高光谱图像分类及端元提取方法研究
|
摘要
高光谱遥感是将目标探测技术与光谱成像技术相结合的多维地物信息获取技术，可以同时获取描述地物分布的二维空间信息与描述地物光谱特征属性的一维光谱信息。随着光谱分辨率的不断增加，人们对地物光谱特征的认知能力也不断深入，许多隐藏在狭窄光谱范围内的地物特性逐渐被人们所发现。“端元”为高光谱数据中可以详尽表示待测地物光谱属性的纯像素，获得的端元向量通常作为高光谱图像处理算法的先验知识，因此得到的端元向量是否可以突出地反映待研究地物的光谱属性信息，对其他高光谱算法的处理精度起到极其重要的作用。相对于多光谱遥感而言，高光谱遥感具有更加丰富的地物光谱信息，可以详尽地反映待测地物细微的光谱属性，其较宽的波谱覆盖范围使得在高光谱数据处理时，可以根据需要选择特定的波段来突显地物特征，为高光谱数据处理算法提供更多的地物原始数据，使地物光谱信息的精确处理与分析成为可能。但是由于高光谱图像具有较高的数据维数，使常规的图像处理方法在处理高光谱图像时有较大的限制。为此本文从分析基本高光谱遥感图像处理理论和现有算法及相关学科技术入手，重点研究了高光谱遥感图像的特征降维、端元提取以及分类处理方法。
在特征降维方面，当训练样本较少时，高光谱数据的分类精度受到严重的影响。通常解决这种现象的办法是对原始数据进行特征降维处理，然而多数特征降维算法无法直接给出最优降维特征数。为此提出利用蒙特卡罗随机实验可以对特征参量进行统计估计的特性，计算高光谱图像的最优降维波段数，并与相关向量机结合，对降维后的数据进行分类。实验结果表明了使用蒙特卡罗算法求解降维波段数的可靠性。相比较原始未降维数据，高光谱图像经过蒙特卡罗特征降维算法处理后，分类精度有较大幅度的提高。
在高光谱端元提取方面，通过对各种算法的分析，主要研究端元提取效果较好的N-FINDR算法。然而样本的排序对该算法的端元提取会造成一定影响，并且传统N-FINDR算法需要根据端元的个数进行降维处理，从而限制了该算法的应用。实际高光谱数据中存在的同一地物在高维空间中非紧密团聚现象也对端元提取增加了难度。为此提出改进的算法停机准则和数据特征预处理方法，并使用支持向量机对提取到的端元进行二次提取。实验结果表明，改进的停机准则进一步增加了由端元向量组组成的凸体体积。数据特征预处理和基于支持向量机的二次端元提取分别提升了数据的可分性和提取到端元的精度。
在高光谱图像分类方面，模糊C-均值聚类算法因算法简单、收敛速度快等优点受到广泛的关注。然而由于高光谱数据的维数较高，其光谱波段的非线性特性使得传统模糊C-均值聚类算法无法在原始空间得到较好的聚类结果。另外，模糊C-均值聚类算法在计算聚类中心时，仅使用了各样本对聚类中心的隶属度，忽略了样本之间固有存在的空间分布特征。为此提出了模糊核加权C-均值聚类算法，在计算模糊核聚类中心时，根据样本的空间分布特征，为每个样本分配不同的权值，使得每个核聚类中心随着样本的不同而各有不同。标准数据和实际高光谱数据的实验结果均表明，相比较传统模糊C-均值聚类算法，模糊核加权C-均值聚类算法在总体分类精度上有较大的提高。
支持向量机被广泛使用在高光谱数据分类中，使用传统支持向量机对高光谱图像进行分类时，认定每个特征波段拥有相同的权值。然而在小样本高光谱图像分类中，由于冗余波段的影响，极容易造成“维数灾难”现象，使得分类精度严重下降。为此提出两种特征加权支持向量机以消除“维数灾难”现象，提高高光谱图像分类的精度：1)ReliefF特征加权支持向量机(RSVM)，2)模糊ReliefF特征加权支持向量机(FRSVM)。实验选取玉米种子图像和公开使用的高光谱数据图像作为实验数据。相对比传统支持向量机，提出的两种加权算法均提高了高光谱图像分类的精度，并且降低了“维数灾难”的影响。
Hyperspectral remote sensing is the multi-dimensional information obtaining technology,which combines target detection and spectral imaging technology together. That is, it couldobtain the two-dimensional object distribution information and one-dimensional spectralfeature characteristic information at the same time. With the increasing of spectral resolution,regarding the spectral characteristic of the objects, people's recognition ability goes deeperthat many characteristics originally hidden in the narrow spectral bands could be discovered."Endmember" is defined as the ideal pure data, which could represent the characteristic of theobjects. Since endmember is generally used as the pre-knowledge of hyperspectral imageryprocessing methods, it plays an important role for the following processing steps， whichdepends on whether the obtained endmembers could represent the characteristic of the objects.Compare with multi-spectral remote sensing, hyperspectral remote sensing provides fruitfulspectral information, which could highlight the tiny spectral characteristic of the objects. Thewide spectral range make it possible for the user to select the special bands to highlight thecharacteristic of the objects, which could provide more original data for the hyperspectralimagery processing methods and make the precise processing of spectral information possible.However, due to the high dimension of the hyperspectral imagery, it has a huge limit whentraditional imagery processing methods are used in hyperspectral imagery. Based on the basichyperspectral remote sensing imagery processing theories and relevant subjects, this studymainly focuses on the feature reduction, endmember extraction, and the classification ofhyperspectral remote sensing imagery.
Regarding feature reduction, due to the high dimension of hyperspectral data, theclassification accuracy is severely affected when there are few training samples. Featurereduction is a common method to deal with this phenomenon. However, most of the featurereduction methods can’t provide optimal feature reduction number. So this study proposes toutilize the statistic estimation characteristic of Monte Carlo random experiments to calculateoptimal feature reduction number and conduct hyperspectral imagery classification withrelevance vector machine. Experiment results show the reliability of the feature reduction number calculated by Monte Carlo method. Compare with the classification of original data,it has a significant improvement on the classification accuracy with the feature reduction data.
Regarding hyperspectral endmember extraction, through the analysis of variousalgorithms, we mainly focus on the N-FINDR endmember extraction algorithm which hasbetter performance. However, the order of the samples has a certain effect on the endmemberextraction, and traditional N-FINDR algorithm also needs to reduce the dimensionality basedon the number of the endmembers, which will limit its application. In the actual hyperspectraldata, the incompact clustering of the same species presented in the high dimensional spacealso increases the difficulty of endmember extraction. So this study proposed an improvedstop rule and the pretreatment of the features, and utilizing Support Vector Machine (SVM) toconduct the second endmember extraction. Experiments show that the improved stop rulefurther increased the volume of the convex polyhedron composed of the endmembers. Thepretreatment of the features and the second SVM based endmember extraction increase theseparability of the data and the precision of the extracted endmembers respectively.
Regarding hyperspectral imagery classification, fuzzy C-means clustering algorithm iswidely utilized for its simpleness and fast convergence rate. Due to the high dimensionality ofhyperspectral data, the nonlinear characteristic of the spectral bands makes it difficult for thetraditional fuzzy C-means clustering algorithm to have good clustering result in the originalspace. Moreover, fuzzy C-means clustering algorithm just uses the membership degree tocalculate the clustering center, which omits the space distribution that intrinsic exists amongthe samples. So this study proposes fuzzy kernel weighted C-means clustering algorithm.When it is used to calculate the fuzzy kernel clustering center, different weights will beassigned to each sample according to the space distribution. As a result, different sampleshave different set of kernel clustering centers. Compare with traditional fuzzy C-meansclustering algorithm, experiments results of both standard data and actual hyperspectral dataprove that the proposed fuzzy kernel weighted C-means clustering algorithm has asignificance improvement on the overall classification accuracy.
Support vector machine is widely used in the classification of hyperspectral reflectancedata. In traditional SVM, features are generated from all or subsets of spectral bands witheach feature contributing equally to the classification. In the classification of smallhyperspectral reflectance data sets, a common challenge is Hughes phenomenon, which is caused by many redundant features and resulting in subsequent poor classification accuracy.In this study, we examined two approaches to assigning weights to SVM features to increaseclassification accuracy and reduce adverse effects of Hughes phenomenon:1)“RSVM” refersto support vector machine with ReliefF feature weighting algorithm, and2)“FRSVM” refersto support vector machine with fuzzy ReliefF feature weighting algorithm. Analyses wereconducted on a reflectance data set of individual maize kernels from three inbred lines and apublic data set with three selected land-cover classes. Both weighting methods increasedclassification accuracy of traditional SVM and therefore reduced adverse effects of Hughesphenomenon.
