高光谱数据降维及端元提取
|
摘要
高光谱数据能提供地物数十至数百个窄波段的光谱信息,通常波段宽度小于10nm,可产生连续的光谱曲线,具有极高光谱分辨率。然而,由于高光谱图像维数众多,数据量巨大,给后续处理带来巨大的挑战。本文以Hyperion图像为对象,研究了高光谱的特征提取和端元提取技术。
论文首先介绍了高光谱数据处理面临的问题,包括高光谱的巨大数据量如何处理,高光谱图像混合像元的产生以及端元的提取,分析了对高光谱图像进行特征提取和端元提取的必要性。
针对PCA特征提取算法反映地物的整体信息,忽略细节信息的缺点,对小波PCA算法进行改进,通过使用不同的小波基,运用小波PCA算法进行特征提取,最终选取最优的小波基。
研究投影寻踪算法中的模拟退火优化算法,将此算法应用于寻找最优的投影向量。
分析PPI端元提取算法,将进行预处理的PCA算法换成小波PCA算法,增加了对地物细节的反映,由此进行端元提取取得了较好的效果。论文还分析了高光谱图像的处理流程,包括图像的几何校正,图像的分类技术(K-mean分类,最大似然分类),高光谱图像镶嵌,并完成了通过高光谱图像提取地物的光谱曲线。
对所开发的各算法的应用与演示,都取得了良好的结果。
Hyperspectral data can provide hundreds of narrow-band spectral information with the band width of less than 10 nm. It can produce a continuous spectrum, with very high spectral resolution. However, due to the hyperspectral image dimension of the large volume of data, we have to deal with these follow-up challenges. In this paper, by studying Hyperion hyperspectral image, feature and endmember extraction technology is developed.
First,the problems of hyperspectral data processing are introduced, including how to deal with huge volume of hyperspectral data, how mixed-pixels developed in hyperspectral image and endmember extraction. the necessity of hyperspectral image feature and endmember extraction is analyzed.
Feature extraction algorithms for PCA reflects the overall features of information, but overlooks details of the feature. Due to this shortcoming, the wavelet PCA algorithm is developed with different wavelet basis;ultimately the best wavelet basis is selected. Simulated annealing in Projection Pursuit algorithm is studied, and then this algorithm is applied to find the optimal projection vector.
By analysis of PPI endmember extraction algorithms, the PCA pre-processing algorithm can be replaced by wavelet PCA algorithm. The wavelet PCA algorithm can increase reflections of the surface feature details, so the PPI endmember extraction algorithm has achieved fairly good results with the improved algorithm.
The paper also analysis the process of hyperspectral data processing,including image geometric correction, classification(K-mean classification, the maximum likelihood classification),hyperspectral image mosaic, and the spectral feature curve extraction from hyperspectral data.
All the algorithms accomplished in this paper carry out excellent results in applications and demonstrations.
论文首先介绍了高光谱数据处理面临的问题,包括高光谱的巨大数据量如何处理,高光谱图像混合像元的产生以及端元的提取,分析了对高光谱图像进行特征提取和端元提取的必要性。
针对PCA特征提取算法反映地物的整体信息,忽略细节信息的缺点,对小波PCA算法进行改进,通过使用不同的小波基,运用小波PCA算法进行特征提取,最终选取最优的小波基。
研究投影寻踪算法中的模拟退火优化算法,将此算法应用于寻找最优的投影向量。
分析PPI端元提取算法,将进行预处理的PCA算法换成小波PCA算法,增加了对地物细节的反映,由此进行端元提取取得了较好的效果。论文还分析了高光谱图像的处理流程,包括图像的几何校正,图像的分类技术(K-mean分类,最大似然分类),高光谱图像镶嵌,并完成了通过高光谱图像提取地物的光谱曲线。
对所开发的各算法的应用与演示,都取得了良好的结果。
Hyperspectral data can provide hundreds of narrow-band spectral information with the band width of less than 10 nm. It can produce a continuous spectrum, with very high spectral resolution. However, due to the hyperspectral image dimension of the large volume of data, we have to deal with these follow-up challenges. In this paper, by studying Hyperion hyperspectral image, feature and endmember extraction technology is developed.
First,the problems of hyperspectral data processing are introduced, including how to deal with huge volume of hyperspectral data, how mixed-pixels developed in hyperspectral image and endmember extraction. the necessity of hyperspectral image feature and endmember extraction is analyzed.
Feature extraction algorithms for PCA reflects the overall features of information, but overlooks details of the feature. Due to this shortcoming, the wavelet PCA algorithm is developed with different wavelet basis;ultimately the best wavelet basis is selected. Simulated annealing in Projection Pursuit algorithm is studied, and then this algorithm is applied to find the optimal projection vector.
