散射中心分布特征提取与核方法分类器关键技术研究
|
摘要
本文以“973”国家安全重大基础研究项目“雷达自动目标识别新机理、新方法研究”为背景,着重从特征提取和分类器设计两个方面研究高分辨雷达目标分类识别的理论与关键技术,主要涉及以下三方面内容:基于散射中心参数估计的分布特征提取、高分辨雷达目标分类中支持向量机参数优化和支持向量数据描述单类分类器向多类分类推广。
(1)基于散射中心参数估计的分布特征提取方面
针对传统估计方法分辨力低、估计的散射点位置特征无法直接用于目标分类的问题,提出了一种基于散射中心径向相对位置估计的分布特征提取方法,该方法首先通过回波预测目标散射中心个数,然后对散射中心的幅度和位置信息进行高分辨估计,并依据估计结果提取散射中心间相对距离信息组成特征向量。经仿真实验验证,所提取的分布特征与传统方法中位置特征相比更稳定,并在一定角度范围内具有旋转和平移不变性,最后基于该特征提取方法实现了对三类目标的有效识别,并与已有方法进行比较,验证了该方法的有效性和稳健性。
(2)高分辨雷达目标分类中支持向量机分类器参数优化方面
针对以往基于遗传算法支持向量机参数优化中适值函数选取与研究背景结合不紧密且缺乏理论指导、参数优化时参数搜索范围靠人工盲目指定、遗传算法不能根据遗传迭代进展情况自适应调整造成收敛过快或过慢等问题,开展了以下几个方面研究:
首先,从估计精度、运算时间代价、稳健性等方面比较了常用泛化误差估计方法的性能,选出了适合高分辨雷达目标分类的泛化误差估计方法,得到了遗传算法的适值函数。
其次,基于高分辨一维距离像,分析了参数与支持向量机泛化能力的关系,为后续参数优化中参数搜索范围确定提供了参考依据,避免了人工指定的盲目性。
最后,提出了一种基于改进遗传算法的支持向量机参数优化和特征提取方法,可同时优化核函数参数、惩罚因子、组合核系数、以及遗传算法的交叉率和变异率。并可以根据遗传算法迭代情况自适应调整交叉变异率。通过对三类目标高分辨一维距离像的分类实验,验证了所提出优化方法的有效性。
(3)在支持向量数据描述单类分类器向多类分类推广研究方面
支持向量数据描述是一种单类分类器,不能直接用于多类分类,针对此问题,开展了以下研究:
首先比较了支持向量数据描述、最近邻数据描述、K近邻数据描述、K均值聚类数据描述等常用数据描述方法的性能,然后简单研究了核函数参数对支持向量数据描述分类边界的影响,最后基于最小距离规则和阈值处理策略提出了一种支持向量数据描述多类分类算法(SVDD_MDRTS),实现了支持向量数据描述从单类分类器向多类分类的推广。
采用仿真实验,分析了核函数参数、噪声等因素对SVDD_MDRTS算法性能的影响,结果表明存在最优参数值使算法性能最佳且算法具有较好的抗噪性能。
采用高频电磁散射数据对SVDD_MDRTS算法的分类性能进行了验证,设定几种测试条件,从识别混淆矩阵和识别ROC曲线等方面验证了不同条件下算法的性能,并将SVDD_MDRTS与SVM、NN多类分类器在识别性能和测试时间两方面进行了比较,结果表明SVDD_MDRTS具有更好的分类性能、测试和训练时间更短。
All the work in the dissertation is supported by the project of“973”National Security Important and Foundational Research, which is focused on investigation of the new mechanism and methods in radar automatic target recognition. Aiming at the background of high-resolution radar target classification, the feature extraction of scattering center parameter estimation and the key technique of classifier based on kernel method are investigated. Three important issues are included, which are feature extraction method based on scattering centers parameter estimation, parameter optimazation of Support Vector Machine (SVM) in high resolution radar target classification and research on generalizing Support Vector Data Description (SVDD) to multi-class classification.
In the aspect of feature extraction of scattering center parameter estimation, one new feature extraction method is proposed. The resolution of traditonal methods is low and the extracted position of scattering centers cannot be directly used in radar target classification. The scattering center number estimation is performed and the amplitude and position of scattering centers are estimated based on one super-resolution method. Then the distributed feature of relative distance between scattering centers is extracted, which is rotationally and translationally invariant compared with position feature itself in a certain aspect scope. Finally by identifying one-dimensional range profiles of three targets, the effectiveness and robustness of the proposed method are testified.
In the aspect of SVM parameters optimazation in high-resolution radar target classification, one parameter optimization method based on improved Gentic algorithm (GA) is proposed.
Aiming at the problems of loose relationship between fitness function selection of GA and research background, blind determination of parameter searching scope and nonautomatic adjustation of GA convergence, several problems have been investigated.
The fitness function selection in high-resolution radar target classification is investigated firstly. Parameter optimization performances of several generalization error estimation methods are analyzed and the estimation precision, computing time cost and robustness are compared respectively. One estimation method fit for SVM parameters optimization in high range resolution radar target classification is chosen as the fitness function of GA.Then the relationship between parameters and performance of SVM classifier is studied using one-dimensional range profiles. The parameter seaching scope is determined by analyzing the relationship. The study here overcomes the blind determination of parameter searching scope. In the end, one parameter optimization and feature extraction algorithm based on improved GA is proposed. Parameters such as kernel function parameter, penalty factor, composite kernel coefficient, the crossover ratio and mutation ratio of GA are all optimized simultaneously. The improved GA can automatically adjust the crossover and mutation ratios according to the state of GA progress, which overcomes the fast-speed or low-speed convergence of GA because of inappropriate crossover and mutation ratio. At last by classifying radar range profiles of three targets, the effectiveness of the algorithm is testified.
In the aspect of research on generalizing SVDD to multi-class classification, one SVDD multi-target classification method based on minimal distance classifier is brought forward.
SVDD is a one-class classifier and it cannot be directly used in multi-class classification. To solve this problem, research has been performed as follows.