在特征降维方面，当训练样本较少时，高光谱数据的分类精度受到严重的影响。通常解决这种现象的办法是对原始数据进行特征降维处理，然而多数特征降维算法无法直接给出最优降维特征数。为此提出利用蒙特卡罗随机实验可以对特征参量进行统计估计的特性，计算高光谱图像的最优降维波段数，并与相关向量机结合，对降维后的数据进行分类。实验结果表明了使用蒙特卡罗算法求解降维波段数的可靠性。相比较原始未降维数据，高光谱图像经过蒙特卡罗特征降维算法处理后，分类精度有较大幅度的提高。
在高光谱端元提取方面，通过对各种算法的分析，主要研究端元提取效果较好的N-FINDR算法。然而样本的排序对该算法的端元提取会造成一定影响，并且传统N-FINDR算法需要根据端元的个数进行降维处理，从而限制了该算法的应用。实际高光谱数据中存在的同一地物在高维空间中非紧密团聚现象也对端元提取增加了难度。为此提出改进的算法停机准则和数据特征预处理方法，并使用支持向量机对提取到的端元进行二次提取。实验结果表明，改进的停机准则进一步增加了由端元向量组组成的凸体体积。数据特征预处理和基于支持向量机的二次端元提取分别提升了数据的可分性和提取到端元的精度。
在高光谱图像分类方面，模糊C-均值聚类算法因算法简单、收敛速度快等优点受到广泛的关注。然而由于高光谱数据的维数较高，其光谱波段的非线性特性使得传统模糊C-均值聚类算法无法在原始空间得到较好的聚类结果。另外，模糊C-均值聚类算法在计算聚类中心时，仅使用了各样本对聚类中心的隶属度，忽略了样本之间固有存在的空间分布特征。为此提出了模糊核加权C-均值聚类算法，在计算模糊核聚类中心时，根据样本的空间分布特征，为每个样本分配不同的权值，使得每个核聚类中心随着样本的不同而各有不同。标准数据和实际高光谱数据的实验结果均表明，相比较传统模糊C-均值聚类算法，模糊核加权C-均值聚类算法在总体分类精度上有较大的提高。
支持向量机被广泛使用在高光谱数据分类中，使用传统支持向量机对高光谱图像进行分类时，认定每个特征波段拥有相同的权值。然而在小样本高光谱图像分类中，由于冗余波段的影响，极容易造成“维数灾难”现象，使得分类精度严重下降。为此提出两种特征加权支持向量机以消除“维数灾难”现象，提高高光谱图像分类的精度：1)ReliefF特征加权支持向量机(RSVM)，2)模糊ReliefF特征加权支持向量机(FRSVM)。实验选取玉米种子图像和公开使用的高光谱数据图像作为实验数据。相对比传统支持向量机，提出的两种加权算法均提高了高光谱图像分类的精度，并且降低了“维数灾难”的影响。
Hyperspectral remote sensing is the multi-dimensional information obtaining technology,which combines target detection and spectral imaging technology together. That is, it couldobtain the two-dimensional object distribution information and one-dimensional spectralfeature characteristic information at the same time. With the increasing of spectral resolution,regarding the spectral characteristic of the objects, people's recognition ability goes deeperthat many characteristics originally hidden in the narrow spectral bands could be discovered."Endmember" is defined as the ideal pure data, which could represent the characteristic of theobjects. Since endmember is generally used as the pre-knowledge of hyperspectral imageryprocessing methods, it plays an important role for the following processing steps， whichdepends on whether the obtained endmembers could represent the characteristic of the objects.Compare with multi-spectral remote sensing, hyperspectral remote sensing provides fruitfulspectral information, which could highlight the tiny spectral characteristic of the objects. Thewide spectral range make it possible for the user to select the special bands to highlight thecharacteristic of the objects, which could provide more original data for the hyperspectralimagery processing methods and make the precise processing of spectral information possible.However, due to the high dimension of the hyperspectral imagery, it has a huge limit whentraditional imagery processing methods are used in hyperspectral imagery. Based on the basichyperspectral remote sensing imagery processing theories and relevant subjects, this studymainly focuses on the feature reduction, endmember extraction, and the classification ofhyperspectral remote sensing imagery.
Regarding feature reduction, due to the high dimension of hyperspectral data, theclassification accuracy is severely affected when there are few training samples. Featurereduction is a common method to deal with this phenomenon. However, most of the featurereduction methods can’t provide optimal feature reduction number. So this study proposes toutilize the statistic estimation characteristic of Monte Carlo random experiments to calculateoptimal feature reduction number and conduct hyperspectral imagery classification withrelevance vector machine. Experiment results show the reliability of the feature reduction number calculated by Monte Carlo method. Compare with the classification of original data,it has a significant improvement on the classification accuracy with the feature reduction data.
Regarding hyperspectral endmember extraction, through the analysis of variousalgorithms, we mainly focus on the N-FINDR endmember extraction algorithm which hasbetter performance. However, the order of the samples has a certain effect on the endmemberextraction, and traditional N-FINDR algorithm also needs to reduce the dimensionality basedon the number of the endmembers, which will limit its application. In the actual hyperspectraldata, the incompact clustering of the same species presented in the high dimensional spacealso increases the difficulty of endmember extraction. So this study proposed an improvedstop rule and the pretreatment of the features, and utilizing Support Vector Machine (SVM) toconduct the second endmember extraction. Experiments show that the improved stop rulefurther increased the volume of the convex polyhedron composed of the endmembers. Thepretreatment of the features and the second SVM based endmember extraction increase theseparability of the data and the precision of the extracted endmembers respectively.
Regarding hyperspectral imagery classification, fuzzy C-means clustering algorithm iswidely utilized for its simpleness and fast convergence rate. Due to the high dimensionality ofhyperspectral data, the nonlinear characteristic of the spectral bands makes it difficult for thetraditional fuzzy C-means clustering algorithm to have good clustering result in the originalspace. Moreover, fuzzy C-means clustering algorithm just uses the membership degree tocalculate the clustering center, which omits the space distribution that intrinsic exists amongthe samples. So this study proposes fuzzy kernel weighted C-means clustering algorithm.When it is used to calculate the fuzzy kernel clustering center, different weights will beassigned to each sample according to the space distribution. As a result, different sampleshave different set of kernel clustering centers. Compare with traditional fuzzy C-meansclustering algorithm, experiments results of both standard data and actual hyperspectral dataprove that the proposed fuzzy kernel weighted C-means clustering algorithm has asignificance improvement on the overall classification accuracy.
Support vector machine is widely used in the classification of hyperspectral reflectancedata. In traditional SVM, features are generated from all or subsets of spectral bands witheach feature contributing equally to the classification. In the classification of smallhyperspectral reflectance data sets, a common challenge is Hughes phenomenon, which is caused by many redundant features and resulting in subsequent poor classification accuracy.In this study, we examined two approaches to assigning weights to SVM features to increaseclassification accuracy and reduce adverse effects of Hughes phenomenon:1)“RSVM” refersto support vector machine with ReliefF feature weighting algorithm, and2)“FRSVM” refersto support vector machine with fuzzy ReliefF feature weighting algorithm. Analyses wereconducted on a reflectance data set of individual maize kernels from three inbred lines and apublic data set with three selected land-cover classes. Both weighting methods increasedclassification accuracy of traditional SVM and therefore reduced adverse effects of Hughesphenomenon.
引文
[1]张兵,高连如.高光谱图像分类与目标探测[M].北京:科学出版社,2011.
[2] Y. Zhong, L. Zhang, B. Hunag. An unsupervised artificial immune classifier formult/hyperspectral remote sensing imagery[J]. IEEE Transactions on Geoscience andRemote Sensing,2006,44(2):420-431.
[3] F. Melgani, L. Bruzzone. Classification of hyperspectral remote sensing images withsupport vector machines[J]. IEEE Transanctions on Geoscience and Remote Sensing,2004,42(8):1778-1790.
[4] A. Sedaghat, M. Mokhtarzade, H. Ebadi. Unifiorm robust scale-invariant featurematching for optical remote sensing images[J]. IEEE Transactions on Geoscience andRemote Sensing,2011,49(11):4516-4527.
[5]童庆禧,张兵,郑兰芬.高光谱遥感-原理技术与应用[M].北京:高等教育出版社,2006.
[6] X. Huang, L. Zhang. An adaptive mean-shift analysis approach for object extractionand classification from urban hyperspectral imagery [J]. IEEE Transanctions onGeoscience and Remote Sensing,2008,46(12):4173-4185.
[7] H. Zhao, G. Jia, N. Li. Transformation from hyperspectral radiance data to data ofother sensors based on spectral superresolution[J]. IEEE Transanctions on Geoscienceand Remote Sensing,2010,48(1):3903-3912.
[8]姚伏天,钱沄涛.高斯过程及其在高光谱图像分类中的应用[J].智能系统学报,2011,6(5):396-404.
[9] Y. Zhong, L. Zhang. An adaptive artificial immune network for supervisedclassification of multi-/hyperspectral remote sensing imagery[J]. IEEE Transactionson Geoscience and Remote Sensing,2012,50(3):894-909.
[10] H. Erives, G. J. Fitzgerald. Automatic subpixel registration for a tunable hyperspectralimaging system [J]. IEEE Transanctions on Geoscience and Remote Sensing,2006,3(3):397-400.
[11]高恒振,万建伟,粘永健.一种基于谱域-空域组合特征支持向量机的高光谱图像分类算法[J].宇航学报,2011,32(4):917-921.
[12]狄文羽,何明一,梅少辉.基于快速非负矩阵分解和RBF网络的高光谱图像分类算法[J].遥感技术与应用,2009,24(3):385-390.
