基于子空间中主成分最优线性预测的高光谱波段选择
Received:February 15, 2017  Revised:June 15, 2017  点此下载全文
引用本文:
Hits: 431
Download times: 48
Author NameAffiliationE-mail
WU Yi-Quan Nanjing University of Aeronautics and Astronautics、Key Laboratory of Spectral Imaging Technology CASXi’an Institute of Optics and Precision Mechanics of CAS nuaaimage@163.com 
ZHOU Yang Nanjing University of Aeronautics and Astronautics 1414032393@qq.com 
SHENG Dong-Hui Nanjing University of Aeronautics and Astronautics  
YE Xiao-Lai Nanjing University of Aeronautics and Astronautics  
基金项目:国家自然科学基金(61573183);中国科学院光谱成像重点实验室开放资助(LSIT201401);江苏高校优势学科建设工程
中文摘要: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。
中文关键词:遥感  高光谱图像  波段选择  主成分  线性预测  子空间追踪  谱聚类
 
Band Selection of Hyperspectral Image Based on Optimal Linear Prediction of Principal Components in Subspace
Abstract:Aiming at the problem 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 is 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, 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 .
keywords:remote sensing  hyperspectral image  band selection  principal component  linear prediction  subspace pursuit  spectral clustering
  View/Add Comment  Download reader

Copyright:《Journal of Infrared And Millimeter Waves》