The performances of SVDD and other data description method are compared and then the relationship between kernel function parameter and SVDD classifier is analyzed. Finally, based on the idea of minimum distance classifier and threshold strategy, one SVDD multi-class classification algorithm is propsed, which is nominated as SVDD_MDRTS. Based on simulation range profiles, the effects of kernel function parameter and noise on SVDD_MDRTS are analyzed. Based on high-frequency electronic and magnetic scattering data, the confusion matrix and Receiver Operating Characteristic (ROC) curve are investigated. Finally the classification performance and testing time of SVDD_MDRTS, SVM and NN are compared, experimental results indicate that SVDD_MDRTS has better perfomance and shorter training and testing time than SVM and NN.
(1)基于散射中心参数估计的分布特征提取方面
针对传统估计方法分辨力低、估计的散射点位置特征无法直接用于目标分类的问题,提出了一种基于散射中心径向相对位置估计的分布特征提取方法,该方法首先通过回波预测目标散射中心个数,然后对散射中心的幅度和位置信息进行高分辨估计,并依据估计结果提取散射中心间相对距离信息组成特征向量。经仿真实验验证,所提取的分布特征与传统方法中位置特征相比更稳定,并在一定角度范围内具有旋转和平移不变性,最后基于该特征提取方法实现了对三类目标的有效识别,并与已有方法进行比较,验证了该方法的有效性和稳健性。
(2)高分辨雷达目标分类中支持向量机分类器参数优化方面
针对以往基于遗传算法支持向量机参数优化中适值函数选取与研究背景结合不紧密且缺乏理论指导、参数优化时参数搜索范围靠人工盲目指定、遗传算法不能根据遗传迭代进展情况自适应调整造成收敛过快或过慢等问题,开展了以下几个方面研究:
首先,从估计精度、运算时间代价、稳健性等方面比较了常用泛化误差估计方法的性能,选出了适合高分辨雷达目标分类的泛化误差估计方法,得到了遗传算法的适值函数。
其次,基于高分辨一维距离像,分析了参数与支持向量机泛化能力的关系,为后续参数优化中参数搜索范围确定提供了参考依据,避免了人工指定的盲目性。
最后,提出了一种基于改进遗传算法的支持向量机参数优化和特征提取方法,可同时优化核函数参数、惩罚因子、组合核系数、以及遗传算法的交叉率和变异率。并可以根据遗传算法迭代情况自适应调整交叉变异率。通过对三类目标高分辨一维距离像的分类实验,验证了所提出优化方法的有效性。
(3)在支持向量数据描述单类分类器向多类分类推广研究方面
支持向量数据描述是一种单类分类器,不能直接用于多类分类,针对此问题,开展了以下研究:
首先比较了支持向量数据描述、最近邻数据描述、K近邻数据描述、K均值聚类数据描述等常用数据描述方法的性能,然后简单研究了核函数参数对支持向量数据描述分类边界的影响,最后基于最小距离规则和阈值处理策略提出了一种支持向量数据描述多类分类算法(SVDD_MDRTS),实现了支持向量数据描述从单类分类器向多类分类的推广。
采用仿真实验,分析了核函数参数、噪声等因素对SVDD_MDRTS算法性能的影响,结果表明存在最优参数值使算法性能最佳且算法具有较好的抗噪性能。
采用高频电磁散射数据对SVDD_MDRTS算法的分类性能进行了验证,设定几种测试条件,从识别混淆矩阵和识别ROC曲线等方面验证了不同条件下算法的性能,并将SVDD_MDRTS与SVM、NN多类分类器在识别性能和测试时间两方面进行了比较,结果表明SVDD_MDRTS具有更好的分类性能、测试和训练时间更短。
All the work in the dissertation is supported by the project of“973”National Security Important and Foundational Research, which is focused on investigation of the new mechanism and methods in radar automatic target recognition. Aiming at the background of high-resolution radar target classification, the feature extraction of scattering center parameter estimation and the key technique of classifier based on kernel method are investigated. Three important issues are included, which are feature extraction method based on scattering centers parameter estimation, parameter optimazation of Support Vector Machine (SVM) in high resolution radar target classification and research on generalizing Support Vector Data Description (SVDD) to multi-class classification.
In the aspect of feature extraction of scattering center parameter estimation, one new feature extraction method is proposed. The resolution of traditonal methods is low and the extracted position of scattering centers cannot be directly used in radar target classification. The scattering center number estimation is performed and the amplitude and position of scattering centers are estimated based on one super-resolution method. Then the distributed feature of relative distance between scattering centers is extracted, which is rotationally and translationally invariant compared with position feature itself in a certain aspect scope. Finally by identifying one-dimensional range profiles of three targets, the effectiveness and robustness of the proposed method are testified.
In the aspect of SVM parameters optimazation in high-resolution radar target classification, one parameter optimization method based on improved Gentic algorithm (GA) is proposed.
Aiming at the problems of loose relationship between fitness function selection of GA and research background, blind determination of parameter searching scope and nonautomatic adjustation of GA convergence, several problems have been investigated.
The fitness function selection in high-resolution radar target classification is investigated firstly. Parameter optimization performances of several generalization error estimation methods are analyzed and the estimation precision, computing time cost and robustness are compared respectively. One estimation method fit for SVM parameters optimization in high range resolution radar target classification is chosen as the fitness function of GA.Then the relationship between parameters and performance of SVM classifier is studied using one-dimensional range profiles. The parameter seaching scope is determined by analyzing the relationship. The study here overcomes the blind determination of parameter searching scope. In the end, one parameter optimization and feature extraction algorithm based on improved GA is proposed. Parameters such as kernel function parameter, penalty factor, composite kernel coefficient, the crossover ratio and mutation ratio of GA are all optimized simultaneously. The improved GA can automatically adjust the crossover and mutation ratios according to the state of GA progress, which overcomes the fast-speed or low-speed convergence of GA because of inappropriate crossover and mutation ratio. At last by classifying radar range profiles of three targets, the effectiveness of the algorithm is testified.
In the aspect of research on generalizing SVDD to multi-class classification, one SVDD multi-target classification method based on minimal distance classifier is brought forward.
SVDD is a one-class classifier and it cannot be directly used in multi-class classification. To solve this problem, research has been performed as follows.