[13] A. C. Jensen, A. S. Solberg. Fast hyperspectral feature reduction using piecewiseconstant fuction approximations[J]. IEEE Gepscoemce and Remote Sensing Letters,2007,4(4):547-551.
[14] B. Mojaradi, H. A. Moghaddam, M. J. Zoej. Dimensionality reduction ofhyperspectral data via spectral feature extraction[J]. IEEE Transanctions onGeoscience and Remote Sensing,2009,47(7):2091-2105.
[15] L. M. Bruce, C. H. Koger, J. Li. Dimensionality reduction of hyperspectral data usingdiscrete wavelet transform feature extraction[J]. IEEE Transanctions on Geoscienceand Remote Sensing,2002,40(10):2331-2338.
[16] L. Zhang, Y. Zhong, B. Huang. Dimensionality reducion based on clonal selection forhyperspectral imagery[J]. IEEE Transanctions on Geoscience and Remote Sensing,2007,45(12):4172-4186.
[17] S. B. Serpico, G. Moser. Extraction of spectral channels from hyperspectral images forclassification purposes[J]. IEEE Transanctions on Geoscience and Remote Sensing,2007,45(2):484-495.
[18] M. Fauvel, J. A. Benediktsson, J. Chanussot. Spectral and spatial classification ofhyperspectral data using SVMs and morphological profiles[J]. IEEE Transanctions onGeoscience and Remote Sensing,2008,46(11):3804-3814.
[19] H. R. Kalluri, S. Prasad, L. M. Bruce. Decision-level fusion of spectral reflectance andderivative information for robust hyperspectral land cover classfication[J]. IEEETransanctions on Geoscience and Remote Sensing,2010,48(11):4047-4058.
[20] K. P. Lin, M. S. Chen. On the design and analysis of the privacy-preserving SVMclassifier[J]. IEEE Transanctions on Knowledge and Data Engineering,2011,23(11):1704-1717.
[21]谭琨,杜培军.基于支持向量机的高光谱遥感图像分类[J].红外与毫米波学报,2008,27(2):123-128.
[22] F. A. Mianji, Y. Zhang. Robust hyperspectral classification using relevance vectormachine[J]. IEEE Transanctions on Geoscience and Remote Sensing,2011,49(6):2100-2112.
[23] L. Wei, Y. Yang, R. M. Nishikawa. Relevance vector machine for automatic detectionof clustered microcalcifications[J]. IEEE Transanctions on Medical Imaging,2005,24(10):1278-1285.
[24] E. Christophe, D. Leger, C. Mailhes. Quality criteria benchmark for hyperspectralimagery[J]. IEEE Transanctions on Geoscience and Remote Sensing,2005,43(9):2103-2114.
[25] H. J. Hyyppa, J. M. Hyyppa. Effects of stand size on the accuracy of remote sensing-based forest inventory[J]. IEEE Transactions on Geoscience and Remote Sensing,2001,39(12):2613-2621.
[26] R. Mayer, R. Priest. Object detection using transformed signatures in multitemporalhyperspectral imagery[J]. IEEE Transanctions on Geoscience and Remote Sensing,2002,40(4):831-840.
[27] H. Bach, W. Mauser. Methods and examples for remote sensing data assimilation inland surface process modeling[J]. IEEE Transactions on Geoscience and RemoteSensing,2003,41(7):1629-1637.
[28] C. Li, T. Sun, K. F. Kelly. A compressive sensing and unmixing scheme forhyperspectral data processing[J]. IEEE Transanctions on Image Processing,2012,21(3):1200-1210.
[29] O. Merlin, A. G. Chehbouni, Y. H. Kerr. A combined modeling and multipectral/multiresolution remote sensing approach for disaggregation of surface soil moisture:application to SMOS configuration[J]. IEEE Transactions on Geoscience and RemoteSensing,2005,43(9):2036-2050.
[30]刘春红.超光谱遥感图像降维及分类方法研究[D].哈尔滨:哈尔滨工程大学博士学位论文,2005.
[31] T. Han, D. G. Goodenough. Investigation of nonlinearity in hyperspectral imageryusing surrogate data methods[J]. IEEE Transanctions on Geoscience and RemoteSensing,2008,46(10):2840-2847.
[32] J. Ma. A single-pixel imaging system for remote sensing by two-step iterative curveletthresholding[J]. IEEE Geoscience and Remote Sensing Letters,2009,6(4):676-680.
[33] J. A. Benediktsson, J. A. Palmason, J. R. Sveinsson. Classification of hyperspectraldata from urban areas based on extended morphological profiles[J]. IEEETransanctions on Geoscience and Remote Sensing,2005,43(3):480-491.
[34] D. R. Fuhrmann, C. Preza, J. A. Osullivan. Spectrum estimation from quantum-limitedinterferograms[J]. IEEE Transanctions on Signal Processing,2004,52(4):950-961.
[35]耿修瑞.高光谱遥感图像目标探测与分类技术研究[D].北京:中国科学院遥感应用研究所博士学位论文,2005.
[36]杨国鹏.基于机器学习方法的高光谱影像分类研究[D].郑州:解放军信息工程大学博士学位论文,2010.
[37]张兵.时空信息辅助下的高光谱数据挖掘[D].北京:中国科学院遥感应用研究所博士学位论文,2002.
[38]杨哲海.高光谱影响分类若干关键技术的研究[D].郑州:解放军信息工程大学博士学位论文,2006.
[39]苏令华.高光谱图像压缩技术研究[D].长沙:国防科技大学博士学位论文,2007.
[40]陈进.高光谱图像分类方法研究[D].长沙:国防科技大学博士学位论文,2010.
[41] L. Zhang, L. Zhang, D. Tao. On combining multiple features for hyperspectral remotesensing image classification[J]. IEEE Transanctions on Geoscience and RemoteSensing,2012,50(3):879-893.
[42] C. Gonzalez, D. Mozos, J. Resano. FPGA implementation of the N-FINDR algorithmfor remotely sensed hyperspectral image analysis[J]. IEEE Transanctions onGeoscience and Remote Sensing,2012,50(2):374-388.
[43]陈万海.基于支持向量机的高光谱图像分类技术研究[D].哈尔滨:哈尔滨工程大学博士学位论文,2008.
[44]陈雨时.基于光谱特性的高光谱图像压缩方法研究[D].哈尔滨:哈尔滨工业大学博士学位论文,2007.
[45] W. Q. Wang. Near-space vehicle-borne SAR with reflector antenna for high-resolution and wide-swath remote sensing[J]. IEEE Transactions on Geoscience andRemote Sensing,2012,50(2):338-348.
[46]储美华.遥感影像的高光谱特征研究[D].青岛:山东科技大学博士学位论文,2007.
[47]崔燕.光谱成像仪定标技术研究[D].西安:中国科学院西安光学精密机械研究所博士学位论文,2009.
[48]戴春妮.高光谱显微图像的特征提取与分类方法及其应用研究[D].上海:华东师范大学博士学位论文,2009.
[49] S. Wang, C. I. Chang. Variable-number variable-band selection for featurecharacterization in hyperspectral signatures[J]. IEEE Transanctions on Geoscience andRemote Sensing,2007,45(9):2979-2992.
[50] Q. Du, C. I. Chang. Linear mixture analysis-based compression for hyperspectralimage analysis[J]. IEEE Transanctions on Geoscience and Remote Sensing,2004,42(4):875-891.
[51] H. Erives, N. B. Targhetta. Implementation of a3-D hyperspectral instrument for skinimaging applications[J]. IEEE Transanctions on Instrumentation and Measurement,2009,58(3):631-638.
[52] J. Li, J. M. Bioucas, A. Plaza. Spectral-spatial hyperspectral image segmentation usingsubspace multinomial logistic regression and markov random fields[J]. IEEETransanctions on Geoscience and Remote Sensing,2012,50(3):809-823.
[53]董延华.超光谱遥感图像处理关键技术研究[D].哈尔滨:哈尔滨理工大学博士学位论文,2006.
[54] A. Ambikapathi, S. H. Chan, W. K. Ma. Chance-constrained robust minimum-volumeenclosing simplex algorithm for hyperspectral unmixing[J]. IEEE Transanctions onGeoscience and Remote Sensing,2011,49(11):4194-4209.
[55] J. M. Bioucas, J. M. Nascimento. Hyperspectral subspace identification[J]. IEEETransanctions on Geoscience and Remote Sensing,2008,46(8):2435-2445.
[56] A. Plaza, Q. Du, Y. L. Chang. High performance computing for hyperspectral remotesensing[J]. IEEE Journal of Selected Topics in Applied Earth Observations andRemote Sensing,2011,4(3):528-544.
[57] C. M. Bachmann, T. L. Ainsworth, R. A. Fusina. Improved manifold coordinaterepresentations of large-scale hyperspectral scenes[J]. IEEE Transanctions onGeoscience and Remote Sensing,2006,44(10):2786-2803.
[58] B. Demir, S. Erturk. Empirical mode decomposition of hyperspectral images forsupport vector machine classification[J]. IEEE Transanctions on Geoscience andRemote Sensing,2010,48(11):4071-4084.
[59] A. A. Farag, R. M. Mohamed, A. E. Baz. A unified framework for MAP estimation inremote sensing image segmentation[J]. IEEE Transactions on Geoscience and RemoteSensing,2005,43(7):1617-1634.
