基于集成学习的高光谱遥感影像分类
|
摘要
伴随着高光谱成像光谱仪像向更高空间和光谱分辨率的快速发展，对现有的数据处理方法提出了新的挑战，需要发展新的数据处理方法。作为机器学习中最有效的方法之一，集成学习使用多个学习机（算法）来解决同一问题，能够显著地提高学习系统的准确性和稳定性。本文从集成学习的理论基础出发，将集成学习引入到高光谱遥感影像分类中，系统研究了旋转森林、监督/半监督特征提取集成分类和光谱-空间特征集成分类等问题。论文主要内容和结论如下：
1）将旋转森林应用于高光谱遥感影像分类，运用了不同特征提取算法，实现了基于独立成分分析、最大噪声分离和局部线性判别分析的旋转森林，并对应用不同特征提取算法的旋转森林分类器的性能进行分析。试验结果表明旋转森林的性能优于常规集成学习算法（Bagging等），基于主成分分析和独立成分分析特征提取算法更为有效。
2）将半监督/监督概率主成分分析用于高光谱遥感影像分类特征提取，从不同数据、不同数量的标记样本和未标记样本、计算时间等方面对半监督/监督概率主成分分析与其他高光谱遥感影像特征提取算法进行比较分析，结果证明半监督/监督概率主成分分析提取出来的分类特征能够得到精度较高的分类结果。在此基础上，提出了通过半监督概率主成分分析、监督概率主成分分析和无参数加权特征提取三种监督/半监督特征提取算法，构建监督/半监督特征提取的并行/串行集成分类策略，分类结果表明基于监督/半监督特征提取集成策略能够有效提高高光谱遥感影像分类精度。
3）建立了基于半监督特征提取和图像分割、马尔科夫模型的光谱-空间特征综合分类模型，图像分割分别选择聚类分割、分水岭分割和Mean-shift分割。马尔科夫模型通过最小化局部能量函数，采用模拟退火算法，能够有效的使邻近像元聚集。光谱-空间特征综合模型使高光谱遥感影像分类精度得到显著改善，并且能够降低遥感影像分类的噪声，分类结果更接近于真实地物分布。
4）结合理论研究，研发了高光谱遥感影像集成学习分类系统，该系统包含遥感影像基本处理、分类、聚类、分割、训练层和特征层集成、并行/串行集成、差异性测度和光谱-空间集成等功能，以城市不透水面层提取分析为例，分析了该系统在实际应用中高光谱遥感信息处理的优势。
With the rapid development of hyperspectral imaging spectroradiometer of highspatial and spectral resolution, the existing data processing methods face a newchallenge and we require new data processin algorithms. As one of the most effectivemethod of machine learning, ensemble learning, which utilizes many learningalgorithms to solve a predition problem, can significantly improve the accuracy andstability of a learning system. This dissertation introduces ensemble learning intohyperspectral remote sensing image classification. The main contributions of ourworks in this dissertation are summarized as follows:
1) Rotation Forest has been applied for hyperspectral remote sensing imageclassification. Based on the framework of original Rotation Forest, we proposeRotation Forest with different feature extraction algorithms, such as independentcomponent analysis, maximum noise fraction and local fisher discriminant analysis.Experimental results indicate that the performances of Rotation Forests are better thanother tradational ensemble learning algorithms (Bagging et al), especially withprincipal component analysis and independent component analysis feature extractionmethods.
2) Supervised/semi-supervised principal component analysis has been used toextract features of hyperspectral remote sensing image classification. Theperformance of supervised/semi-supervised principal component analysis is comparedwith other traditional feature extraction methods and evaluated based on severalcriterias: different datasets, different number of labeled/unlabeled samples and thecomputation time. Experimental results revealed that supervised/semi-supervisedprincipal component analysis can extract more reliable features for hyperspectralimage classification than other feature extraction algorithms. Futhurmore,parallel/concatenation ensemble methods based on three powful supervised/semi-supervised feature extraction, including supervised/semi-supervised principalcomponent analysis, non-parametric weighted feature extraction, are proposed forhyperspectral image classification. Experimental results show that the ensemblemethods can improve the accuracy of hyperspectral remote sensing imageclassification.
3) Spectral-spatial ensemble method based on semi-supervised feature extractionand image segmentation or markov random field are developed for hyperspectral image classification. We choose clustering, watershed transformation and mean-shiftas image segmentation techniques. The clustering algorithms include K-means,ISODATA, Fuzzy K-means, Kernel K-means and EM. Markov random field adoptsthe simulated annealing algorithm to make the adjacent pixels aggregation byminimizing the local energy function. The above spectral-spatial methods significantltimprove the classification accuracies and reduce the noise of classification results.
4) Based on the theoretical research, hyperspectral image ensemble learningclassification system (HIELCS), which includes image preprocessing, classification,clustering, segmentation, train/feature stage ensemble learning, spectral-spatial,diversity measures et al, is implememted. Urban impervious surface area extractionusing HIELCS sytem shows the advantages of hyperspectral remote sensinginformation processing in practical application.
1）将旋转森林应用于高光谱遥感影像分类，运用了不同特征提取算法，实现了基于独立成分分析、最大噪声分离和局部线性判别分析的旋转森林，并对应用不同特征提取算法的旋转森林分类器的性能进行分析。试验结果表明旋转森林的性能优于常规集成学习算法（Bagging等），基于主成分分析和独立成分分析特征提取算法更为有效。
2）将半监督/监督概率主成分分析用于高光谱遥感影像分类特征提取，从不同数据、不同数量的标记样本和未标记样本、计算时间等方面对半监督/监督概率主成分分析与其他高光谱遥感影像特征提取算法进行比较分析，结果证明半监督/监督概率主成分分析提取出来的分类特征能够得到精度较高的分类结果。在此基础上，提出了通过半监督概率主成分分析、监督概率主成分分析和无参数加权特征提取三种监督/半监督特征提取算法，构建监督/半监督特征提取的并行/串行集成分类策略，分类结果表明基于监督/半监督特征提取集成策略能够有效提高高光谱遥感影像分类精度。
3）建立了基于半监督特征提取和图像分割、马尔科夫模型的光谱-空间特征综合分类模型，图像分割分别选择聚类分割、分水岭分割和Mean-shift分割。马尔科夫模型通过最小化局部能量函数，采用模拟退火算法，能够有效的使邻近像元聚集。光谱-空间特征综合模型使高光谱遥感影像分类精度得到显著改善，并且能够降低遥感影像分类的噪声，分类结果更接近于真实地物分布。
4）结合理论研究，研发了高光谱遥感影像集成学习分类系统，该系统包含遥感影像基本处理、分类、聚类、分割、训练层和特征层集成、并行/串行集成、差异性测度和光谱-空间集成等功能，以城市不透水面层提取分析为例，分析了该系统在实际应用中高光谱遥感信息处理的优势。
With the rapid development of hyperspectral imaging spectroradiometer of highspatial and spectral resolution, the existing data processing methods face a newchallenge and we require new data processin algorithms. As one of the most effectivemethod of machine learning, ensemble learning, which utilizes many learningalgorithms to solve a predition problem, can significantly improve the accuracy andstability of a learning system. This dissertation introduces ensemble learning intohyperspectral remote sensing image classification. The main contributions of ourworks in this dissertation are summarized as follows:
1) Rotation Forest has been applied for hyperspectral remote sensing imageclassification. Based on the framework of original Rotation Forest, we proposeRotation Forest with different feature extraction algorithms, such as independentcomponent analysis, maximum noise fraction and local fisher discriminant analysis.Experimental results indicate that the performances of Rotation Forests are better thanother tradational ensemble learning algorithms (Bagging et al), especially withprincipal component analysis and independent component analysis feature extractionmethods.
2) Supervised/semi-supervised principal component analysis has been used toextract features of hyperspectral remote sensing image classification. Theperformance of supervised/semi-supervised principal component analysis is comparedwith other traditional feature extraction methods and evaluated based on severalcriterias: different datasets, different number of labeled/unlabeled samples and thecomputation time. Experimental results revealed that supervised/semi-supervisedprincipal component analysis can extract more reliable features for hyperspectralimage classification than other feature extraction algorithms. Futhurmore,parallel/concatenation ensemble methods based on three powful supervised/semi-supervised feature extraction, including supervised/semi-supervised principalcomponent analysis, non-parametric weighted feature extraction, are proposed forhyperspectral image classification. Experimental results show that the ensemblemethods can improve the accuracy of hyperspectral remote sensing imageclassification.
3) Spectral-spatial ensemble method based on semi-supervised feature extractionand image segmentation or markov random field are developed for hyperspectral image classification. We choose clustering, watershed transformation and mean-shiftas image segmentation techniques. The clustering algorithms include K-means,ISODATA, Fuzzy K-means, Kernel K-means and EM. Markov random field adoptsthe simulated annealing algorithm to make the adjacent pixels aggregation byminimizing the local energy function. The above spectral-spatial methods significantltimprove the classification accuracies and reduce the noise of classification results.
4) Based on the theoretical research, hyperspectral image ensemble learningclassification system (HIELCS), which includes image preprocessing, classification,clustering, segmentation, train/feature stage ensemble learning, spectral-spatial,diversity measures et al, is implememted. Urban impervious surface area extractionusing HIELCS sytem shows the advantages of hyperspectral remote sensinginformation processing in practical application.
引文
[1]赵英时．遥感应用分析原理与方法[M]．北京：科学出版社，2003
[2]童庆禧,张兵,郑兰芬.高光谱遥感—原理、技术及应用[M].北京:高等教育出版社.2006.
[3] Cocks T, Jenssen R, Stewart A et al. The HyMap airborne hyperspectral sensor: the system,calibration and performance [C]. In Proc. of First EARSEL Workshop on ImagingSpectroscopy, October1998.
[4] Gamba P. A collection of data for urban area characterization [C]. In Proc. of IGARSS’04,Anchorage, USA, Sept.2004.
[5] Green R O, Eastwood M L, Sarture C M et al. Imaging spectroscopy and the AirborneVisible Infrared Imaging Spectrometer (AVIRIS)[J]. Remote Sensing of Environment,1998,65(3):227-248.