The performances of SVDD and other data description method are compared and then the relationship between kernel function parameter and SVDD classifier is analyzed. Finally, based on the idea of minimum distance classifier and threshold strategy, one SVDD multi-class classification algorithm is propsed, which is nominated as SVDD_MDRTS. Based on simulation range profiles, the effects of kernel function parameter and noise on SVDD_MDRTS are analyzed. Based on high-frequency electronic and magnetic scattering data, the confusion matrix and Receiver Operating Characteristic (ROC) curve are investigated. Finally the classification performance and testing time of SVDD_MDRTS, SVM and NN are compared, experimental results indicate that SVDD_MDRTS has better perfomance and shorter training and testing time than SVM and NN.
引文
[1]裴炳南.高分辩雷达自动目标识别方法研究.博士学位论文,西安:西安电子科技大学,2002.
[2]M.W.Roth . Survey of Neural Network Technology for Automatic Target Recognition.IEEE Trans. on Neural Network,1990,1 (1) :28-43.
[3]Wang Yunhong and Liu Guosui.A fuzzy radial basis function neural network for radar target recognition.Proceeding of SPIE on application and science of ANN III,1997,3077:670~677.
[4]H.H.Szu Brian Telfer and shubha Kadambe.Neural Network Adaptive Wavelets for Signal Representation and classification.Optical Engineering,1992,31(9):1907-1916.
[5]朱永生.支持向量机及其在机械故障模式识别中的应用.博士学位论文,西安:西安交通大学,2003.
[6]周志明.基于小波和支持向量机的人脸识别技术.计算机工程与应用,2004,12:52-54.
[7]杨俊,谢寿生.基于模糊支持向量机的飞机动作识别.航空学报,2005,26(6): 738-742.
[8]王国燕,田娜.基于SVM超平面的水中目标分类识别技术.水雷战与舰船防护,2003,1:42-45.
[9]高学,金连文.一种基于支持向量机的手写汉字识别方法.电子学报,2002,30(5):651-654.
[10] Mike.Bryant.SVM classifier applied to MSTAR public data set.SPIE Conference on SAR imagery,1999,4(3721):355-360.
[11] D.Tax and R.Duin.Outliers and data despcription.Seventh Annual Conference of the Advanced School for Computing and Imaging,Delft,2001.
[12] Yan Liu and Srikanth Gururajan.Validating an Online Adaptive System Using SVDD.Proceedings of the 15th IEEE International Conference on Tools with Artificial Intelligence (ICTAI’03),2003.
[13] D.Tax and R.Duin.Support vector domain description.Pattern Recognition Letters,1999,20(11-13):1191-1199.
[14]廖学军.基于高分辨距离像的雷达目标识别.博士学位论文,西安:西安电子科技大学,1999.
[15]何松华.高距离分辨毫米波雷达目标识别理论与应用.博士学位论文,长沙:国防科技大学,1993.
[16] D.F.Fuller and R.Willimas.Approach to object classification using dispersive scattering centers.IEE Proc.-radar sonar navig,2004,151(2):85-90.
[17] Carriere and Moses.High resolution radar target modeling using a modified Prony estimatior.IEEE Trans. on antennas propagation,1992,40:13-18.
[18] Q.S. Ren and A.J. Willis.Fast root MUSIC algorithm.Electron letter,1997,33(6): 450-451.
[19] B.H. Kwon.New high resolution techniques and their performance analysis forAOA estimation.Ph.D.Disertaion,Polytechnique University,1989.
[20] Y.Hua and T.K Sarkar.Matrix pencil method for estimating parameters of exponentially damped/undamped sinusoids in noise.IEEE Trans. on Acoust,Speech Signal process,1990,38(5):814-824.
[21] Richard.Roy and Thomas.Kailath.ESPRIT-Estimation of Signal Parameters via Rotational Invariance Techniques.IEEE Trans.on acoustics speech and Signal Processing,1989,37(7):984-995.
[22] Octavia and Emauel.Radoi . Adavances in subspace eigenanalysisi based algorithms:from 1D to 3D uperresolution techniques.TELSIKS,2001:547-554.
[23]张贤达.现代信号处理(第二版).北京:清华大学出版社,2002.
[24] Emanuel.Radoi and Felix-Costinel.Some radar imagery results using supper resolution technique.IEEE Trans.on antennas and propagation,2004,52(5):1230-1243.
[25] A Moghaddar and E.K Waltion.Estimating the time delay and frequency decay parameter of scattering components using a modified MUSIC algorithmn.IEEE Trans. on antennas propagation,1994,42(10):1412-1418.
[26] M. Mcclure and L. Carin.On the superresolution identification of observables from swept-frequency scattering data.IEEE Trans. on antennas propagation,1997,45(4):631-641.
[27] K.-T Kim and Seo D.-K.Radar target identification using one-dimensional scattering centers.IEE Pro.-Radar Sonar Navig,2001,48(5):285-296.
[28]姜卫东,陈曾平,庄钊文.雷达目标高分辨距离像特征提取及其识别方法.国防科技大学学报,1999,21(3):55-58.
[29] Chapelle.O and V.vapnik.Choosing Multiple Parameters for Support Vector Machines.Machine Learning,2002,46(1):131-159.
[30]宋晓峰.优生演进优化和统计学习建模.博士学位论文,浙江:浙江大学,2003.
[31] V.N.Vapnik.张学工译:统计学习理论的本质.北京:清华大学出版社,2000.
[32] Vapnik.Nature of Statistical Learning Theory.New York:SpringerVerlag,2000.
[33]冯汉中,陈永义.处理非线性分类和回归问题的一种新方法-支持向量机方法在天气预报中的应用.应用气象学报,2004,15(3):355-365.
[34] Edmundo Bonilla Huerta.A Hybrid GA/SVM Approach for Gene Selection and Classification of Microarray Data.EvoWorkshops 2006,LNCS 3907,2006:34–44.
[35] Hong Guo and Lindsay B. Jack.Feature Generation Using Genetic Programming With Application to Fault Classification.IEEE Trans. on Systems, Man and Cybernetics—Part B:Cybernetics,2005,35(1):89-99.
[36]周伟达,张莉,焦李成.一种改进的推广能力衡量准则.计算机学报,2003,26(5):558-604.
[37] B. Samanta.Gear fault detection using artificial neural networks and support vector machines with genetic algorithms.Mechanical Systems and Signal Processing,2004,18:625-644.
