文档介绍：1007-4619 (2011) 04-659-21 Journal of Remote Sensing 遥感学报
The development parison of endmember extraction
algorithms using hyperspectral imagery
LI Ersen1,2, ZHU Shulong1, ZHOU Xiaoming1, YU Wenjie1
1. Institute of Surveying and Mapping,Information Engineering University, Henan Zhengzhou 450052, China;
2. Key Laboratory of Mine Spatial Information Technologies, Henan Jiaozuo 454000, China
Abstract: The mixels in the hypersepectral images not only infl uence the accuracy of target detection and classifi cation, but
also greatly hinder the development of quantitative remote sensing. The typical endmember extraction algorithms now available
are analyzed and summarized. These algorithms can be classifi ed into two types based on the hypothesis of the existence of the
pure pixels: endmember identifi cation algorithm and endmember generation algorithm. Six endmember extraction algorithms, in-
cluding N-FINDR, VCA, SGA, OSP, ICE and MVC-NMF, are introduced pared using experimental data, which further
show their advantages and disadvantages. With results of various methods, the future perspective is proposed for further study.
Key words: hyperspectral image, mixel, linear spectral mixing model, endmember
CLC number: TP79 Document code: A
Citation format: Li E S, Zhu S L, Zhou X M and Yu W J. 2011. The development parison of endmember extraction algorithms
using hyperspectral imagery. Journal of Remote Sensing, 15(4): 659–679
1 INTRODUCTION (Gruninger, et al., 2004), ponent Analysis(VCA)
(Nascimento, et al., 2005; Nascimento, 2006), Simplex Grow-
The mixels exist abroadly in the hyperspectral images, and ing Algorithm(SGA) (Chang, et al., 2006), Orthogonal Subspace
the mixed pixel position is the effective approach to solve Projection(OSP) (Harsanyi & Chang, 1994), and the Sequential
this problem to realize the sub-pixel classifi cation (Kumar, et al., Projection Algorithm(SPA)(Zhang, et al., 2008). The EGAs mainly
2008). The Linear