By analysis of PPI endmember extraction algorithms, the PCA pre-processing algorithm can be replaced by wavelet PCA algorithm. The wavelet PCA algorithm can increase reflections of the surface feature details, so the PPI endmember extraction algorithm has achieved fairly good results with the improved algorithm.
The paper also analysis the process of hyperspectral data processing,including image geometric correction, classification(K-mean classification, the maximum likelihood classification),hyperspectral image mosaic, and the spectral feature curve extraction from hyperspectral data.
All the algorithms accomplished in this paper carry out excellent results in applications and demonstrations.
引文
[1] GF.Hughes. On the mean accuracy of statistical pattern recognizers[J]. IEEE Transactions.Inform.Theory,vol.IT-14,1968
[2] [英] A.R.Webb著,王萍,杨培龙,罗颖听译. 统计模式识别(第二版)[M].北京.电子工业出版社,2004
[3] A A Green,M berman,P Switzer. A transformation for ordering multispectral data in terms of iamge quality with implications for noise removal[J]. IEEE Transactions on Geoscience and Remote Sensing,1998,26(1):65-74
[4] C.Jutten,J.Herault. Independent component ana1ysis versus principal component analysis[C]. Proc. European Symposium on Signal Processing Conference (EUSIPCO’88),Grenob1e,1988:643-646.
[5] J.M.P.Nascimento,J.M.B.Dias. Does independent component analysis play a role in unmixing hyperspectral data?[J]. IEEE Transactions.Geoscience and Remote Sensing,2005,43(1):175-187
[6] 曾生根,王小敏,范瑞斌等. 基于独立量分析的遥感图像分类技术[J]. 遥感学报,2004,8(2):150-157
[7] R.A.Fisher. The use of multiple measurements in taxonomic problems[J]. Annual of Eugenics,1936,7(1):179-188
[8] Lee C,Landgrebe D A. Decision Boundary Feature Selection for Non-Parametric Classifications [J]. IEEE Transactions on System , Man , and Cybernetics ,1993 ,23 ( 2): 433-444
[9] Luis O,Jimenez,David A.Landgrebe,Hyperspectral data anaysis and supervised feature reduction via projection pursuit[J]. Geoscience and Remote Sensing,IEEE Transactions,1999,37(6):2653-2667
[10] 张连蓬,柳钦火等. 多方向投影寻踪与高光谱遥感图像特征提取[J]. 2007,32(2):155-158
[11] 王秀朋. 基于投影寻踪的高光谱图像降维算法研究[D]. 西安:西北工业大学,2006
[12] James Theiler,Dominique D.Lavenier,Neal R.Harvey,et al.Using blocks of skewers for the faster computation of pixel purity index[J]. 2000,4132:61-71
[13] 薛绮. 基于线性混合模型的高光谱图像端元提取方法研究[D]. 长沙: 国防科技大学,2004
[14] 薛绮等. 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
[15] 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,2003
[16] M.E.Winter, N-FINDR:An Algorithm for Fast Autonomous Spectral End-Member Determination in Hyperspectral Data SPIE Proceedings in Imaging Spectrometry[J] 1999,3753:266-277
[17] 王立国. 高光谱图像混合像素处理技术研究[D]. 哈尔滨:哈尔滨工业大学,2005
[18] R.A.Neville and K.Staenz,T.Szeredi and J.Lefebvre,P.Hauff, Automatic Endmember Extraction From Hyperspectral Data For Mineral Exploration[C],The Four International Airborne Sensing Conference and E xhibition/21 Canadian Symposium on Remote Sensing,Ottawa, Ontario,Canada,June 1999:21-24
[19] 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,2006.