[38] Chunhong Zheng, Guiwen Zheng and Licheng Jiao . Heuristic Genetic Algorithm-Based Support Vector Classifier for Recognition of Remote SensingImages.Springer-Verlag Berlin Heidelberg,2004:629-635.
[39] Cheng-Lung Huang and Chieh-Jen Wang.A GA-based feature selection and parameters optimization for support vector machines . Expert Systems with Applications,2006,31:231-240.
[40]陆从德,张太镒,胡金燕.基于乘性规则的支持向量域分类器.计算机学报, 2004,27(5):691-694.
[41] XinDong, WuZhaohui, and ZhangWanfeng.Support vector domain description for speaker recognition.2001 IEEE Signal Processing Society Workshop,Falmouth,2001.
[42]陆从德,张太镒.基于支持向量域描述的学习分类器.微电子学与计算机,2005,22(11):75-81.
[43] Woo-Sung Kang and Jin Hee Na . Face Recognition System by Fast and Incremental Learning Method.SICE-ICASE International Joint Conference 2006,Bexco, Busan, Korea,2006.
[44] Woo-Sung Kang, Ki Hong Im, and Jin Young Choi.SVDD-Based Method for Fast Training of Multi-class Support Vector Classifier.ISNN 2006, LNCS 3971, Springer-Verlag Berlin Heidelberg,2006:991-996.
[45] Woo–Sung Kang and Jin Young Choi.SVDD-based method for Face Recognition System.SCIS&ISIS2006,Tokyo, Japan,2006:1302-1306.
[46]David M.J. Tax and Robert P.W. Duin.Combining One-Class Classiers.MCS 2001, LNCS 2096, Springer-Verlag Berlin Heidelberg,2001:299-308.
[47] Thomas C.W. Landgrebe and David M.J. Tax.One-class and Multi-class classier combining for ill-defined problems.Elsevier Science,2004:1-21.
[48] Potter and Gerry.A GTD-based parametric model for radar scattering.IEEE Trans. on antennas propagation,1995,45:1058-1067.
[49] C.Potter Lee and Da ming chiang.A GTD-based parametric model for radar scattering.IEEE Trans. on antennas & propagation,1995,43:1058-1067.
[50] Michael.Burrows.Two-dimensional ESPRIT with Tracking for radar imaging and feature extraction.IEEE Trans. on antennas & propagation,2004,52(2):524-532.
[51] Douglas.B.willimas.Counting the degrees of freedom when using AIC and MDL to detect signals.IEEE transaction on Signal Processing,1994,42(11):3282-3284.
[52]王开文,张济民.基于AIC信息准则法的摆式列车倾摆伺服系统建模研究.中国铁道科学,2003,24(6):6-9.
[53]施建良,余松煜.基于最短描述长度的序列图像运动分割.计算机学报,1999,22(8):809-815.
[54]袁曾任.人工神经网络及其应用.北京:清华大学出版社,1999.
[55] K R Muller, S M ika.G, and Ratsc.An Introduction to Kernel_Based Learning Algorithms.IEEE Trans. on Neural Network,2001,12(2):181-201.
[56]赵玲玲,翁苏明.模式分析的核方法.北京:机械工业出版社,2006.
[57] R. Courant and D.Hilbert.Method of Mathematical Physics.Volume I:Springer Verlag,1953.
[58]肖健华.人脸识别中的核方法.五邑大学学报(自然科学版),2005,19(2):7-12.
[59]张学工.关于统计学习理论与支持向量机.自动化学报,2000,126(1):32-41.
[60] Li Zhang, Weida Zhou, and Licheng Jiao.Wavelet Support Vector Machine.IEEE Trans. on System, Man and Cybertics-Part B:Cybernetcis,2004,34(1):34-39.
[61]王国胜,钟义信.支持向量机若干新进展.电子学报,2001,129(10):1397-1400.
[62]杜树新,吴铁军.模式识别中的支持向量机方法.浙江大学学报(工学版),2003,137(5):521-527.
[63] Johan.A.K. Suykens.Benchmarking Least Squares Support Vector Machine Classifiers.Machine Learning,2004,54:5-32.
[64] Johannes TAX.One-class classification.Ph.D.Disertaion,Technische Universiteit Delft,2001.
[65]张周锁,李凌均,何正嘉.基于支持向量机故障分类器的参数优化研究.西安交通大学学报,2003,37(11):1101-1109.
[66] Zheng.Chunhong and Jiao.Licheng.Automatic parameters Selection for SVM based on GA.Proceedings of the 5" World Congress on Intelligent Control and Automation,Hangzhou, P.R. China,June 15-19 2004:1869-1872.
[67] S.Sathiya and Keerthi . Efficient Tuning of SVM Hyperparameters Using Radius/Margin Bound and Iterative Algorithms . IEEE Trans. On. Neural Networks,2002,13(5):1225-1229.
[68] Martin M. S. Lee and S. Sathiya Keerthi.An Efficient Method for Computing Leave-One-Out Error in Support Vector Machines With Gaussian Kernels.IEEE Transactions on Neural Networks,2004,15(3):750-757.
[69] T.S. Jaakkola and D.Haussler.Probabilistic kernel regression models.Proceeding of the 1999 Conference on AI and Statistics,1999.
[70] Kai-min Chung and Tony Suh.Radius margin bounds for support vector machine with the RBF kernel.Proceeding of th 9th international conference on Neural Information Processing,2002,2:893-897.
[71]任靖,李春平.最小距离分类器的改进算法——加权最小距离分类器.计算机应用,2005,25(5):992-994.
[72]李凌均,张周锁,何正嘉.基于支持向量数据描述的机械故障诊断研究.西安交通大学学报,2003,37(9):910-913.
[73]张浩然.支持向量机算法及应用研究.博士学位论文,上海:上海交通大学,2003.
[74] G.F smits.Improved SVM Regression Using Mixtures of Kernels.Proc of the 2002 International Joint Conf on Neural Networks,Honolulu,Hi,USA,2002:2785-2790.
[75]朱家元,张恒喜.支持向量机的多层动态自适应参数优化.控制与决策,2004,19(2):223-225.
[76]封筠,郝卫东.一种SVM分类器自动模型选择方法.北京科技大学学报,2006,28(1):88-92.