[60] W. Xia, X. Liu, B. Wang. Independent component analysis for blind unmixing ofhyperspectral imagery with additional constraints[J]. IEEE Transanctions onGeoscience and Remote Sensing,2011,49(6):2165-2179.
[61] S. Lopez, P. Horstrand, G. M. Callico. A low-computational-complexity algorithm forhyperspectral endmember extraction: modified vertex component analysis[J]. IEEEGeoscience and Remote Sensing Letters,2012,9(3):502-506.
[62] T. H. Chan, W. K. Ma, A. Amnikapathi. A simplex volume maximization frameworkfor hyperspectral endmember extraction[J]. IEEE Transanctions on Geoscience andRemote Sensing,2011,49(11):4177-4193.
[63] S. Inamdar, F. Bovolo, L. Bruzzone. Multidimensional probability density functionmatching for preprocessing of multitemporal remote sensing images[J]. IEEETransactions on Geoscience and Remote Sensing,2008,46(4):1243-1252.
[64]郭俊先.基于高光谱成像技术的棉花杂志检测方法的研究[D].杭州:浙江大学博士学位论文,2011.
[65]黄睿.高光谱遥感数据特征约简技术研究[D].西安:西北工业大学博士学位论文,2005.
[66]贾森.非监督的高光谱图像解混技术研究[D].杭州:浙江大学博士学位论文,2007.
[67]李智勇.高光谱图像异常检测方法研究[D].长沙:国防科学技术大学博士学位论文,2004.
[68]刘伟东.高光谱遥感土壤信息提取与挖掘研究[D].北京:中国科学院遥感应用研究所博士学位论文,2002.
[69] X. Chang, L. Jiao, F. Liu. Multicontourlet-based adaptive fusion of infrared andvisible remote sensing images[J]. IEEE Geoscience and Remote Sensing Letters,2010,7(3):549-553.
[70]刘翔.基于光谱维变换的高光谱图像目标探测研究[D].北京:中国科学院遥感应用研究所博士学位论文,2008.
[71]刘志刚.支撑向量机在光谱遥感影像分类中的若干问题研究[D].武汉:武汉大学博士学位论文,2004.
[72]路威.面向目标探测的高光谱影响特征提取与分类技术研究[D].郑州:解放军信息工程大学博士学位论文,2005.
[73] G. J. Briem, J. A. Benediktsson, J. R. Sveinsson. Multiple classifiers applied tomultisource remote sensing data[J]. IEEE Transactions on Geoscience and RemoteSensing,2002,40(10):2291-2299.
[74] X. Liu, X. Li, L. Liu. An innovative method to classifiy remote-sensing images usingant colony optimization[J]. IEEE Transactions on Geoscience and Remote Sensing,2008,46(12):4198-4208.
[75] M. T. Eismann, R. C. Hardie. Application of the stochastic mixing model tohyperspectral resolution enhancement[J]. IEEE Transanctions on Geoscience andRemote Sensing,2004,42(9):1924-1933.
[76] Y. Tarabalka, M. Fauvel, J. Chanussot. SVM-and MRF-based method for accurateclassification of hyperspectral images[J]. IEEE Geoscience and Remote SensingLetters,2010,7(4):736-740.
[77]冯燕.高光谱图像压缩技术研究[D].西安:西北工业大学博士学位论文,2006.
[78] P. E. Lehner, L. Adelman, R. J. Distasio. Confirmation bias in the analysis of remotesensing data[J]. IEEE Transactions on Systems, Man and Cybernetics,2009,39(1):218-226.
[79]姚伏天.基于高斯过程的高光谱图像分类研究[D].杭州:浙江大学博士学位论文,2011.
[80] X. Jia, J. A. Richards. Efficient maximum likehood classification for imagingspectrometer data sets[J]. IEEE Transactions on Geoscience and Remote Sensing,1994,32(2):274-281.
[81] C. H. Chen, T. M. Tu. Computation reduction of the maximum likehood classifierusing the Winograd identity[J]. Pattern Recognition,1996,29(7):1213-1220.
[82] T. M. Tu, C. H. Chen, J. L. Wu. A fast two-stage classification method for high-dimensional remote sensing data[J]. IEEE Transactions on Geoscience and RemoteSensing,1998,36(1):182-191.
[83] G. Baudat, F. Anouar. Generalized discriminant analysis using a kernel approach[J].Neural Computation,2000,12(10):2385-2404.
[84] D. A. Clausi. K-means iterative fisher (KIF) unsupervised clustering algorithm appliedto image texture segmentation[J]. Pattern Recognition,2002,35(9):1959-1972.
[85] K. L. Wu, J. Yu, M. S. Yang. A novel fuzzy clustering algorithm based on a fuzzyscatter matrix with optimality tests[J]. Pattern Recognition Letters,2005,26(4):639-652.
[86] M. S. Kamel, S. Z. Selim. A thresholded fuzzy c-means algorithm for semi-fuzzyclustering[J]. Pattern Recognition,1991,24(9):825-833.
[87] S. Z. Selim, M. A. Ismail. Soft clustering of multidimensional data: a semi-fuzzyapproach[J]. Pattern Recognition,1984,17(5):568-599.
[88]魏立梅,谢维信.对手抑制式模糊C-均值算法[J].电子学报,2000,28(7):63-66.
[89] Z. Yin, Y. Tang, F. Sun. Fuzzy clustering with novel separable criterion[J]. TsinghuaScience&Technology,2006,11(2):50-53.
[90]支晓斌,范九伦.基于模糊最大散度差别准则的自适应特征提取模糊聚类算法[J].电子学报,2011,39(6):1358-1363.
[91] V. Vapnik. The Nature of Statistical Learning Theory[M]. New York: Springer Verlag,1995.
[92]范昕伟,杜树新,吾铁军.可补偿类别差异的加权支持向量机算法[J].中国图像图形学报,2003,8(9):1037-1042.
[93]赵晖,荣莉莉.支持向量机组合分类及其在文本分类中的应用[J].小型微型计算机系统,2005,26(10):1816-1820.
[94]张翔,肖小玲,徐光右.基于样本之间紧密度的模糊支持向量机方法[J].软件学报,2006,17(5):951-958.
[95]王立国,赵春晖,乔玉龙.高光谱图像分类的全面加权方法研究[J].红外与毫米波学报,2008,27(6):442-446.
[96]汪延华,田盛丰,黄厚宽.特征加权支持向量机[J].电子与信息学报,2009,31(3):514-518.
[97] M. E. Tipping. Sparse Bayesian learning and relevance vector machine[J]. J. Mach.Learn. Res.,2001,1(3):211-244.
[98]杨国鹏,周欣,余旭初.基于相关向量机的高光谱影像混合像元分解[J].电子学报,2010,38(12):2751-2756.
[99] C. I. Chang, C. C. Wu, C. S. Lo. Real-time simplex growing algorithms forhyperspectral endmember extraction[J]. IEEE Transactions on Geoscience andRemote Sensing,2010,49(11):4177-4193.
[100] X. Tao, B. Wang, L. Zhang. Orthogonal bases approach for the decomposition ofmixed pixels in hyperpsectral imagery[J]. IEEE Geoscience and Remote SensingLetters,2009,6(2):219-223.
[101] C. I. Chang, C. C. Wu, W. M. Liu. A new growing method for simplex-basedendmember extraction algorithm[J]. IEEE Transactions on Geoscience and RemoteSensing,2006,44(10):2804-2819.
[102]耿修瑞,赵永超,周冠华.一种利用单形体体积自动提取高光谱图像端元的算法[J].自然科学进展,2006,16(9):1196-1200.
[103] M. Zortea, A. Plaza. A quantitative and comparative analysis of differentimplementations of N-FINDR: a fast endmember extraction algorithm[J]. IEEEGeoscience and Remote Sensing Letters,2009,6(4):787-791.
[104]王立国,邓禄群,张晶.基于线性最小二乘支持向量机的光谱端元选择算法[J].光谱学与光谱分析,2010,30(3):743-747.
[105] W. Xiong, C. I. Chang, C. C. Wu. Fast algorithms to implement N-FINDR forhyperspectral endmember extraction[J]. IEEE Journal of Selected Topics in AppliedEarth Observations and Remote Sensing,2011,4(3):545-564.
[106]邓乃扬,田英杰.支持向量机-理论算法与拓展[M].北京:科学出版社,2009.
[107]罗林开,叶凌君,彭洪.给定经验风险水平的支持向量回归机[J].华中科技大学学报,2010,38(10):47-51.
[108]陈果.一种实现结构风险最小化思想的结构自适应神经网络模型[J].仪器仪表学报,2007,28(10):1874-1879.
[109]邱根胜,李动锋.关于凸规划对偶模型的讨论[J].大学数学,2008,24(2):91-93.
[110]高学,金连文,尹俊勋.一种基于支持向量机的手写汉字识别方法[J].电子学报,2002,30(5):651-654.
[111]皋军,王士同.基于矩阵模式的最小类内散度支持向量机[J].电子学报,2009,37(5):1051-1057.
[112]刘涵,郭勇,郑岗.基于最小二乘支持向量机的图像边缘检测研究[J].电子学报,2006,34(7):1275-1279.