[20] Clark,R.N.,King,T.V.V.,Kleijwa,M.,Swayze.High spectral resolution reflectance spectroscopy of minerals[J].Geophysics,1990, 95:12653-12680
[21] 杨诸胜. 高光谱图像降维及分割研究[D]. 西安:西北工业大学,2006
[22] Jia Xiuping,J.A.Richards. Segmented principal components transformation for efficient hyperspectral remote-sensing image display and classification [J]. IEEE Geoscience and Remote Sensing,1999,37(1):538-542
[23] A A Green,M berman,P Switzer. A transformation for ordering multispectral data in terms of iamge quality with implications for noise removal[J]. Geoscience and Remote Sensing,IEEE Transactions,1998,26(1):65-74
[24] 万余庆,谭克龙,周日平. 高光谱遥感应用研究[M]. 北京:科学出版社,2006
[25] 顾海燕,李海涛,杨景辉. 基于最小噪声分离变换的遥感影响融合方法[J] 国土资源遥感,2007 2:53-55
[26] 杜鹏,赵慧洁. 基于抗噪声 ICA 的高光谱数据特征提取方法[J]. 北京航空航天大学学报 2005,31(10):1101-1105
[27] 沈渊婷. 油气勘探高光谱遥感图像处理技术研究[D]. 北京:北京理工大学,2006
[28] 舒先标. 高光谱油气勘探空间空间信息系统研究与高光谱库研建[D]. 北京:北京理工大学,2007
[29] 浦瑞良,宫鹏.高光谱遥感及应用 [M].北京:高等教育出版社,2000
[30] 童庆禧.遥感科学技术进展 [J].地理学报. 1994, 49: 616～624
[31] 赵英时等.遥感应用分析原理与方法 [M].北京:科学出版社,2003
[32] 张良培,张立福. 高光谱遥感[M]. 武汉:武汉大学出版社,2005
[33] 张纯,胡振东,仲政.车桥耦合振动分析的 Harr 小波方法[J]. 振动与冲击,2007,26(4):77-80
[34] 吴传庆,童庆禧,郑兰芬. 基于小波变换的高光谱图像消噪[J]. 遥感信息理论研究.2005.4,10-13
[35] 朱峰,蔡碧野,李峰. 基于 Symlets 小波的虹膜特征提取[J]. 长沙理工大学学报(自然科学版).2005,2(4):66-69
[36] 薛白,刘文耀,王金涛,左坤隆. 虹膜识别的身份认证系统[J]. 光电工程. 2004,31(1):69-72
[37] 王强. 模拟退火算法的改进及其应用[J]. 应用数学.1993,6(4):392-397
[38] 朱桂凤,毛淑贤,白志富. 基于 VC 的模拟退火算法的实现[J]. 计算机工程与设计. 2007,28(7):1514-1516
[39] 闫立梅,焦德杰. 一种改进的特征提取模型及其算法[J]. 德州学院学报.2007,23(4):15-22
[40] 谢云. 模拟退火算法综述[J]. 微计算机信息.1998,14(5):66-68
[41] 赵晶,唐焕文,朱训芝. 模拟退火算法的一种改进及其应用研究[J].2006,46(5):775-780
[42] http://eo1.usgs.gov.EO-1 User Guide.2003
[43] Van der Meer, F.D.. Analysis of spectral absorption features in hyperspectral imagery[J] . International Journal of Applied Earth Observation and Geo information, 2004.5: 55-68
[44] Bowman, W.D.. The relationships between leaf water status, gas exchange , and spectral reflectance in cotton leaves[J] . Rem. Sens. Environ. 1989, 30: 49-255
[45] Cloutis, E.A.. Spectral reflectance properties of hydrocarbons[J]. remote-sensing implications. Science, 1989, 245: 165-168
[2] [英] A.R.Webb著,王萍,杨培龙,罗颖听译. 统计模式识别(第二版)[M].北京.电子工业出版社,2004
[3] A A Green,M berman,P Switzer. A transformation for ordering multispectral data in terms of iamge quality with implications for noise removal[J]. IEEE Transactions on Geoscience and Remote Sensing,1998,26(1):65-74
[4] C.Jutten,J.Herault. Independent component ana1ysis versus principal component analysis[C]. Proc. European Symposium on Signal Processing Conference (EUSIPCO’88),Grenob1e,1988:643-646.
[5] J.M.P.Nascimento,J.M.B.Dias. Does independent component analysis play a role in unmixing hyperspectral data?[J]. IEEE Transactions.Geoscience and Remote Sensing,2005,43(1):175-187
[6] 曾生根,王小敏,范瑞斌等. 基于独立量分析的遥感图像分类技术[J]. 遥感学报,2004,8(2):150-157
[7] R.A.Fisher. The use of multiple measurements in taxonomic problems[J]. Annual of Eugenics,1936,7(1):179-188
[8] Lee C,Landgrebe D A. Decision Boundary Feature Selection for Non-Parametric Classifications [J]. IEEE Transactions on System , Man , and Cybernetics ,1993 ,23 ( 2): 433-444
[9] Luis O,Jimenez,David A.Landgrebe,Hyperspectral data anaysis and supervised feature reduction via projection pursuit[J]. Geoscience and Remote Sensing,IEEE Transactions,1999,37(6):2653-2667
[10] 张连蓬,柳钦火等. 多方向投影寻踪与高光谱遥感图像特征提取[J]. 2007,32(2):155-158
[11] 王秀朋. 基于投影寻踪的高光谱图像降维算法研究[D]. 西安:西北工业大学,2006
[12] James Theiler,Dominique D.Lavenier,Neal R.Harvey,et al.Using blocks of skewers for the faster computation of pixel purity index[J]. 2000,4132:61-71
[13] 薛绮. 基于线性混合模型的高光谱图像端元提取方法研究[D]. 长沙: 国防科技大学,2004
[14] 薛绮等. 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
[15] 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,2003
[16] M.E.Winter, N-FINDR:An Algorithm for Fast Autonomous Spectral End-Member Determination in Hyperspectral Data SPIE Proceedings in Imaging Spectrometry[J] 1999,3753:266-277
[17] 王立国. 高光谱图像混合像素处理技术研究[D]. 哈尔滨:哈尔滨工业大学,2005
[18] R.A.Neville and K.Staenz,T.Szeredi and J.Lefebvre,P.Hauff, Automatic Endmember Extraction From Hyperspectral Data For Mineral Exploration[C],The Four International Airborne Sensing Conference and E xhibition/21 Canadian Symposium on Remote Sensing,Ottawa, Ontario,Canada,June 1999:21-24
[19] 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,2006.