[77] Benchamar datasets. avialable at: http://ida.first.gmd.de/~raetsch/data/benchamarks.htm
[78] C.L.Blake and C.J.Merz.UCI Repository of machine learning databases,avialable at:http://www.ics.uci.edu/~mlearn/MLRepository.html.University of California,Department of Information and Computer Science,1998.
[79] D.M.J. Tax . A Matlab toolbox for data description , outlier and noveltydetection.July 26, 2006.
[80] B.Dubuisson and M.Masson.A statistical decision rule with incomplete knowledge about classes.Pattern Recognition,1993,26(1):155-165.
[81]刘瑞元.加权欧氏距离及其应用.青海师专学报(自然科学),2001,6:1-4.
[82]蒋慧峰,蔡光兴.一种新的距离分类函数.湖北工业大学学报,2006,21(1): 78-79.
[83]甄蜀春,冯有前,卜雪妮.基于距离分类准则的雷达目标识别方法.空军工程大学学报,2004,5(6):18-20.
[84]金连文,梁宇杰.一种新的距离分类方法及其应用.计算机应用,2003,25(8):30-31.
[85] Chih-Wei Hsu and Chih-Jen Lin.A Comparison of Methods for Multiclass Support Vector Machines.IEEE Trans. on Neural Network,2002,13(2):415-425.
[86]Markus Peich and Timo Kempf . ATR Perfomance at Extended Operating Conditions for highly resolved ISAR-images of relocatable targets.International Conference on Radar Systems,Toulouse,2004.
[87] John C. Mossing and Timothy D.Ross.An evaluation of SAR ATR alogorithm performance sensitivity to MSTAR Extended Opertaing Conditions . SPIE Conferece on alorithm for SAR imagery V,Orlando,Florida,1998:554-565.
[88] Harris Drucker and Vladimir N. Vapnik.Support Vector Machines for Spam Categorization.IEEE Trans. on neural networks,1999,10(5):1048-1054.
[89] L.J. Cao and Zhang Jingqing.An empirical study of feature selection in support vector machines.Neural Network World,2006,16(5):433-453.
[90] Vladimir Cherkassky and Vladimir N. Vapnik.Model Complexity Control for Regression Using VC Generalization Bounds.IEEE Trans. on neural networks,1999,10(5):1075-1089.
[91] Zheng chunhong and Jiao Licheng.Fuzzy pre-extracting method for support vector machine.Proceeding of the first international conference on machine learning and cybernetics,2002:2026-2030.
[92] Kie B. Eom and Rama Chellappa.Noncooperative Target Classification Using Hierarchical Modeling of High-Range Resolution Radar Signatures.IEEE Trans. on signal processing,1997,45 (9) :2318-2327.
[93] Isabelle Guyon and Vladimir Vapnik.What Size Test Set Gives Good Error Rate Estimates?.IEEE Trans. on pattern analysis and machine intelligence,1998,20 (1) :52-64.
[94] Huang Jing-Tao and Ma Long-Hua . Improved support vector machine for multi-class classification problems.Journal of Zhejiang University (Engineering Science),2004,38(12):1633-1636.
[95] Kuri-Morales and Angel.Determination of the regularization parameter for support vector machines via Vasconcelos' genetic algorithm.WSEAS Transactions on Circuits and Systems,2005,4(4):281-286.
[96] Sung-Hwan Min, Jumin Lee b, and Ingoo Han.Hybrid genetic algorithms and support vector machines for bankruptcy prediction . Expert Systems with Applications,2006,31:652-660.
[97] Syng-Yup Ohn and Ha-Nam Nguyen.Determining Optimal Decision Model for Support Vector Machine by Genetic Algorithm . CIS 2004,LNCS 3314 , Springer-Verlag Berlin Heidelberg,2004:895-902.
[98] Xu Peng and Chan Andrew K.Support Vector Machines for Multi-class Signal Classification with Unbalanced Samples.Proceedings of the International Joint Conference on Neural Networks,2003,2:1116-1119.
[99] Xu Qi-Hua and Shi Jun.New multi-class support vector algorithm and its application in fault diagnosis.Journal of System Simulation,2005,17(11):2766-2768+2784.
[100]Wang Shang-Fei and Wang Xu-Fa.An improved Interactive Genetic Algorithm Incorporating Relevant Feedback.2005 International Conference on Machine Learning and Cybernetics,2005,2005:2996-3001.
[101]Ohn SY and Nguyen HN.Determining optimal decision model for support vector machine by genetic algorithm.Proceedings lecture notes in computer science,2004,3314:895-902.
[102]Vladimir N. Vapnik.An Overview of Statistical Learning Theory.IEEE Trans. on neural networks,1999,10(5):988-999.
[103]Eugene Wong.A statistical approach to incomplete information in database systems.ACM trans. on database system,1982,7(3):470-488.
[104]Chih-Hung Wu and Gwo-Hshiung Tzeng.A real-valued genetic algorithm to optimize the parameters of support vector machine for predicting bankruptcy.Expert Systems with Applications,2007,32:397-408.
[105]Wang Xin-Feng and Qiu Jing.Joint optimization of the fault feature and classifier.Journal of National University of Defense Technology,2005,27(2):92-95.
[106]Sheng-Fa Yuana and Fu-Lei Chu.Fault diagnostics based on particle swarm optimisation and support vector machines . Mechanical Systems and Signal Processing,2007,21:1787-1798.
[107]Youshen Xia and Jun Wang.A One-Layer Recurrent Neural Network for Support Vector Machine Learning.IEEE Trans. on system,man and cybernetics-part B:Cybernetics, 2004,34(2):1261-1269.
[108]Brian M. Huether and Steven C. Gustafson.Wavelet preprocessing for high range resolution radar classification.IEEE Trans on aerospace and electronic systems,2001,37(4):1321-1331.
[109]David L. Donoho.De-nosing by soft-thresholding.IEEE Trans on information theory,1995,41(3):613-627.
[110]YangMing.Hsuan and N.Ahuja . Face Recognition Using Kernel Eigenfaces.Proceedings of International Conference on Image Processing,2000,1:37-40.
[111]Himaanshu Gupta and Amit K Agrawal.An Experimental Evaluation of Linear and Kernel-Based Methods for Face Recognition.Proceeding of the sixth IEEE workshop on application of computer vision(WACV'02),2002.
[112]Peter Sollich.Probabilistic interpretations and bayesian methods for support vectormachine.IEE Conference on Artificial Neural Networks,1999,470:91-96.
