A Convex Optimization-Based Coupled Nonnegative Matrix Factorization Algorithm for Hyperspectral and Multispectral Data Fusion

被引:92
|
作者
Lin, Chia-Hsiang [1 ]
Ma, Fei [2 ]
Chi, Chong-Yung [1 ]
Hsieh, Chih-Hsiang [1 ]
机构
[1] Natl Tsing Hua Univ, Inst Commun Engn, Hsinchu 30013, Taiwan
[2] Liaoning Tech Univ, Sch Elect & Informat Engn, Huludao 125105, Peoples R China
来源
关键词
Alternating direction method of multipliers (ADMM); convex optimization; coupled nonnegative matrix factorization (CNMF); data fusion; hyperspectral data; IMAGES; RESOLUTION; IDENTIFIABILITY; REGULARIZATION; MINIMIZATION; FORMULATION; CHALLENGES; REGRESSION; CRITERION; SPARSITY;
D O I
10.1109/TGRS.2017.2766080
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
Fusing a low-spatial-resolution hyperspectral data with a high-spatial-resolution (HSR) multispectral data has been recognized as an economical approach for obtaining HSR hyperspectral data, which is important to accurate identification and classification of the underlying materials. A natural and promising fusion criterion, called coupled nonnegative matrix factorization (CNMF), has been reported that can yield highquality fused data. However, the CNMF criterion amounts to an ill-posed inverse problem, and hence, advisable regularization can be considered for further upgrading its fusion performance. Besides the commonly used sparsity-promoting regularization, we also incorporate the well-known sum-of-squared-distances regularizer, which serves as a convex surrogate of the volume of the simplex of materials' spectral signature vectors (i.e., endmembers), into the CNMF criterion, thereby leading to a convex formulation of the fusion problem. Then, thanks to the biconvexity of the problem nature, we decouple it into two convex subproblems, which are then, respectively, solved by two carefully designed alternating direction method of multipliers (ADMM) algorithms. Closed-form expressions for all the ADMM iterates are derived via convex optimization theories (e.g., Karush-Kuhn-Tucker conditions), and furthermore, some matrix structures are employed to obtain alternative expressions with much lower computational complexities, thus suitable for practical applications. Some experimental results are provided to demonstrate the superior fusion performance of the proposed algorithm over state-of-the-art methods.
引用
收藏
页码:1652 / 1667
页数:16
相关论文
共 50 条
  • [1] Coupled Nonnegative Matrix Factorization Unmixing for Hyperspectral and Multispectral Data Fusion
    Yokoya, Naoto
    Yairi, Takehisa
    Iwasaki, Akira
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2012, 50 (02): : 528 - 537
  • [2] Hyperspectral and multispectral data fusion based on linear-quadratic nonnegative matrix factorization
    Benhalouche, Fatima Zohra
    Karoui, Moussa Sofiane
    Deville, Yannick
    Ouamri, Abdelaziz
    [J]. JOURNAL OF APPLIED REMOTE SENSING, 2017, 11
  • [3] JOINT NONNEGATIVE MATRIX FACTORIZATION FOR HYPERSPECTRAL AND MULTISPECTRAL REMOTE SENSING DATA FUSION
    Karoui, Moussa Sofiane
    Deville, Yannick
    Kreri, Sarah
    [J]. 2013 5TH WORKSHOP ON HYPERSPECTRAL IMAGE AND SIGNAL PROCESSING: EVOLUTION IN REMOTE SENSING (WHISPERS), 2013,
  • [4] Alternating Direction Iterative Nonnegative Matrix Factorization Unmixing for Multispectral and Hyperspectral Data Fusion
    Zhou, Xinyu
    Zhang, Ye
    Zhang, Junping
    Shi, Shaoqi
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2020, 13 : 5223 - 5232
  • [5] Conic optimization-based algorithms for nonnegative matrix factorization
    Leplat, Valentin
    Nesterov, Yurii
    Gillis, Nicolas
    Glineur, Francois
    [J]. OPTIMIZATION METHODS & SOFTWARE, 2023, 38 (04): : 837 - 859
  • [6] HYPERSPECTRAL UNMIXING ALGORITHM BASED ON NONNEGATIVE MATRIX FACTORIZATION
    Bao, Wenxing
    Li, Qin
    Xin, Liping
    Qu, Kewen
    [J]. 2016 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS), 2016, : 6982 - 6985
  • [7] HYPERSPECTRAL AND MULTISPECTRAL IMAGE FUSION VIA NONNEGATIVE MATRIX FACTORIZATION AND DEEP PRIOR REGULARIZATION
    Lin, Baihong
    Zhang, Yiqi
    Lin, Zengrong
    Wang, Xiaoqing
    Huang, Haifeng
    [J]. 2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022), 2022, : 1440 - 1443
  • [8] Blind Fusion of Hyperspectral Multispectral Images Based on Matrix Factorization
    Long, Jian
    Peng, Yuanxi
    [J]. REMOTE SENSING, 2021, 13 (21)
  • [9] Nonnegative matrix factorization-based hyperspectral and panchromatic image fusion
    Zhou Zhang
    Zhenwei Shi
    [J]. Neural Computing and Applications, 2013, 23 : 895 - 905
  • [10] Nonnegative matrix factorization-based hyperspectral and panchromatic image fusion
    Zhang, Zhou
    Shi, Zhenwei
    [J]. NEURAL COMPUTING & APPLICATIONS, 2013, 23 (3-4): : 895 - 905