基于机器学习的土地覆盖遥感信息提取方法研究
|
摘要
在遥感信息提取技术中,对多光谱遥感图像进行特征提取与分类是进行土地覆盖信息提取的主要环节。然而,由于多光谱遥感图像固有的特点,传统的信息提取方法已不能满足遥感信息提取对计算精度、计算速度与扩展性的要求。本文主要研究以现有遥感特征提取与分类算法:主成分分析、吸引子传播算法为基础,融合模糊统计学、半监督学习、增量学习理论的机器学习新理论与新算法,从而提高信息提取与识别的水平。围绕这一主线,对现有算法存在的问题进行深入地研究,并在总结以往研究成果的基础上,提出了三种新的基于机器学习的特征提取与分类算法。主要工作有以下四个方面:
1.针对主成分分析(Principal Component Analysis,PCA)的不足和遥感数据具有模糊性和随机性的特点,提出了基于模糊统计学的一元和多元图像统计特征,将其引入PCA中,提出了基于模糊统计学的主成分分析算法(Fuzzy Statistics-Based Principal Component Analysis,FS-PCA)。应用于多光谱遥感图像特征提取中。
2.鉴于吸引子传播算法(Affinity Propagation,AP)快速高效的特点和遥感图像中地物分界具有的模糊性和随机性等特点,提出了模糊统计学相似性度量(Fuzzy Statistics Similarity Measure,FSS)的概念,将其引入到AP中,提出了基于模糊统计学的吸引子传播算法(Fuzzy Statistics-Based Affinity Propagation,FS-AP)。应用到多光谱遥感图像分类中,提高了分类的精度和效率。
3.鉴于传统的遥感图像聚类算法对聚类结果缺乏指导性,提出半监督增量学习策略,将其引入吸引子传播算法中,提出了增量式半监督吸引子传播算法(Incremental Semi-Supervised Affinity Propagation,IS-AP)。进一步提高多光谱遥感分类算法的精度。
4.算法综合性应用。将本文提出的新理论和新算法应用于吉林省北部土地覆盖遥感信息提取中,取得了很好的效果。
Land is one of the country's most important natural resources. Objective, accurate and real land resources is the important basis for scientific formulation of land management policies and implementation of farmland protection policy. At the same time, it is the fundamental basis for countries to strengthen supervision of land and national macroeconomic control. Aviation, remote sensing technology and global positioning system improve the efficiency and accuracy of land resources information access. Remote sensing image has been used more and more to extract local, regional and global scale land cover information.
Obtaining remote sensing land cover information requires advanced information extraction technology. The innovation of information extraction technology comes from the exploration of mechanism and changes in the concept of design ideas. Remote sensing information extraction can not be separated from mathematical methods and algorithms, as well as computer technology and programme. Machine learning is the important frontier of computer science and information science. It is more than one interdisciplinary research, including mathematics, statistics, artificial intelligence, control theory, philosophy, information science and cognitive science, and so many disciplines. Its research content and applications is extremely broad, covering almost all of human cognition field. Machine learning methods are the most commonly used for multispectral remote sensing image feature extraction and classification to get land cover information. However, due to a high number of spectral channels, a large information quantity and uncertainty in remote sensing image, and the insufficiency of existing machine learning methods; If only simply use or getting a little knowledge of a subject, it is difficult to achieve high accuracy and tackle complex geological problems.
The purpose of this research is to design and implementation of new machine learning algorithm which is based on analyzing the characteristics of remote sensing image and the insufficiency of existing machine learning algorithms. Applying the proposed algorithms to extract land cover information, thus increase the level of land cover remote sensing information extraction. Around this theme, conduct in-depth study to the existing problems of information extraction algorithms and previous studies. I propose three novel machine learning algorithms. To verify proposed method, these algorithms are applied to three kinds of multispectral images-Landsat-7 ETM+, Quickbird, and moderate resolution imaging spectroradiometer (MODIS). The following are the main aspects.
1. Proposed fuzzy-statistics-based principal component analysis (FS-PCA) algorithm which is applied to remote sensing image feature extraction. Considering principal component analysis (PCA) is sensitive to outliers and missing data. Fuzziness and randomicity are just the important characteristics of remote sensing images. By introducing fuzzy statistics variables into PCA, a novel method called fuzzy-statistics-based PCA (FS-PCA) is proposed. To verify proposed method, the FS-PCA is applied to the multispectral data for image feature extraction. It can be explained that if fuzzy statistics is applied into PCA by making fuzzy sets participate in decision making, it can overcome the insufficiency of PCA, and effectively extract image feature.
2. Proposed fuzzy-statistics-based affinity propagation (FS-AP) algorithm which is applied to remote sensing image classification. Considering affinity propagation (AP) exhibits a fast execution speed and finds clusters with small error, and the characteristics of remote sensing images. I propose a novel clustering method, called fuzzy statistics-based AP (FS-AP) which is based on a fuzzy statistical similarity measure (FSS). Results obtained on three kinds of multispectral images-Landsat-7 ETM+, Quickbird, and MODIS by comparing the proposed technique with K-means, fuzzy K-means, and AP based on Euclidean distance (ED-AP) demonstrate the good efficiency and high accuracy of FS-AP.
3. Proposed Incremental semi-supervised affinity propagation (IS-AP) algorithm which is applied to remote sensing image classification. Considering clustering methods is lack of instruction for results. To address this problem, I developed a novel semi-supervised clustering method of incorporating a semi-supervised incremental learning principle into AP, which called incremental semi-supervised AP (IS-AP). Three kinds of multispectral images-Landsat-7 ETM+, Quickbird, and MODIS is applied to comparing the proposed semi-supervised technique with seed K-means, constrained K-means, and semi-supervised affinity propagation (SAP). Experimental results show that the accuracy is further improved.
4. Integrated applications. New theory and algorithms proposed in the thesis are applied to northern Jilin Province land cover information extraction. By introducing FSS into IS-AP, a novel method called fuzzy-statistics-based IS-AP (FIS-AP) is proposed. Select northern Jilin Province as study area, Landsat 7 ETM+ and fuzzy principal component as data source. IS-AP and FIS-AP are used to classification. Experimental results show that our proposed new theory and algorithms achieve good results and improve effectiveness of land cover information extraction.
Due to the characteristics of remote sensing images, the traditional machine learning methods cannot meet the requirements of information extraction. This thesis studies the existing information extraction algorithms (PCA and AP), integrates fuzzy statistics, semi-supervised learning and incremental learning to propose new theories and algorithms. Experimental results show that the introduction and integration of them improves the accuracy, efficiency and scalability of these algorithms. It provides and enriches the theoretical aspects of the machine learning theory and has strong theoretical and practical value in remote sensing information extraction.
1.针对主成分分析(Principal Component Analysis,PCA)的不足和遥感数据具有模糊性和随机性的特点,提出了基于模糊统计学的一元和多元图像统计特征,将其引入PCA中,提出了基于模糊统计学的主成分分析算法(Fuzzy Statistics-Based Principal Component Analysis,FS-PCA)。应用于多光谱遥感图像特征提取中。
2.鉴于吸引子传播算法(Affinity Propagation,AP)快速高效的特点和遥感图像中地物分界具有的模糊性和随机性等特点,提出了模糊统计学相似性度量(Fuzzy Statistics Similarity Measure,FSS)的概念,将其引入到AP中,提出了基于模糊统计学的吸引子传播算法(Fuzzy Statistics-Based Affinity Propagation,FS-AP)。应用到多光谱遥感图像分类中,提高了分类的精度和效率。
3.鉴于传统的遥感图像聚类算法对聚类结果缺乏指导性,提出半监督增量学习策略,将其引入吸引子传播算法中,提出了增量式半监督吸引子传播算法(Incremental Semi-Supervised Affinity Propagation,IS-AP)。进一步提高多光谱遥感分类算法的精度。
4.算法综合性应用。将本文提出的新理论和新算法应用于吉林省北部土地覆盖遥感信息提取中,取得了很好的效果。
Land is one of the country's most important natural resources. Objective, accurate and real land resources is the important basis for scientific formulation of land management policies and implementation of farmland protection policy. At the same time, it is the fundamental basis for countries to strengthen supervision of land and national macroeconomic control. Aviation, remote sensing technology and global positioning system improve the efficiency and accuracy of land resources information access. Remote sensing image has been used more and more to extract local, regional and global scale land cover information.
Obtaining remote sensing land cover information requires advanced information extraction technology. The innovation of information extraction technology comes from the exploration of mechanism and changes in the concept of design ideas. Remote sensing information extraction can not be separated from mathematical methods and algorithms, as well as computer technology and programme. Machine learning is the important frontier of computer science and information science. It is more than one interdisciplinary research, including mathematics, statistics, artificial intelligence, control theory, philosophy, information science and cognitive science, and so many disciplines. Its research content and applications is extremely broad, covering almost all of human cognition field. Machine learning methods are the most commonly used for multispectral remote sensing image feature extraction and classification to get land cover information. However, due to a high number of spectral channels, a large information quantity and uncertainty in remote sensing image, and the insufficiency of existing machine learning methods; If only simply use or getting a little knowledge of a subject, it is difficult to achieve high accuracy and tackle complex geological problems.