[113]Bernardete Ribeiro.Support Vector Machines for Quality Monitoring in a Plastic Injection Molding Process.IEEE Trans on system, man and cybernetics,part c:applications and reviews,2005,35(3):401-410.
[2]M.W.Roth . Survey of Neural Network Technology for Automatic Target Recognition.IEEE Trans. on Neural Network,1990,1 (1) :28-43.
[3]Wang Yunhong and Liu Guosui.A fuzzy radial basis function neural network for radar target recognition.Proceeding of SPIE on application and science of ANN III,1997,3077:670~677.
[4]H.H.Szu Brian Telfer and shubha Kadambe.Neural Network Adaptive Wavelets for Signal Representation and classification.Optical Engineering,1992,31(9):1907-1916.
[5]朱永生.支持向量机及其在机械故障模式识别中的应用.博士学位论文,西安:西安交通大学,2003.
[6]周志明.基于小波和支持向量机的人脸识别技术.计算机工程与应用,2004,12:52-54.
[7]杨俊,谢寿生.基于模糊支持向量机的飞机动作识别.航空学报,2005,26(6): 738-742.
[8]王国燕,田娜.基于SVM超平面的水中目标分类识别技术.水雷战与舰船防护,2003,1:42-45.
[9]高学,金连文.一种基于支持向量机的手写汉字识别方法.电子学报,2002,30(5):651-654.
[10] Mike.Bryant.SVM classifier applied to MSTAR public data set.SPIE Conference on SAR imagery,1999,4(3721):355-360.
[11] D.Tax and R.Duin.Outliers and data despcription.Seventh Annual Conference of the Advanced School for Computing and Imaging,Delft,2001.
[12] Yan Liu and Srikanth Gururajan.Validating an Online Adaptive System Using SVDD.Proceedings of the 15th IEEE International Conference on Tools with Artificial Intelligence (ICTAI’03),2003.
[13] D.Tax and R.Duin.Support vector domain description.Pattern Recognition Letters,1999,20(11-13):1191-1199.
[14]廖学军.基于高分辨距离像的雷达目标识别.博士学位论文,西安:西安电子科技大学,1999.
[15]何松华.高距离分辨毫米波雷达目标识别理论与应用.博士学位论文,长沙:国防科技大学,1993.
[16] D.F.Fuller and R.Willimas.Approach to object classification using dispersive scattering centers.IEE Proc.-radar sonar navig,2004,151(2):85-90.
[17] Carriere and Moses.High resolution radar target modeling using a modified Prony estimatior.IEEE Trans. on antennas propagation,1992,40:13-18.
[18] Q.S. Ren and A.J. Willis.Fast root MUSIC algorithm.Electron letter,1997,33(6): 450-451.
[19] B.H. Kwon.New high resolution techniques and their performance analysis forAOA estimation.Ph.D.Disertaion,Polytechnique University,1989.
[20] Y.Hua and T.K Sarkar.Matrix pencil method for estimating parameters of exponentially damped/undamped sinusoids in noise.IEEE Trans. on Acoust,Speech Signal process,1990,38(5):814-824.
[21] Richard.Roy and Thomas.Kailath.ESPRIT-Estimation of Signal Parameters via Rotational Invariance Techniques.IEEE Trans.on acoustics speech and Signal Processing,1989,37(7):984-995.
[22] Octavia and Emauel.Radoi . Adavances in subspace eigenanalysisi based algorithms:from 1D to 3D uperresolution techniques.TELSIKS,2001:547-554.
[23]张贤达.现代信号处理(第二版).北京:清华大学出版社,2002.
[24] Emanuel.Radoi and Felix-Costinel.Some radar imagery results using supper resolution technique.IEEE Trans.on antennas and propagation,2004,52(5):1230-1243.
[25] A Moghaddar and E.K Waltion.Estimating the time delay and frequency decay parameter of scattering components using a modified MUSIC algorithmn.IEEE Trans. on antennas propagation,1994,42(10):1412-1418.
[26] M. Mcclure and L. Carin.On the superresolution identification of observables from swept-frequency scattering data.IEEE Trans. on antennas propagation,1997,45(4):631-641.
[27] K.-T Kim and Seo D.-K.Radar target identification using one-dimensional scattering centers.IEE Pro.-Radar Sonar Navig,2001,48(5):285-296.
[28]姜卫东,陈曾平,庄钊文.雷达目标高分辨距离像特征提取及其识别方法.国防科技大学学报,1999,21(3):55-58.
[29] Chapelle.O and V.vapnik.Choosing Multiple Parameters for Support Vector Machines.Machine Learning,2002,46(1):131-159.
[30]宋晓峰.优生演进优化和统计学习建模.博士学位论文,浙江:浙江大学,2003.
[31] V.N.Vapnik.张学工译:统计学习理论的本质.北京:清华大学出版社,2000.
[32] Vapnik.Nature of Statistical Learning Theory.New York:SpringerVerlag,2000.
[33]冯汉中,陈永义.处理非线性分类和回归问题的一种新方法-支持向量机方法在天气预报中的应用.应用气象学报,2004,15(3):355-365.
[34] Edmundo Bonilla Huerta.A Hybrid GA/SVM Approach for Gene Selection and Classification of Microarray Data.EvoWorkshops 2006,LNCS 3907,2006:34–44.
[35] Hong Guo and Lindsay B. Jack.Feature Generation Using Genetic Programming With Application to Fault Classification.IEEE Trans. on Systems, Man and Cybernetics—Part B:Cybernetics,2005,35(1):89-99.
[36]周伟达,张莉,焦李成.一种改进的推广能力衡量准则.计算机学报,2003,26(5):558-604.
[37] B. Samanta.Gear fault detection using artificial neural networks and support vector machines with genetic algorithms.Mechanical Systems and Signal Processing,2004,18:625-644.
[38] Chunhong Zheng, Guiwen Zheng and Licheng Jiao . Heuristic Genetic Algorithm-Based Support Vector Classifier for Recognition of Remote SensingImages.Springer-Verlag Berlin Heidelberg,2004:629-635.
[39] Cheng-Lung Huang and Chieh-Jen Wang.A GA-based feature selection and parameters optimization for support vector machines . Expert Systems with Applications,2006,31:231-240.
