基于流形学习的高光谱图像非线性降维算法
|
摘要
针对高光谱图像同一像元内存在多种地物种类,且地物之间具有多重反射,导致高光谱数据的非线性,采用传统的线性降维算法效果不佳等问题,提出利用流形学习的方法来寻找嵌入在高维观测数据空间的低维光滑流形,实现高光谱数据的非线性光谱降维。模拟和真实高光谱遥感数据实验结果表明,与传统的线性降维方法 PCA相比,经过等距映射、局部切空间排列等流行学习算法降维后的高光谱图像具有更好的光谱端元可分性。
Hyperspectral data is nonlinear,which is caused by the existence of a variety of features and multiple reflectance in a pixel.Therefore that traditional linear reduction algorithms has little effect.According to the nonlinear structure of hyperspectral data,a manifold learning method is presented to find out the low-dimensional smooth manifold embedded in high-dimensional data space,so that nonlinear spectral dimensionality reduction can be carried out.Experimental results on simulated and real hyperspectral data show that machine learning dimensionality reduction methods based on isometric mapping(ISOMAP) or local tangent space alignment(LTSA) outperform principal component analysis(PCA) algorithm with the advantage of better endmember divisibility in hyperspectral images.
Hyperspectral data is nonlinear,which is caused by the existence of a variety of features and multiple reflectance in a pixel.Therefore that traditional linear reduction algorithms has little effect.According to the nonlinear structure of hyperspectral data,a manifold learning method is presented to find out the low-dimensional smooth manifold embedded in high-dimensional data space,so that nonlinear spectral dimensionality reduction can be carried out.Experimental results on simulated and real hyperspectral data show that machine learning dimensionality reduction methods based on isometric mapping(ISOMAP) or local tangent space alignment(LTSA) outperform principal component analysis(PCA) algorithm with the advantage of better endmember divisibility in hyperspectral images.
引文
[1]KESHAVN A N,MUSTARD J F.Spectral unmixing[J].IEEE Signal Processing Magazine,2002,19(1):44-57.
[2]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.TONG Q X,ZHANG B,ZHENG L F.Hyperspectral remote sensing-principle,technology and application[M].Beijing:Higher Education Press,2006.
[3]BACHMANN C M,AINSWORTH T L,FUSINA R A.Exploiting manifold geometry in hyperspectral imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2005,43(3):441-454.
[4]TENENBAUM J B,DE SILVA V,LANGFORD J C.A global geometric framework for nonlinear dimensionality reduction[J].Science,2000,290:2319-2322.
[5]ROWEIS S T,SAUL L K.Nonlinear dimensionality reduction by locally linear embedding[J].Science,2000,290:2323-2326.
[6]王靖.流型学习的理论与方法研究[D].杭州:浙江大学,2006.WANG J.Research on manifold learning:theories and approaches[D].Hangzhou:Zhejiang University,2006
[7]ROWEIS S T,SAUL L K.Nonlinear dimensionality reduction by locally linear embedding[J].Science,2000,290:2323-2326.
[8]BELKIN M,NIYOGI P.Laplacian eigenmaps and spectral techniques for embedding and clustering[C]//[s.n.].NIPS.[s.l.]:[s.n.],2001:585-591.
[9]杜培军,王小美,谭琨,等.利用流形学习进行高光谱遥感影像的降维与特征提取[J].武汉大学学报(信息科学版),2011,36(2):148-152.DU P J,WANG X M,TAN K,et al.Dimensionality reduction and feature extraction from hyperspectral remote sensing imagery based on manifold learning[J].Geomatics and Information Science of Wuhan University,2011,36(2):148-152.
[10]CHEN Y,CRAWFORD M M,GHOSH J.Applying nonlinear manifold learning to hyperspectral data for land cover classification[C]//[s.n.].IGARSS.[s.l.]:[s.n.],2005:24-29.
[11]杨磊,刘尚争.基于图像欧式距离和拉普拉斯特征映射的端元提取算法[J].电光与控制,2016,23(4):48-52.YANG L,LIU S Z.Endmember extraction based on image euclidean distance and laplacian eigenmaps[J].Electronics Optics&Control,2016,23(4):48-52.
[12]谷瑞军.基于流形学习的高维空间分类器研究[D].镇江:江南大学,2008.GU R J.A study on classifiers in high-dimensional space based on manifold learning[D].Zhenjiang:Jiangnan University,2008.
