基于子空间中主成分最优线性预测的高光谱波段选择

吴一全; 周杨; 盛东慧; 叶骁来

doi:doi:10.11972/j.issn.1001-9014.2018.01.021

红外与毫米波学报, 2018, 37 (1): 119, 网络出版: 2018-03-14

基于子空间中主成分最优线性预测的高光谱波段选择

Band selection of hyperspectral image based on optimal linear prediction of principal components in subspace

论文大纲

吴一全 ^1,2,*周杨 ¹盛东慧 ¹叶骁来 ¹

作者单位

¹ 南京航空航天大学电子信息工程学院，江苏南京 211106

² 中国科学院西安光学精密机械研究所中科院光谱成像技术重点实验室，陕西西安 710119

摘要

针对高光谱遥感图像的异常检测问题，为了使高光谱降维数据能更完整地保留其光谱信息，提出了基于子空间中主成分最优线性预测的波段选择方法.采用改进相关性度量的谱聚类方法将高光谱波段划分为不同的子空间，并对各子空间中的波段进行主成分分析(PCA)，选择主要分量作为重构目标；以子空间追踪法为搜索策略，从各子空间中选择数个波段对其重构目标进行联合最优线性预测；合并各子空间中的所选波段得到最佳波段子集.实验结果表明，该方法选择的波段子集可以较完整地重构原始数据，与原始数据以及自适应波段选择(ABS)方法、线性预测(LP)方法、最大方差主成分分析(MVPCA)方法、自相关矩阵波段选择(ACMBS)方法、组合因子最优波段选择(OCFBS)方法得到的波段子集相比，其波段子集具有更好的异常检测性能.

Abstract

In the case of hyperspectral anomaly detection, in order to make hyperspectral low-dimensional data preserve the spectral information more completely, a band selection method based on the optimal linear prediction of principal components in subspace was proposed. Hyperspectral bands are divided into different subspaces by spectral clustering with the improved correlation measure. The principal component analysis (PCA) of bands is presented in each subspace, and main components are selected as the reconstructed targets. The subspace tracking method serves as the search strategy, and several bands are selected from each subspace to perform the joint optimal linear prediction of reconstructed targets. The selected bands in each subspace are combined to obtain the optimal band subset. Experimental results show that, the proposed method can reconstruct the original data more completely. Compared with original data, and the band subsets obtained by adaptive band selection (ABS) method, linear prediction (LP) method, maximum-variance principal component analysis (MVPCA) method, auto correlation matrix-based band selection (ACMBS) method and optimal combination factors-based band selection (OCFBS) method, the band subset of proposed method has superior performance of anomaly detection.

参考文献

[1] Wang S, Chang C I. Variable-number variable-band selection for feature characterization in hyperspectral signatures ［J］. IEEE Transactions on Geoscience and Remote Sensing, 2007,45(9):2979-2992.

[2] Jia S, Ji Z, Qian Y. Unsupervised band selection for hyperspectral imagery classification without manual band removal ［J］. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2012, 5(2): 531-543.

[3] LIU Xue-Song, GE Liang, WANG Bin, et al. An unsupervised band selection algorithm for hyperspectral imagery based on maximal information ［J］. J. Infrared Millim. Waves(刘雪松, 葛亮, 王斌, 等. 基于最大信息量的高光谱遥感图像无监督波段选择方法. 红外与毫米波学报), 2012, 31(2):166-170.

[4] Pal M, Foody G M. Feature selection for classification of hyperspectral data by SVM ［J］. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(5):2297-2307.

[5] LIU Chun-Hong, ZHAO Chun-Hui, ZHANG Ling-Yan. A new method of hyperspectral remote sensing image dimensional reduction ［J］. Journal of Image and Graphics(刘春红, 赵春晖, 张凌雁. 一种新的高光谱遥感图像降维方法. 中国图象图形学报), 2005, 10(2):218-222.

[6] Yang H, Du Q, Chen G. Unsupervised hyperspectral band selection using graphics processing units ［J］. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2011, 4(3):660-668.

[7] Chang C I, Du Q, Sun T L, et al. A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classification ［J］. IEEE Transactions on Geoscience and Remote Sensing, 1999, 37(6):2631-2641.

[8] Sun K, Geng X, Ji L. An efficient unsupervised band selection method based on an autocorrelation matrix for a hyperspectral image ［J］. International Journal of Remote Sensing, 2014, 35(21):7458-7476.

[9] FANG Shuai, QU Chneg-Jia, YANG Xue-Zhi, et al. Linear prediction band selection based on optimal combination factors［J］. Journal of Image and Graphics(方帅, 瞿成佳, 杨学志, 等. 组合因子最优的线性预测波段选择. 中国图象图形学报), 2016, 21(2):255-262.

[10] Dai W, Milenkovic O. Subspace pursuit for compressive sensing signal reconstruction［J］. IEEE Transactions on Information Theory, 2009, 55(5):2230-2249.

[11] Von L U. A tutorial on spectral clustering［J］. Statistics and Computing, 2007, 17(4):395-416.

[12] QIN Fang-Pu, ZHANG Ai-Wu, WANG Shu-Min, et al. Hyperspectral band selection based on spectral clustering and inter-class separability factor［J］. Spectroscopy and Spectral Analysis(秦方普, 张爱武, 王书民, 等. 基于谱聚类与类间可分性因子的高光谱波段选择. 光谱学与光谱分析), 2015, 35(5):1357-1364.

[13] TIAN Yei, ZHAO Chun-Hui, JI Ya-Xin. The principal component analysis applied to hyperspectral remote sensing image dimensional reduction［J］. Natural Sciences Journal of Harbin Normal University(田野, 赵春晖, 季亚新. 主成分分析在高光谱遥感图像降维中的应用. 哈尔滨师范大学自然科学学报), 2007, 23(5): 58-60.

[14] CHEN Hong-Da, PU Han-Ye, WANG Bin, et al. Image Euclidean distance-based manifold dimensionality reduction algorithm for hyperspectral imagery［J］. J. Infrared Millim. Waves(陈宏达,普晗晔,王斌, 等. 基于图像欧氏距离的高光谱图像流形降维算法. 红外与毫米波学报), 2013, 32(5):450-455.

[15] ZHAO Chun-Hui, YAO Zhe-Feng. Local kernel RX algorithm-based hyperspectral real-time detection［J］. J. Infrared Millim. Waves(赵春晖，姚浙峰. 基于局部核RX算法的高光谱实时检测. 红外与毫米波学报), 2016, 35(6): 708-714.

吴一全, 周杨, 盛东慧, 叶骁来. 基于子空间中主成分最优线性预测的高光谱波段选择[J]. 红外与毫米波学报, 2018, 37(1): 119. WU Yi-Quan, ZHOU Yang, SHENG Dong-Hui, YE Xiao-Lai. Band selection of hyperspectral image based on optimal linear prediction of principal components in subspace[J]. Journal of Infrared and Millimeter Waves, 2018, 37(1): 119.

基于子空间中主成分最优线性预测的高光谱波段选择

关于本站 Cookie 的使用提示

全站搜索

基于子空间中主成分最优线性预测的高光谱波段选择

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索