[6] Pearlman J, Segal C, Liao L et al. Development and operations of theEO-1Hyperionimaging spectrometer [C]. In Proc. of Earth Observing Systems V, SPIE, volume4135, pages243–253,2000.
[7] Plaza A, Benediktsson J A, Boardman J et al. Recent advances in techniques forhyperspectral image processing [J]. Remote Sensing of Environment,2009,113(S1):110-122.
[8] Rickard L, Basedow W, Zalewski E et al. HYDICE: An airborne system for hyperspectralimaging [C]. In Proc. of Imaging Spectrometry of the Terrestrial Environment, SPIE,19931937:173–179.
[9] Staenz K. Terrestrial imaging spectroscopy-some future perspectives [C]. In Proc. of the6thEARSeL SIG IS workshop,2009:1-12.
[10] Stevenson B, Connor R, Kendall W et al. The civil air patrol ARCHER hyperspectral sensorsystem [C]. In Proc. of Airborne ISR Systems and Applications II, SPIE,2005:5787:17–28,
[11]李红波，舒嵘，薛永祺．PHI超光谱成像系统及其海洋遥感应用前景分析[J]．红外与毫米波学报，2002，21(6)：429-433．
[12]刘银年，薛永祺.实用型模块化成像光谱仪[J]．红外与毫米波学报，2002，21(1)：9-13.
[13]中国资源卫星应用中心．http://www.cresda.com.cn/n16/index.html
[14] http://www.ieee-whispers.com/
[15] Gao B C, Montes M J, Davis C O et al. Atmospheric correction algorithms for hyperspectralremote sensing data of land and ocean [J]. Remote Sensing of Environment,2009,113(S1):S17-S24.
[16] Cairns B, Carlson B E, Ruoxian Y et al. Atmospheric correction and its application to ananalysis of Hyperion data [J]. IEEE Transactions on Geoscience and Remote Sensing,2003,41(6):1232-1245.
[17] Malpica J A, Rejas J G, Alonso M C. A projection pursuit algorithm for anomaly detection inhyperspectral imagery [J]. Pattern Recognition,2008,41(11):3313-3327.
[18] Chen C M, Hepner G F, Forster R R. Fusion of hyperspectral and radar data using the HIStransformation to enhance urban surface features [J]. ISPRS Journal of Photogrammetry andRemote Sensing,2003,58(1-2):19-30.
[19] Pu R, Gong P. Wavelet transform applied to EO-1hyperspectral data for forest LAI andcrown closure mapping [J]. Remote Sensing of Environment,2004,91(2):212-224.
[20] Du Q. Unsupervised real-time constrained linear discriminant analysis to hyperspectral imageclassification [J].Pattern Recognition,2007,40(5):1510-1519.
[21] Tarabalka Y. Classification of hyperspectral data using spectral-spatial approach [D]. PhDthesis, joint by University of Iceland, Iceland and University of Grenoble, France.
[22] Mahesh P, Mather P M. Some issues in the classification of DAIS hyperspectral data [J].International Journal of Remote Sensing,2006,27(14):2895–2916.
[23]浦瑞良,宫鹏.高光谱遥感及其应用[M].北京:高等教育出版社.2000.
[24]张良培,张立福.高光谱遥感[M].武汉:武汉大学出版社.2005.
[25] Chang C I. Hyperspectral imaging: techniques for spectral detection and classification [M].Kluwer Academic&Plenum Publishers.2003
[26] Pramod V K, Arora M K. Advanced image processing techniques for remotely sensedhyperspectral data [M]. Springer-Verlag Berlin and Heidelberg GmbH&Co. K.2004.
[27] Melgani, F, Bruzzone L. Classification of hyperspectral remote sensing images with supportvector machines [J]. IEEE Transactions on Geoscience and Remote Sensing,2004,42(8):1778-1790.
[28] Benediktsson J A, Palmason J A, Sveinsson J R. Classification of hyperspectral data fromurban areas based on extended morphological profiles [J]. IEEE Transactions on Geoscienceand Remote Sensing,2005,43(3):480-491.
[29] Hughes G. On the mean accuracy of statistical pattern recognizers [J]. IEEE Transactions onInformation Theory,1968,14(1):55-63.
[30]熊桢,童庆禧,郑兰芬.用于高光谱遥感图象分类的一种高阶神经网络算法[J].中国图象图形学报,2000,5A(3):196-201
[31]王晋年,张兵,刘建贵等.以地物识别和分类为目标的高光谱数据挖掘[J].中国图象图形学报,1999,4A(11):957-964
[32] Lu D, Weng Q. A survey of image classification methods and techniques forimproving classification performance [J]. International Journal of Remote Sensing,2007(28):823-870.
[33] Campus-Valls G C, Bruzzone L. Kernel-based methods for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing,2005,43(6):1351-1362.
[34] Beguim D, Sarp E. Hyperspectral image classification using relevance vector machines [J].IEEE Geoscience and Remote Sensing Letters,2007,4(4):586-590.
[35] Mahesh P, Foody G M. Evaluation of SVM, RVM and SMLR for accurate imageclassification with limited ground data [J]. IEEE Journal of Selected Topics in Applied EarthObservations and Remote Sensing,2012,5(5):1344-1355.
[36] Du P, Xia J, Zhang W et al. Multiple classifier System for Remote Sensing ImageClassification: A Review [J]. Sensors,2012;12(4):4764-4792.
[37] Vapnik, V N. The Nature of Statistical Learning Theory [M]. Springer-Verlag,1995.
[38] Mianji, F A, Zhang Y. Robust hyperspectral classification using relevance vector machine [J].IEEE Transactions on Geoscience and Remote Sensing,2011,49(6):2100-2122.
[39] Suykens J, Horvath G, Basu S et al. Advances in learning theory: methods, models andapplications, computer and systems sciences [M]. IOS Press,2003.
[40] Kuncheva L I. Combining pattern classifiers: methods and algorithms [M].Wiley-Interscience,2004.
[41] Briem G J, Benediktsson J A, Sveinsson J R. Use of multiple classifiers in classification ofdata from multiple data sources [J]. IEEE Geoscience and Remote Sensing Symposium,2001.IGARSS, Volume2, pp882–884.
[42] Doan H T, Foody G M. Increasing soft classification accuracy through the use of anensemble of classifiers [J]. International Journal of Remote Sensing,2007,28(20):4606-4623.
[43] Benediktsson J A, Chanussot J, Fauvel M. Multiple classifier systems in remote sensing:From basics to recent developments [C]. In Multiple classifier systems,2007,4472:501–512.
[44] Rokach L. Ensemble-based classifiers [J]. Artificial Intelligence Review.2010,33(1-2).
[45] Du P, Liu S, Gamba P et al. Fusion of difference images for change detection over urbanareas [J]. IEEE Journal of Selected Topics in Applied Earth Observations and RemoteSensing,2012,5(4):1076-1086.
[46] Gholizadeh H Z, Mohammad J V, Mojaradi, B. A decision fusion approach for clusteringof hyperspectral data using spectral unmixing methods [C]. IEEE Aerospace Conference,2012.
[47] Dragi K, Saso D, Matt D W. Using single-and multi-target regression trees and ensembles tomodel a compound index of vegetation condition [J]. Ecological Modelling,2009,220(8):1159-1168
[48] Wolpert, D H. The supervised learning no-free-lunch theorems [C]. Proceedings of the6thOnline World Conference on Soft Computing in Industrial Applications,2001.
[49] Roli F, Giacinto G, Vernazza G. Comparison and combination of statistical and neuralnetwork algorithms for remote sensing image classification[C]. Neurocomputation in RemoteSensing Data Analysis. Berlin, Heidelberg: Springer-Verlag,1997.
[50] Gincinto G, Roli F. Ensembles of neural networks for soft classification of remote sensingimages[C]. Proceedings of the European Symposium on Intelligent Techniques. Bari, Italy,1997:166-170.
[51] Xu L, Krzyzak A, Suen CY. Methods of combining multiple classifiers and their applicationsto handwriting recognition [J]. IEEE Transactions on Systems, Man, and Cybernetics,1992,22(3):418–435.
[52]谢宗霞,于达仁,胡清华.多SVM分类器融合的传感器故障容错研究[J].哈尔滨工程大学学报,2006,27:389-342
[53]董火明,高隽,汪荣贵.多分类器融合的人脸识别与身份认证[J].系统仿真学报,2004,16(8):1849-1853
[54]廖闻剑,成瑜.基于多分类器决策融合的印鉴真伪鉴别方法[J].南京航空航天大学学报,2002,34(4):368-371
[55] Brian M S. Combining multiple classifiers: an application using spatial and remotely sensedinformation for land cover type mapping [J]. Remote Sensing of Environment,2000, No.74,pp545–556
[56] Tobias K, Volker C R, Kajetan P, Patrick H. Cross-border comparison of land cover andlandscape pattern in Eastern Europe using a hybrid classification technique [J]. RemoteSensing of Environment,2006,103:449–464.
[57]柏延臣,王劲峰.结合多分类器的遥感数据专题分类方法研究[J].遥感学报,2005,9(5):555-563.
[58] Hansen L K, Salamon P. Neural network ensembles [J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,1990,12(10):993-1001.
[59] Optiz D, Maclin R. Popular ensemble methods: an empirical study [J]. Journal of ArtificialIntelligence Research,1999,11:169-198.
[60] Kuncheva L I, Whitaker C J. Measures of diversity in classifier ensembles and theirrelationship with the ensemble accuracy [J]. Machine Learning,2003,51(2):181-207.
[61] Windeatt T. Diversity measures for multiple classifier system analysis and design [J].Information Fusion,2005,6(1):21-36.
[62] Banfield R E, Hall L, Bowyer K W et al. Ensemble diversity measures and their applicationto thinning [J]. Information Fusion,2005,6(1):49-62.
[63] Ruta D, Gabrys B. Classifier selection for majority voting [J]. Information Fusion,2005,6(1):63-81
[64] Tsymbal A, Pechenizkiy M, Cunningham P. Diversity in search strategies for ensemblefeature selection [J]. Information Fusion,2005,6(1):83-98.
