ADMM-Based Hyperspectral Unmixing Networks for Abundance and Endmember Estimation
- 16 December 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60 (01962892) , 1-18
Abstract
Hyperspectral image (HSI) unmixing is an increasingly studied problem in various areas, including remote sensing. It has been tackled using both physical model-based approaches and more recently machine learning-based ones. In this article, we propose a new HSI unmixing algorithm combining both model- and learning-based techniques, based on algorithm unrolling approaches, delivering improved unmixing performance. Our approach unrolls the alternating direction method of multipliers (ADMMs) solver of a constrained sparse regression problem underlying a linear mixture model. We then propose a neural network structure for abundance estimation that can be trained using supervised learning techniques based on a new composite loss function. We also propose another neural network structure for blind unmixing that can be trained using unsupervised learning techniques. Our proposed networks are also shown to possess a lighter and richer structure containing less learnable parameters and more skip connections compared with other competing architectures. Extensive experiments show that the proposed methods can achieve much faster convergence and better performance even with a very small training dataset size when compared with other unmixing methods, such as model-inspired neural network for abundance estimation (MNN-AE), model-inspired neural network for blind unmixing (MNN-BU), unmixing using deep image prior (UnDIP), and endmember-guided unmixing network (EGU-Net).Keywords
Funding Information
- University College London (UCL) Project, Art Through the ICT Lens
- Engineering and Physical Sciences Research Council (EP/R032785/1)
- China Scholarship Council
This publication has 44 references indexed in Scilit:
- Convergence of an inertial proximal method for l1-regularized least-squares2015
- Spectral Unmixing via Data-Guided SparsityIEEE Transactions on Image Processing, 2014
- Spectral Unmixing of Hyperspectral Imagery Using Multilayer NMFIEEE Geoscience and Remote Sensing Letters, 2014
- Structured Sparse Method for Hyperspectral UnmixingISPRS Journal of Photogrammetry and Remote Sensing, 2014
- Remote Sensing Image ProcessingSynthesis Lectures on Image, Video, and Multimedia Processing, 2011
- Hyperspectral Unmixing via $L_{1/2}$ Sparsity-Constrained Nonnegative Matrix FactorizationIEEE Transactions on Geoscience and Remote Sensing, 2011
- Signal processing for hyperspectral image exploitationIEEE Signal Processing Magazine, 2002
- Spectral unmixingIEEE Signal Processing Magazine, 2002
- Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imageryIEEE Transactions on Geoscience and Remote Sensing, 2001
- Search CD450 database: USGS library: (one of the three databases on the Search CD450 Earth Sciences CD-ROM disc). Dublin, OH: OCLC Online Computer Library Center, Inc. CD-ROM software. 1975- present. Updated quarterly. 130,000 records as of June 1989. $350/year (non-members); $300/year (members)Government Publications Review, 1990