[113]杨树仁,沈洪远.基于相关向量机的机器学习算法研究与应用[J].计算技术与自动化,2010,29(1):43-47.
[114]邬俊,鲁明羽,刘闯.基于混合学习框架的SVM反馈算法研究[J].电子学报,2010,38(9):2101-2106.
[115]李昆仑,黄厚宽,田盛丰.模糊多类SVM模型[J].电子学报,2004,32(5):830-832.
[116]夏威,王斌,张立明.基于独立分量分析的高光谱遥感图像混合像元盲分解[J].红外与毫米波学报,2011,30(2):131-136.
[117] M. Pal, G. M. Foody. Feature selection for classification of hyperspectral data bySVM[J]. IEEE Transactions on Geoscience and Remote Sensing,2010,48(5):2297-2307.
[118] P. H. Hsu. Feature extraction of hyperspectral images using wavelet and matchingpursuit[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2007,62(2):78-92.
[119]石吉勇,邹小波,赵杰文. BiPLS结合模拟退火算法的近红外光谱特征波长选择[J].红外与毫米波学报,2011,30(5):458-462.
[120]苏红军,盛业华.基于正交投影散度的高光谱遥感波段特征选择[J].光谱学与光谱分析,2011,31(5):1309-1313.
[121] H. Chan, Y. Chi, M. Huang. A convex analysis-based minimum-volume enclosingsimplex algorithm for hyperspectral unmixing[J]. IEEE T. Signal Proces,2009,57(11):4418-4432.
[122] C. Miesch, L. Poutier, V. Achard. Direct and inverse radiative transfer solutions forvisible and near-infrared hyperspectral imagery[J]. IEEE Transactions on Geoscienceand Remote Sensing,2005,43(7):1552-1562.
[123] I. Guyon, J. Weston, S. Barnhill. Gene selection for cancer classification using supportvector machines[J]. Machine Learning,2002,46(1):389-422.
[124]刘翔,张兵,高连如.一种改进高光谱图像噪声评估的MNF变换算法[J].中国科学,2009,39(12):1305-1313.
[125] J. Via, D. P. Palomar, L. Vielva. Quaternion ICA from second-order statistics[J]. IEEETransactions on Signal Processing,2011,59(4):1586-1600.
[126]王泽,朱贻盛,李音.独立分量分析在混沌信号分析中的应用[J].电子学报,2002,30(10):1505-1507.
[127] J. Kim, H. Choi, J. Yi. Effective representation using ICA for face recognition robustto local distortion and partial occlusion[J]. IEEE Transactions of Pattern Analysis andMachine Intelligence,2005,27(12):1977-1981.
[128] A. Plaze, C. I. Chang. An improved N-FINDR algorithm in implementation[J].Algorithms and Technologies for Multispectral, Hyperspectral, and UltraspectralImagery XI,2005,5806(1):298-306.
[129]王立国,张晶,刘丹凤.从端元选择到光谱解混的距离测算方法[J].红外与毫米波学报,2010,29(6):471-475.
[130]陶剑文,王士同.具有磁场效应的大间隔支持向量机[J].电子与信息学报,2011,33(5):1055-1061.
[131]皋军,王士同,邓赵红.基于全局和局部保持的半监督支持向量机[J].电子学报,2010,38(7):1626-1633.
[132] C. I. Chang, C. C. Wu, C. T. Tsai. Random N-Finder (N-FINDR) endmemberextraction algorithms for hyperspectral imagery[J]. IEEE Transactions on ImageProcessing,2011,20(3):641-656.
[133] C. I. Chang, M. Hsueh, W. Liu. A pyramid-based block of skewers for pixel purityindex for endmember extraction in hyperspectral imagery[J]. International Journal ofHigh Speed Electronics and Systems,2008,18(2):469-482.
[134] C. I. Chang, A. Plaza. A fast iterative algorithm for implementation of pixel purityindex[J]. IEEE Geoscience and Remote Sensing Letters,2006,3(1):63-67.
[135] A. Ifarraguerri, C. I. Chang. Multispectral and hyperspectral image analysis withconvex cones[J]. IEEE Transactions on Geoscience and Remote Sensing,1999,37(2):756-770.
[136]李智勇,郁文贤,赵和鹏.迭代误差分析方法在高光谱异常检测中的应用[J].系统工程与电子技术,2008,30(12):2340-2344.
[137]薛绮,匡纲要,李智勇.基于线性混合模型的高光谱图像端元提取[J].遥感技术与应用,2004,19(3):197-201.
[138]李慧,王云鹏,李岩.基于形态学和支持向量的遥感图像混合像元分解[J].遥感技术与应用,2009,24(1):114-119.
[139] J. M. Nascimento, J. M. Dias. Vertex component analysis: a fast algorithm to unmixhyperspectral data[J]. IEEE Transactions on Geoscience and Remote Sensing,2005,43(4):898-910.
[140] C. C. Hung, S. Kulkarni, B. C. Kuo. A new weighted fuzzy C-means clusteringalgorithm for remotely sensed image classification[J]. IEEE Journal of SelectedTopics in Signal Processing,2011,5(3):543-553.
[141]刘伟东,冯伍法,任利华.改进相似性测度的高光谱影像FCM聚类分析[J].测绘工程,2008,17(5):29-33.
[142] K. Canham, A. Schlamm, A. Ziemann. Spatially adaptive hyperspectral unmixing[J].IEEE Transactions on Geoscience and Remote Sensing,2011,49(11):4248-4262.
[143] J. Jin, B. Wang, L. Zhang. A novel approach based on fisher discriminant null spacefor decomposition of mixed pixels in hyperspectral imagery[J]. IEEE Geoscience andRemote Sensing Letters,2010,7(8):699-703.
[144] J. Zhang, B. Rivard, A. S. Azofeifa. Derivative spectral unmixing of hyperspectraldata applied to mixtures of lichen and rock[J]. IEEE Transactions on Geoscience andRemote Sensing,2004,42(9):1934-1940.
[145] X. Du, H. Ying, F. Lin. Theory of extended fuzzy discrete-event systems for handlingranges of knowledge uncertainties and subjectivity[J]. IEEE Transactions on FuzzySystems,2009,17(2):316-328.
[146]乔玉龙,潘正祥,孙圣和.一种改进的快速k-近邻分类算法[J].电子学报,2005,33(6):1146-1149.
[147]徐海霞,刘国海,周大为.基于改进核模糊聚类算法的软测量建模研究[J].仪器仪表学报,2009,30(10):2226-2231.
[148]伍忠东,高新波,谢维信.基于核方法的模糊聚类算法[J].西安电子科技大学学报,2004,31(4):533-537.
[149]李洁,高新波,焦李成.基于特征加权的模糊聚类新算法[J].电子学报,2006,34(1):89-92.
[150]皋军,王士同.基于最大散度差别判别准则的模糊聚类算法[J].软件学报,2009,20(11):2939-2949.
[151] B. C. Kuo, D. A. Landgrebe. Nonparametric weighted feature extraction forclassification[J]. IEEE Transactions on Geoscience and Remote Sensing,2004,42(5):1096-1105.
[152]王宏伟,詹荣开,贺汉根.基于模糊聚类的改进模糊辨识方法[J].电子学报,2001,29(4):436-438.
[153]高新波,裴继红,谢维信.模糊c-均值聚类算法中加权指数m的研究[J].电子学报,2000,28(4):80-83.
[154]邢宗义,张永,侯远龙.基于模糊聚类和遗传算法的具备解释性和精确性的模糊分类系统设计[J].电子学报,2006,34(1):83-88.
[155]高新波,李洁,姬红兵.基于模糊c均值聚类与统计检验指导的多阈值图像自动分割算法[J].电子学报,2004,32(4):661-664.
[156]周世兵,徐振源,唐旭清.新的K-均值算法最佳聚类数确定方法[J].计算机工程与应用,2010,46(16):27-31.
[157] J. R. Quinlan. Induction of decision trees[J]. Machine Learning,1986,1(1):81-106.
[158] C. I. Chang, B. Ji. Weighted abundance-constrained linear spectral mixure analysis[J].IEEE Transactions on Geoscience and Remote Sensing,2006,44(2):378-388.
[2] Y. Zhong, L. Zhang, B. Hunag. An unsupervised artificial immune classifier formult/hyperspectral remote sensing imagery[J]. IEEE Transactions on Geoscience andRemote Sensing,2006,44(2):420-431.
[3] F. Melgani, L. Bruzzone. Classification of hyperspectral remote sensing images withsupport vector machines[J]. IEEE Transanctions on Geoscience and Remote Sensing,2004,42(8):1778-1790.
[4] A. Sedaghat, M. Mokhtarzade, H. Ebadi. Unifiorm robust scale-invariant featurematching for optical remote sensing images[J]. IEEE Transactions on Geoscience andRemote Sensing,2011,49(11):4516-4527.
[5]童庆禧,张兵,郑兰芬.高光谱遥感-原理技术与应用[M].北京:高等教育出版社,2006.
[6] X. Huang, L. Zhang. An adaptive mean-shift analysis approach for object extractionand classification from urban hyperspectral imagery [J]. IEEE Transanctions onGeoscience and Remote Sensing,2008,46(12):4173-4185.