[65]张春霞.集成学习中有关算法的研究[D].西安交通大学博士论文,2010.
[66] Breiman L. Bagging predictors [J]. Machine Learning,1996,24(2):123–140.
[67] Freund Y. Boosting a weak learning algorithm by majority [J]. Proceedings of the ThirdAnnual Workshop on Computational Learning Theory,1990.
[68] Richards J A, Jia X. Remote Sensing Digital Image Analysis: An Introduction [M].Springer-Verlag Berlin Heidelberg,2006.
[69] Ho T K. The random subspace method for constructing decision forest [J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,1998,20(8):832-844.
[70] Breiman L. Random forest [J]. Machine Learning,2001,45(1):5–32.
[71]王旭红,周明全,耿国华.―Dempster-Shafer‖证据理论在智能化遥感分类中应用研究[J].计算机应用与软件,2004,9:28-29.
[72]刘安斐,李弼程,张先飞.基于数据融合的多特征遥感图像分类[J].数据采集与处理，2006,12:463-467.
[73]孙怀江,胡钟山,杨静宇.基于证据理论的多分类器融合方法研究[J].计算机学报,2001.3:231-235.
[74]邓文胜,邵晓莉,刘海等.基于证据理论的遥感图像分类方法探讨[J].遥感学报,2007,7:568-573.
[75]石爱业,徐立中,杨先一等.基于神经网络-证据理论的遥感图像,数据融合与湖泊水质状况识别[J].中国图象图形学报,2005,3:372-377.
[76]石绥祥,夏登文,于戈.基于D-S理论的卫星遥感海表面温度和同化数据融合方法[J].海洋学报,2005,7:31-37.
[77]林剑,鲍光淑,林强.基于模糊分析的多光谱遥感图像的纹理特征[J],中南大学学报,2004.4,281-284.
[78]林剑,王润生,鲍光淑等.基于空间模糊纹理光谱的多光谱遥感图像分类方法[J].中国图形图像学报,2006,2:186-190.
[79]肖刚,敬忠良,李建勋等.一种基于模糊积分的图像最优融合方法[J].上海交通大学学报,2005,8:1312-1316.
[80] Benediktsson J A, Philip H S. Consensus theoretic classification methods [J]. IEEETransactions on Systems, Man, and Cybernetics,1992,22(4):688-704.
[81] Benediktsson J A, Sveinsson J R. Consensus based classification of multisource remotesensing data. Multiple Classifier Systems2000:280-289
[82] Smits P C, Multiple classifier systems for supervised remote sensing image classificationbased on dynamic classifier selection [J]. IEEE Transactions of Geosciences and remotesensing,2002,40(4),801-813.
[83] Zhou Z H, Wu J X, Tang W. Ensembling neural networks: many could be better than all [J].Artificial Intelligence,2002,137(1-2):239-263.
[84]张春霞，张讲社.选择性集成学习算法综述[J].计算机学报,2011,34(8):1399-1410.
[85] Briem G J, Benediktsson J A, Sveinsson J R. Multiple classifiers applied to multisourceremote sensing data [J]. IEEE Transactions on Geoscience and Remote Sensing,2002,40(10):2291-2299
[86] Briem G J, Benediktsson J A, Sveinsson J R. Boosting, bagging, and consensus basedclassification of multisource remote sensing data [J]. Multiple Classifier Systems,2001,2096:279-288
[87] Foody G M, Boyd D S, Sanchez-Hernandez C. Mapping a specific class with an ensemble ofclassifiers [J], International Journal of Remote Sensing,2007,28(8):1733-1746.
[88] Olivier D, Patrice L, Isabelle V D S. Remote sensing classification of spectral, spatial andcontextual data using multiple classifier systems [J]. Image Anal Sterol,2001,20(Suppl1):584-589.
[89] Waske B, Braun M. Classifier ensembles for land cover mapping using multitemporal SARimagery [J]. ISPRS Journal of Photogrammetry and Remote Sensing,2009,64:450-457.
[90] Nemmour H, Chibani Y. Multiple support vector machines for land cover change detection:an application for mapping urban extensions [J]. ISPRS Journal of Photogrammetry andRemote Sensing,2006,61:25-33.
[91] Mustard J F, Li L, He G. Nolinear spectral mixture modeling of lunar multispectral data:implications for lateral transport [J]. Journal of Geophysical Research,1998,103(E8):19419-19425.
[92] Jolliffe T. Principal component analysis [M]. NewYork: Springer-Verlag,1986.
[93] Green A A, Berman M, Switzer P, et al. A transformation for ordering multispectral data interms of image quality with implication for noise removal [J]. IEEE Transactions onGeoscience and Remote Sensing,1988,26(1):65-74.
[94] Jimenez L O, Landgrede D A. Hyperspectral data analysis and supervised feature reductionvia projection pursuit [J]. IEEE Transactions on Geoscience and Remote Sensing,1999,37(6):2653-2667.
[95] Bayliss J, Gualtieri J A, Cromp R. Analysing hyperspectral data with independent componentanalysis [C]. Proc. SPIE,1997,3240:133-143.
[96] Kumar S, Ghosh J, Crawford M M. Best-bases feature extraction algorithms for classificationof hyperspectral data [J]. IEEE Transactions on Geoscience and Remote Sensing,2001,39(7):1368-1379.
[97] Kegl B, Krzyzak A, Linder T. Learning and design of principal curves [J]. IEEE Transactionon Pattern Analysis and Machine Intelligence,2000,22(3):281-297.
[98] Duda R O, Hart P E, Stork D G. Pattern Classification [M]. New York: Wiley Press,2001.
[99] Nie F P, Xu D, Tsang I W et al. Flexible manifold embedding: a framework forsemi-supervised and unsupervised dimension reduction [J]. IEEE Transactions on ImageProcessing,2010,19(7):1921–1932.
[100] Liao W, Pizurica A, Scheunders P et al. Semisupervised local discriminant analysis forfeature extraction in hyperspectral images. IEEE Transactions on Geoscience and RemoteSensing,2013,51(1):184–198.
[101] Xia J, Chanussot J, Du P et al. Semi-supervised dimensionality reduction for hyperspectralremote sensing image classification. In Proceedings of WHISPERS,2012, pp.1–4.
[102] Gualtieri J A, Cromp R F. Support vector machine for hyperspectral remote sensingclassification [C]. In Proceeding of the SPIE,27thAIPR Workshop,1998,3584:221-232.
[103] Campus-Valls G, Martin-Guerrero J D, Rojo-Alvarez J L et al. Fuzzy sigmoid kernel forsupport vector classifiers [J]. Neurocomputing,2004,62(6):501-506.
[104] Campus-Valls G, Gomez-Chova L, Munoz-Mari J et al. Composite kernels forhyperspectral image classification [J]. IEEE Geoscience and Remote Sensing Letters,2006,3(1):93-97.
[105] Bruzzone L, Chi M, Marconcini M. A novel transductive SVM for semi-supervisedclassification of remote sensing images [J]. IEEE Transactions on Geoscience and RemoteSensing,2006,44(11):3363-3373.
[106] Mountrakis G, Im J, Ogole C. Support vector machines in remote sensing: A review [J].ISPRS Journal of Photogrammetry and Remote Sensing,2011,66(3):247-259.
[107] Bishop C M. Pattern Recognition and Machine Learning [M]. Springer,2007.
[108] Waske B, Linden S V D, Benediktsson J A et al. Sensitivity of support vector machines torandom feature selection in classification of hyperspectral data [J]. IEEE Transactions onGeoscience and Remote Sensing.2010,48(7):2880-2889.
[109] Gislason P, Benediktsson J A, Sveinsson J R. Random forests for land cover classification[J]. Pattern Recognition Letters,2006,27(4):294–300.
[110] Ham J, Chen Y, Crawford M M et al. Investigation of the random forest framework forclassification of hyperspectral data [J]. IEEE Transactions on Geoscience and RemoteSensing,2005,43(3):492–501.
[111] Lawrence R L, Wood S D, Sheley R L. Mapping invasive plants using hyperspectralimagery and Breiman Cutler classifications (random Forest)[J]. Remote Sensing ofEnvironment,2006,100(3):356-362.
[112] Chan J C, Paelinckx D. Evaluation of Random Forest and Adaboost tree-based ensembleclassification and spectral band selection for ecotope mapping using airborne hyperspectralimagery [J]. Remote Sensing of Environment,2008,112:2999-3011.
[113] Tarabalka, Y, Benediktsson J A, Chanussot, J, Tilton, J. Multiple spectral-spatialclassification approach for hyperspectral data [J]. IEEE Transactions on Geoscience andRemote Sensing,2010,48(11):4122–4132.
[114] Favuel M. Classification of urban areas using high resolution satellite remote seningimages [M]. PhD thesis. Joint by University of Iceland, Iceland and University of Grenoble,France.
[115] Kettig R L, Landgrebe D A. Classification of multispectral image data by extraction andclassification of homogeneous objects [J]. IEEE Transactions on Geoscience Electronics,1976,14(1):19–26.
[116] Pony O, Descombres X, Zerubia J. Classification d’images satellitaires hyperspectrales enzone rurale et periurbane. Technical report, Sophia Antipolis, France, Sept.2000.
[117] Jackson Q, Landgrebe D A. Adaptive bayesian contextual classification based on Markovrandom fields [J]. IEEE Transactions on Geoscience and Remote Sensing,2000,40(11):2454–2463.
[118] Tsai F, Chang C L, Liu G R. Texture analysis for three dimension remote sensing data3DGLCM [C]. In Proc. of27th Asian Conference on Remote Sensing, pages1–6, August2006.
[119] Farag A A, Mohamed R M, El-Baz A. A unified framework for MAP estimation in remotesensing image segmentation [J]. IEEE Transactions on Geoscience and Remote Sensing,2005,43(7):1617–1634.
[120] Dell’Acqua F, Gamba P, Ferrari A et al. Exploiting spectral and spatial information inhyperspectral urban data with high resolution [J]. IEEE Geoscience and Remote SensingLetters,2004,1(4):322–326.