The purpose of this research is to design and implementation of new machine learning algorithm which is based on analyzing the characteristics of remote sensing image and the insufficiency of existing machine learning algorithms. Applying the proposed algorithms to extract land cover information, thus increase the level of land cover remote sensing information extraction. Around this theme, conduct in-depth study to the existing problems of information extraction algorithms and previous studies. I propose three novel machine learning algorithms. To verify proposed method, these algorithms are applied to three kinds of multispectral images-Landsat-7 ETM+, Quickbird, and moderate resolution imaging spectroradiometer (MODIS). The following are the main aspects.
1. Proposed fuzzy-statistics-based principal component analysis (FS-PCA) algorithm which is applied to remote sensing image feature extraction. Considering principal component analysis (PCA) is sensitive to outliers and missing data. Fuzziness and randomicity are just the important characteristics of remote sensing images. By introducing fuzzy statistics variables into PCA, a novel method called fuzzy-statistics-based PCA (FS-PCA) is proposed. To verify proposed method, the FS-PCA is applied to the multispectral data for image feature extraction. It can be explained that if fuzzy statistics is applied into PCA by making fuzzy sets participate in decision making, it can overcome the insufficiency of PCA, and effectively extract image feature.
2. Proposed fuzzy-statistics-based affinity propagation (FS-AP) algorithm which is applied to remote sensing image classification. Considering affinity propagation (AP) exhibits a fast execution speed and finds clusters with small error, and the characteristics of remote sensing images. I propose a novel clustering method, called fuzzy statistics-based AP (FS-AP) which is based on a fuzzy statistical similarity measure (FSS). Results obtained on three kinds of multispectral images-Landsat-7 ETM+, Quickbird, and MODIS by comparing the proposed technique with K-means, fuzzy K-means, and AP based on Euclidean distance (ED-AP) demonstrate the good efficiency and high accuracy of FS-AP.
3. Proposed Incremental semi-supervised affinity propagation (IS-AP) algorithm which is applied to remote sensing image classification. Considering clustering methods is lack of instruction for results. To address this problem, I developed a novel semi-supervised clustering method of incorporating a semi-supervised incremental learning principle into AP, which called incremental semi-supervised AP (IS-AP). Three kinds of multispectral images-Landsat-7 ETM+, Quickbird, and MODIS is applied to comparing the proposed semi-supervised technique with seed K-means, constrained K-means, and semi-supervised affinity propagation (SAP). Experimental results show that the accuracy is further improved.
4. Integrated applications. New theory and algorithms proposed in the thesis are applied to northern Jilin Province land cover information extraction. By introducing FSS into IS-AP, a novel method called fuzzy-statistics-based IS-AP (FIS-AP) is proposed. Select northern Jilin Province as study area, Landsat 7 ETM+ and fuzzy principal component as data source. IS-AP and FIS-AP are used to classification. Experimental results show that our proposed new theory and algorithms achieve good results and improve effectiveness of land cover information extraction.
Due to the characteristics of remote sensing images, the traditional machine learning methods cannot meet the requirements of information extraction. This thesis studies the existing information extraction algorithms (PCA and AP), integrates fuzzy statistics, semi-supervised learning and incremental learning to propose new theories and algorithms. Experimental results show that the introduction and integration of them improves the accuracy, efficiency and scalability of these algorithms. It provides and enriches the theoretical aspects of the machine learning theory and has strong theoretical and practical value in remote sensing information extraction.
引文
[1] Antonio D G, Louisa J M, Jansen. Landcover classification system (LCCS) [M]. FAO. 2000.
[2]王静等.土地资源遥感监测与评价方法[M].北京:科学出版社,2006.
[3]叶树华,任志远,遥感概论[M].陕西:陕西科学技术出版社,1993.
[4]汤国安,张友顺,刘咏梅,谢元礼,杨昕,刘爱利.遥感数字图像处理[M].北京:科学出版社,2004.
[5] Lee W T. The face of the earth as seen from the air: A study in the application of airplane photography to geography [M]. New York: American Geographical Society, Special Publication, 1922.
[6] Tucker C J, Townshend J R G, Goff T E. Africa land cover classification using satellite data [J]. Science, 1985, 227: 369-375.
[7] Townshend J R G, Justice C O, Kalb V T. Characterization and classification of South America land covertypes using satellite data [J]. International Journal of Remote Sensing, 1987, 8: 1189-1207.
[8] Loveland T R, Reed B C, Brown J F, et al. Development of a global land cover characteristics database and IGBP DISCover from 1km AVHRR data [J]. Internation al Journal of Remote Sensing, 2000, 21(6&7): 1303-1330.
[9] Lee C and Landgrebe D. Feature extraction and classification algorithms for high dimension data: [PhD thesis]. USA: Purdue University.1993.
[10] Samuel A. Some studies in machine learning using the game of checkers [J]. IBM Journal of Research and Development, 1959, 3: 211-229.
[11] Mitchell, T. Machine Learning [M]. McGraw-Hill, 1997.
[12]周志华.机器学习及其挑战[PPT].南京大学计算机软件新技术国家重点实验室网站http://cs.nju.edu.cn/people/zhouzh/. 2003.
[13] Mitchell, T M著,曾华军,张银奎等译.机器学习[M].北京:机械工业出版社, 2003.
[14] Baldi P, Brunak著,张东晖等译.生物信息学——机器学习方法[M].北京:中信出版社, 2003.
[15]王珏.机器学习研究回顾与趋势[PPT].中科院自动化研究所模式识别国家重点实验室网站http://www.intsci.ac.cn/research/wangj04.ppt, 2004.
[16]曾志远.卫星遥感图像计算机分类与地学应用研究[M].北京:科学出版社, 2004.
[17] Zhao G and Maclean A L. A comparison of canonical discriminant analysis and principal component analysis for spectral transformation [J]. Photogrammetric Engineering&Remote Sensing, 2000, 66(7): 841-847.
[18] Almeida-Fiho R and Shimabukuro Y E. Digital processing of a Landset TM time series for mappingand monitoring degraded areas caused by independent gold miners, Roraima State, Brazilian Amazon [J]. Remote Sensing of Environment, 2002, 79(1): 42-50.
[19] Mitternicht G I and Zinck J A. Remote sensing of soil salinity: potentials and constraints [J]. Remote Sensing of Environment, 2003, 85: 1-20.
[20] Reese H M, Lillesand T M, Nagel D E, Stewart J S, Gold-mann R A, Simmons T E, Chipman J W and Tessar P A. Statewide land cover derived from multisensonal Landset TM data: a retrospective of the WISCLAND project [J]. Remote Sensing of Environment, 2002, 82: 224-237.
[21] Sarbu C, Pop H F. Principal component analysis versus fuzzy principal component analysis a case study: the quality of Danube water (1985-1996) [J]. Talanta65, 2005: 1215-1220.
[22] POP H F. Principal components analysis based on a fuzzy sets approach [J]. Studia Univ. Babes-bolyai, Informatica, 2001, 46(2): 45-52.
[23]杨翠芬,田村正行.差分主成分分析在辽河三角洲景观变换中的应用[J].地理学报, 2004, 59(4): 592-598.
[24] Colwell R N. Manual of Remote Sensing: 2ndEd [M]. Bethesda: American Society for Photogrammetry&Remote Sensing, 1983.
[25] Colwell R N. Manual of Photographic Interpretation: 2ndEd [M]. Bethesda: American Society for Photogrammetry&Remote Sensing, 1997.
[26] MacQueen J. Some methods for classification and analysis of multivariate observations [C]. In Proc. of 5th Berkeley Symp. on Math. Stat. and Prob., 1967, 1: 281–297.
[27] Thitimajshima P. A new modified fuzzy c-means algorithm for multispectral satellite images segmentation [C]. In Proc. IGARSS, 2000, 4: 1684-1686.
[28] Zhong Y F, Zhang L P, Huang B and Li P X. An unsupervised artificial immune classifier for multi/hyperspectral remote sensing imagery [J]. IEEE Trans. Geosci. Remote Sens., 44(2): 420-431, 2006.
[29] Ding, Z J, Yu J and Zhang Y Q. A new improved k-means algorithm with penalized term [C]. IEEE Int. Conf. Granular Computing, 2007.
[30] Wang Y and Mo J. Fuzzy logic applied in remote sensing image classification [C]. In Proc. Int. Conf. Systems, Man and Cybernetics, 2004, 6378-6382.
[31] Dunn J C. A fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters [J]. J. Cybernet, 1974, 3(3): 32-57.
[32] Bezdek J C. Pattern Recognition with Fuzzy Objective Function Algorithms [M]. Plenum. New York. 1981.
[33] Liu P Y, Jia K B and Zhang P Z. An effective method of image retrieval based on modified fuzzy c-means clustering scheme [C]. IEEE Int. Conf. Sign Processing, 2006.
[34]钱乐详等.遥感数字影像处理与地理特征提取[M].北京:科学出版社, 2005.
[35]承继成,郭华东,史文中.遥感数据的不确定性问题[M].北京:科学出版社, 2004.
[36] Jawahar C V and Ray A K. Fuzzy statistics of digital images [J]. IEEE Signal Process. Lett., 1996, 3(8): 225-227.
[37] Yang C, Lu L J, Lin H P, Guan R C, Shi X H and. Liang Y C. A fuzzy-statistics-based principal component analysis (FS-PCA) method for multispectral image enhancement and display [J]. IEEE Trans. Geosci. Remote Sensing, 2008, 46(11): 3937-3947.