[7] H. Zhao, G. Jia, N. Li. Transformation from hyperspectral radiance data to data ofother sensors based on spectral superresolution[J]. IEEE Transanctions on Geoscienceand Remote Sensing,2010,48(1):3903-3912.
[8]姚伏天,钱沄涛.高斯过程及其在高光谱图像分类中的应用[J].智能系统学报,2011,6(5):396-404.
[9] Y. Zhong, L. Zhang. An adaptive artificial immune network for supervisedclassification of multi-/hyperspectral remote sensing imagery[J]. IEEE Transactionson Geoscience and Remote Sensing,2012,50(3):894-909.
[10] H. Erives, G. J. Fitzgerald. Automatic subpixel registration for a tunable hyperspectralimaging system [J]. IEEE Transanctions on Geoscience and Remote Sensing,2006,3(3):397-400.
[11]高恒振,万建伟,粘永健.一种基于谱域-空域组合特征支持向量机的高光谱图像分类算法[J].宇航学报,2011,32(4):917-921.
[12]狄文羽,何明一,梅少辉.基于快速非负矩阵分解和RBF网络的高光谱图像分类算法[J].遥感技术与应用,2009,24(3):385-390.
[13] A. C. Jensen, A. S. Solberg. Fast hyperspectral feature reduction using piecewiseconstant fuction approximations[J]. IEEE Gepscoemce and Remote Sensing Letters,2007,4(4):547-551.
[14] B. Mojaradi, H. A. Moghaddam, M. J. Zoej. Dimensionality reduction ofhyperspectral data via spectral feature extraction[J]. IEEE Transanctions onGeoscience and Remote Sensing,2009,47(7):2091-2105.
[15] L. M. Bruce, C. H. Koger, J. Li. Dimensionality reduction of hyperspectral data usingdiscrete wavelet transform feature extraction[J]. IEEE Transanctions on Geoscienceand Remote Sensing,2002,40(10):2331-2338.
[16] L. Zhang, Y. Zhong, B. Huang. Dimensionality reducion based on clonal selection forhyperspectral imagery[J]. IEEE Transanctions on Geoscience and Remote Sensing,2007,45(12):4172-4186.
[17] S. B. Serpico, G. Moser. Extraction of spectral channels from hyperspectral images forclassification purposes[J]. IEEE Transanctions on Geoscience and Remote Sensing,2007,45(2):484-495.
[18] M. Fauvel, J. A. Benediktsson, J. Chanussot. Spectral and spatial classification ofhyperspectral data using SVMs and morphological profiles[J]. IEEE Transanctions onGeoscience and Remote Sensing,2008,46(11):3804-3814.
[19] H. R. Kalluri, S. Prasad, L. M. Bruce. Decision-level fusion of spectral reflectance andderivative information for robust hyperspectral land cover classfication[J]. IEEETransanctions on Geoscience and Remote Sensing,2010,48(11):4047-4058.
[20] K. P. Lin, M. S. Chen. On the design and analysis of the privacy-preserving SVMclassifier[J]. IEEE Transanctions on Knowledge and Data Engineering,2011,23(11):1704-1717.
[21]谭琨,杜培军.基于支持向量机的高光谱遥感图像分类[J].红外与毫米波学报,2008,27(2):123-128.
[22] F. A. Mianji, Y. Zhang. Robust hyperspectral classification using relevance vectormachine[J]. IEEE Transanctions on Geoscience and Remote Sensing,2011,49(6):2100-2112.
[23] L. Wei, Y. Yang, R. M. Nishikawa. Relevance vector machine for automatic detectionof clustered microcalcifications[J]. IEEE Transanctions on Medical Imaging,2005,24(10):1278-1285.
[24] E. Christophe, D. Leger, C. Mailhes. Quality criteria benchmark for hyperspectralimagery[J]. IEEE Transanctions on Geoscience and Remote Sensing,2005,43(9):2103-2114.
[25] H. J. Hyyppa, J. M. Hyyppa. Effects of stand size on the accuracy of remote sensing-based forest inventory[J]. IEEE Transactions on Geoscience and Remote Sensing,2001,39(12):2613-2621.
[26] R. Mayer, R. Priest. Object detection using transformed signatures in multitemporalhyperspectral imagery[J]. IEEE Transanctions on Geoscience and Remote Sensing,2002,40(4):831-840.
[27] H. Bach, W. Mauser. Methods and examples for remote sensing data assimilation inland surface process modeling[J]. IEEE Transactions on Geoscience and RemoteSensing,2003,41(7):1629-1637.
[28] C. Li, T. Sun, K. F. Kelly. A compressive sensing and unmixing scheme forhyperspectral data processing[J]. IEEE Transanctions on Image Processing,2012,21(3):1200-1210.
[29] O. Merlin, A. G. Chehbouni, Y. H. Kerr. A combined modeling and multipectral/multiresolution remote sensing approach for disaggregation of surface soil moisture:application to SMOS configuration[J]. IEEE Transactions on Geoscience and RemoteSensing,2005,43(9):2036-2050.
[30]刘春红.超光谱遥感图像降维及分类方法研究[D].哈尔滨:哈尔滨工程大学博士学位论文,2005.
[31] T. Han, D. G. Goodenough. Investigation of nonlinearity in hyperspectral imageryusing surrogate data methods[J]. IEEE Transanctions on Geoscience and RemoteSensing,2008,46(10):2840-2847.
[32] J. Ma. A single-pixel imaging system for remote sensing by two-step iterative curveletthresholding[J]. IEEE Geoscience and Remote Sensing Letters,2009,6(4):676-680.
[33] J. A. Benediktsson, J. A. Palmason, J. R. Sveinsson. Classification of hyperspectraldata from urban areas based on extended morphological profiles[J]. IEEETransanctions on Geoscience and Remote Sensing,2005,43(3):480-491.
[34] D. R. Fuhrmann, C. Preza, J. A. Osullivan. Spectrum estimation from quantum-limitedinterferograms[J]. IEEE Transanctions on Signal Processing,2004,52(4):950-961.
[35]耿修瑞.高光谱遥感图像目标探测与分类技术研究[D].北京:中国科学院遥感应用研究所博士学位论文,2005.
[36]杨国鹏.基于机器学习方法的高光谱影像分类研究[D].郑州:解放军信息工程大学博士学位论文,2010.
[37]张兵.时空信息辅助下的高光谱数据挖掘[D].北京:中国科学院遥感应用研究所博士学位论文,2002.
[38]杨哲海.高光谱影响分类若干关键技术的研究[D].郑州:解放军信息工程大学博士学位论文,2006.
[39]苏令华.高光谱图像压缩技术研究[D].长沙:国防科技大学博士学位论文,2007.
[40]陈进.高光谱图像分类方法研究[D].长沙:国防科技大学博士学位论文,2010.
[41] L. Zhang, L. Zhang, D. Tao. On combining multiple features for hyperspectral remotesensing image classification[J]. IEEE Transanctions on Geoscience and RemoteSensing,2012,50(3):879-893.
[42] C. Gonzalez, D. Mozos, J. Resano. FPGA implementation of the N-FINDR algorithmfor remotely sensed hyperspectral image analysis[J]. IEEE Transanctions onGeoscience and Remote Sensing,2012,50(2):374-388.
[43]陈万海.基于支持向量机的高光谱图像分类技术研究[D].哈尔滨:哈尔滨工程大学博士学位论文,2008.
[44]陈雨时.基于光谱特性的高光谱图像压缩方法研究[D].哈尔滨:哈尔滨工业大学博士学位论文,2007.
[45] W. Q. Wang. Near-space vehicle-borne SAR with reflector antenna for high-resolution and wide-swath remote sensing[J]. IEEE Transactions on Geoscience andRemote Sensing,2012,50(2):338-348.
[46]储美华.遥感影像的高光谱特征研究[D].青岛:山东科技大学博士学位论文,2007.
[47]崔燕.光谱成像仪定标技术研究[D].西安:中国科学院西安光学精密机械研究所博士学位论文,2009.
[48]戴春妮.高光谱显微图像的特征提取与分类方法及其应用研究[D].上海:华东师范大学博士学位论文,2009.
[49] S. Wang, C. I. Chang. Variable-number variable-band selection for featurecharacterization in hyperspectral signatures[J]. IEEE Transanctions on Geoscience andRemote Sensing,2007,45(9):2979-2992.
[50] Q. Du, C. I. Chang. Linear mixture analysis-based compression for hyperspectralimage analysis[J]. IEEE Transanctions on Geoscience and Remote Sensing,2004,42(4):875-891.
[51] H. Erives, N. B. Targhetta. Implementation of a3-D hyperspectral instrument for skinimaging applications[J]. IEEE Transanctions on Instrumentation and Measurement,2009,58(3):631-638.
[52] J. Li, J. M. Bioucas, A. Plaza. Spectral-spatial hyperspectral image segmentation usingsubspace multinomial logistic regression and markov random fields[J]. IEEETransanctions on Geoscience and Remote Sensing,2012,50(3):809-823.
[53]董延华.超光谱遥感图像处理关键技术研究[D].哈尔滨:哈尔滨理工大学博士学位论文,2006.
[54] A. Ambikapathi, S. H. Chan, W. K. Ma. Chance-constrained robust minimum-volumeenclosing simplex algorithm for hyperspectral unmixing[J]. IEEE Transanctions onGeoscience and Remote Sensing,2011,49(11):4194-4209.