[20] Clark,R.N.,King,T.V.V.,Kleijwa,M.,Swayze.High spectral resolution reflectance spectroscopy of minerals[J].Geophysics,1990, 95:12653-12680
[21] 杨诸胜. 高光谱图像降维及分割研究[D]. 西安:西北工业大学,2006
[22] Jia Xiuping,J.A.Richards. Segmented principal components transformation for efficient hyperspectral remote-sensing image display and classification [J]. IEEE Geoscience and Remote Sensing,1999,37(1):538-542
[23] A A Green,M berman,P Switzer. A transformation for ordering multispectral data in terms of iamge quality with implications for noise removal[J]. Geoscience and Remote Sensing,IEEE Transactions,1998,26(1):65-74
[24] 万余庆,谭克龙,周日平. 高光谱遥感应用研究[M]. 北京:科学出版社,2006
[25] 顾海燕,李海涛,杨景辉. 基于最小噪声分离变换的遥感影响融合方法[J] 国土资源遥感,2007 2:53-55
[26] 杜鹏,赵慧洁. 基于抗噪声 ICA 的高光谱数据特征提取方法[J]. 北京航空航天大学学报 2005,31(10):1101-1105
[27] 沈渊婷. 油气勘探高光谱遥感图像处理技术研究[D]. 北京:北京理工大学,2006
[28] 舒先标. 高光谱油气勘探空间空间信息系统研究与高光谱库研建[D]. 北京:北京理工大学,2007
[29] 浦瑞良,宫鹏.高光谱遥感及应用 [M].北京:高等教育出版社,2000
[30] 童庆禧.遥感科学技术进展 [J].地理学报. 1994, 49: 616～624
[31] 赵英时等.遥感应用分析原理与方法 [M].北京:科学出版社,2003
[32] 张良培,张立福. 高光谱遥感[M]. 武汉:武汉大学出版社,2005
[33] 张纯,胡振东,仲政.车桥耦合振动分析的 Harr 小波方法[J]. 振动与冲击,2007,26(4):77-80
[34] 吴传庆,童庆禧,郑兰芬. 基于小波变换的高光谱图像消噪[J]. 遥感信息理论研究.2005.4,10-13
[35] 朱峰,蔡碧野,李峰. 基于 Symlets 小波的虹膜特征提取[J]. 长沙理工大学学报(自然科学版).2005,2(4):66-69
[36] 薛白,刘文耀,王金涛,左坤隆. 虹膜识别的身份认证系统[J]. 光电工程. 2004,31(1):69-72
[37] 王强. 模拟退火算法的改进及其应用[J]. 应用数学.1993,6(4):392-397
[38] 朱桂凤,毛淑贤,白志富. 基于 VC 的模拟退火算法的实现[J]. 计算机工程与设计. 2007,28(7):1514-1516
[39] 闫立梅,焦德杰. 一种改进的特征提取模型及其算法[J]. 德州学院学报.2007,23(4):15-22
[40] 谢云. 模拟退火算法综述[J]. 微计算机信息.1998,14(5):66-68
[41] 赵晶,唐焕文,朱训芝. 模拟退火算法的一种改进及其应用研究[J].2006,46(5):775-780
[42] http://eo1.usgs.gov.EO-1 User Guide.2003
[43] Van der Meer, F.D.. Analysis of spectral absorption features in hyperspectral imagery[J] . International Journal of Applied Earth Observation and Geo information, 2004.5: 55-68
[44] Bowman, W.D.. The relationships between leaf water status, gas exchange , and spectral reflectance in cotton leaves[J] . Rem. Sens. Environ. 1989, 30: 49-255
[45] Cloutis, E.A.. Spectral reflectance properties of hydrocarbons[J]. remote-sensing implications. Science, 1989, 245: 165-168
© 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 |