[38] Taheri S M. Trends in fuzzy statistics [M]. Austrian J. Stat., 2003, 32: 239-257.
[39] Frey B J, Dueck D, Clustering by passing messages between data points [J]. Science, 2007, 315: 972-976.
[40] Jiang W, Ding F, Xiang Q L. An affinity propagation based method for vector quantization [J]. Eprint arXiv: 0710.2037.
[41] Frey B J, Dueck D. Non-metric affinity propagation for unsupervised image categorization [C]. In the ICCV 2007, Rio de Janeiro, Brazil, 2007.
[42] Brusco M J and Kohn H F. Comment on‘clustering by passing messages between data points’[J]. Science, 2008, 319(5864): 726c.
[43] Frey B J, Dueck D. Response to comment on‘clustering by passing messages between data points’[J]. Science, 2008, 319(5864): 726d.
[44] Jensen J R, Qiu F, Ji M. Predictive modelling of coniferous forest age using statistical and artificial neural network approaches applied to remote sensor data [J]. International Journal of Remote Sensing. 1999, 20(14): 2805-2822.
[45] Foody G M, Arora M K. An evaluation of some factors affecting the accuracy of classification by an artificial neural network [J]. International Journal of Remote Sensing, 1997, 18(4): 799-810.
[46] Kuo B and David A L. A covariance estimator for small size classification problem and its application to feature extraction [J]. IEEE Trans. Geosci. Remote Sens., 2002, 40(4): 814-819.
[47] Jeon B and Landgrebe D. Partially supervised classification using weighted unsupervised Clustering [J]. IEEE Trans. Geosci. Remote Sens., 2002, 37(2): 1073-1079.
[48] Zhang J, Zhang Y and Zhou T. Classification of hyperspectral data using support vector machine [C]. In: IEEE International Conference on Image Processing, 2001, 882-885.
[49] Huang C, Davis L and Townshend J. An assessment of support vector machines for land cover classification [J]. Int. remote sensing, 2002, 23(4): 725-749.
[50] Melgani F, Bruzzone L. Classification of hyperspectral remote-sensing images with support vector machines [J]. IEEE Trans. Geosci. Remote Sens., 2004, 42(8): 1778-1790.
[51]骆剑承,周成虎,梁怡等.支撑向量机及其遥感影像空间特征提取和分类的应用研究[J].遥感学报, 2002, 6(1): 50-55.
[52]何灵敏.支持向量机集成及在遥感分类中的应用[D].浙江大学, 2006.
[53]祁享年,杨建刚,方陆明.基于多类支持向量机的遥感图像分类及其半监督式改进策略[J].复旦学报, 2004, 43(5): 158-162.
[54]刘志刚,史文中,李德仁,秦前清.一种基于支撑向量机的遥感影像不完全监督分类新方法[J].遥感学报, 2005, 9(4): 158-162.
[55] Baraldi A, Bruzzone L, Blonda P. A multiscale expectation-maximization semisupervised classifiersuitable for badly posed image classification [J]. IEEE Transactions on Image Processing, 2006, 5(8): 2208-2225.
[56] Bruzzone L, Chi M M, Marconcini M. A novel transductive SVM for the semisupervised classification of remote-sensing images [J]. IEEE Trans. Geosci. Remote Sensing, 2006, 44(11): 3363-3373, 2006.
[57] Chi M M, Bruzzone L. Classification of hyperspectral data by continuation semi-supervised SVM [C]. IEEE Int. Conf. Geosci. Remote Sensing Symposium, IGARSS, 2007, 3794-3797.
[58] Santiteban A and Munoz L. Principal component of a mutilspectral image: application to a geokogical problem [J]. IBM Journal of Research and Development, 1978, 22(5): 444-454.
[59] Giddings L E, Soto M, and Angulo M J. The use of landsat data in mapping tropical vegetation monitoring [C]. In Processing of the 14th International Symposium on Remote Sensing of Environment, 1980, 3: 1383-1387.
[60]孙家炳.遥感原理与应用[M],武汉:武汉大学出版社. 2003.
[61]高新波.模糊聚类分析及其应用[M],西安:西安电子科技大学出版社. 2004.
[62] Richards J A and Jia X P. Remote Sensing Digital Image Analysis: An Introduction: 4th Ed [M]. Berlin, Germany: Springer-Verlag, 2006.
[63] Jensen J R. Introductory Digital Image Processing: A Remote Sensing Perspective [M]. Upper Saddle River, NJ: Prentice-Hall, 2005.
[64] Bryant J. On displaying multispectral imagery [J]. Photogramm. Eng. Remote Sens., 1988, 54(12): 1739-1743.
[65] Du Q and Fowler J E. Hyperspectral image compression using JPEG2000 and principal component analysis [J]. IEEE Geosci. Remote Sens. Letter, 2007, 4(2): 201–205.
[66] Farrell M D and Mersereau R M. On the impact of PCA dimension reduction for hyperspectral detection of difficult targets [J]. IEEE Geosci. Remote Sens. Letter, 2005, 2(2): 192-195.
[67] Wang J and Chang C I. Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis [J]. IEEE Geosci. Remote Sens. Letter, 2006, 44(6): 1586-1600.
[68] Moussaoui S. On the decomposition of Mars hyperspectral data by ICA and Bayesian positive source separation [J]. Neurocomputing, 2008, 71(10-12): 2194-2208.
[69] Zubko V, Kaufman Y J, Burg R I, and Martins J V. Principal component analysis of remote sensing of aerosols over oceans [J]. IEEE Trans. Geosci. Remote Sens., 2007, 45(3): 730-745.
[70] Li R, Mersereau M, and Simske S. Atmospheric turbulencedegraded image restoration using principal components analysis [J]. IEEE Geosci. Remote Sens. Letter, 2007, 4(3): 340-344.
[71] Jimenez-Rodriguez L O, Arzuaga-Cruz E, and Velez-Reyes M. Unsupervised linear feature-extraction methods and their effects in the classification of high-dimensional data [J]. IEEE Trans. Geosci. Remote Sens., 2007, 45(2): 469-483.
[72] Fauvel M, Chanussot J, and Benediktsson J A. Kernel principal component analysis for feature reduction in hyperspectrale images analysis [C]. In Proc. 7th NORSIG, 2006, 238-241.
[73] Ricotta C, Avena G C, and Volpe F. The influence of principal component analysis on the spatial structure of a multispectral dataset [J]. Int. J. Remote Sens., 1999, 20(17): 3367-3376.
[74] Liang Y C, Lee H P, Lim S P, Lin W Z, Lee K H, and Wu C G. Proper orthogonal decomposition and its applications—Part I: Theory [J]. J. Sound Vib., 2002, 252(3): 527–544.
[75] Yao H B and Tian L. A genetic-algorithm-based selective principal component analysis (GA-SPCA) method for high-dimensional data feature extraction [J]. IEEE Trans. Geosci. Remote Sens., 2003, 41(6): 1469-1478.
[76] Hubert M, Rousseeuw P J, and Verboven S A. Fast method for robust principal components with applications to chemometrics [J]. Chemom. Intell. Lab. Syst., 2002. 60(1): 101-111.
[77] Yu X and Cheng X. Research of independent component analysis [C]. In Proc. IEEE Int. Conf. Syst., Man Cybern., 2004, 4804-4809.
[78] Hyvarinen A, Karhunen J, and Oja E. Independent Component Analysis [M]. Hoboken, NJ: Wiley, 2001.
[79] Sun Z L, Huang D S, Cheung Y M, Liu J, and Huang G B. Using FCMC, FVS, and PCA techniques for feature extraction of multispectral images [J]. IEEE Geosci. Remote Sens. Letter, 2005, 2(2): 108-112.
[80] Serge A, Ludovic R, Yannick C, and Alain B. A fuzzy-possibilistic scheme of study for objects with indeterminate boundaries: Application to French Polynesian reefscapes [J]. IEEE Trans. Geosci. Remote Sens., 2000, 38(1): 257-270.
[81] Zadeh L A. Fuzzy sets [J]. Inf. Control, 1965, 8: 338-353.
[82] Wang F. Improving remote sensing image analysis through fuzzy information representation [J]. Photogramm. Eng. Remote Sens., 1990. 56(8): 1160-1169.
[83] Chanussot J, Mauris G, and Lambert P. Fuzzy fusion techniques for linear features detection in multitemporal SAR images [J]. IEEE Trans. Geosci. Remote Sens., 1999, 37(3): 1292-1305.
[84] Filippi A. M and Jensen J R. Effect of continuum removal on hyperspectral coastal vegetation classification using a fuzzy learning vector quantizer [J]. IEEE Trans. Geosci. Remote Sens., 2007, 45(6): 1857-1869.
[85] Vlag D E and Stein A. Incorporating uncertainty via hierarchical classification using fuzzy decision trees [J]. IEEE Trans. Geosci. Remote Sens., 2007, 45(1): 237-245.
[86] Chanussot J, Benediktsson J A, and Fauvel M. Classification of remote sensing images from urban areas using a fuzzy possibilistic model [J]. IEEE Geosci. Remote Sens. Letter, 2006, 3(1): 40-44.
[87] Cheng Q M, Jing L H, and Panahi A. Principal component analysis with optimum order sample correlation coefficient for image enhancement [J]. Int. J. Remote Sens., 2006, 27(16): 3387-3401.