[40]陆从德,张太镒,胡金燕.基于乘性规则的支持向量域分类器.计算机学报, 2004,27(5):691-694.
[41] XinDong, WuZhaohui, and ZhangWanfeng.Support vector domain description for speaker recognition.2001 IEEE Signal Processing Society Workshop,Falmouth,2001.
[42]陆从德,张太镒.基于支持向量域描述的学习分类器.微电子学与计算机,2005,22(11):75-81.
[43] Woo-Sung Kang and Jin Hee Na . Face Recognition System by Fast and Incremental Learning Method.SICE-ICASE International Joint Conference 2006,Bexco, Busan, Korea,2006.
[44] Woo-Sung Kang, Ki Hong Im, and Jin Young Choi.SVDD-Based Method for Fast Training of Multi-class Support Vector Classifier.ISNN 2006, LNCS 3971, Springer-Verlag Berlin Heidelberg,2006:991-996.
[45] Woo–Sung Kang and Jin Young Choi.SVDD-based method for Face Recognition System.SCIS&ISIS2006,Tokyo, Japan,2006:1302-1306.
[46]David M.J. Tax and Robert P.W. Duin.Combining One-Class Classiers.MCS 2001, LNCS 2096, Springer-Verlag Berlin Heidelberg,2001:299-308.
[47] Thomas C.W. Landgrebe and David M.J. Tax.One-class and Multi-class classier combining for ill-defined problems.Elsevier Science,2004:1-21.
[48] Potter and Gerry.A GTD-based parametric model for radar scattering.IEEE Trans. on antennas propagation,1995,45:1058-1067.
[49] C.Potter Lee and Da ming chiang.A GTD-based parametric model for radar scattering.IEEE Trans. on antennas & propagation,1995,43:1058-1067.
[50] Michael.Burrows.Two-dimensional ESPRIT with Tracking for radar imaging and feature extraction.IEEE Trans. on antennas & propagation,2004,52(2):524-532.
[51] Douglas.B.willimas.Counting the degrees of freedom when using AIC and MDL to detect signals.IEEE transaction on Signal Processing,1994,42(11):3282-3284.
[52]王开文,张济民.基于AIC信息准则法的摆式列车倾摆伺服系统建模研究.中国铁道科学,2003,24(6):6-9.
[53]施建良,余松煜.基于最短描述长度的序列图像运动分割.计算机学报,1999,22(8):809-815.
[54]袁曾任.人工神经网络及其应用.北京:清华大学出版社,1999.
[55] K R Muller, S M ika.G, and Ratsc.An Introduction to Kernel_Based Learning Algorithms.IEEE Trans. on Neural Network,2001,12(2):181-201.
[56]赵玲玲,翁苏明.模式分析的核方法.北京:机械工业出版社,2006.
[57] R. Courant and D.Hilbert.Method of Mathematical Physics.Volume I:Springer Verlag,1953.
[58]肖健华.人脸识别中的核方法.五邑大学学报(自然科学版),2005,19(2):7-12.
[59]张学工.关于统计学习理论与支持向量机.自动化学报,2000,126(1):32-41.
[60] Li Zhang, Weida Zhou, and Licheng Jiao.Wavelet Support Vector Machine.IEEE Trans. on System, Man and Cybertics-Part B:Cybernetcis,2004,34(1):34-39.
[61]王国胜,钟义信.支持向量机若干新进展.电子学报,2001,129(10):1397-1400.
[62]杜树新,吴铁军.模式识别中的支持向量机方法.浙江大学学报(工学版),2003,137(5):521-527.
[63] Johan.A.K. Suykens.Benchmarking Least Squares Support Vector Machine Classifiers.Machine Learning,2004,54:5-32.
[64] Johannes TAX.One-class classification.Ph.D.Disertaion,Technische Universiteit Delft,2001.
[65]张周锁,李凌均,何正嘉.基于支持向量机故障分类器的参数优化研究.西安交通大学学报,2003,37(11):1101-1109.
[66] Zheng.Chunhong and Jiao.Licheng.Automatic parameters Selection for SVM based on GA.Proceedings of the 5" World Congress on Intelligent Control and Automation,Hangzhou, P.R. China,June 15-19 2004:1869-1872.
[67] S.Sathiya and Keerthi . Efficient Tuning of SVM Hyperparameters Using Radius/Margin Bound and Iterative Algorithms . IEEE Trans. On. Neural Networks,2002,13(5):1225-1229.
[68] Martin M. S. Lee and S. Sathiya Keerthi.An Efficient Method for Computing Leave-One-Out Error in Support Vector Machines With Gaussian Kernels.IEEE Transactions on Neural Networks,2004,15(3):750-757.
[69] T.S. Jaakkola and D.Haussler.Probabilistic kernel regression models.Proceeding of the 1999 Conference on AI and Statistics,1999.
[70] Kai-min Chung and Tony Suh.Radius margin bounds for support vector machine with the RBF kernel.Proceeding of th 9th international conference on Neural Information Processing,2002,2:893-897.
[71]任靖,李春平.最小距离分类器的改进算法——加权最小距离分类器.计算机应用,2005,25(5):992-994.
[72]李凌均,张周锁,何正嘉.基于支持向量数据描述的机械故障诊断研究.西安交通大学学报,2003,37(9):910-913.
[73]张浩然.支持向量机算法及应用研究.博士学位论文,上海:上海交通大学,2003.
[74] G.F smits.Improved SVM Regression Using Mixtures of Kernels.Proc of the 2002 International Joint Conf on Neural Networks,Honolulu,Hi,USA,2002:2785-2790.
[75]朱家元,张恒喜.支持向量机的多层动态自适应参数优化.控制与决策,2004,19(2):223-225.
[76]封筠,郝卫东.一种SVM分类器自动模型选择方法.北京科技大学学报,2006,28(1):88-92.
[77] Benchamar datasets. avialable at: http://ida.first.gmd.de/~raetsch/data/benchamarks.htm
[78] C.L.Blake and C.J.Merz.UCI Repository of machine learning databases,avialable at:http://www.ics.uci.edu/~mlearn/MLRepository.html.University of California,Department of Information and Computer Science,1998.
[79] D.M.J. Tax . A Matlab toolbox for data description , outlier and noveltydetection.July 26, 2006.
