Awesome resources on Hyperspectral Image Unmixing

A list of hyperspectral image unmixing resources collected by Xiuheng Wang ([email protected]) and Min Zhao ([email protected]). For more details, please refer to our paper: Integration of Physics-Based and Data-Driven Models for Hyperspectral Image Unmixing: A summary of current methods. [Paper]. If you find any important resources are not included, please feel free to contact us.

@article{chen2022integration,
  title={Integration of Physics-Based and Data-Driven Models for Hyperspectral Image Unmixing: A summary of current methods}, 
  author={Chen, Jie and Zhao, Min and Wang, Xiuheng and Richard, Cédric and Rahardja, Susanto},
  journal={IEEE Signal Processing Magazine},
  volume={40},
  number={2},
  pages={61--74},
  year={2023},
  publisher={IEEE}
}

Contents

• Physics-based mixture model
  • Survey
  • Examples
• Classic unmixing methods
• Integrating of physics-based models in DNN design
• Prior information learning with data-driven approaches
• Integrating loss learnt from data into physics-based inverse problems
• Databases
• Toolbox

Physics-based mixture model

Survey

1. Advances in hyperspectral image and signal processing: A comprehensive overview of the state of the art. GRSM 2017, P. Ghamisi et al. [Paper]
2. Nonlinear unmixing of hyperspectral images: Models and algorithms. SPM 2013, N. Dobigeon et al. [Paper]
3. A review of nonlinear hyperspectral unmixing methods. JSTARS 2014, R. Heylen et al. [Paper]
4. Spectral variability in hyperspectral data unmixing: A comprehensive review. GRSM 2021, R. Borsoi et al. [Paper] [Code] 🔥

Examples

1. Nonlinear unmixing of hyperspectral images using a generalized bilinear model. TGRS 2011, A. Halimi et al. [Paper]
2. Theory of reflectance and emittance spectroscopy. Cambridge university press 2012, B. Hapke. [Paper]

Classic unmixing methods

1. Fully Constrained Least Squares Linear Spectral Mixture Analysis Method for Material Quantification in Hyperspectral Imagery. TGRS 2001, D. C. Heinz et al. [Paper]
2. Endmember extraction from highly mixed data using minimum volume constrained nonnegative matrix factorization. TGRS 2007, L Miao et al. [Paper]
3. Low-rank and sparse representation for hyperspectral image processing: A review. GRSM 2021, J Peng et al. [Paper]
4. Nonlinear estimation of material abundances in hyperspectral images with ℓ1-norm spatial regularization. TGRS 2013, J. Chen et al. [Paper] [Code] 🔥
5. Kernel-based nonlinear spectral unmixing with dictionary pruning. RS 2019, Z. Li et al. [Paper] 🔥
6. Non-linear spectral unmixing by geodesic simplex volume maximization. JSTSP 2010, R. Heylen et al. [Paper]
7. Fully constrained least squares spectral unmixing by simplex projection. TGRS 2011, R. Heylen et al. [Paper] [Code]
8. A distance geometric framework for nonlinear hyperspectral unmixing. JSTARS 2014, R. Heylen et al. [Paper] [Code]

Integrating of physics-based models in DNN design.

1. EndNet: Sparse autoencoder network for endmember extraction and hyperspectral unmixing. TGRS 2018, S. Ozkan et al. [Paper] [Code]
2. uDAS: An untied denoising autoencoder with sparsity for spectral unmixing. TGRS 2018, Y. Qu et al. [Paper] [Code]
3. DAEN: Deep autoencoder networks for hyperspectral unmixing. TGRS 2019, Y. Su et al. [Paper]
4. Hyperspectral unmixing using a neural network autoencoder. IEEE Access 2018, B. Palsson et al. [Paper] [Code]
5. Convolutional autoencoder for spectral–spatial hyperspectral unmixing. TGRS 2020, B. Palsson et al. [Paper] [Code]
6. Hyperspectral unmixing using deep convolutional autoencoders in a supervised scenario. JSTARS 2020, F. Khajehrayeni et al. [Paper]
7. Nonlinear unmixing of hyperspectral data via deep autoencoder network. GRSL 2019, M. Wang et al. [Paper] [Code] 🔥
8. LSTM-DNN based autoencoder network for nonlinear hyperspectral image unmixing. JSTSP 2021, M. Zhao et al. [Paper] 🔥
9. Hyperspectral unmixing for additive nonlinear models with a 3-D-CNN autoencoder network. TGRS 2022, M. Zhao et al. [Paper] [Code] 🔥
10. Deep autoencoders with multitask learning for bilinear hyperspectral unmixing. TGRS 2020, Y. Su et al. [Paper]
11. Unsupervised hyperspectral unmixing via nonlinear autoencoders. TGRS 2021, K. T. Shahid et al. [Paper] [Code]
12. TANet: An unsupervised two-stream autoencoder network for hyperspectral unmixing. TGRS 2021, Q. Jin et al. [Paper] [Code]
13. Probabilistic generative model for hyperspectral unmixing accounting for endmember variability. TGRS 2022, S. Shi et al. [Paper] [Code] 🔥
14. Deep generative endmember modeling: An application to unsupervised spectral unmixing. TCI 2019, R. Borsoi et al. [Paper] [Code]
15. Deep half-siamese networks for hyperspectral unmixing. GRSL 2020, Z. Han et al. [Paper]
16. Dual branch autoencoder network for spectral-spatial hyperspectral unmixing. GRSL 2021, Z. Hua et al. [Paper]
17. Cycu-net: Cycle-consistency unmixing network by learning cascaded autoencoders. TGRS 2021, L. Gao et al. [Paper] [Code]
18. UnDIP: Hyperspectral unmixing using deep image prior. TGRS 2021, B. Rasti et al. [Paper] [Code]
19. SSCU-Net: Spatial–spectral collaborative unmixing network for hyperspectral images. TGRS 2022, L. Qi et al. [Paper]
20. Multimodal hyperspectral unmixing: Insights from attention networks. TGRS 2022, Z. Han et al. [Paper] [Code]

Prior information learning with data-driven approaches

1. A plug-and-play priors framework for hyperspectral unmixing. TGRS 2021, M. Zhao et al. [Paper] [Code] 🔥
2. Hyperspectral unmixing via nonnegative matrix factorization with handcrafted and learned priors. GRSL 2021, M. Zhao et al. [Paper] 🔥
3. Hyperspectral nonlinear unmixing by using plug-and-play prior for abundance maps RS 2020, Z. Wang et al. [Paper]
4. An ADMM based network for hyperspectral unmixing tasks. ICASSP 2021, C. Zhou et al. [Paper]
5. SNMF-Net: Learning a deep alternating neural network for hyperspectral unmixing. TGRS 2021, F. Xiong et al. [Paper] [Code]

Integrating loss learnt from data into physics-based inverse problems

1. JMnet: Joint metric neural network for hyperspectral unmixing. TGRS 2021, A. Min et al. [Paper]
2. Adversarial autoencoder network for hyperspectral unmixing. TNNLS 2021, Q. Jin et al. [Paper] [Code]

Databases

1. A laboratory-created dataset with ground truth for hyperspectral unmixing evaluation. JSTARS 2019, M. Zhao et al. [Paper] [Data] 🔥

Toolbox

1. Blind hyperspectral unmixing using autoencoders: A critical comparison. JSTARS 2022, B. Palsson et al. [Paper] [Code]