[88] Narumalani S, Hlady J and Jensen J R. Information extraction from remotely sensed data [J], Manual of Geospatial Science and Technology, 2002, 299-324.
[89] Davis J C. Statistics and data analysis in geology: 3rd ed. [M]. New York: John Wiley & Sons, 2002.
[90] Freud R J and Wilson W J. Statistical Methods: 2nd ed. [M]. New York: Academic Press, 2003.
[91] Samuels M L and Witmer J A. Statistics for the Life Science [M]. Upper Saddle River, NJ:Prentice-Hall, 2003.
[92] Zadeh L A. Outline of a new approach to the analysis of complex systems and decision process [J]. IEEE Trans. Syst.,Man, Cybern., 1973, 3(1): 28-44, 1973.
[93] Kandel A and Byatt W J. Fuzzy sets, fuzzy algebra, and fuzzy statistics [J]. Proc. IEEE, 1978, 66(12): 1619-1639.
[94] Bezdek J C and Pal S K. Fuzzy Models for Pattern Recognition [M]. New York: IEEE, 1992.
[95] Chen B, Chen Y, and Hsu W. Automatic histogram specification based on fuzzy set operations for image enhancement [J]. IEEE Signal Process. Letter, 1995, 2(2): 37-40.
[96] Yang X and Toh P S. Adaptive fuzzy multilevel median filter [J]. IEEE Trans. Image Process., 1995, 4(5): 680-682.
[97]马建文,李启青等.遥感数据智能处理方法与程序设计[M].北京:科学出版社, 2005.
[98] Townshend J R G and Justice C O. Towards operational monitoring of terrestrial systems by moderate-resolution remote sensing [J]. Remote Sensing of Environment, 2002, 83: 351-359.
[99] Lunetta R S and Lyons J G. Geospatial Data Accuracy Assessment [M]. Las Vegas: Environmental Protection Agency, 2003.
[100] Loveland T R, Zhiliang Z, Ohlen D O, Brown J F, Reed B C and Yang L. An analysis of the IGBP global landcover characterization process [J]. Photogrammetric Engineering & Remote Sensing, 1999, 65(9): 1021-1032.
[101] Huang K. A synergistic automatic clustering technique for multispectral image analysis [J]. Photogrammetric Engineering & Remote Sensing, 2002, 68(1): 33-44.
[102] Bandyopadhyay S, Maulik U, and Mukhopadhyay A. Multiobjective genetic clustering for pixel classification in remote sensing imagery [J]. IEEE Trans. Geosci. Remote Sensing, 2007, 45(5): 1506-1511.
[103] Chi M M, Qian Q, and Benediktsson J A. Cluster-based ensemble classification for hyperspectral remote sensing [C]. IEEE Int. Conf. Geosci. Remote Sensing Symposium, IGARSS, 2008, 209-212.
[104] Maulik U and Saha I. Modified differential evolution based fuzzy clustering for pixel classification in remote sensing imagery [J]. Pattern Recognition, 2009, 42(9): 2135-2149.
[105] Duda R O, Hart P E and Stork D G. Pattern Classification [M], New York: John Wiley & Sons, 2001.
[106] Robert A S. Remote Sensing: Models and Methods for Image Processing: 3rd ed. [M]. Elsevier Inc. USA, 2007.
[107] Aldenderfer M S and Blashfield R K. Cluster Analysis [M]. Sage Publications, Beverly Hills. USA, 1984.
[108] Chehata N and Bretar F. Terrain modeling from lidar data: Hierarchical K-means filtering and Markovian regularization [C]. IEEE Int. Conf. Image Processing, ICIP, 2008, 1900-1903.
[109] Zheng J, Cui Z Z, Liu A F, and Jia Y. A k-means remote sensing image classification method based on AdaBoost [C]. IEEE Int. Conf. Natural Computation, ICNC, 2008, 27-32.
[110] Maheshwary P and Srivastav N. Retrieving similar image using color moment feature detector andk-means clustering of remote sensing images [C]. IEEE Int. Conf. Computer and Electrical Engineering, 2008, 20-22.
[111] Witten L H and Frank E. Data mining: practical machine learning tools and techniques: 2nd ed. [M]. Elsevier Inc. USA, 2005.
[112] Wu X D, Kumar V, and Quinlan J R. Top 10 algorithms in data mining [J]. Knowledge and Information Systems, 2008, 14(1):1-37.
[113] Dinesh M S, Gowda K G, and Nagabhuehan P. Unsupervised classification for remotely sensed data using fuzzy set theory [C]. IEEE Int. Geosci. Remote Sensing- A Scientific Vision for Sustainable Development, IGARSS, 1997, 1:521-523.
[114] Borasca B, Bruzzone L, Carlin L, and Zusi M. A fuzzy-input fuzzy-output SVM technique for classification of hyperspectral remote sensing images [C]. Signal Processing Symposium, Processing of the 7th Nordic, NORSIG, 2006, 2-5.
[115] Altman D. Efficient fuzzy clustering of multi-spectral images [C]. IEEE Int. Conf. Geoscience and Remote Sensing Symposium, IGARSS, 1999, 3: 1594-1596.
[116] Gorsevski P V, Gessler P E, and Jankowski P. Integrating a fuzzy k-means classification and a Bayesian approach for spatial prediction of landslide hazard [J]. Journal of Geographical Systems, 2003, 5(3): 223-251.
[117] Liu X F, Li X W, Zhang Y, Yang C J, Xu W B, Li M, and Luo H M. Remote sensing image classification based on dot density function weighted FCM clustering algorithm [C]. IEEE Int. Conf. Geosci. Remote Sensing Symposium, IGARSS, 2007, 2010-2013.
[118] Hung C C, Liu W P, and Kuo B C. A new adaptive fuzzy clustering algorithm for remotely sensed images [C]. Int. Conf. Geoscience and Remote Sensing Symposium, IGARSS, 2008, 2: 863-866.
[119] Fan J C, Han M, and Wang J. Single point iterative weighted fuzzy C-means clustering algorithm for remote sensing image segmentation [J]. Pattern Recognition, 2009, 42(11): 2527-2540.
[120] Zhang H Y, Wu Q T, and Pu J X. A novel fuzzy kernel clustering algorithm for outlier detection [C], IEEE Int. Conf. Mechatronics and Automation, ICMA, 2007, 2378-2382.
[121] Wu K L and Yang M S. Alternative c-means clustering algorithms [J]. Pattern Recognition, 2002, 35(10): 2267-2278.
[122] Gath I and Geva A B. Unsupervised Optimal Fuzzy Clustering [J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 1989, 11: 773-781.
[123] Nasser S, Alkhaldi R and Vert G. A modified fuzzy k-means clustering using expectation maximization [C]. IEEE Int. Conf. Fuzzy Systems, 2006, 231-235.
[124] Li M J, Ng M K, and Cheung Y. Agglomerative fuzzy k-means clustering algorithm with selection of number of clusters [J]. IEEE Trans. Knowledge and Data Engineering, 2008, 20(11): 1519-1534.
[125] Frey B J and Dueck D. Mixture modeling by affinity propagation [J]. In Advances in Neural Information Processing Systems 18, MIT Press, 2006, 2, 3, 7.
[126] http://www.psi.toronto.edu/affinitypropagation/faq.html.
[127] Schowengerdt R A. Remote sensing models and methods for image processing: 2nd Edition [M].San Diego London Bobton New York Sydney Tokyo Toronto: Academic Press, 1997.
[128] Sun M C and Chou C H. A modified version of the K-means algorithm with a distance based on cluster symmetry [J]. IEEE Tran. Pattern Analysis and Machine Intelligence, 2001, 23(6): 674-680.
[129] Sweet J N. The spectral similarity scale and its application to the classification of hyperspectral remote sensing data [J]. IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2004, 92-99.
[130] Chintalapudi K K and Kam M. A noise-resistant fuzzy C-means algorithm for clustering [C]. Fuzzy Systems Proceedings. IEEE World Congress on Computational Intelligence, 1998, 2: 1458-1463.
[131] Fung T and Drew E L. The determination of optimal threshold levels for change detection using various accuracy indices [J]. Photogrammetric Eng. and Remote Sensing, 1998, 54: 1449-1454.
[132] Liu C, Frazier P, and Kumar L. Comparative assessment of the measures of thematic classification accuracy [J]. Photogramm. Eng. and Remote Sensing, 2007, 107(4): 606-616.
[133] Short N M. The Landsat Tutorial Workbook Basics of Satellite Remote Sensing [M]. Greenbelt, Md., Goddard Space Flight Center, NASA Reference Publication 1078, 1982.
[134] Rosenfield G H and Fitzpatrick-Lins K. A coefficient of agreement as a measure of thematic classification accuracy [J]. Photogramm. Eng. and Remote Sensing, 1986, 52(2): 223-227.
[135] Maulik U and Bandyopadhyay S. Fuzzy partitioning using a real-coded variable- length genetic algorithm for pixel classification [J]. IEEE Trans. Geosci. Remote Sensing, 2003, 41(5): 1075-1081.
[136] Tran T N, Wehrens R, Hoekman D H and Buydens L M C. Initialization of markov random field clustering of large remote sensing images [J]. IEEE Trans. Geosci. Remote Sensing, 2005, 43(8): 1912-1919.