[55] J. M. Bioucas, J. M. Nascimento. Hyperspectral subspace identification[J]. IEEETransanctions on Geoscience and Remote Sensing,2008,46(8):2435-2445.
[56] A. Plaza, Q. Du, Y. L. Chang. High performance computing for hyperspectral remotesensing[J]. IEEE Journal of Selected Topics in Applied Earth Observations andRemote Sensing,2011,4(3):528-544.
[57] C. M. Bachmann, T. L. Ainsworth, R. A. Fusina. Improved manifold coordinaterepresentations of large-scale hyperspectral scenes[J]. IEEE Transanctions onGeoscience and Remote Sensing,2006,44(10):2786-2803.
[58] B. Demir, S. Erturk. Empirical mode decomposition of hyperspectral images forsupport vector machine classification[J]. IEEE Transanctions on Geoscience andRemote Sensing,2010,48(11):4071-4084.
[59] A. A. Farag, R. M. Mohamed, A. E. Baz. A unified framework for MAP estimation inremote sensing image segmentation[J]. IEEE Transactions on Geoscience and RemoteSensing,2005,43(7):1617-1634.
[60] W. Xia, X. Liu, B. Wang. Independent component analysis for blind unmixing ofhyperspectral imagery with additional constraints[J]. IEEE Transanctions onGeoscience and Remote Sensing,2011,49(6):2165-2179.
[61] S. Lopez, P. Horstrand, G. M. Callico. A low-computational-complexity algorithm forhyperspectral endmember extraction: modified vertex component analysis[J]. IEEEGeoscience and Remote Sensing Letters,2012,9(3):502-506.
[62] T. H. Chan, W. K. Ma, A. Amnikapathi. A simplex volume maximization frameworkfor hyperspectral endmember extraction[J]. IEEE Transanctions on Geoscience andRemote Sensing,2011,49(11):4177-4193.
[63] S. Inamdar, F. Bovolo, L. Bruzzone. Multidimensional probability density functionmatching for preprocessing of multitemporal remote sensing images[J]. IEEETransactions on Geoscience and Remote Sensing,2008,46(4):1243-1252.
[64]郭俊先.基于高光谱成像技术的棉花杂志检测方法的研究[D].杭州:浙江大学博士学位论文,2011.
[65]黄睿.高光谱遥感数据特征约简技术研究[D].西安:西北工业大学博士学位论文,2005.
[66]贾森.非监督的高光谱图像解混技术研究[D].杭州:浙江大学博士学位论文,2007.
[67]李智勇.高光谱图像异常检测方法研究[D].长沙:国防科学技术大学博士学位论文,2004.
[68]刘伟东.高光谱遥感土壤信息提取与挖掘研究[D].北京:中国科学院遥感应用研究所博士学位论文,2002.
[69] X. Chang, L. Jiao, F. Liu. Multicontourlet-based adaptive fusion of infrared andvisible remote sensing images[J]. IEEE Geoscience and Remote Sensing Letters,2010,7(3):549-553.
[70]刘翔.基于光谱维变换的高光谱图像目标探测研究[D].北京:中国科学院遥感应用研究所博士学位论文,2008.
[71]刘志刚.支撑向量机在光谱遥感影像分类中的若干问题研究[D].武汉:武汉大学博士学位论文,2004.
[72]路威.面向目标探测的高光谱影响特征提取与分类技术研究[D].郑州:解放军信息工程大学博士学位论文,2005.
[73] G. J. Briem, J. A. Benediktsson, J. R. Sveinsson. Multiple classifiers applied tomultisource remote sensing data[J]. IEEE Transactions on Geoscience and RemoteSensing,2002,40(10):2291-2299.
[74] X. Liu, X. Li, L. Liu. An innovative method to classifiy remote-sensing images usingant colony optimization[J]. IEEE Transactions on Geoscience and Remote Sensing,2008,46(12):4198-4208.
[75] M. T. Eismann, R. C. Hardie. Application of the stochastic mixing model tohyperspectral resolution enhancement[J]. IEEE Transanctions on Geoscience andRemote Sensing,2004,42(9):1924-1933.
[76] Y. Tarabalka, M. Fauvel, J. Chanussot. SVM-and MRF-based method for accurateclassification of hyperspectral images[J]. IEEE Geoscience and Remote SensingLetters,2010,7(4):736-740.
[77]冯燕.高光谱图像压缩技术研究[D].西安:西北工业大学博士学位论文,2006.
[78] P. E. Lehner, L. Adelman, R. J. Distasio. Confirmation bias in the analysis of remotesensing data[J]. IEEE Transactions on Systems, Man and Cybernetics,2009,39(1):218-226.
[79]姚伏天.基于高斯过程的高光谱图像分类研究[D].杭州:浙江大学博士学位论文,2011.
[80] X. Jia, J. A. Richards. Efficient maximum likehood classification for imagingspectrometer data sets[J]. IEEE Transactions on Geoscience and Remote Sensing,1994,32(2):274-281.
[81] C. H. Chen, T. M. Tu. Computation reduction of the maximum likehood classifierusing the Winograd identity[J]. Pattern Recognition,1996,29(7):1213-1220.
[82] T. M. Tu, C. H. Chen, J. L. Wu. A fast two-stage classification method for high-dimensional remote sensing data[J]. IEEE Transactions on Geoscience and RemoteSensing,1998,36(1):182-191.
[83] G. Baudat, F. Anouar. Generalized discriminant analysis using a kernel approach[J].Neural Computation,2000,12(10):2385-2404.
[84] D. A. Clausi. K-means iterative fisher (KIF) unsupervised clustering algorithm appliedto image texture segmentation[J]. Pattern Recognition,2002,35(9):1959-1972.
[85] K. L. Wu, J. Yu, M. S. Yang. A novel fuzzy clustering algorithm based on a fuzzyscatter matrix with optimality tests[J]. Pattern Recognition Letters,2005,26(4):639-652.
[86] M. S. Kamel, S. Z. Selim. A thresholded fuzzy c-means algorithm for semi-fuzzyclustering[J]. Pattern Recognition,1991,24(9):825-833.
[87] S. Z. Selim, M. A. Ismail. Soft clustering of multidimensional data: a semi-fuzzyapproach[J]. Pattern Recognition,1984,17(5):568-599.
[88]魏立梅,谢维信.对手抑制式模糊C-均值算法[J].电子学报,2000,28(7):63-66.
[89] Z. Yin, Y. Tang, F. Sun. Fuzzy clustering with novel separable criterion[J]. TsinghuaScience&Technology,2006,11(2):50-53.
[90]支晓斌,范九伦.基于模糊最大散度差别准则的自适应特征提取模糊聚类算法[J].电子学报,2011,39(6):1358-1363.
[91] V. Vapnik. The Nature of Statistical Learning Theory[M]. New York: Springer Verlag,1995.
[92]范昕伟,杜树新,吾铁军.可补偿类别差异的加权支持向量机算法[J].中国图像图形学报,2003,8(9):1037-1042.
[93]赵晖,荣莉莉.支持向量机组合分类及其在文本分类中的应用[J].小型微型计算机系统,2005,26(10):1816-1820.
[94]张翔,肖小玲,徐光右.基于样本之间紧密度的模糊支持向量机方法[J].软件学报,2006,17(5):951-958.
[95]王立国,赵春晖,乔玉龙.高光谱图像分类的全面加权方法研究[J].红外与毫米波学报,2008,27(6):442-446.
[96]汪延华,田盛丰,黄厚宽.特征加权支持向量机[J].电子与信息学报,2009,31(3):514-518.
[97] M. E. Tipping. Sparse Bayesian learning and relevance vector machine[J]. J. Mach.Learn. Res.,2001,1(3):211-244.
[98]杨国鹏,周欣,余旭初.基于相关向量机的高光谱影像混合像元分解[J].电子学报,2010,38(12):2751-2756.
[99] C. I. Chang, C. C. Wu, C. S. Lo. Real-time simplex growing algorithms forhyperspectral endmember extraction[J]. IEEE Transactions on Geoscience andRemote Sensing,2010,49(11):4177-4193.
[100] X. Tao, B. Wang, L. Zhang. Orthogonal bases approach for the decomposition ofmixed pixels in hyperpsectral imagery[J]. IEEE Geoscience and Remote SensingLetters,2009,6(2):219-223.
[101] C. I. Chang, C. C. Wu, W. M. Liu. A new growing method for simplex-basedendmember extraction algorithm[J]. IEEE Transactions on Geoscience and RemoteSensing,2006,44(10):2804-2819.
[102]耿修瑞,赵永超,周冠华.一种利用单形体体积自动提取高光谱图像端元的算法[J].自然科学进展,2006,16(9):1196-1200.
[103] M. Zortea, A. Plaza. A quantitative and comparative analysis of differentimplementations of N-FINDR: a fast endmember extraction algorithm[J]. IEEEGeoscience and Remote Sensing Letters,2009,6(4):787-791.
[104]王立国,邓禄群,张晶.基于线性最小二乘支持向量机的光谱端元选择算法[J].光谱学与光谱分析,2010,30(3):743-747.