[121] Fauvel M, Chanussot J, Benediktsson J A, Sveinsson J R. Spectral and spatial classificationof hyperspectral data using SVMs and morphological profiles [J]. IEEE Transactions onGeoscience and Remote Sensing,2008,46(11):3804–3814.
[122] Linden S V D, Janz A, Waske B et al. Classifying segmented hyperspectral data from aheterogeneous urban environment using support vector machines [J]. Journal of AppliedRemote Sensing,2007,1(1):013543.
[123]刘志刚,李德仁,秦前清等.支持向量机在多类分类问题中的推广[J].计算机工程与应用,2004,7:10-13.
[124]沈照庆,舒宁,陶建斌.一种基于NPA的加权―1v M‖SVM高光谱影像分类算法[J].武汉大学学报（信息科学版）,2009,12:1444-1447.
[125]杨国鹏,余旭初,周欣等.基于相关向量机的高光谱影像分类研究[J].测绘学报,2010,39(6):572-578.
[126]宋相法,焦李成.基于稀疏表示及光谱信息的高光谱遥感图像分类[J].电子与信息学报,2012,34(2):268-272.
[127]谭琨,杜培军.基于再生核Hilbert空间小波核函数支持向量机的高光谱遥感影像分类[J].测绘学报,2011,40(2):572-578.
[128]吴婉澜,皮亦鸣,何强. AdaBoost算法实现全极化SAR图像分类[J].信号处理,2009,25(10),1594-1597.
[129]杨艺,韩崇昭,韩德强.利用特征子空间评价与多分类器融合的高光谱图像分类[J].西安交通大学学报,2010,8:20-24.
[130]孙亮,韩崇昭,沈建京.集成特征选择的广义粗集方法与多分类器融合[J].自动化学报,2008,34(3):298-304
[131] Woods K, Kegelmeyer W P, Bowyer K. Combination of multiple classifiers using localaccuracy estimates [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19,405–410.
[132] Czyz J, Kittler J, Vandendorpe L. Multiple classifier combination for face-based identityverification [J]. Pattern Recognition,2003,37,1459–1469.
[133] Rodriguez J J, Kuncheva L I. Rotation forest: A new classifier ensemble method [J]. IEEETransactions on Pattern Analysis and Machine Intelligence.2006,28:1619–1630.
[134] Benediktsson J A, Kanellopoulos I. Classification of multisource and hyperspectral databased on decision fusion [J]. IEEE Transactions on Geoscience and Remote Sensing,1999,37:1367–1377.
[135] Fauvel M, Chanussot J, Benediktsson J A. Decision fusion for the classification of urbanremote sensing images [J]. IEEE Transactions on Geoscience and Remote Sensing.2006,44:2828–2838.
[136] Zhou Z H. When Semi-supervised learning meets ensemble learning [C]. In Proceedings ofthe8thInternational Workshop on Multiple Classifier Systems Reykjavik, Iceland, June10-12,2009; Volume5, pp.529–538.
[137] Brown, G. Ensemble learning [C]. In Encyclopedia of Machine Learning; Webb, G.I.,Sammut, C., Eds.; Springer: New York, NY, USA,2010.
[138] Dietterich T G. Ensemble Methods in Machine Learning [C].2000. In J. Kittler and F. Roli(Ed.) First International Workshop on Multiple Classifier Systems, Lecture Notes inComputer Science (pp.1-15). New York: Springer Verlag.
[139] Dietterich T G. Ensemble learning [C]. In the Handbook of Brain Theory and NeuralNetworks, Second Edition,2002.
[140] Valentini G, Masulli F. Ensembles of learning machines [C]. In Neural Nets WIRNVietri-02, Series Lecture Notes in Computer Sciences, Marinaro M and Tagliaferri R, Eds.:Springer-Verlag, Heidelberg (Germany),2002,
[141] Seung H S, Opper M, Sompolinsky H. Query by committee [C]. In Proceedings of the FifthWorkshop on Computaional Learning Theory, pages287--294, San Mateo, California,1992.Morgan Kaufmann.
[142] Kittler J, Hatef M, Duin R P, Matas J. On combining classifiers [J]. IEEE Transactions onPattern Analysis and Machine Intelligence,1998,20(3):226-239.
[143] Kearns M, Vazirani U. An introduction to computational learning theory [M]. MIT Press,1994.
[144] Kearns M J, Valiant L G. Learning Boolean formulae or factoring. Aiken ComputationLaboratory, Harvard University, Cambridge, MA, Technical Report: TR-1488,1988.
[145] Valiant L G. A theory of the learnable [J]. Communications of the ACM,1984,27(11):1134-1142.
[146] Schapire R E. The strength of weak learnability [J]. Machine Learning,5(2):197-227,1990.
[147] Freund Y, Schapire R E. A decision-theoretic generalization of on-line learning and anapplication to boosting [J]. Journal of Computer and System Sciences,1997,55(1):119–139.
[148] Schapire R E, Singer Y. Improved boosting algorithms using confidence-rated predictions[J]. Machine Learning,37(3):297-336,1999.
[149] Demiriz A, Bennett K P, Taylor J H. Linear programming boosting via column generation[J]. Machine Learning,46, pages225–254,2002.
[150] Freund Y. An adaptive version of the boost by majority algorithm [J]. Machine Learning,43(3):293–318,2001.
[151] Domingos P, Pazzani M. On the optimality of the simple bayesian classifier under zero-oneloss [J]. Machine Learning,1997,29(2-3):103-130.
[152] Holte R C. Very simple classification rules perform well on most commonly used datasets[J]. Machine Learning,1993,11(1):63-90.
[153] Geman S, Bienenstock E, Doursat R. Neural networks and the bias/variance dilemma [J].Neural Computation,1992,4(1):1-58.
[154] Wahba G, Lin X, Gao F et al. The bias-variance trade of and the randomized gacv [C]. InProceedings of the1998conference on Advances in neural information processing systems II,pages620-626,1999.
[155] Brown G, Wyatt J, Tino P. Managing diversity in regression ensembles [J]. Journal ofMachine Learning Research,2005,6:1621-1650.
[156] Islam M M, Yao X, Murase K. A constructive algorithm for training cooperative neuralnetwork ensembles [J]. IEEE Transaction on Neural Networks,2003,14(4):820-834,2003.
[157] Liu Y, Yao X. Simultaneous training of negatively correlated neural networks in anensemble [J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics,1999a,29(6):716-725.
[158] Liu Y, Yao X. Ensemble learning via negative correlation [J]. Neural Networks,1999b,12(10):1399-1404,
[159] Ueda P, Nakano R. Generalization error of ensemble estimators [C]. In Proceedings ofInternational Conference on Neural Networks, pages90-95,1996.
[160]焦李成,公茂国,王爽等.自然计算、机器学习与图像理解前沿[M].西安电子科技大学出版社,西安,2008.
[161] Ricardo V, Youssef D.: A perspective view and survey of meta-learning [J]. ArtificialIntelligence Review,18(2):77–95,2002.
[162] Kearns M, Mansour Y. On the boosting ability of top-down decision tree learningalgorithms [C]. In Proceedings of the Twenty-Eighth Annual ACM Symposium on theTheory of Computing,1996.
[163] Kuncheva L I, Rodriguez J J. An experimental study on rotation forest ensembles [C]. InProceedings of the7th International Workshop on Multiple Classifier Systems,2007, pp.459–468.
[164] Ozcift A, Gulten A. Classifier ensemble construction with rotation forest to improvemedical diagnosis performance of machine learning algorithms [J]. Computer Methods andPrograms in Biomedicine,2011,104(3):443–451.
[165] Breiman L, Friedman J, Olshen R et al. Classification and regression trees [M]. Boca Raton,FL: CRC Press,1984.
[166] Loh W Y. Classification and regression tree methods [C]. In Encyclopedia of Statistics inQuality and Reliability, Ruggeri, Kenett and Faltin (eds.),2008, pp.315–323.
[167] Jutten C, Herault J. Independent component analysis verus principal component analysis[C]. In Proceedings of Europe Signal Processing Conf. EUSIPC088.1988.
[168]易尧华，余长慧，秦前清．基于独立分量分析的遥感影像非监督分类方法[J]．武汉大学学报(信息科学版)，2005，30(1)：19-23．
[169]曾生根，王小敏，范瑞彬．基于独立分量分析的遥感图像分类技术[J]．遥感学报，2004，8(2)：150-157.
[170] Sugiyama M. Dimensionality reduction of multimodal labeled data by local Fisherdiscriminant analysis [J]. Journal of Machine Learning Research,2007,8,1027–1061.
[171] Fukunaga K. Introduction to statistical pattern recognition [M]. Boston: Academic Press,Inc. Second edition.1990.
[172] He X, Niyogi P. Locality preserving projections. Advances in Neural InformationProcessing Systems, Cambridge, MA: MIT Press.2004.
[173] Li J, Bioucas-Dias J, Plaza A. Spectral-spatial classification of hyperspectral data usingloopy belief propagation and active learning [J]. IEEE Transaction on Geoscience andRemote Sensing, in press
[174] Chang C, Lin C. LIBSVM: A library for Support Vector Machines. Software available athttp://www.csie.ntu.edu.tw/cjlin/libsvm,2010.
[175] Bachmann M, Ainsworth T, Fusina R. Exploiting manifold geometry in hyperspectralimagery [J]. IEEE Transaction Geoscience and Remote Sensing,2005,43(3):441-454
[176]陈诗国,张道强.非监督降维方法的实验比较[J].软件学报,2011,22(1):28~42
[177] Kim K, Jung K, Kim H et al. Face recognition using kernel principal component analysis[J]. IEEE Image Processing Letters,2002,9(2):40-42
[178] Kuo B C, Landgrebe D A. Nonparametric Weighted Feature extraction for classification [J].IEEE Transaction on Geoscience and Remote Sensing,2004,42(5):1096-1105.
[179] Zhang D Q, Zhou Z H, Chen S C. Semi-supervised dimensionality reduction [C]. InProceedings of the7th SIAM International Conference on Data Mining,2007, pp.387–393.