[137] Grira N, Crucianu M and Boujemaa N. Unsupervised and semi-supervised clustering: a brief survey [C]. In A Review of Machine Learning Techniques for Processing Multimedia Content, Report of the MUSCLE European Network of Excellence (6th Framework Programme), 2005.
[138] Nguyen N and Caruana R. Classification with partial labels [C]. In proc. of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, 2008, 551-559.
[139] Wagsta K and Cardie C. Clustering with instance-level constraints [C]. In proc. of the 17th International Conference on Machine Learning, 2000, 1103-1110.
[140] Givoni I E and Frey B J. Semi-supervised affinity propagation with instance-level constraints [C]. In proc. of the 12th International Conference on Artificial Intelligence and Statistics (AISTATS), 2009, 161-168,
[141] Zhu X. Semi-supervised learning literature survey [C]. Technical Report 1530, Computer Sciences, University of Wisconsin-Madison, 2005.
[142] Cozman F, Cohen I, and Cirelo M. Semi-supervised learning of mixture models and bayesian networks [C]. In Proc. of the 12th International Conference of Machine Learning, 2003.
[143] Karakoulas G and Salakhutdinov R. Semi-supervised mixture-of-experts classification [C]. In Proc. of the 4th IEEE International Conference on Data Mining, 2004, 138-145.
[144] Blum A and Mitchell T. Combining labeled and unlabeled data with co-training [C]. In Proc. of the11th Conference on Computational Learning Theory, 1998, 92-100.
[145] Nigam K and Ghani R. Analyzing the effectiveness and applicability of co-training [C]. In Proc. of the 9th International Conference on Information and Knowledge Management, 2000, 86-93.
[146] Moreno P J and Agarwal S. An experimental study of EM-based algorithms for semi-supervised learning in audio classification [C]. In Proc. of ICML-2003 Workshop on Continuum from Labeled to Unlabeled Data, 2003.
[147] Rosenberg C, Hebert M and Schneiderman H. Semi-supervised self-training of Object Detection Models [C]. In Proc. of 7th IEEE Workshop on Applications of Computer Vision, 2005, 29-36.
[148] Ghanramani Z and Jordan M I. Supervised learning from incomplete data via the EM approach [J]. Advances in Neural Information Processing Systems, 1994, 120-127.
[149] Wagstaff K, Cardie C, Rogers S, and Schroedl S. Constrained K-Means clustering with background knowledge [C]. In Proceedings of 18th International Conference on Machine Learning (ICML), 2001.
[150] Basu S, Bilenko M and Mooney R J. Comparing and unifying search-based and similarity-based approaches to semi-supervised clustering [C]. In Proc. of the ICML-2003 Workshop on the Continuum from Labeled to Unlabeled Data in Machine Learning and Data Mining System, 2003.
[151] Demiriz A, Bennett K P, and Embrechts M J. Semi-supervised clustering using genetic algorithms [C]. ANNIE'99 (Articial Neural Networks in Engineering), 1999.
[152] Bilenko M, Basu S, and Mooney R J. Integrating constraints and metric learning in semi-supervised clustering [C]. In Proceedings of 21th International Conference on Machine Learning, 2004.
[153] Globerson A and Roweis S. Metric learning by collapsing classes [J]. In Weiss, Y.; Sch¨olkopf, B.; and Platt, J., eds., Advances in Neural Information Processing Systems, Cambridge, MA: MIT Press, 2006, 18: 451-458.
[154] Ghodsi A, Wilkinson D F, and Southey F. Improving embeddings by flexible exploitation of side information [C]. In Veloso M M, ed., International Joint Conference on Artificial Intelligence, 2007, 810-816.
[155] Bilenko M, Basu S, and Mooney R J. Semi-supervised clustering by seeding [C]. In Proceedings of 19th International Conference on Machine Learning, 2002.
[156] Laird J E, Rosenbloom P S and Newel A. Chunking in SOAR: the anatomy of a general learning mechanism [J]. Machine Learning, 1986, 1: 11-46.
[157] Sutton R. Learning to precdict by the methods of temporal differences [J]. Machine Learning, 1988, 3: 9-44.
[158] Utgoff P E. Incremental induction of decision trees [J]. Machine Learning, 1989, 4(2):161-186.
[159] Utgoff P E. An improved algorithm for incremental induction of decision trees [C]. In Processings of the 10th International Machine Learning Workshop, Rutgers University, New Brunswick, NJ, 1993.
[160] Shen W M. Complementary discrimination learning: A duality between generalization and discrimination [C]. In Proceedings of 8th National Conference on Artificial Intelligence. MIT Press,1990.
[161] Shen W M. Complementary discrimination learning with decision lists [C]. In Proceedings of 8th National Conference on Artificial Intelligence. MIT Press, 1992.
[162] Shen W M. Learning finite state automata using local distinguishing experiments [C]. In Proceedings of IJCAI-93, Chambery, France, 1993.
[163] Shen W M. Efficient Incremental Induction of Decision Lists - Can Incremental Learning Outperform Non-Incremental Learning [J]. 1996.
[164] Ozawa S, Toh S L, Abe S, Pang S N and Kasabov N. Incremental learning of feature space and classifier for face recognition [J]. Neural Networks, 2005, 18 (5-6): 575-584.
[165] Shen F R, Tomotaka O, and Osamu H. An enhanced self-organizing incremental neural network for online unsupervised learning [J]. Neural Networks, 2007, 20(8): 893-903.
[166] Deng X S and Wang X Z. Incremental learning of dynamic fuzzy neural networks for accurate system modeling [J]. Fuzzy Sets and Systems, 2009, 160(7): 972-987.
[167] Zou L, Zhang T, and Cao Z. An incremental learning algorithm based on Support Vector Machine for pattern recognition [C].In Proc. SPIE, 2009, 7496.
[168] Liang Z Z and Li Y F. Incremental support vector machine learning in the primal and applications [J]. Neurocomputing, 2009, 72 (10): 2249-2258.
[169] Givoni I E, Frey B J. Semi Supervised Affinity Propagation with Instance Level [C]. In Proceedings of the 12th International Conference on Articial Intelligence and Statistics (AISTATS) 2009, Clearwater Beach, Florida, USA, 2009, 5: 161-168.
[170] Rand W M. Objective criteria for the evaluation of clustering methods [J]. J. American Stat. Assoc., 1971, 6(336): 846-850.
[2]王静等.土地资源遥感监测与评价方法[M].北京:科学出版社,2006.
[3]叶树华,任志远,遥感概论[M].陕西:陕西科学技术出版社,1993.
[4]汤国安,张友顺,刘咏梅,谢元礼,杨昕,刘爱利.遥感数字图像处理[M].北京:科学出版社,2004.
[5] Lee W T. The face of the earth as seen from the air: A study in the application of airplane photography to geography [M]. New York: American Geographical Society, Special Publication, 1922.
[6] Tucker C J, Townshend J R G, Goff T E. Africa land cover classification using satellite data [J]. Science, 1985, 227: 369-375.
[7] Townshend J R G, Justice C O, Kalb V T. Characterization and classification of South America land covertypes using satellite data [J]. International Journal of Remote Sensing, 1987, 8: 1189-1207.
[8] Loveland T R, Reed B C, Brown J F, et al. Development of a global land cover characteristics database and IGBP DISCover from 1km AVHRR data [J]. Internation al Journal of Remote Sensing, 2000, 21(6&7): 1303-1330.
[9] Lee C and Landgrebe D. Feature extraction and classification algorithms for high dimension data: [PhD thesis]. USA: Purdue University.1993.
[10] Samuel A. Some studies in machine learning using the game of checkers [J]. IBM Journal of Research and Development, 1959, 3: 211-229.
[11] Mitchell, T. Machine Learning [M]. McGraw-Hill, 1997.
[12]周志华.机器学习及其挑战[PPT].南京大学计算机软件新技术国家重点实验室网站http://cs.nju.edu.cn/people/zhouzh/. 2003.
[13] Mitchell, T M著,曾华军,张银奎等译.机器学习[M].北京:机械工业出版社, 2003.
[14] Baldi P, Brunak著,张东晖等译.生物信息学——机器学习方法[M].北京:中信出版社, 2003.
[15]王珏.机器学习研究回顾与趋势[PPT].中科院自动化研究所模式识别国家重点实验室网站http://www.intsci.ac.cn/research/wangj04.ppt, 2004.
[16]曾志远.卫星遥感图像计算机分类与地学应用研究[M].北京:科学出版社, 2004.
[17] Zhao G and Maclean A L. A comparison of canonical discriminant analysis and principal component analysis for spectral transformation [J]. Photogrammetric Engineering&Remote Sensing, 2000, 66(7): 841-847.
[18] Almeida-Fiho R and Shimabukuro Y E. Digital processing of a Landset TM time series for mappingand monitoring degraded areas caused by independent gold miners, Roraima State, Brazilian Amazon [J]. Remote Sensing of Environment, 2002, 79(1): 42-50.
[19] Mitternicht G I and Zinck J A. Remote sensing of soil salinity: potentials and constraints [J]. Remote Sensing of Environment, 2003, 85: 1-20.
[20] Reese H M, Lillesand T M, Nagel D E, Stewart J S, Gold-mann R A, Simmons T E, Chipman J W and Tessar P A. Statewide land cover derived from multisensonal Landset TM data: a retrospective of the WISCLAND project [J]. Remote Sensing of Environment, 2002, 82: 224-237.
