Fully Constrained Least Squares Spectral Unmixing by Simplex Projection

被引:172
作者
Heylen, Rob [1 ]
Burazerovic, Dzevdet [1 ]
Scheunders, Paul [1 ]
机构
[1] Univ Antwerp, Dept Phys, Interdisciplinary Inst Broadband Technol IBBT, Visionlab, B-2610 Antwerp, Belgium
来源
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 2011年 / 49卷 / 11期
关键词
Hyperspectral imaging; multidimensional signal processing; spectral analysis; NONNEGATIVE MATRIX FACTORIZATION; ENDMEMBER EXTRACTION; HYPERSPECTRAL DATA; CLASSIFICATION; ALGORITHM; MIXTURE;
D O I
10.1109/TGRS.2011.2155070
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
We present a new algorithm for linear spectral mixture analysis, which is capable of supervised unmixing of hyperspectral data while respecting the constraints on the abundance coefficients. This simplex-projection unmixing algorithm is based upon the equivalence of the fully constrained least squares problem and the problem of projecting a point onto a simplex. We introduce several geometrical properties of high-dimensional simplices and combine them to yield a recursive algorithm for solving the simplex-projection problem. A concrete implementation of the algorithm for large data sets is provided, and the algorithm is benchmarked against well-known fully constrained least squares unmixing (FCLSU) techniques, on both artificial data sets and real hyperspectral data collected over the Cuprite mining region. Unlike previous algorithms for FCLSU, the presented algorithm possesses no optimization steps and is completely analytical, severely reducing the required processing power.
引用
收藏
页码:4112 / 4122
页数:11
相关论文
共 38 条
[1]  
[Anonymous], 2010, Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), 2010 2nd Workshop on, DOI DOI 10.1109/WHISPERS.2010.5594929
[2]  
[Anonymous], 1974, Solving least squares problems
[3]  
Bajorski P, 2004, INT GEOSCI REMOTE SE, P3207
[4]  
BOARDMAN JW, 1994, INT GEOSCI REMOTE SE, P2369, DOI 10.1109/IGARSS.1994.399740
[5]   A new growing method for simplex-based endmember extraction algorithm [J].
Chang, Chein-I ;
Wu, Chao-Cheng ;
Liu, Wei-min ;
Ouyang, Yen-Chieh .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2006, 44 (10) :2804-2819
[6]   Constrained band selection for hyperspectral imagery [J].
Chang, Chein-I ;
Wang, Su .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2006, 44 (06) :1575-1585
[7]   Estimation of subpixel target size for remotely sensed imagery [J].
Chang, CI ;
Ren, H ;
Chang, CC ;
D'Amico, F ;
Jensen, JO .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2004, 42 (06) :1309-1320
[8]   Estimation of number of spectrally distinct signal sources in hyperspectral imagery [J].
Chang, CI ;
Du, Q .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2004, 42 (03) :608-619
[9]   Least squares subspace projection approach to mixed pixel classification for hyperspectral images [J].
Chang, CI ;
Zhao, XL ;
Althouse, MLG ;
Pan, JJ .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 1998, 36 (03) :898-912
[10]   Generalization of Subpixel Analysis for Hyperspectral Data With Flexibility in Spectral Similarity Measures [J].
Chen, Jin ;
Jia, Xiuping ;
Yang, Wei ;
Matsushita, Bunkei .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2009, 47 (07) :2165-2171