[180] Sugiyama M, Id T, Nakajima S et al. Semi-supervised local fisher discriminant analysis fordimensionality reduction [J]. Machine Learning,2010,78:35–61.
[181] Yu S P, Yu K, Tresp V et al. Supervised probabilistic principal component analysis [C]. InProceedings of the12th ACM SIGKDD international conference on Knowledge discoveryand data mining,2006, pp.464–473
[182] Tipping M E, Bishop C M. Probabilistic principal component analysis [J]. BMCBioinformatics,1999,61(3):611–622.
[183] Paterlini S, Krink T. Differential evolution and particle swarm optimisation in partitionalclustering [J]. Computational Statistics and Data Analysis,2006,50(5):1220–1247.
[184] Pal N R, Pal S K. A review on image segmentation techniques [J]. Pattern Recognition,1993,26(9):1277–1294.
[185] Fu K S, Mui J K. A survey on image segmentation [J]. Pattern Recognition,1981,13(1):3-16.
[186] MacQueen J B. Some methods for classification and analysis of multivariate observations
[C]. In Proc. of the fifth Berkeley Symposium on Mathematical Statistics and Probability, vol.1,1967, pp.281–297.
[187] Jain A K. Data clustering:50years beyond k-means [J]. Pattern Recognition Letters,2010,31(8):651–666.
[188] Dunn J C. A fuzzy relative of the ISODATA process and its use in detecting compactwell-separated clusters [J]. Cybernetics and Systems,1973,3(3):32–57.
[189] Bezdek J, Ehrlich R, Full W. FCM: The fuzzy c-means clustering algorithm Computers andGeosciences [J].1984,10(2-3):191–203.
[190] Gath I, Geva A B. Unsupervised optimal fuzzy clustering [J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,1989,11(7):773-780.
[191] Dhillon I S, Guan Y, Kulis B. Kernel k-means: spectral clustering and normalized cuts [C].In Proceedings of the Tenth ACM SIGKDD International Conference on KnowledgeDiscovery and Data Mining, Seattle, Washington, USA, August22-25,2004,2004.
[192] Shawe-Taylor J S, Cristianini N. Kernel Methods for Pattern Analysis [M]. New York, NY,USA: Cambridge University Press,2004.
[193] Ball G B, Hall D J. ISODATA, a novel method of data analysis and pattern classification.Technical report, Stanford, Stanford Research Institute,1965.
[194] Tarabalka Y, Benediktsson J A, Chanussot J. Spectral-spatial classification ofhyperspectral imagery based on partitional clustering techniques [J]. IEEE Transactions onGeoscience and Remote Sensing,2009,47(8):2973–2987.
[195] Canty M. Image Analysis, Classification and Change Detection in Remote Sensing, withAlgorithms for ENVI/IDL, Second Revised Edition [M]. Taylor&Francis, CRC Press2010
[196] Vincent L, Soille P. Watersheds in digital spaces: an efficient algorithm based onimmersion simulations [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1991,13(6):583–598.
[197] Osma-Ruiz V, Godino-Llorente J I, Soenz-Lechn N et al. An improved watershedalgorithm based on efficient computation of shortest paths [J]. Pattern Recognition,2007,40(3):1078–1090.
[198] Serra J. Image Analysis and Mathematical Morphology [J]. New York, NY, USA:Academic Press,1982
[199] Tarabalka Y, Chanussot J, Benediktsson J A. Segmentation and classification ofhyperspectral images using watershed transformation [J]. Pattern Recognition,2010,43(7):2367–2379.
[200] Soille P. Morphological Image Analysis, second edition [M]. Berlin, Germany: Springer,2003.
[201] Comaniciu D, Meer P. Mean shift: a robust approach toward feature space analysis [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603–619.
[202] Kirkpatrick S, Gelatt C D, Vecch M P. Optimization by simmulated annealing [J]. Science,1983,220(4598):671-680.
[203] Yuan F，Bauer M E．Comparison of impervious surface area and normalized differencevegetation index as indicators of surface urban heat island effects in Landsat imagery[J]．Remote sensing of Environment，2007(16)：375-386．
[204] Ridd M K．Exploring a V-I-S (vegetation-impervious surface-soil) model for urbanecosystem analysis through remote sensing：comparative anatomy for cities [J]．InternationalJournal of Remote Sensing，1995，16(12)：2165-2185．
[205] Small C．Estimation and vicarious validation of urban vegetation abundance by spectralmixture analysis [J]．International Journal of Remote Sensing，2001，22(7)：1305-1334．
[206] Weng Q．Remote Sensing of Impervious Surfaces[M]．CRC Press，2008．
[207]夏俊士，杜培军，逄云峰等.基于高光谱数据的城市不透水层提取与分析[J].中国矿业大学学报,2011,4:660-666.
[2]童庆禧,张兵,郑兰芬.高光谱遥感—原理、技术及应用[M].北京:高等教育出版社.2006.
[3] Cocks T, Jenssen R, Stewart A et al. The HyMap airborne hyperspectral sensor: the system,calibration and performance [C]. In Proc. of First EARSEL Workshop on ImagingSpectroscopy, October1998.
[4] Gamba P. A collection of data for urban area characterization [C]. In Proc. of IGARSS’04,Anchorage, USA, Sept.2004.
[5] Green R O, Eastwood M L, Sarture C M et al. Imaging spectroscopy and the AirborneVisible Infrared Imaging Spectrometer (AVIRIS)[J]. Remote Sensing of Environment,1998,65(3):227-248.
[6] Pearlman J, Segal C, Liao L et al. Development and operations of theEO-1Hyperionimaging spectrometer [C]. In Proc. of Earth Observing Systems V, SPIE, volume4135, pages243–253,2000.
[7] Plaza A, Benediktsson J A, Boardman J et al. Recent advances in techniques forhyperspectral image processing [J]. Remote Sensing of Environment,2009,113(S1):110-122.
[8] Rickard L, Basedow W, Zalewski E et al. HYDICE: An airborne system for hyperspectralimaging [C]. In Proc. of Imaging Spectrometry of the Terrestrial Environment, SPIE,19931937:173–179.
[9] Staenz K. Terrestrial imaging spectroscopy-some future perspectives [C]. In Proc. of the6thEARSeL SIG IS workshop,2009:1-12.
[10] Stevenson B, Connor R, Kendall W et al. The civil air patrol ARCHER hyperspectral sensorsystem [C]. In Proc. of Airborne ISR Systems and Applications II, SPIE,2005:5787:17–28,
[11]李红波，舒嵘，薛永祺．PHI超光谱成像系统及其海洋遥感应用前景分析[J]．红外与毫米波学报，2002，21(6)：429-433．
[12]刘银年，薛永祺.实用型模块化成像光谱仪[J]．红外与毫米波学报，2002，21(1)：9-13.
[13]中国资源卫星应用中心．http://www.cresda.com.cn/n16/index.html
[14] http://www.ieee-whispers.com/
[15] Gao B C, Montes M J, Davis C O et al. Atmospheric correction algorithms for hyperspectralremote sensing data of land and ocean [J]. Remote Sensing of Environment,2009,113(S1):S17-S24.
[16] Cairns B, Carlson B E, Ruoxian Y et al. Atmospheric correction and its application to ananalysis of Hyperion data [J]. IEEE Transactions on Geoscience and Remote Sensing,2003,41(6):1232-1245.
[17] Malpica J A, Rejas J G, Alonso M C. A projection pursuit algorithm for anomaly detection inhyperspectral imagery [J]. Pattern Recognition,2008,41(11):3313-3327.
[18] Chen C M, Hepner G F, Forster R R. Fusion of hyperspectral and radar data using the HIStransformation to enhance urban surface features [J]. ISPRS Journal of Photogrammetry andRemote Sensing,2003,58(1-2):19-30.
[19] Pu R, Gong P. Wavelet transform applied to EO-1hyperspectral data for forest LAI andcrown closure mapping [J]. Remote Sensing of Environment,2004,91(2):212-224.
[20] Du Q. Unsupervised real-time constrained linear discriminant analysis to hyperspectral imageclassification [J].Pattern Recognition,2007,40(5):1510-1519.
[21] Tarabalka Y. Classification of hyperspectral data using spectral-spatial approach [D]. PhDthesis, joint by University of Iceland, Iceland and University of Grenoble, France.
[22] Mahesh P, Mather P M. Some issues in the classification of DAIS hyperspectral data [J].International Journal of Remote Sensing,2006,27(14):2895–2916.
[23]浦瑞良,宫鹏.高光谱遥感及其应用[M].北京:高等教育出版社.2000.
[24]张良培,张立福.高光谱遥感[M].武汉:武汉大学出版社.2005.
[25] Chang C I. Hyperspectral imaging: techniques for spectral detection and classification [M].Kluwer Academic&Plenum Publishers.2003
[26] Pramod V K, Arora M K. Advanced image processing techniques for remotely sensedhyperspectral data [M]. Springer-Verlag Berlin and Heidelberg GmbH&Co. K.2004.
[27] Melgani, F, Bruzzone L. Classification of hyperspectral remote sensing images with supportvector machines [J]. IEEE Transactions on Geoscience and Remote Sensing,2004,42(8):1778-1790.
[28] Benediktsson J A, Palmason J A, Sveinsson J R. Classification of hyperspectral data fromurban areas based on extended morphological profiles [J]. IEEE Transactions on Geoscienceand Remote Sensing,2005,43(3):480-491.
[29] Hughes G. On the mean accuracy of statistical pattern recognizers [J]. IEEE Transactions onInformation Theory,1968,14(1):55-63.
[30]熊桢,童庆禧,郑兰芬.用于高光谱遥感图象分类的一种高阶神经网络算法[J].中国图象图形学报,2000,5A(3):196-201
[31]王晋年,张兵,刘建贵等.以地物识别和分类为目标的高光谱数据挖掘[J].中国图象图形学报,1999,4A(11):957-964
[32] Lu D, Weng Q. A survey of image classification methods and techniques forimproving classification performance [J]. International Journal of Remote Sensing,2007(28):823-870.