[105] W. Xiong, C. I. Chang, C. C. Wu. Fast algorithms to implement N-FINDR forhyperspectral endmember extraction[J]. IEEE Journal of Selected Topics in AppliedEarth Observations and Remote Sensing,2011,4(3):545-564.
[106]邓乃扬,田英杰.支持向量机-理论算法与拓展[M].北京:科学出版社,2009.
[107]罗林开,叶凌君,彭洪.给定经验风险水平的支持向量回归机[J].华中科技大学学报,2010,38(10):47-51.
[108]陈果.一种实现结构风险最小化思想的结构自适应神经网络模型[J].仪器仪表学报,2007,28(10):1874-1879.
[109]邱根胜,李动锋.关于凸规划对偶模型的讨论[J].大学数学,2008,24(2):91-93.
[110]高学,金连文,尹俊勋.一种基于支持向量机的手写汉字识别方法[J].电子学报,2002,30(5):651-654.
[111]皋军,王士同.基于矩阵模式的最小类内散度支持向量机[J].电子学报,2009,37(5):1051-1057.
[112]刘涵,郭勇,郑岗.基于最小二乘支持向量机的图像边缘检测研究[J].电子学报,2006,34(7):1275-1279.
[113]杨树仁,沈洪远.基于相关向量机的机器学习算法研究与应用[J].计算技术与自动化,2010,29(1):43-47.
[114]邬俊,鲁明羽,刘闯.基于混合学习框架的SVM反馈算法研究[J].电子学报,2010,38(9):2101-2106.
[115]李昆仑,黄厚宽,田盛丰.模糊多类SVM模型[J].电子学报,2004,32(5):830-832.
[116]夏威,王斌,张立明.基于独立分量分析的高光谱遥感图像混合像元盲分解[J].红外与毫米波学报,2011,30(2):131-136.
[117] M. Pal, G. M. Foody. Feature selection for classification of hyperspectral data bySVM[J]. IEEE Transactions on Geoscience and Remote Sensing,2010,48(5):2297-2307.
[118] P. H. Hsu. Feature extraction of hyperspectral images using wavelet and matchingpursuit[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2007,62(2):78-92.
[119]石吉勇,邹小波,赵杰文. BiPLS结合模拟退火算法的近红外光谱特征波长选择[J].红外与毫米波学报,2011,30(5):458-462.
[120]苏红军,盛业华.基于正交投影散度的高光谱遥感波段特征选择[J].光谱学与光谱分析,2011,31(5):1309-1313.
[121] H. Chan, Y. Chi, M. Huang. A convex analysis-based minimum-volume enclosingsimplex algorithm for hyperspectral unmixing[J]. IEEE T. Signal Proces,2009,57(11):4418-4432.
[122] C. Miesch, L. Poutier, V. Achard. Direct and inverse radiative transfer solutions forvisible and near-infrared hyperspectral imagery[J]. IEEE Transactions on Geoscienceand Remote Sensing,2005,43(7):1552-1562.
[123] I. Guyon, J. Weston, S. Barnhill. Gene selection for cancer classification using supportvector machines[J]. Machine Learning,2002,46(1):389-422.
[124]刘翔,张兵,高连如.一种改进高光谱图像噪声评估的MNF变换算法[J].中国科学,2009,39(12):1305-1313.
[125] J. Via, D. P. Palomar, L. Vielva. Quaternion ICA from second-order statistics[J]. IEEETransactions on Signal Processing,2011,59(4):1586-1600.
[126]王泽,朱贻盛,李音.独立分量分析在混沌信号分析中的应用[J].电子学报,2002,30(10):1505-1507.
[127] J. Kim, H. Choi, J. Yi. Effective representation using ICA for face recognition robustto local distortion and partial occlusion[J]. IEEE Transactions of Pattern Analysis andMachine Intelligence,2005,27(12):1977-1981.
[128] A. Plaze, C. I. Chang. An improved N-FINDR algorithm in implementation[J].Algorithms and Technologies for Multispectral, Hyperspectral, and UltraspectralImagery XI,2005,5806(1):298-306.
[129]王立国,张晶,刘丹凤.从端元选择到光谱解混的距离测算方法[J].红外与毫米波学报,2010,29(6):471-475.
[130]陶剑文,王士同.具有磁场效应的大间隔支持向量机[J].电子与信息学报,2011,33(5):1055-1061.
[131]皋军,王士同,邓赵红.基于全局和局部保持的半监督支持向量机[J].电子学报,2010,38(7):1626-1633.
[132] C. I. Chang, C. C. Wu, C. T. Tsai. Random N-Finder (N-FINDR) endmemberextraction algorithms for hyperspectral imagery[J]. IEEE Transactions on ImageProcessing,2011,20(3):641-656.
[133] C. I. Chang, M. Hsueh, W. Liu. A pyramid-based block of skewers for pixel purityindex for endmember extraction in hyperspectral imagery[J]. International Journal ofHigh Speed Electronics and Systems,2008,18(2):469-482.
[134] C. I. Chang, A. Plaza. A fast iterative algorithm for implementation of pixel purityindex[J]. IEEE Geoscience and Remote Sensing Letters,2006,3(1):63-67.
[135] A. Ifarraguerri, C. I. Chang. Multispectral and hyperspectral image analysis withconvex cones[J]. IEEE Transactions on Geoscience and Remote Sensing,1999,37(2):756-770.
[136]李智勇,郁文贤,赵和鹏.迭代误差分析方法在高光谱异常检测中的应用[J].系统工程与电子技术,2008,30(12):2340-2344.
[137]薛绮,匡纲要,李智勇.基于线性混合模型的高光谱图像端元提取[J].遥感技术与应用,2004,19(3):197-201.
[138]李慧,王云鹏,李岩.基于形态学和支持向量的遥感图像混合像元分解[J].遥感技术与应用,2009,24(1):114-119.
[139] J. M. Nascimento, J. M. Dias. Vertex component analysis: a fast algorithm to unmixhyperspectral data[J]. IEEE Transactions on Geoscience and Remote Sensing,2005,43(4):898-910.
[140] C. C. Hung, S. Kulkarni, B. C. Kuo. A new weighted fuzzy C-means clusteringalgorithm for remotely sensed image classification[J]. IEEE Journal of SelectedTopics in Signal Processing,2011,5(3):543-553.
[141]刘伟东,冯伍法,任利华.改进相似性测度的高光谱影像FCM聚类分析[J].测绘工程,2008,17(5):29-33.
[142] K. Canham, A. Schlamm, A. Ziemann. Spatially adaptive hyperspectral unmixing[J].IEEE Transactions on Geoscience and Remote Sensing,2011,49(11):4248-4262.
[143] J. Jin, B. Wang, L. Zhang. A novel approach based on fisher discriminant null spacefor decomposition of mixed pixels in hyperspectral imagery[J]. IEEE Geoscience andRemote Sensing Letters,2010,7(8):699-703.
[144] J. Zhang, B. Rivard, A. S. Azofeifa. Derivative spectral unmixing of hyperspectraldata applied to mixtures of lichen and rock[J]. IEEE Transactions on Geoscience andRemote Sensing,2004,42(9):1934-1940.
[145] X. Du, H. Ying, F. Lin. Theory of extended fuzzy discrete-event systems for handlingranges of knowledge uncertainties and subjectivity[J]. IEEE Transactions on FuzzySystems,2009,17(2):316-328.
[146]乔玉龙,潘正祥,孙圣和.一种改进的快速k-近邻分类算法[J].电子学报,2005,33(6):1146-1149.
[147]徐海霞,刘国海,周大为.基于改进核模糊聚类算法的软测量建模研究[J].仪器仪表学报,2009,30(10):2226-2231.
[148]伍忠东,高新波,谢维信.基于核方法的模糊聚类算法[J].西安电子科技大学学报,2004,31(4):533-537.
[149]李洁,高新波,焦李成.基于特征加权的模糊聚类新算法[J].电子学报,2006,34(1):89-92.
[150]皋军,王士同.基于最大散度差别判别准则的模糊聚类算法[J].软件学报,2009,20(11):2939-2949.
[151] B. C. Kuo, D. A. Landgrebe. Nonparametric weighted feature extraction forclassification[J]. IEEE Transactions on Geoscience and Remote Sensing,2004,42(5):1096-1105.
[152]王宏伟,詹荣开,贺汉根.基于模糊聚类的改进模糊辨识方法[J].电子学报,2001,29(4):436-438.
[153]高新波,裴继红,谢维信.模糊c-均值聚类算法中加权指数m的研究[J].电子学报,2000,28(4):80-83.
[154]邢宗义,张永,侯远龙.基于模糊聚类和遗传算法的具备解释性和精确性的模糊分类系统设计[J].电子学报,2006,34(1):83-88.
[155]高新波,李洁,姬红兵.基于模糊c均值聚类与统计检验指导的多阈值图像自动分割算法[J].电子学报,2004,32(4):661-664.
[156]周世兵,徐振源,唐旭清.新的K-均值算法最佳聚类数确定方法[J].计算机工程与应用,2010,46(16):27-31.
[157] J. R. Quinlan. Induction of decision trees[J]. Machine Learning,1986,1(1):81-106.
[158] C. I. Chang, B. Ji. Weighted abundance-constrained linear spectral mixure analysis[J].IEEE Transactions on Geoscience and Remote Sensing,2006,44(2):378-388.
© 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 |