[21] Sarbu C, Pop H F. Principal component analysis versus fuzzy principal component analysis a case study: the quality of Danube water (1985-1996) [J]. Talanta65, 2005: 1215-1220.
[22] POP H F. Principal components analysis based on a fuzzy sets approach [J]. Studia Univ. Babes-bolyai, Informatica, 2001, 46(2): 45-52.
[23]杨翠芬,田村正行.差分主成分分析在辽河三角洲景观变换中的应用[J].地理学报, 2004, 59(4): 592-598.
[24] Colwell R N. Manual of Remote Sensing: 2ndEd [M]. Bethesda: American Society for Photogrammetry&Remote Sensing, 1983.
[25] Colwell R N. Manual of Photographic Interpretation: 2ndEd [M]. Bethesda: American Society for Photogrammetry&Remote Sensing, 1997.
[26] MacQueen J. Some methods for classification and analysis of multivariate observations [C]. In Proc. of 5th Berkeley Symp. on Math. Stat. and Prob., 1967, 1: 281–297.
[27] Thitimajshima P. A new modified fuzzy c-means algorithm for multispectral satellite images segmentation [C]. In Proc. IGARSS, 2000, 4: 1684-1686.
[28] Zhong Y F, Zhang L P, Huang B and Li P X. An unsupervised artificial immune classifier for multi/hyperspectral remote sensing imagery [J]. IEEE Trans. Geosci. Remote Sens., 44(2): 420-431, 2006.
[29] Ding, Z J, Yu J and Zhang Y Q. A new improved k-means algorithm with penalized term [C]. IEEE Int. Conf. Granular Computing, 2007.
[30] Wang Y and Mo J. Fuzzy logic applied in remote sensing image classification [C]. In Proc. Int. Conf. Systems, Man and Cybernetics, 2004, 6378-6382.
[31] Dunn J C. A fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters [J]. J. Cybernet, 1974, 3(3): 32-57.
[32] Bezdek J C. Pattern Recognition with Fuzzy Objective Function Algorithms [M]. Plenum. New York. 1981.
[33] Liu P Y, Jia K B and Zhang P Z. An effective method of image retrieval based on modified fuzzy c-means clustering scheme [C]. IEEE Int. Conf. Sign Processing, 2006.
[34]钱乐详等.遥感数字影像处理与地理特征提取[M].北京:科学出版社, 2005.
[35]承继成,郭华东,史文中.遥感数据的不确定性问题[M].北京:科学出版社, 2004.
[36] Jawahar C V and Ray A K. Fuzzy statistics of digital images [J]. IEEE Signal Process. Lett., 1996, 3(8): 225-227.
[37] Yang C, Lu L J, Lin H P, Guan R C, Shi X H and. Liang Y C. A fuzzy-statistics-based principal component analysis (FS-PCA) method for multispectral image enhancement and display [J]. IEEE Trans. Geosci. Remote Sensing, 2008, 46(11): 3937-3947.
[38] Taheri S M. Trends in fuzzy statistics [M]. Austrian J. Stat., 2003, 32: 239-257.
[39] Frey B J, Dueck D, Clustering by passing messages between data points [J]. Science, 2007, 315: 972-976.
[40] Jiang W, Ding F, Xiang Q L. An affinity propagation based method for vector quantization [J]. Eprint arXiv: 0710.2037.
[41] Frey B J, Dueck D. Non-metric affinity propagation for unsupervised image categorization [C]. In the ICCV 2007, Rio de Janeiro, Brazil, 2007.
[42] Brusco M J and Kohn H F. Comment on‘clustering by passing messages between data points’[J]. Science, 2008, 319(5864): 726c.
[43] Frey B J, Dueck D. Response to comment on‘clustering by passing messages between data points’[J]. Science, 2008, 319(5864): 726d.
[44] Jensen J R, Qiu F, Ji M. Predictive modelling of coniferous forest age using statistical and artificial neural network approaches applied to remote sensor data [J]. International Journal of Remote Sensing. 1999, 20(14): 2805-2822.
[45] Foody G M, Arora M K. An evaluation of some factors affecting the accuracy of classification by an artificial neural network [J]. International Journal of Remote Sensing, 1997, 18(4): 799-810.
[46] Kuo B and David A L. A covariance estimator for small size classification problem and its application to feature extraction [J]. IEEE Trans. Geosci. Remote Sens., 2002, 40(4): 814-819.
[47] Jeon B and Landgrebe D. Partially supervised classification using weighted unsupervised Clustering [J]. IEEE Trans. Geosci. Remote Sens., 2002, 37(2): 1073-1079.
[48] Zhang J, Zhang Y and Zhou T. Classification of hyperspectral data using support vector machine [C]. In: IEEE International Conference on Image Processing, 2001, 882-885.
[49] Huang C, Davis L and Townshend J. An assessment of support vector machines for land cover classification [J]. Int. remote sensing, 2002, 23(4): 725-749.
[50] Melgani F, Bruzzone L. Classification of hyperspectral remote-sensing images with support vector machines [J]. IEEE Trans. Geosci. Remote Sens., 2004, 42(8): 1778-1790.
[51]骆剑承,周成虎,梁怡等.支撑向量机及其遥感影像空间特征提取和分类的应用研究[J].遥感学报, 2002, 6(1): 50-55.
[52]何灵敏.支持向量机集成及在遥感分类中的应用[D].浙江大学, 2006.
[53]祁享年,杨建刚,方陆明.基于多类支持向量机的遥感图像分类及其半监督式改进策略[J].复旦学报, 2004, 43(5): 158-162.
[54]刘志刚,史文中,李德仁,秦前清.一种基于支撑向量机的遥感影像不完全监督分类新方法[J].遥感学报, 2005, 9(4): 158-162.
[55] Baraldi A, Bruzzone L, Blonda P. A multiscale expectation-maximization semisupervised classifiersuitable for badly posed image classification [J]. IEEE Transactions on Image Processing, 2006, 5(8): 2208-2225.
[56] Bruzzone L, Chi M M, Marconcini M. A novel transductive SVM for the semisupervised classification of remote-sensing images [J]. IEEE Trans. Geosci. Remote Sensing, 2006, 44(11): 3363-3373, 2006.
[57] Chi M M, Bruzzone L. Classification of hyperspectral data by continuation semi-supervised SVM [C]. IEEE Int. Conf. Geosci. Remote Sensing Symposium, IGARSS, 2007, 3794-3797.
[58] Santiteban A and Munoz L. Principal component of a mutilspectral image: application to a geokogical problem [J]. IBM Journal of Research and Development, 1978, 22(5): 444-454.
[59] Giddings L E, Soto M, and Angulo M J. The use of landsat data in mapping tropical vegetation monitoring [C]. In Processing of the 14th International Symposium on Remote Sensing of Environment, 1980, 3: 1383-1387.
[60]孙家炳.遥感原理与应用[M],武汉:武汉大学出版社. 2003.
[61]高新波.模糊聚类分析及其应用[M],西安:西安电子科技大学出版社. 2004.
[62] Richards J A and Jia X P. Remote Sensing Digital Image Analysis: An Introduction: 4th Ed [M]. Berlin, Germany: Springer-Verlag, 2006.
[63] Jensen J R. Introductory Digital Image Processing: A Remote Sensing Perspective [M]. Upper Saddle River, NJ: Prentice-Hall, 2005.
[64] Bryant J. On displaying multispectral imagery [J]. Photogramm. Eng. Remote Sens., 1988, 54(12): 1739-1743.
[65] Du Q and Fowler J E. Hyperspectral image compression using JPEG2000 and principal component analysis [J]. IEEE Geosci. Remote Sens. Letter, 2007, 4(2): 201–205.
[66] Farrell M D and Mersereau R M. On the impact of PCA dimension reduction for hyperspectral detection of difficult targets [J]. IEEE Geosci. Remote Sens. Letter, 2005, 2(2): 192-195.
[67] Wang J and Chang C I. Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis [J]. IEEE Geosci. Remote Sens. Letter, 2006, 44(6): 1586-1600.
[68] Moussaoui S. On the decomposition of Mars hyperspectral data by ICA and Bayesian positive source separation [J]. Neurocomputing, 2008, 71(10-12): 2194-2208.
[69] Zubko V, Kaufman Y J, Burg R I, and Martins J V. Principal component analysis of remote sensing of aerosols over oceans [J]. IEEE Trans. Geosci. Remote Sens., 2007, 45(3): 730-745.
[70] Li R, Mersereau M, and Simske S. Atmospheric turbulencedegraded image restoration using principal components analysis [J]. IEEE Geosci. Remote Sens. Letter, 2007, 4(3): 340-344.
[71] Jimenez-Rodriguez L O, Arzuaga-Cruz E, and Velez-Reyes M. Unsupervised linear feature-extraction methods and their effects in the classification of high-dimensional data [J]. IEEE Trans. Geosci. Remote Sens., 2007, 45(2): 469-483.
[72] Fauvel M, Chanussot J, and Benediktsson J A. Kernel principal component analysis for feature reduction in hyperspectrale images analysis [C]. In Proc. 7th NORSIG, 2006, 238-241.