[13]PLAZA A,MARTINEZ P,PEREZ R,et al.A quantitative and comparative analysis of endmember extraction algorithms from hyperspectral data[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(3):650-663.
[14]唐晓燕,高昆,倪国强,等.基于流形学习和空间信息的改进N-FINDR端元提取算法[J].光谱学与光谱分析,2013,33(9):2519-2524.TANG X Y,GAO K,NI G Q,et al.An improved NFINDR endmember extraction algorithm based on manifold learning and spatial information[J].Spectroscopy and Spectral Analysis,2013,33(9):2519-2524.
[15]CHANG C I,DU Q.Estimation of number of spectrally distinct signal sources in hyperspectral imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(3):608-619.
[16]SWAYZE G,CLARK R,KRUSE F,et al.Ground-truthing AVIRIS mineral mapping at Cuprite,Nevada[C]//JPL.Summaries of the Third Annual JPL Airborne Geoscience Workshop.[s.l.]:[s.n.],1992.
[2]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.TONG Q X,ZHANG B,ZHENG L F.Hyperspectral remote sensing-principle,technology and application[M].Beijing:Higher Education Press,2006.
[3]BACHMANN C M,AINSWORTH T L,FUSINA R A.Exploiting manifold geometry in hyperspectral imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2005,43(3):441-454.
[4]TENENBAUM J B,DE SILVA V,LANGFORD J C.A global geometric framework for nonlinear dimensionality reduction[J].Science,2000,290:2319-2322.
[5]ROWEIS S T,SAUL L K.Nonlinear dimensionality reduction by locally linear embedding[J].Science,2000,290:2323-2326.
[6]王靖.流型学习的理论与方法研究[D].杭州:浙江大学,2006.WANG J.Research on manifold learning:theories and approaches[D].Hangzhou:Zhejiang University,2006
[7]ROWEIS S T,SAUL L K.Nonlinear dimensionality reduction by locally linear embedding[J].Science,2000,290:2323-2326.
[8]BELKIN M,NIYOGI P.Laplacian eigenmaps and spectral techniques for embedding and clustering[C]//[s.n.].NIPS.[s.l.]:[s.n.],2001:585-591.
[9]杜培军,王小美,谭琨,等.利用流形学习进行高光谱遥感影像的降维与特征提取[J].武汉大学学报(信息科学版),2011,36(2):148-152.DU P J,WANG X M,TAN K,et al.Dimensionality reduction and feature extraction from hyperspectral remote sensing imagery based on manifold learning[J].Geomatics and Information Science of Wuhan University,2011,36(2):148-152.
[10]CHEN Y,CRAWFORD M M,GHOSH J.Applying nonlinear manifold learning to hyperspectral data for land cover classification[C]//[s.n.].IGARSS.[s.l.]:[s.n.],2005:24-29.
[11]杨磊,刘尚争.基于图像欧式距离和拉普拉斯特征映射的端元提取算法[J].电光与控制,2016,23(4):48-52.YANG L,LIU S Z.Endmember extraction based on image euclidean distance and laplacian eigenmaps[J].Electronics Optics&Control,2016,23(4):48-52.
[12]谷瑞军.基于流形学习的高维空间分类器研究[D].镇江:江南大学,2008.GU R J.A study on classifiers in high-dimensional space based on manifold learning[D].Zhenjiang:Jiangnan University,2008.
[13]PLAZA A,MARTINEZ P,PEREZ R,et al.A quantitative and comparative analysis of endmember extraction algorithms from hyperspectral data[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(3):650-663.
[14]唐晓燕,高昆,倪国强,等.基于流形学习和空间信息的改进N-FINDR端元提取算法[J].光谱学与光谱分析,2013,33(9):2519-2524.TANG X Y,GAO K,NI G Q,et al.An improved NFINDR endmember extraction algorithm based on manifold learning and spatial information[J].Spectroscopy and Spectral Analysis,2013,33(9):2519-2524.
[15]CHANG C I,DU Q.Estimation of number of spectrally distinct signal sources in hyperspectral imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(3):608-619.
[16]SWAYZE G,CLARK R,KRUSE F,et al.Ground-truthing AVIRIS mineral mapping at Cuprite,Nevada[C]//JPL.Summaries of the Third Annual JPL Airborne Geoscience Workshop.[s.l.]:[s.n.],1992.