[80] B.Dubuisson and M.Masson.A statistical decision rule with incomplete knowledge about classes.Pattern Recognition,1993,26(1):155-165.
[81]刘瑞元.加权欧氏距离及其应用.青海师专学报(自然科学),2001,6:1-4.
[82]蒋慧峰,蔡光兴.一种新的距离分类函数.湖北工业大学学报,2006,21(1): 78-79.
[83]甄蜀春,冯有前,卜雪妮.基于距离分类准则的雷达目标识别方法.空军工程大学学报,2004,5(6):18-20.
[84]金连文,梁宇杰.一种新的距离分类方法及其应用.计算机应用,2003,25(8):30-31.
[85] Chih-Wei Hsu and Chih-Jen Lin.A Comparison of Methods for Multiclass Support Vector Machines.IEEE Trans. on Neural Network,2002,13(2):415-425.
[86]Markus Peich and Timo Kempf . ATR Perfomance at Extended Operating Conditions for highly resolved ISAR-images of relocatable targets.International Conference on Radar Systems,Toulouse,2004.
[87] John C. Mossing and Timothy D.Ross.An evaluation of SAR ATR alogorithm performance sensitivity to MSTAR Extended Opertaing Conditions . SPIE Conferece on alorithm for SAR imagery V,Orlando,Florida,1998:554-565.
[88] Harris Drucker and Vladimir N. Vapnik.Support Vector Machines for Spam Categorization.IEEE Trans. on neural networks,1999,10(5):1048-1054.
[89] L.J. Cao and Zhang Jingqing.An empirical study of feature selection in support vector machines.Neural Network World,2006,16(5):433-453.
[90] Vladimir Cherkassky and Vladimir N. Vapnik.Model Complexity Control for Regression Using VC Generalization Bounds.IEEE Trans. on neural networks,1999,10(5):1075-1089.
[91] Zheng chunhong and Jiao Licheng.Fuzzy pre-extracting method for support vector machine.Proceeding of the first international conference on machine learning and cybernetics,2002:2026-2030.
[92] Kie B. Eom and Rama Chellappa.Noncooperative Target Classification Using Hierarchical Modeling of High-Range Resolution Radar Signatures.IEEE Trans. on signal processing,1997,45 (9) :2318-2327.
[93] Isabelle Guyon and Vladimir Vapnik.What Size Test Set Gives Good Error Rate Estimates?.IEEE Trans. on pattern analysis and machine intelligence,1998,20 (1) :52-64.
[94] Huang Jing-Tao and Ma Long-Hua . Improved support vector machine for multi-class classification problems.Journal of Zhejiang University (Engineering Science),2004,38(12):1633-1636.
[95] Kuri-Morales and Angel.Determination of the regularization parameter for support vector machines via Vasconcelos' genetic algorithm.WSEAS Transactions on Circuits and Systems,2005,4(4):281-286.
[96] Sung-Hwan Min, Jumin Lee b, and Ingoo Han.Hybrid genetic algorithms and support vector machines for bankruptcy prediction . Expert Systems with Applications,2006,31:652-660.
[97] Syng-Yup Ohn and Ha-Nam Nguyen.Determining Optimal Decision Model for Support Vector Machine by Genetic Algorithm . CIS 2004,LNCS 3314 , Springer-Verlag Berlin Heidelberg,2004:895-902.
[98] Xu Peng and Chan Andrew K.Support Vector Machines for Multi-class Signal Classification with Unbalanced Samples.Proceedings of the International Joint Conference on Neural Networks,2003,2:1116-1119.
[99] Xu Qi-Hua and Shi Jun.New multi-class support vector algorithm and its application in fault diagnosis.Journal of System Simulation,2005,17(11):2766-2768+2784.
[100]Wang Shang-Fei and Wang Xu-Fa.An improved Interactive Genetic Algorithm Incorporating Relevant Feedback.2005 International Conference on Machine Learning and Cybernetics,2005,2005:2996-3001.
[101]Ohn SY and Nguyen HN.Determining optimal decision model for support vector machine by genetic algorithm.Proceedings lecture notes in computer science,2004,3314:895-902.
[102]Vladimir N. Vapnik.An Overview of Statistical Learning Theory.IEEE Trans. on neural networks,1999,10(5):988-999.
[103]Eugene Wong.A statistical approach to incomplete information in database systems.ACM trans. on database system,1982,7(3):470-488.
[104]Chih-Hung Wu and Gwo-Hshiung Tzeng.A real-valued genetic algorithm to optimize the parameters of support vector machine for predicting bankruptcy.Expert Systems with Applications,2007,32:397-408.
[105]Wang Xin-Feng and Qiu Jing.Joint optimization of the fault feature and classifier.Journal of National University of Defense Technology,2005,27(2):92-95.
[106]Sheng-Fa Yuana and Fu-Lei Chu.Fault diagnostics based on particle swarm optimisation and support vector machines . Mechanical Systems and Signal Processing,2007,21:1787-1798.
[107]Youshen Xia and Jun Wang.A One-Layer Recurrent Neural Network for Support Vector Machine Learning.IEEE Trans. on system,man and cybernetics-part B:Cybernetics, 2004,34(2):1261-1269.
[108]Brian M. Huether and Steven C. Gustafson.Wavelet preprocessing for high range resolution radar classification.IEEE Trans on aerospace and electronic systems,2001,37(4):1321-1331.
[109]David L. Donoho.De-nosing by soft-thresholding.IEEE Trans on information theory,1995,41(3):613-627.
[110]YangMing.Hsuan and N.Ahuja . Face Recognition Using Kernel Eigenfaces.Proceedings of International Conference on Image Processing,2000,1:37-40.
[111]Himaanshu Gupta and Amit K Agrawal.An Experimental Evaluation of Linear and Kernel-Based Methods for Face Recognition.Proceeding of the sixth IEEE workshop on application of computer vision(WACV'02),2002.
[112]Peter Sollich.Probabilistic interpretations and bayesian methods for support vectormachine.IEE Conference on Artificial Neural Networks,1999,470:91-96.
[113]Bernardete Ribeiro.Support Vector Machines for Quality Monitoring in a Plastic Injection Molding Process.IEEE Trans on system, man and cybernetics,part c:applications and reviews,2005,35(3):401-410.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |