1301.
[75]成功,赵巍,毛士艺.基于快速提升KLDA准则的MSTAR SAR目标特征提取与识别研究[J].航空学报,2007,28(3):667-672.
[76]汪玲.逆合成孔径雷达成像关键技术研究[D].南京:南京航空航天大学,2006.
[77]庄钊文,袁乃昌.雷达散射截面测量一紧凑场理论与技术[M].长沙:国防科技大学出版社,2000.
[78]王胜廷.高分辨率SAR仿真图像生成技术及其在目标识别中的应用[D].长沙:国防科学技术大学,2011.
[79]王乘,周均清,李利军. creator可视化仿真建模技术[M].武汉:华中科技大学出版社,2000.
[80] Blasband C, Jafolla J. RadBaseTM: A Physics-Based Radar Database GenerationToolkit. http://www.surfaceoptics.com/data/pdf/RadBase2_tech_paper.PDF.2001.
[81]孙即祥.数字图像处理[M].石家庄:河北教育出版社,1993.
[82]王润生.图像理解[M].长沙:国防科技大学出版社,1995.
[83]吴京.信号与系统分析[M].长沙:国防科技大学出版社,1999.
[84] Zhang X G, Bai Y C. ISAR Target2D Images Pre-processing in ATR[C].2009World Congress on Computer Science and Information Engineering, Los Angeles,2009:542-544.
[85]计科峰. SAR图像目标特征提取与分类方法研究[D].长沙:国防科学技术大学,2003.
[86] Wehner D R. High Resolution Radar,2nd ed[M]. Norwood, Massachusetts: Ar-tech House,1995.
[87]高贵,何娟,匡纲要等. SAR图像目标方位角估计方法综述[J].信号处理,2008,24(3):438-443.
[88]辛宁,王国宏,张静.改进的SAR图像车辆目标姿态角估计方法[J].火力与指挥控制,2008,33(5):57-60.
[89]程鹏飞,闫浩文,韩振辉.一个求解多边形最小面积外接矩形的算法[J].工程图学学报,2008,29(1):122-126.
[90]孙广家.计算机图形学(第三版)[M].北京:清华大学出版社,1998.
[91] Chiang H C. Feature based classification with Application to Synthetic ApertureRadar[D]. Columbus: The Ohio State University,1999.
[92]孙季丰,方少元.几何型哈什法和图像的快速合成[J].华南理工大学学报,1998,26(2):75-80.
[93] Wolfson H J, Rigoutsos I. Geometric hashing: an overview[J]. IEEE Computa-tional Science and Engineering,1997,4(4):10-21.
[94] Lamdan Y, Wolfson H J. Geometric hashing:A general and efficient model-basedcognition scheme[C].2nd IEEE International Conference on Comput Vision,Tamp FL,1988:238-249.
[95] Lamdan Y, Schualtz J T, Wolfson H J. Affine invariant model-based objectrecognition[J]. IEEE Trans on Robotics and Automation,1990,6(5):576-587.
[96] Lamdan Y, Schualtz J T, Wolfson H J. Object recognition by Affine invariantmatching[C]. Proc IEEE Conf Comput Vision Pattern, Recog., Ann Anbor, MI,1988:335-344.
[97] Chum O, Perdoch M, Matas J. Geometric min-Hashing Finding a (thick) needle ina haystack[C]. Proc.IEEE Conf.on Computer Vision and Pattern Recognition,Miami, FL, USA,2009:10-21.
[98] Zhe J, Teoh A B J, Thian S O, et al. Secure minutiae-based fingerprint templatesusing random triangle hashing[C]. Proceedings of the1st International Visual In-formatics Conference on Visual Informatics: Bridging Research and Practice,Kuala Lumpur, Malaysia,2009:521-531.
[99] He Z, Lu J, Kuang G. A fast SAR target recognition approach using PCA fea-tures[C]. Proceeding of the Fourth International Conference on Image andGraphics, Sichuan,2007:580-585.
[100]Huang C W, Lee K C. Application of ica technique to PCA based radar targetrecognition[J]. Progress In Electromagnetics Research,2010,105:157-170.
[101]Terry C. Visual perception, theory and practice[M]. New York: Pergamon,1981.
[102]Marr D. Vision: A Computational Investigation into the Human Representationand Processing of Visual Information[M]. New York: Freeman,1982.
[103]Lee D D, Seung H S. Learning the parts of objects by non-negative matrix factor-ization[J]. Nature,1999,401(21):788-792.
[104]Guillametd D, Vitria J. Non-negative matrix factorization for face recognition[C].Proceeding of the5th Catalonian Conference on AI, Berlin,2002:336-344.
