Spatial Purity Based Endmember Extraction for Spectral Mixture Analysis

被引:80
作者
Mei, Shaohui [1 ]
He, Mingyi [1 ]
Wang, Zhiyong [2 ]
Feng, Dagan [2 ]
机构
[1] Northwestern Polytech Univ, Dept Elect & Informat Engn, Xian 710129, Peoples R China
[2] Univ Sydney, Sch Informat Technol, Sydney, NSW 2006, Australia
来源
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 2010年 / 48卷 / 09期
基金
美国国家科学基金会;
关键词
Endmember extraction (EE); endmember identification; hyperspectral remote sensing; spatial purity (SP) index; spectral mixture analysis (SMA); spectral-spatial; INDEPENDENT COMPONENT ANALYSIS; ALGORITHM; QUANTIFICATION; CLASSIFICATION; SELECTION;
D O I
10.1109/TGRS.2010.2046671
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
Spectral mixture analysis (SMA) has been widely utilized to address the mixed-pixel problem in the quantitative analysis of hyperspectral remote sensing images, in which endmember extraction (EE) plays an extremely important role. In this paper, a novel algorithm is proposed to integrate both spectral similarity and spatial context for EE. The spatial context is exploited from two aspects. At first, initial endmember candidates are identified by determining the spatial purity (SP) of pixels in their spatial neighborhoods (SNs). Several SP measurements are investigated at both intensity level and feature level. In order to alleviate local spectra variability, the average of the pixels in pure SNs are voted as endmember candidates. Then, the spatial connectivity is utilized to merge spatially related endmember candidates by finding connection paths in a graph so that the number of endmember candidates is further reduced, which results in computational efficiency and better performance in SMA by alleviating global spectral variability. Experimental results on both synthetic and real hyperspectral images demonstrate that the proposed SP based EE (SPEE) algorithm outperforms the other popular EE algorithms. It is also observed that feature-level SP measurements are more distinguishable than intensity-level SP measurements to discriminate pure SNs from mixed SNs.
引用
收藏
页码:3434 / 3445
页数:12
相关论文
共 55 条
[1]  
AGUILAR P, 2000, P INT C AUT ROB COMP, P58
[2]  
[Anonymous], 12TH P IGARSS 89 CAN
[3]  
[Anonymous], 1995, 5 ANN JPL AIRB EARTH
[4]  
BALAKRISHNAN R, 2004, GRAPH THEORY ITS APP
[5]  
Ball, 1965, ISODATA NOVEL METHOD
[6]  
Banerjee A, 2007, INT GEOSCI REMOTE SE, P3817
[7]   A method for manual endmember selection and spectral unmixing [J].
Bateson, A ;
Curtiss, B .
REMOTE SENSING OF ENVIRONMENT, 1996, 55 (03) :229-243
[8]   Endmember bundles: A new approach to incorporating endmember variability into spectral mixture analysis [J].
Bateson, CA ;
Asner, GP ;
Wessman, CA .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2000, 38 (02) :1083-1094
[9]   ICE: A statistical approach to identifying endmembers in hyperspectral images [J].
Berman, M ;
Kiiveri, H ;
Lagerstrom, R ;
Ernst, A ;
Dunne, R ;
Huntington, JF .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2004, 42 (10) :2085-2095
[10]   Hyperspectral subspace identification [J].
Bioucas-Dias, Jose M. ;
Nascimento, Jose M. P. .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2008, 46 (08) :2435-2445