[33] Campus-Valls G C, Bruzzone L. Kernel-based methods for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing,2005,43(6):1351-1362.
[34] Beguim D, Sarp E. Hyperspectral image classification using relevance vector machines [J].IEEE Geoscience and Remote Sensing Letters,2007,4(4):586-590.
[35] Mahesh P, Foody G M. Evaluation of SVM, RVM and SMLR for accurate imageclassification with limited ground data [J]. IEEE Journal of Selected Topics in Applied EarthObservations and Remote Sensing,2012,5(5):1344-1355.
[36] Du P, Xia J, Zhang W et al. Multiple classifier System for Remote Sensing ImageClassification: A Review [J]. Sensors,2012;12(4):4764-4792.
[37] Vapnik, V N. The Nature of Statistical Learning Theory [M]. Springer-Verlag,1995.
[38] Mianji, F A, Zhang Y. Robust hyperspectral classification using relevance vector machine [J].IEEE Transactions on Geoscience and Remote Sensing,2011,49(6):2100-2122.
[39] Suykens J, Horvath G, Basu S et al. Advances in learning theory: methods, models andapplications, computer and systems sciences [M]. IOS Press,2003.
[40] Kuncheva L I. Combining pattern classifiers: methods and algorithms [M].Wiley-Interscience,2004.
[41] Briem G J, Benediktsson J A, Sveinsson J R. Use of multiple classifiers in classification ofdata from multiple data sources [J]. IEEE Geoscience and Remote Sensing Symposium,2001.IGARSS, Volume2, pp882–884.
[42] Doan H T, Foody G M. Increasing soft classification accuracy through the use of anensemble of classifiers [J]. International Journal of Remote Sensing,2007,28(20):4606-4623.
[43] Benediktsson J A, Chanussot J, Fauvel M. Multiple classifier systems in remote sensing:From basics to recent developments [C]. In Multiple classifier systems,2007,4472:501–512.
[44] Rokach L. Ensemble-based classifiers [J]. Artificial Intelligence Review.2010,33(1-2).
[45] Du P, Liu S, Gamba P et al. Fusion of difference images for change detection over urbanareas [J]. IEEE Journal of Selected Topics in Applied Earth Observations and RemoteSensing,2012,5(4):1076-1086.
[46] Gholizadeh H Z, Mohammad J V, Mojaradi, B. A decision fusion approach for clusteringof hyperspectral data using spectral unmixing methods [C]. IEEE Aerospace Conference,2012.
[47] Dragi K, Saso D, Matt D W. Using single-and multi-target regression trees and ensembles tomodel a compound index of vegetation condition [J]. Ecological Modelling,2009,220(8):1159-1168
[48] Wolpert, D H. The supervised learning no-free-lunch theorems [C]. Proceedings of the6thOnline World Conference on Soft Computing in Industrial Applications,2001.
[49] Roli F, Giacinto G, Vernazza G. Comparison and combination of statistical and neuralnetwork algorithms for remote sensing image classification[C]. Neurocomputation in RemoteSensing Data Analysis. Berlin, Heidelberg: Springer-Verlag,1997.
[50] Gincinto G, Roli F. Ensembles of neural networks for soft classification of remote sensingimages[C]. Proceedings of the European Symposium on Intelligent Techniques. Bari, Italy,1997:166-170.
[51] Xu L, Krzyzak A, Suen CY. Methods of combining multiple classifiers and their applicationsto handwriting recognition [J]. IEEE Transactions on Systems, Man, and Cybernetics,1992,22(3):418–435.
[52]谢宗霞,于达仁,胡清华.多SVM分类器融合的传感器故障容错研究[J].哈尔滨工程大学学报,2006,27:389-342
[53]董火明,高隽,汪荣贵.多分类器融合的人脸识别与身份认证[J].系统仿真学报,2004,16(8):1849-1853
[54]廖闻剑,成瑜.基于多分类器决策融合的印鉴真伪鉴别方法[J].南京航空航天大学学报,2002,34(4):368-371
[55] Brian M S. Combining multiple classifiers: an application using spatial and remotely sensedinformation for land cover type mapping [J]. Remote Sensing of Environment,2000, No.74,pp545–556
[56] Tobias K, Volker C R, Kajetan P, Patrick H. Cross-border comparison of land cover andlandscape pattern in Eastern Europe using a hybrid classification technique [J]. RemoteSensing of Environment,2006,103:449–464.
[57]柏延臣,王劲峰.结合多分类器的遥感数据专题分类方法研究[J].遥感学报,2005,9(5):555-563.
[58] Hansen L K, Salamon P. Neural network ensembles [J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,1990,12(10):993-1001.
[59] Optiz D, Maclin R. Popular ensemble methods: an empirical study [J]. Journal of ArtificialIntelligence Research,1999,11:169-198.
[60] Kuncheva L I, Whitaker C J. Measures of diversity in classifier ensembles and theirrelationship with the ensemble accuracy [J]. Machine Learning,2003,51(2):181-207.
[61] Windeatt T. Diversity measures for multiple classifier system analysis and design [J].Information Fusion,2005,6(1):21-36.
[62] Banfield R E, Hall L, Bowyer K W et al. Ensemble diversity measures and their applicationto thinning [J]. Information Fusion,2005,6(1):49-62.
[63] Ruta D, Gabrys B. Classifier selection for majority voting [J]. Information Fusion,2005,6(1):63-81
[64] Tsymbal A, Pechenizkiy M, Cunningham P. Diversity in search strategies for ensemblefeature selection [J]. Information Fusion,2005,6(1):83-98.
[65]张春霞.集成学习中有关算法的研究[D].西安交通大学博士论文,2010.
[66] Breiman L. Bagging predictors [J]. Machine Learning,1996,24(2):123–140.
[67] Freund Y. Boosting a weak learning algorithm by majority [J]. Proceedings of the ThirdAnnual Workshop on Computational Learning Theory,1990.
[68] Richards J A, Jia X. Remote Sensing Digital Image Analysis: An Introduction [M].Springer-Verlag Berlin Heidelberg,2006.
[69] Ho T K. The random subspace method for constructing decision forest [J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,1998,20(8):832-844.
[70] Breiman L. Random forest [J]. Machine Learning,2001,45(1):5–32.
[71]王旭红,周明全,耿国华.―Dempster-Shafer‖证据理论在智能化遥感分类中应用研究[J].计算机应用与软件,2004,9:28-29.
[72]刘安斐,李弼程,张先飞.基于数据融合的多特征遥感图像分类[J].数据采集与处理，2006,12:463-467.
[73]孙怀江,胡钟山,杨静宇.基于证据理论的多分类器融合方法研究[J].计算机学报,2001.3:231-235.
[74]邓文胜,邵晓莉,刘海等.基于证据理论的遥感图像分类方法探讨[J].遥感学报,2007,7:568-573.
[75]石爱业,徐立中,杨先一等.基于神经网络-证据理论的遥感图像,数据融合与湖泊水质状况识别[J].中国图象图形学报,2005,3:372-377.
[76]石绥祥,夏登文,于戈.基于D-S理论的卫星遥感海表面温度和同化数据融合方法[J].海洋学报,2005,7:31-37.
[77]林剑,鲍光淑,林强.基于模糊分析的多光谱遥感图像的纹理特征[J],中南大学学报,2004.4,281-284.
[78]林剑,王润生,鲍光淑等.基于空间模糊纹理光谱的多光谱遥感图像分类方法[J].中国图形图像学报,2006,2:186-190.
[79]肖刚,敬忠良,李建勋等.一种基于模糊积分的图像最优融合方法[J].上海交通大学学报,2005,8:1312-1316.
[80] Benediktsson J A, Philip H S. Consensus theoretic classification methods [J]. IEEETransactions on Systems, Man, and Cybernetics,1992,22(4):688-704.
[81] Benediktsson J A, Sveinsson J R. Consensus based classification of multisource remotesensing data. Multiple Classifier Systems2000:280-289
[82] Smits P C, Multiple classifier systems for supervised remote sensing image classificationbased on dynamic classifier selection [J]. IEEE Transactions of Geosciences and remotesensing,2002,40(4),801-813.
[83] Zhou Z H, Wu J X, Tang W. Ensembling neural networks: many could be better than all [J].Artificial Intelligence,2002,137(1-2):239-263.
[84]张春霞，张讲社.选择性集成学习算法综述[J].计算机学报,2011,34(8):1399-1410.
[85] Briem G J, Benediktsson J A, Sveinsson J R. Multiple classifiers applied to multisourceremote sensing data [J]. IEEE Transactions on Geoscience and Remote Sensing,2002,40(10):2291-2299
[86] Briem G J, Benediktsson J A, Sveinsson J R. Boosting, bagging, and consensus basedclassification of multisource remote sensing data [J]. Multiple Classifier Systems,2001,2096:279-288
[87] Foody G M, Boyd D S, Sanchez-Hernandez C. Mapping a specific class with an ensemble ofclassifiers [J], International Journal of Remote Sensing,2007,28(8):1733-1746.
[88] Olivier D, Patrice L, Isabelle V D S. Remote sensing classification of spectral, spatial andcontextual data using multiple classifier systems [J]. Image Anal Sterol,2001,20(Suppl1):584-589.
[89] Waske B, Braun M. Classifier ensembles for land cover mapping using multitemporal SARimagery [J]. ISPRS Journal of Photogrammetry and Remote Sensing,2009,64:450-457.
[90] Nemmour H, Chibani Y. Multiple support vector machines for land cover change detection:an application for mapping urban extensions [J]. ISPRS Journal of Photogrammetry andRemote Sensing,2006,61:25-33.
[91] Mustard J F, Li L, He G. Nolinear spectral mixture modeling of lunar multispectral data:implications for lateral transport [J]. Journal of Geophysical Research,1998,103(E8):19419-19425.
[92] Jolliffe T. Principal component analysis [M]. NewYork: Springer-Verlag,1986.
[93] Green A A, Berman M, Switzer P, et al. A transformation for ordering multispectral data interms of image quality with implication for noise removal [J]. IEEE Transactions onGeoscience and Remote Sensing,1988,26(1):65-74.