[73] Ricotta C, Avena G C, and Volpe F. The influence of principal component analysis on the spatial structure of a multispectral dataset [J]. Int. J. Remote Sens., 1999, 20(17): 3367-3376.
[74] Liang Y C, Lee H P, Lim S P, Lin W Z, Lee K H, and Wu C G. Proper orthogonal decomposition and its applications—Part I: Theory [J]. J. Sound Vib., 2002, 252(3): 527–544.
[75] Yao H B and Tian L. A genetic-algorithm-based selective principal component analysis (GA-SPCA) method for high-dimensional data feature extraction [J]. IEEE Trans. Geosci. Remote Sens., 2003, 41(6): 1469-1478.
[76] Hubert M, Rousseeuw P J, and Verboven S A. Fast method for robust principal components with applications to chemometrics [J]. Chemom. Intell. Lab. Syst., 2002. 60(1): 101-111.
[77] Yu X and Cheng X. Research of independent component analysis [C]. In Proc. IEEE Int. Conf. Syst., Man Cybern., 2004, 4804-4809.
[78] Hyvarinen A, Karhunen J, and Oja E. Independent Component Analysis [M]. Hoboken, NJ: Wiley, 2001.
[79] Sun Z L, Huang D S, Cheung Y M, Liu J, and Huang G B. Using FCMC, FVS, and PCA techniques for feature extraction of multispectral images [J]. IEEE Geosci. Remote Sens. Letter, 2005, 2(2): 108-112.
[80] Serge A, Ludovic R, Yannick C, and Alain B. A fuzzy-possibilistic scheme of study for objects with indeterminate boundaries: Application to French Polynesian reefscapes [J]. IEEE Trans. Geosci. Remote Sens., 2000, 38(1): 257-270.
[81] Zadeh L A. Fuzzy sets [J]. Inf. Control, 1965, 8: 338-353.
[82] Wang F. Improving remote sensing image analysis through fuzzy information representation [J]. Photogramm. Eng. Remote Sens., 1990. 56(8): 1160-1169.
[83] Chanussot J, Mauris G, and Lambert P. Fuzzy fusion techniques for linear features detection in multitemporal SAR images [J]. IEEE Trans. Geosci. Remote Sens., 1999, 37(3): 1292-1305.
[84] Filippi A. M and Jensen J R. Effect of continuum removal on hyperspectral coastal vegetation classification using a fuzzy learning vector quantizer [J]. IEEE Trans. Geosci. Remote Sens., 2007, 45(6): 1857-1869.
[85] Vlag D E and Stein A. Incorporating uncertainty via hierarchical classification using fuzzy decision trees [J]. IEEE Trans. Geosci. Remote Sens., 2007, 45(1): 237-245.
[86] Chanussot J, Benediktsson J A, and Fauvel M. Classification of remote sensing images from urban areas using a fuzzy possibilistic model [J]. IEEE Geosci. Remote Sens. Letter, 2006, 3(1): 40-44.
[87] Cheng Q M, Jing L H, and Panahi A. Principal component analysis with optimum order sample correlation coefficient for image enhancement [J]. Int. J. Remote Sens., 2006, 27(16): 3387-3401.
[88] Narumalani S, Hlady J and Jensen J R. Information extraction from remotely sensed data [J], Manual of Geospatial Science and Technology, 2002, 299-324.
[89] Davis J C. Statistics and data analysis in geology: 3rd ed. [M]. New York: John Wiley & Sons, 2002.
[90] Freud R J and Wilson W J. Statistical Methods: 2nd ed. [M]. New York: Academic Press, 2003.
[91] Samuels M L and Witmer J A. Statistics for the Life Science [M]. Upper Saddle River, NJ:Prentice-Hall, 2003.
[92] Zadeh L A. Outline of a new approach to the analysis of complex systems and decision process [J]. IEEE Trans. Syst.,Man, Cybern., 1973, 3(1): 28-44, 1973.
[93] Kandel A and Byatt W J. Fuzzy sets, fuzzy algebra, and fuzzy statistics [J]. Proc. IEEE, 1978, 66(12): 1619-1639.
[94] Bezdek J C and Pal S K. Fuzzy Models for Pattern Recognition [M]. New York: IEEE, 1992.
[95] Chen B, Chen Y, and Hsu W. Automatic histogram specification based on fuzzy set operations for image enhancement [J]. IEEE Signal Process. Letter, 1995, 2(2): 37-40.
[96] Yang X and Toh P S. Adaptive fuzzy multilevel median filter [J]. IEEE Trans. Image Process., 1995, 4(5): 680-682.
[97]马建文,李启青等.遥感数据智能处理方法与程序设计[M].北京:科学出版社, 2005.
[98] Townshend J R G and Justice C O. Towards operational monitoring of terrestrial systems by moderate-resolution remote sensing [J]. Remote Sensing of Environment, 2002, 83: 351-359.
[99] Lunetta R S and Lyons J G. Geospatial Data Accuracy Assessment [M]. Las Vegas: Environmental Protection Agency, 2003.
[100] Loveland T R, Zhiliang Z, Ohlen D O, Brown J F, Reed B C and Yang L. An analysis of the IGBP global landcover characterization process [J]. Photogrammetric Engineering & Remote Sensing, 1999, 65(9): 1021-1032.
[101] Huang K. A synergistic automatic clustering technique for multispectral image analysis [J]. Photogrammetric Engineering & Remote Sensing, 2002, 68(1): 33-44.
[102] Bandyopadhyay S, Maulik U, and Mukhopadhyay A. Multiobjective genetic clustering for pixel classification in remote sensing imagery [J]. IEEE Trans. Geosci. Remote Sensing, 2007, 45(5): 1506-1511.
[103] Chi M M, Qian Q, and Benediktsson J A. Cluster-based ensemble classification for hyperspectral remote sensing [C]. IEEE Int. Conf. Geosci. Remote Sensing Symposium, IGARSS, 2008, 209-212.
[104] Maulik U and Saha I. Modified differential evolution based fuzzy clustering for pixel classification in remote sensing imagery [J]. Pattern Recognition, 2009, 42(9): 2135-2149.
[105] Duda R O, Hart P E and Stork D G. Pattern Classification [M], New York: John Wiley & Sons, 2001.
[106] Robert A S. Remote Sensing: Models and Methods for Image Processing: 3rd ed. [M]. Elsevier Inc. USA, 2007.
[107] Aldenderfer M S and Blashfield R K. Cluster Analysis [M]. Sage Publications, Beverly Hills. USA, 1984.
[108] Chehata N and Bretar F. Terrain modeling from lidar data: Hierarchical K-means filtering and Markovian regularization [C]. IEEE Int. Conf. Image Processing, ICIP, 2008, 1900-1903.
[109] Zheng J, Cui Z Z, Liu A F, and Jia Y. A k-means remote sensing image classification method based on AdaBoost [C]. IEEE Int. Conf. Natural Computation, ICNC, 2008, 27-32.
[110] Maheshwary P and Srivastav N. Retrieving similar image using color moment feature detector andk-means clustering of remote sensing images [C]. IEEE Int. Conf. Computer and Electrical Engineering, 2008, 20-22.
[111] Witten L H and Frank E. Data mining: practical machine learning tools and techniques: 2nd ed. [M]. Elsevier Inc. USA, 2005.
[112] Wu X D, Kumar V, and Quinlan J R. Top 10 algorithms in data mining [J]. Knowledge and Information Systems, 2008, 14(1):1-37.
[113] Dinesh M S, Gowda K G, and Nagabhuehan P. Unsupervised classification for remotely sensed data using fuzzy set theory [C]. IEEE Int. Geosci. Remote Sensing- A Scientific Vision for Sustainable Development, IGARSS, 1997, 1:521-523.
[114] Borasca B, Bruzzone L, Carlin L, and Zusi M. A fuzzy-input fuzzy-output SVM technique for classification of hyperspectral remote sensing images [C]. Signal Processing Symposium, Processing of the 7th Nordic, NORSIG, 2006, 2-5.
[115] Altman D. Efficient fuzzy clustering of multi-spectral images [C]. IEEE Int. Conf. Geoscience and Remote Sensing Symposium, IGARSS, 1999, 3: 1594-1596.
[116] Gorsevski P V, Gessler P E, and Jankowski P. Integrating a fuzzy k-means classification and a Bayesian approach for spatial prediction of landslide hazard [J]. Journal of Geographical Systems, 2003, 5(3): 223-251.
[117] Liu X F, Li X W, Zhang Y, Yang C J, Xu W B, Li M, and Luo H M. Remote sensing image classification based on dot density function weighted FCM clustering algorithm [C]. IEEE Int. Conf. Geosci. Remote Sensing Symposium, IGARSS, 2007, 2010-2013.
[118] Hung C C, Liu W P, and Kuo B C. A new adaptive fuzzy clustering algorithm for remotely sensed images [C]. Int. Conf. Geoscience and Remote Sensing Symposium, IGARSS, 2008, 2: 863-866.
[119] Fan J C, Han M, and Wang J. Single point iterative weighted fuzzy C-means clustering algorithm for remote sensing image segmentation [J]. Pattern Recognition, 2009, 42(11): 2527-2540.
[120] Zhang H Y, Wu Q T, and Pu J X. A novel fuzzy kernel clustering algorithm for outlier detection [C], IEEE Int. Conf. Mechatronics and Automation, ICMA, 2007, 2378-2382.
