RCMF: Robust Constrained Matrix Factorization for Hyperspectral Unmixing

被引:26
作者
Akhtar, Naveed [1 ]
Mian, Ajmal [2 ]
机构
[1] Australian Natl Univ, Coll Engn & Comp Sci, Canberra, ACT 2601, Australia
[2] Univ Western Australia, Sch Comp Sci & Software Engn, Crawley, WA 6009, Australia
来源
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 2017年 / 55卷 / 06期
基金
澳大利亚研究理事会;
关键词
Blind source separation; hyperspectral unmixing; robust matrix factorization; sparse representation; unsupervised unmixing; ENDMEMBER EXTRACTION; SPARSE REGRESSION; MATCHING PURSUIT; SIGNAL RECOVERY; ALGORITHM; NMF; DICTIONARIES;
D O I
10.1109/TGRS.2017.2669991
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 propose a constrained matrix factorization approach for linear unmixing of hyperspectral data. Our approach factorizes a hyperspectral cube into its constituent endmembers and their fractional abundances such that the endmembers are sparse nonnegative linear combinations of the observed spectra themselves. The association between the extracted endmembers and the observed spectra is explicitly noted for physical interpretability. To ensure reliable unmixing, we make the matrix factorization procedure robust to outliers in the observed spectra. Our approach simultaneously computes the endmembers and their abundances in an efficient and unsupervised manner. The extracted endmembers are nonnegative quantities, whereas their abundances additionally follow the sum-to-one constraint. We thoroughly evaluate our approach using synthetic data with white and correlated noise as well as real hyperspectral data. Experimental results establish the effectiveness of our approach.
引用
收藏
页码:3354 / 3366
页数:13
相关论文
共 69 条
[1]   K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation [J].
Aharon, Michal ;
Elad, Michael ;
Bruckstein, Alfred .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2006, 54 (11) :4311-4322
[2]  
Akhtar N, 2015, PROC CVPR IEEE, P3631, DOI 10.1109/CVPR.2015.7298986
[3]   SUnGP: A greedy sparse approximation algorithm for hyperspectral unmixing [J].
Akhtar, Naveed ;
Shafait, Faisal ;
Mian, Ajmal .
2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2014, :3726-3731
[4]   Futuristic Greedy Approach to Sparse Unmixing of Hyperspectral Data [J].
Akhtar, Naveed ;
Shafait, Faisal ;
Mian, Ajmal .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2015, 53 (04) :2157-2174
[5]  
Akhtar N, 2014, IEEE WINT CONF APPL, P953, DOI 10.1109/WACV.2014.6836001
[6]   Hyperspectral Data Geometry-Based Estimation of Number of Endmembers Using p-Norm-Based Pure Pixel Identification Algorithm [J].
Ambikapathi, ArulMurugan ;
Chan, Tsung-Han ;
Chi, Chong-Yung ;
Keizer, Kannan .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2013, 51 (05) :2753-2769
[7]  
Andersen R., 2008, SAGE UNI PAPER SERIE, P152
[8]  
[Anonymous], 1993, SUMM 4 ANN JPL AIRB
[9]  
[Anonymous], 1995, 5 ANN JPL AIRB EARTH
[10]   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