[105]Karvonen J. Non-negative matrix factorization with sparseness constraints for seaice SAR feature extraction[C].7th European Conference on Synthetic ApertureRadar, Friedrichshafen,2008:1-4.
[106]龙泓琳,皮亦鸣,曹宗杰.基于非负矩阵分解的SAR图像目标识别[J].电子学报,2010,38(6):1-5.
[107]Li S Z, Hou X, Zhang H, et al. Learning spatially localized, part-based representa-tion[J]. Computer Vision and Pattern Recognition,2001,1(8):207-212.
[108]Hoyer P O. Non-negative Sparse Coding[J]. Neural Networks for Signal Pro-cessing,2002:557-565.
[109]Shastri B J, Levine M D. Face recognition using localized features based onnon-negative sparse coding[J]. Machine Vision And Applications,2007,18:107-122.
[110]陈霖.知觉过程是从哪里开始的?[C].中国科学院第十二次院士大会学术报告,2004.
[111]刘维湘,郑南宁,游屈波.非负矩阵分解及其在模式识别中的应用[J].科学通报,2006,51(3):241-250.
[112]郑南宁.认知与计算机视觉[C].国家杰出青年科学基金实施十周年学术报告,2004.
[113]Buciu I, Pitas I. NMF, LNMF, and DNMF modeling of neural receptive fields in-volved in human facial expression perception[J]. J. Vis. Commun. Image Rep.,2006,17(5):958-969.
[114]Hoyer P O. Modeling receptive fields with non-negative sparse coding[J]. Com-putational Neuroscience,2003,52/54(0):547-552.
[115]Knight B W, Omurtag A, Sirovich L. The approach of a neuron population firingrate to a new equilibrium: an exact theoretical result[J]. Neural Computation,2000,12(5):1045-1055.
[116]Guillamet D, Vitria J Chiele B. Introducing a weighted non-negative matrix fac-torization for image classification[J]. Pattern Recognition Letters,2003,24(14):2447-2454.
[117]罗四维.视觉信息认知计算理论[M].北京:科学出版社,2010.
[118]Paatero P, Tapper U. Positive matrix factorization: A non-negative factor modelwith optimal utilization of error estimates of data values[J]. Environmetrics,1994,5:111-126.
[119]Lee D D, Seung H S. Algorithms for non-negative matrix factorization[J]. Ad-vances Neural Information Processing Systems,2001,13:556-562.
[120]Sauer T, Schroth A. Robust range alignment algorithm via hough transform in anISAR imaging system[J]. Ieee Transactions On Aerospace And Electronic Sys-tems,1995,31:1173-1177.
[121]Olshausen B A, Field D J. Emergency of simple-cell receptive-field properties bylearning a sparse code for natural images[J]. Nature,1996,381:607-609.
[122]Olshausen B A, Field D J. Sparse coding with an overcomplete basis set：a strat-egy employed by V1[J]. Vision Research,1997,37(23):3311-3325.
[123]Yang J C, Yu K, Gong Y H, et al. Linear Spatial Pyramid Matching Using SparseCoding for Image Classification[J]. Computer Vision and Pattern Recognition,2009:1794-1801.
[124]Sivaram G S V S, Nemala S K, Elhilali M, et al. Sparse Coding for SpeechRecognition[J]. Acoustics Speech and Signal Processing,2010:4346-4349.
[125]Ghanem B, Ahuja N. Sparse Coding of Linear Dynamical Systems with an Appli-cation to Dynamic Texture Recognition[C].20th International Conference on Pat-tern Recognition,2010:987-990.
[126]Hoyer P O. Non-negative Matrix Factorization with Sparseness Constraints[J].Journal Of Machine Learning Research,2004,5:1457-1469.
[127]Chen B, Liu H, Bao Z. A kernel optimization method based on the localized ker-nel fisher criterion[J]. Pattern Recognition,2008,41(3):1098-1109.
[128]Kim T K, Kittler J. Locally linear discriminant analysis for multimodally distrib-uted classes for face recognition with a single model image[J]. IEEE Trans. Pat-tern Analysis and Machine Intelligence,2005,27(3):318-327.
[129]Fukunaga K. Introduction to statistical pattern recognition[M]. Boston: AcademicPress Inc.,1990.
[130]Chen X, Huang T. Facial Expression Recognition: a Clustering-Based Ap-proach[J]. Pattern Recognition Letters,2003,24(9-10):1295-1302.
[131]Dempster A P, Laird N M, Rubin D B. Maximum Likelihood from IncompleteData via the EM Algorithm[J]. Royal Stat. Soc.,2001,39:1-38.
[132]Likas A, Vlassis N, Verbeek J J. The global k-means clustering algorithm[J]. Pat-tern Recognition,2003,36:451-461.