[94] Jimenez L O, Landgrede D A. Hyperspectral data analysis and supervised feature reductionvia projection pursuit [J]. IEEE Transactions on Geoscience and Remote Sensing,1999,37(6):2653-2667.
[95] Bayliss J, Gualtieri J A, Cromp R. Analysing hyperspectral data with independent componentanalysis [C]. Proc. SPIE,1997,3240:133-143.
[96] Kumar S, Ghosh J, Crawford M M. Best-bases feature extraction algorithms for classificationof hyperspectral data [J]. IEEE Transactions on Geoscience and Remote Sensing,2001,39(7):1368-1379.
[97] Kegl B, Krzyzak A, Linder T. Learning and design of principal curves [J]. IEEE Transactionon Pattern Analysis and Machine Intelligence,2000,22(3):281-297.
[98] Duda R O, Hart P E, Stork D G. Pattern Classification [M]. New York: Wiley Press,2001.
[99] Nie F P, Xu D, Tsang I W et al. Flexible manifold embedding: a framework forsemi-supervised and unsupervised dimension reduction [J]. IEEE Transactions on ImageProcessing,2010,19(7):1921–1932.
[100] Liao W, Pizurica A, Scheunders P et al. Semisupervised local discriminant analysis forfeature extraction in hyperspectral images. IEEE Transactions on Geoscience and RemoteSensing,2013,51(1):184–198.
[101] Xia J, Chanussot J, Du P et al. Semi-supervised dimensionality reduction for hyperspectralremote sensing image classification. In Proceedings of WHISPERS,2012, pp.1–4.
[102] Gualtieri J A, Cromp R F. Support vector machine for hyperspectral remote sensingclassification [C]. In Proceeding of the SPIE,27thAIPR Workshop,1998,3584:221-232.
[103] Campus-Valls G, Martin-Guerrero J D, Rojo-Alvarez J L et al. Fuzzy sigmoid kernel forsupport vector classifiers [J]. Neurocomputing,2004,62(6):501-506.
[104] Campus-Valls G, Gomez-Chova L, Munoz-Mari J et al. Composite kernels forhyperspectral image classification [J]. IEEE Geoscience and Remote Sensing Letters,2006,3(1):93-97.
[105] Bruzzone L, Chi M, Marconcini M. A novel transductive SVM for semi-supervisedclassification of remote sensing images [J]. IEEE Transactions on Geoscience and RemoteSensing,2006,44(11):3363-3373.
[106] Mountrakis G, Im J, Ogole C. Support vector machines in remote sensing: A review [J].ISPRS Journal of Photogrammetry and Remote Sensing,2011,66(3):247-259.
[107] Bishop C M. Pattern Recognition and Machine Learning [M]. Springer,2007.
[108] Waske B, Linden S V D, Benediktsson J A et al. Sensitivity of support vector machines torandom feature selection in classification of hyperspectral data [J]. IEEE Transactions onGeoscience and Remote Sensing.2010,48(7):2880-2889.
[109] Gislason P, Benediktsson J A, Sveinsson J R. Random forests for land cover classification[J]. Pattern Recognition Letters,2006,27(4):294–300.
[110] Ham J, Chen Y, Crawford M M et al. Investigation of the random forest framework forclassification of hyperspectral data [J]. IEEE Transactions on Geoscience and RemoteSensing,2005,43(3):492–501.
[111] Lawrence R L, Wood S D, Sheley R L. Mapping invasive plants using hyperspectralimagery and Breiman Cutler classifications (random Forest)[J]. Remote Sensing ofEnvironment,2006,100(3):356-362.
[112] Chan J C, Paelinckx D. Evaluation of Random Forest and Adaboost tree-based ensembleclassification and spectral band selection for ecotope mapping using airborne hyperspectralimagery [J]. Remote Sensing of Environment,2008,112:2999-3011.
[113] Tarabalka, Y, Benediktsson J A, Chanussot, J, Tilton, J. Multiple spectral-spatialclassification approach for hyperspectral data [J]. IEEE Transactions on Geoscience andRemote Sensing,2010,48(11):4122–4132.
[114] Favuel M. Classification of urban areas using high resolution satellite remote seningimages [M]. PhD thesis. Joint by University of Iceland, Iceland and University of Grenoble,France.
[115] Kettig R L, Landgrebe D A. Classification of multispectral image data by extraction andclassification of homogeneous objects [J]. IEEE Transactions on Geoscience Electronics,1976,14(1):19–26.
[116] Pony O, Descombres X, Zerubia J. Classification d’images satellitaires hyperspectrales enzone rurale et periurbane. Technical report, Sophia Antipolis, France, Sept.2000.
[117] Jackson Q, Landgrebe D A. Adaptive bayesian contextual classification based on Markovrandom fields [J]. IEEE Transactions on Geoscience and Remote Sensing,2000,40(11):2454–2463.
[118] Tsai F, Chang C L, Liu G R. Texture analysis for three dimension remote sensing data3DGLCM [C]. In Proc. of27th Asian Conference on Remote Sensing, pages1–6, August2006.
[119] Farag A A, Mohamed R M, El-Baz A. A unified framework for MAP estimation in remotesensing image segmentation [J]. IEEE Transactions on Geoscience and Remote Sensing,2005,43(7):1617–1634.
[120] Dell’Acqua F, Gamba P, Ferrari A et al. Exploiting spectral and spatial information inhyperspectral urban data with high resolution [J]. IEEE Geoscience and Remote SensingLetters,2004,1(4):322–326.
[121] Fauvel M, Chanussot J, Benediktsson J A, Sveinsson J R. Spectral and spatial classificationof hyperspectral data using SVMs and morphological profiles [J]. IEEE Transactions onGeoscience and Remote Sensing,2008,46(11):3804–3814.
[122] Linden S V D, Janz A, Waske B et al. Classifying segmented hyperspectral data from aheterogeneous urban environment using support vector machines [J]. Journal of AppliedRemote Sensing,2007,1(1):013543.
[123]刘志刚,李德仁,秦前清等.支持向量机在多类分类问题中的推广[J].计算机工程与应用,2004,7:10-13.
[124]沈照庆,舒宁,陶建斌.一种基于NPA的加权―1v M‖SVM高光谱影像分类算法[J].武汉大学学报（信息科学版）,2009,12:1444-1447.
[125]杨国鹏,余旭初,周欣等.基于相关向量机的高光谱影像分类研究[J].测绘学报,2010,39(6):572-578.
[126]宋相法,焦李成.基于稀疏表示及光谱信息的高光谱遥感图像分类[J].电子与信息学报,2012,34(2):268-272.
[127]谭琨,杜培军.基于再生核Hilbert空间小波核函数支持向量机的高光谱遥感影像分类[J].测绘学报,2011,40(2):572-578.
[128]吴婉澜,皮亦鸣,何强. AdaBoost算法实现全极化SAR图像分类[J].信号处理,2009,25(10),1594-1597.
[129]杨艺,韩崇昭,韩德强.利用特征子空间评价与多分类器融合的高光谱图像分类[J].西安交通大学学报,2010,8:20-24.
[130]孙亮,韩崇昭,沈建京.集成特征选择的广义粗集方法与多分类器融合[J].自动化学报,2008,34(3):298-304
[131] Woods K, Kegelmeyer W P, Bowyer K. Combination of multiple classifiers using localaccuracy estimates [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19,405–410.
[132] Czyz J, Kittler J, Vandendorpe L. Multiple classifier combination for face-based identityverification [J]. Pattern Recognition,2003,37,1459–1469.
[133] Rodriguez J J, Kuncheva L I. Rotation forest: A new classifier ensemble method [J]. IEEETransactions on Pattern Analysis and Machine Intelligence.2006,28:1619–1630.
[134] Benediktsson J A, Kanellopoulos I. Classification of multisource and hyperspectral databased on decision fusion [J]. IEEE Transactions on Geoscience and Remote Sensing,1999,37:1367–1377.
[135] Fauvel M, Chanussot J, Benediktsson J A. Decision fusion for the classification of urbanremote sensing images [J]. IEEE Transactions on Geoscience and Remote Sensing.2006,44:2828–2838.
[136] Zhou Z H. When Semi-supervised learning meets ensemble learning [C]. In Proceedings ofthe8thInternational Workshop on Multiple Classifier Systems Reykjavik, Iceland, June10-12,2009; Volume5, pp.529–538.
[137] Brown, G. Ensemble learning [C]. In Encyclopedia of Machine Learning; Webb, G.I.,Sammut, C., Eds.; Springer: New York, NY, USA,2010.
[138] Dietterich T G. Ensemble Methods in Machine Learning [C].2000. In J. Kittler and F. Roli(Ed.) First International Workshop on Multiple Classifier Systems, Lecture Notes inComputer Science (pp.1-15). New York: Springer Verlag.
[139] Dietterich T G. Ensemble learning [C]. In the Handbook of Brain Theory and NeuralNetworks, Second Edition,2002.
[140] Valentini G, Masulli F. Ensembles of learning machines [C]. In Neural Nets WIRNVietri-02, Series Lecture Notes in Computer Sciences, Marinaro M and Tagliaferri R, Eds.:Springer-Verlag, Heidelberg (Germany),2002,
[141] Seung H S, Opper M, Sompolinsky H. Query by committee [C]. In Proceedings of the FifthWorkshop on Computaional Learning Theory, pages287--294, San Mateo, California,1992.Morgan Kaufmann.
[142] Kittler J, Hatef M, Duin R P, Matas J. On combining classifiers [J]. IEEE Transactions onPattern Analysis and Machine Intelligence,1998,20(3):226-239.
[143] Kearns M, Vazirani U. An introduction to computational learning theory [M]. MIT Press,1994.
[144] Kearns M J, Valiant L G. Learning Boolean formulae or factoring. Aiken ComputationLaboratory, Harvard University, Cambridge, MA, Technical Report: TR-1488,1988.
[145] Valiant L G. A theory of the learnable [J]. Communications of the ACM,1984,27(11):1134-1142.
[146] Schapire R E. The strength of weak learnability [J]. Machine Learning,5(2):197-227,1990.