[121] Wu K L and Yang M S. Alternative c-means clustering algorithms [J]. Pattern Recognition, 2002, 35(10): 2267-2278.
[122] Gath I and Geva A B. Unsupervised Optimal Fuzzy Clustering [J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 1989, 11: 773-781.
[123] Nasser S, Alkhaldi R and Vert G. A modified fuzzy k-means clustering using expectation maximization [C]. IEEE Int. Conf. Fuzzy Systems, 2006, 231-235.
[124] Li M J, Ng M K, and Cheung Y. Agglomerative fuzzy k-means clustering algorithm with selection of number of clusters [J]. IEEE Trans. Knowledge and Data Engineering, 2008, 20(11): 1519-1534.
[125] Frey B J and Dueck D. Mixture modeling by affinity propagation [J]. In Advances in Neural Information Processing Systems 18, MIT Press, 2006, 2, 3, 7.
[126] http://www.psi.toronto.edu/affinitypropagation/faq.html.
[127] Schowengerdt R A. Remote sensing models and methods for image processing: 2nd Edition [M].San Diego London Bobton New York Sydney Tokyo Toronto: Academic Press, 1997.
[128] Sun M C and Chou C H. A modified version of the K-means algorithm with a distance based on cluster symmetry [J]. IEEE Tran. Pattern Analysis and Machine Intelligence, 2001, 23(6): 674-680.
[129] Sweet J N. The spectral similarity scale and its application to the classification of hyperspectral remote sensing data [J]. IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2004, 92-99.
[130] Chintalapudi K K and Kam M. A noise-resistant fuzzy C-means algorithm for clustering [C]. Fuzzy Systems Proceedings. IEEE World Congress on Computational Intelligence, 1998, 2: 1458-1463.
[131] Fung T and Drew E L. The determination of optimal threshold levels for change detection using various accuracy indices [J]. Photogrammetric Eng. and Remote Sensing, 1998, 54: 1449-1454.
[132] Liu C, Frazier P, and Kumar L. Comparative assessment of the measures of thematic classification accuracy [J]. Photogramm. Eng. and Remote Sensing, 2007, 107(4): 606-616.
[133] Short N M. The Landsat Tutorial Workbook Basics of Satellite Remote Sensing [M]. Greenbelt, Md., Goddard Space Flight Center, NASA Reference Publication 1078, 1982.
[134] Rosenfield G H and Fitzpatrick-Lins K. A coefficient of agreement as a measure of thematic classification accuracy [J]. Photogramm. Eng. and Remote Sensing, 1986, 52(2): 223-227.
[135] Maulik U and Bandyopadhyay S. Fuzzy partitioning using a real-coded variable- length genetic algorithm for pixel classification [J]. IEEE Trans. Geosci. Remote Sensing, 2003, 41(5): 1075-1081.
[136] Tran T N, Wehrens R, Hoekman D H and Buydens L M C. Initialization of markov random field clustering of large remote sensing images [J]. IEEE Trans. Geosci. Remote Sensing, 2005, 43(8): 1912-1919.
[137] Grira N, Crucianu M and Boujemaa N. Unsupervised and semi-supervised clustering: a brief survey [C]. In A Review of Machine Learning Techniques for Processing Multimedia Content, Report of the MUSCLE European Network of Excellence (6th Framework Programme), 2005.
[138] Nguyen N and Caruana R. Classification with partial labels [C]. In proc. of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, 2008, 551-559.
[139] Wagsta K and Cardie C. Clustering with instance-level constraints [C]. In proc. of the 17th International Conference on Machine Learning, 2000, 1103-1110.
[140] Givoni I E and Frey B J. Semi-supervised affinity propagation with instance-level constraints [C]. In proc. of the 12th International Conference on Artificial Intelligence and Statistics (AISTATS), 2009, 161-168,
[141] Zhu X. Semi-supervised learning literature survey [C]. Technical Report 1530, Computer Sciences, University of Wisconsin-Madison, 2005.
[142] Cozman F, Cohen I, and Cirelo M. Semi-supervised learning of mixture models and bayesian networks [C]. In Proc. of the 12th International Conference of Machine Learning, 2003.
[143] Karakoulas G and Salakhutdinov R. Semi-supervised mixture-of-experts classification [C]. In Proc. of the 4th IEEE International Conference on Data Mining, 2004, 138-145.
[144] Blum A and Mitchell T. Combining labeled and unlabeled data with co-training [C]. In Proc. of the11th Conference on Computational Learning Theory, 1998, 92-100.
[145] Nigam K and Ghani R. Analyzing the effectiveness and applicability of co-training [C]. In Proc. of the 9th International Conference on Information and Knowledge Management, 2000, 86-93.
[146] Moreno P J and Agarwal S. An experimental study of EM-based algorithms for semi-supervised learning in audio classification [C]. In Proc. of ICML-2003 Workshop on Continuum from Labeled to Unlabeled Data, 2003.
[147] Rosenberg C, Hebert M and Schneiderman H. Semi-supervised self-training of Object Detection Models [C]. In Proc. of 7th IEEE Workshop on Applications of Computer Vision, 2005, 29-36.
[148] Ghanramani Z and Jordan M I. Supervised learning from incomplete data via the EM approach [J]. Advances in Neural Information Processing Systems, 1994, 120-127.
[149] Wagstaff K, Cardie C, Rogers S, and Schroedl S. Constrained K-Means clustering with background knowledge [C]. In Proceedings of 18th International Conference on Machine Learning (ICML), 2001.
[150] Basu S, Bilenko M and Mooney R J. Comparing and unifying search-based and similarity-based approaches to semi-supervised clustering [C]. In Proc. of the ICML-2003 Workshop on the Continuum from Labeled to Unlabeled Data in Machine Learning and Data Mining System, 2003.
[151] Demiriz A, Bennett K P, and Embrechts M J. Semi-supervised clustering using genetic algorithms [C]. ANNIE'99 (Articial Neural Networks in Engineering), 1999.
[152] Bilenko M, Basu S, and Mooney R J. Integrating constraints and metric learning in semi-supervised clustering [C]. In Proceedings of 21th International Conference on Machine Learning, 2004.
[153] Globerson A and Roweis S. Metric learning by collapsing classes [J]. In Weiss, Y.; Sch¨olkopf, B.; and Platt, J., eds., Advances in Neural Information Processing Systems, Cambridge, MA: MIT Press, 2006, 18: 451-458.
[154] Ghodsi A, Wilkinson D F, and Southey F. Improving embeddings by flexible exploitation of side information [C]. In Veloso M M, ed., International Joint Conference on Artificial Intelligence, 2007, 810-816.
[155] Bilenko M, Basu S, and Mooney R J. Semi-supervised clustering by seeding [C]. In Proceedings of 19th International Conference on Machine Learning, 2002.
[156] Laird J E, Rosenbloom P S and Newel A. Chunking in SOAR: the anatomy of a general learning mechanism [J]. Machine Learning, 1986, 1: 11-46.
[157] Sutton R. Learning to precdict by the methods of temporal differences [J]. Machine Learning, 1988, 3: 9-44.
[158] Utgoff P E. Incremental induction of decision trees [J]. Machine Learning, 1989, 4(2):161-186.
[159] Utgoff P E. An improved algorithm for incremental induction of decision trees [C]. In Processings of the 10th International Machine Learning Workshop, Rutgers University, New Brunswick, NJ, 1993.
[160] Shen W M. Complementary discrimination learning: A duality between generalization and discrimination [C]. In Proceedings of 8th National Conference on Artificial Intelligence. MIT Press,1990.
[161] Shen W M. Complementary discrimination learning with decision lists [C]. In Proceedings of 8th National Conference on Artificial Intelligence. MIT Press, 1992.
[162] Shen W M. Learning finite state automata using local distinguishing experiments [C]. In Proceedings of IJCAI-93, Chambery, France, 1993.
[163] Shen W M. Efficient Incremental Induction of Decision Lists - Can Incremental Learning Outperform Non-Incremental Learning [J]. 1996.
[164] Ozawa S, Toh S L, Abe S, Pang S N and Kasabov N. Incremental learning of feature space and classifier for face recognition [J]. Neural Networks, 2005, 18 (5-6): 575-584.
[165] Shen F R, Tomotaka O, and Osamu H. An enhanced self-organizing incremental neural network for online unsupervised learning [J]. Neural Networks, 2007, 20(8): 893-903.
[166] Deng X S and Wang X Z. Incremental learning of dynamic fuzzy neural networks for accurate system modeling [J]. Fuzzy Sets and Systems, 2009, 160(7): 972-987.
[167] Zou L, Zhang T, and Cao Z. An incremental learning algorithm based on Support Vector Machine for pattern recognition [C].In Proc. SPIE, 2009, 7496.
[168] Liang Z Z and Li Y F. Incremental support vector machine learning in the primal and applications [J]. Neurocomputing, 2009, 72 (10): 2249-2258.
[169] Givoni I E, Frey B J. Semi Supervised Affinity Propagation with Instance Level [C]. In Proceedings of the 12th International Conference on Articial Intelligence and Statistics (AISTATS) 2009, Clearwater Beach, Florida, USA, 2009, 5: 161-168.
[170] Rand W M. Objective criteria for the evaluation of clustering methods [J]. J. American Stat. Assoc., 1971, 6(336): 846-850.