[133]Thomas M, Jason C, Robert H, et al. An analysis of fully polarimetric W-BandISAR imagery on seven scale model main battle tanks for use in target recogni-tion[C].2002:17-26.
[134]Lu H, Fainman Y, Robert H. Image manifolds[C]. Proc.SPIE,1998:52-63.
[135]Seung H S, Lee D D. The manifold ways of perception[J]. Science,2000,290(5500):2268-2269.
[136]Belkin M, Niyogi P. Laplacian eigenmaps and spectral techniques for embeddingand clustering[J]. Adv Neural Inform Proc Syst,2002,14:585-591.
[137]Roweis S T, Saul L K. Nonlinear dimensionality reduction by locally linear em-bedding[J]. Science,2000,290(5500):2323-2326.
[138]Tenenbaum J B, Silva V D, Langford J C. A global geometric framework for non-linear dimensionality reduction[J]. Science,2000,290(5500):2319-2323.
[139]He X F, Niyogi P. Locality preserving projections[J]. Advan Neural Inform ProcSyst.,2004,16:153-160.
[140]He X F, Cai D, Yan S. neighborhood preserving embedding[C]. Proceedings ofthe10th IEEE international conference on computer vision, Beijing,China,2005:1208-1213.
[141]Sugiyama M. Dimensionality reduction of multimodal labeled data by local fisherdiscriminant analysis[J]. Journal Of Machine Learning Research,2007,8:1027-1061.
[142]Yan S, Xu D, Zhang B. Graph embedding and extensions：A general frameworkfor dimensionality reduction[J]. IEEE transaction on pattern analysis and machineintelligence,2007,29(1):40-51.
[143]罗磊.基于流形学习的毫米波探测器目标识别方法研究[D].南京理工大学,2010.
[144]Belkin M, Niyogi P. Laplacian eigenmaps for dimensionality reduction and datarepresentation[J]. Neural Computation,2003,15(6):1373-1396.
[145]Hu H F. Orthogonal neighborhood preserving discriminant analysis for facerecognition[J]. Pattern Recognition,2008,41(6):2045-2054.
[146]Zhu M, Martinez A M. Subclass discriminant analysis[J]. IEEE transaction onpattern analysis and machine intelligence,2006,28(8):1274-1286.
[147]谢钧,刘剑.一种新的局部判别投影方法[J].计算机学报,2011,34(11):2243-2250.
[148]He X, Yan S, Hu Y. Face recognition using Laplacianfaces[J]. IEEE transactionon pattern analysis and machine intelligence,2005,27(3):328-340.
[149]Fukunaga K. introduction to statistical pattern recognition[M]. boston: academicpress,1990.
[150]Yang M H. kernel eigenfaces vs. kernel fisherfaces：face recognition using kernelmethods[C]. Proceedings of the fifth IEEE international conference on automaticface and gesture recognition, Washington, DC, USA,2002:215-220.
[151]祝磊,朱善安. KSLPP:新的人脸识别算法[J].浙江大学学报(工学版),2007,41(7):1066-1069.
[152]庞彦伟,俞能海,沈道义.基于核邻域保持投影的人脸识别[J].电子学报,2006,34(8):1542-1544.
[153]Chen B, Liu H, Bao Z. Speeding up SVM in Test Phase：Application to RadarHRRP ATR[C]. The Proc．of ICONIP06, Hongkong, China,2006.
[154]John S T, Nello C. Kernel methods for pattern analysis[M]. Cambridge UniversityPress,2004.
[155]Vapnik V N. the nature of statistical learning theory[M]. new york: springerverlag,1995.
[156]陈渤.基于核方法的雷达高分辨距离像目标识别技术研究[D].西安:西安电子科技大学,2008.
[157]Liu S, Ruan Q. orthogonal tensor neighborhood preserving embedding for facialexpression recognition[J]. Pattern Recognition,44(7):1497-1513.
[158]Zheng D, Du X, Cui L. tensor locality preserving projections for face recogni-tion[C]. IEEE international conference on systems man and cybernetics(SMC),Istanbul,2010:2347-2350.
[159]He X, Cai D, Niyogi P. Tensor subspace analysis[C]. Proceeding of the Confer-ence on Advances in the Neural Information Processing Systems, Vancouver,Canada,2005:249-254.
[160]Ye J P, Janardan R, Li Q. Two-dimensional linear discriminant analysis[C]. Proc.of Advances in Neural Information Processing Systems, Vancouver, Canada,2004:1-7.
[161]王兵,黄钰林,杨建宇等.基于最大异类距离特征提取的SAR目标识别方法[J].中国科学,2011,41(10):1388-1392.