[147] Freund Y, Schapire R E. A decision-theoretic generalization of on-line learning and anapplication to boosting [J]. Journal of Computer and System Sciences,1997,55(1):119–139.
[148] Schapire R E, Singer Y. Improved boosting algorithms using confidence-rated predictions[J]. Machine Learning,37(3):297-336,1999.
[149] Demiriz A, Bennett K P, Taylor J H. Linear programming boosting via column generation[J]. Machine Learning,46, pages225–254,2002.
[150] Freund Y. An adaptive version of the boost by majority algorithm [J]. Machine Learning,43(3):293–318,2001.
[151] Domingos P, Pazzani M. On the optimality of the simple bayesian classifier under zero-oneloss [J]. Machine Learning,1997,29(2-3):103-130.
[152] Holte R C. Very simple classification rules perform well on most commonly used datasets[J]. Machine Learning,1993,11(1):63-90.
[153] Geman S, Bienenstock E, Doursat R. Neural networks and the bias/variance dilemma [J].Neural Computation,1992,4(1):1-58.
[154] Wahba G, Lin X, Gao F et al. The bias-variance trade of and the randomized gacv [C]. InProceedings of the1998conference on Advances in neural information processing systems II,pages620-626,1999.
[155] Brown G, Wyatt J, Tino P. Managing diversity in regression ensembles [J]. Journal ofMachine Learning Research,2005,6:1621-1650.
[156] Islam M M, Yao X, Murase K. A constructive algorithm for training cooperative neuralnetwork ensembles [J]. IEEE Transaction on Neural Networks,2003,14(4):820-834,2003.
[157] Liu Y, Yao X. Simultaneous training of negatively correlated neural networks in anensemble [J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics,1999a,29(6):716-725.
[158] Liu Y, Yao X. Ensemble learning via negative correlation [J]. Neural Networks,1999b,12(10):1399-1404,
[159] Ueda P, Nakano R. Generalization error of ensemble estimators [C]. In Proceedings ofInternational Conference on Neural Networks, pages90-95,1996.
[160]焦李成,公茂国,王爽等.自然计算、机器学习与图像理解前沿[M].西安电子科技大学出版社,西安,2008.
[161] Ricardo V, Youssef D.: A perspective view and survey of meta-learning [J]. ArtificialIntelligence Review,18(2):77–95,2002.
[162] Kearns M, Mansour Y. On the boosting ability of top-down decision tree learningalgorithms [C]. In Proceedings of the Twenty-Eighth Annual ACM Symposium on theTheory of Computing,1996.
[163] Kuncheva L I, Rodriguez J J. An experimental study on rotation forest ensembles [C]. InProceedings of the7th International Workshop on Multiple Classifier Systems,2007, pp.459–468.
[164] Ozcift A, Gulten A. Classifier ensemble construction with rotation forest to improvemedical diagnosis performance of machine learning algorithms [J]. Computer Methods andPrograms in Biomedicine,2011,104(3):443–451.
[165] Breiman L, Friedman J, Olshen R et al. Classification and regression trees [M]. Boca Raton,FL: CRC Press,1984.
[166] Loh W Y. Classification and regression tree methods [C]. In Encyclopedia of Statistics inQuality and Reliability, Ruggeri, Kenett and Faltin (eds.),2008, pp.315–323.
[167] Jutten C, Herault J. Independent component analysis verus principal component analysis[C]. In Proceedings of Europe Signal Processing Conf. EUSIPC088.1988.
[168]易尧华，余长慧，秦前清．基于独立分量分析的遥感影像非监督分类方法[J]．武汉大学学报(信息科学版)，2005，30(1)：19-23．
[169]曾生根，王小敏，范瑞彬．基于独立分量分析的遥感图像分类技术[J]．遥感学报，2004，8(2)：150-157.
[170] Sugiyama M. Dimensionality reduction of multimodal labeled data by local Fisherdiscriminant analysis [J]. Journal of Machine Learning Research,2007,8,1027–1061.
[171] Fukunaga K. Introduction to statistical pattern recognition [M]. Boston: Academic Press,Inc. Second edition.1990.
[172] He X, Niyogi P. Locality preserving projections. Advances in Neural InformationProcessing Systems, Cambridge, MA: MIT Press.2004.
[173] Li J, Bioucas-Dias J, Plaza A. Spectral-spatial classification of hyperspectral data usingloopy belief propagation and active learning [J]. IEEE Transaction on Geoscience andRemote Sensing, in press
[174] Chang C, Lin C. LIBSVM: A library for Support Vector Machines. Software available athttp://www.csie.ntu.edu.tw/cjlin/libsvm,2010.
[175] Bachmann M, Ainsworth T, Fusina R. Exploiting manifold geometry in hyperspectralimagery [J]. IEEE Transaction Geoscience and Remote Sensing,2005,43(3):441-454
[176]陈诗国,张道强.非监督降维方法的实验比较[J].软件学报,2011,22(1):28~42
[177] Kim K, Jung K, Kim H et al. Face recognition using kernel principal component analysis[J]. IEEE Image Processing Letters,2002,9(2):40-42
[178] Kuo B C, Landgrebe D A. Nonparametric Weighted Feature extraction for classification [J].IEEE Transaction on Geoscience and Remote Sensing,2004,42(5):1096-1105.
[179] Zhang D Q, Zhou Z H, Chen S C. Semi-supervised dimensionality reduction [C]. InProceedings of the7th SIAM International Conference on Data Mining,2007, pp.387–393.
[180] Sugiyama M, Id T, Nakajima S et al. Semi-supervised local fisher discriminant analysis fordimensionality reduction [J]. Machine Learning,2010,78:35–61.
[181] Yu S P, Yu K, Tresp V et al. Supervised probabilistic principal component analysis [C]. InProceedings of the12th ACM SIGKDD international conference on Knowledge discoveryand data mining,2006, pp.464–473
[182] Tipping M E, Bishop C M. Probabilistic principal component analysis [J]. BMCBioinformatics,1999,61(3):611–622.
[183] Paterlini S, Krink T. Differential evolution and particle swarm optimisation in partitionalclustering [J]. Computational Statistics and Data Analysis,2006,50(5):1220–1247.
[184] Pal N R, Pal S K. A review on image segmentation techniques [J]. Pattern Recognition,1993,26(9):1277–1294.
[185] Fu K S, Mui J K. A survey on image segmentation [J]. Pattern Recognition,1981,13(1):3-16.
[186] MacQueen J B. Some methods for classification and analysis of multivariate observations
[C]. In Proc. of the fifth Berkeley Symposium on Mathematical Statistics and Probability, vol.1,1967, pp.281–297.
[187] Jain A K. Data clustering:50years beyond k-means [J]. Pattern Recognition Letters,2010,31(8):651–666.
[188] Dunn J C. A fuzzy relative of the ISODATA process and its use in detecting compactwell-separated clusters [J]. Cybernetics and Systems,1973,3(3):32–57.
[189] Bezdek J, Ehrlich R, Full W. FCM: The fuzzy c-means clustering algorithm Computers andGeosciences [J].1984,10(2-3):191–203.
[190] Gath I, Geva A B. Unsupervised optimal fuzzy clustering [J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,1989,11(7):773-780.
[191] Dhillon I S, Guan Y, Kulis B. Kernel k-means: spectral clustering and normalized cuts [C].In Proceedings of the Tenth ACM SIGKDD International Conference on KnowledgeDiscovery and Data Mining, Seattle, Washington, USA, August22-25,2004,2004.
[192] Shawe-Taylor J S, Cristianini N. Kernel Methods for Pattern Analysis [M]. New York, NY,USA: Cambridge University Press,2004.
[193] Ball G B, Hall D J. ISODATA, a novel method of data analysis and pattern classification.Technical report, Stanford, Stanford Research Institute,1965.
[194] Tarabalka Y, Benediktsson J A, Chanussot J. Spectral-spatial classification ofhyperspectral imagery based on partitional clustering techniques [J]. IEEE Transactions onGeoscience and Remote Sensing,2009,47(8):2973–2987.
[195] Canty M. Image Analysis, Classification and Change Detection in Remote Sensing, withAlgorithms for ENVI/IDL, Second Revised Edition [M]. Taylor&Francis, CRC Press2010
[196] Vincent L, Soille P. Watersheds in digital spaces: an efficient algorithm based onimmersion simulations [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1991,13(6):583–598.
[197] Osma-Ruiz V, Godino-Llorente J I, Soenz-Lechn N et al. An improved watershedalgorithm based on efficient computation of shortest paths [J]. Pattern Recognition,2007,40(3):1078–1090.
[198] Serra J. Image Analysis and Mathematical Morphology [J]. New York, NY, USA:Academic Press,1982
[199] Tarabalka Y, Chanussot J, Benediktsson J A. Segmentation and classification ofhyperspectral images using watershed transformation [J]. Pattern Recognition,2010,43(7):2367–2379.
[200] Soille P. Morphological Image Analysis, second edition [M]. Berlin, Germany: Springer,2003.
[201] Comaniciu D, Meer P. Mean shift: a robust approach toward feature space analysis [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603–619.
[202] Kirkpatrick S, Gelatt C D, Vecch M P. Optimization by simmulated annealing [J]. Science,1983,220(4598):671-680.
[203] Yuan F，Bauer M E．Comparison of impervious surface area and normalized differencevegetation index as indicators of surface urban heat island effects in Landsat imagery[J]．Remote sensing of Environment，2007(16)：375-386．
[204] Ridd M K．Exploring a V-I-S (vegetation-impervious surface-soil) model for urbanecosystem analysis through remote sensing：comparative anatomy for cities [J]．InternationalJournal of Remote Sensing，1995，16(12)：2165-2185．
[205] Small C．Estimation and vicarious validation of urban vegetation abundance by spectralmixture analysis [J]．International Journal of Remote Sensing，2001，22(7)：1305-1334．
[206] Weng Q．Remote Sensing of Impervious Surfaces[M]．CRC Press，2008．
[207]夏俊士，杜培军，逄云峰等.基于高光谱数据的城市不透水层提取与分析[J].中国矿业大学学报,2011,4:660-666.
© 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 |