Deep learning based unmixing methods have received great attention in recent years and achieve remarkable performance. These methods employ a data-driven approach to extract structure features from hyperspectral image, however, they tend to be less physical interpretable. Conventional unmixing methods are with much more interpretability, whereas they require manually designing regularization and choosing penalty parameters. To overcome these limitations, we propose a novel unmixing method by unrolling the plug-and-play unmixing algorithm to conduct the deep architecture. Our method integrates both inner and outer priors. The carefully designed unfolding deep architecture is used to learn the spectral and spatial information from the hyperspectral image, which we refer to as inner priors. Additionally, our approach incorporates deep denoisers that have been pretrained on a large volume of image data to leverage the outer priors. Secondly, we design a dynamic convolution to model the multiscale information. Different scales are fused using an attention module. Experimental results of both synthetic and real datasets demonstrate that our method outperforms compared methods.
Hyperspectral Unmixing Based on Nonnegative Matrix Factorization: A Comprehensive Review
Southwest Jiaotong University（西南交通大学）；Mississippi State University（密西西比州立大学）；University of New South Wales（新南威尔士大学）；University of Extremadura（埃斯特雷马杜拉大学）
Transformer for Multitemporal Hyperspectral Image Unmixing
College of Computer Science, Nankai University（南开大学计算机学院）；Haihe Lab（海河实验室）；School of Computer Science and Technology, Tiangong University（天津工业大学计算机科学与技术学院）；School of Astronautics, Beihang University（北京航空航天大学宇航学院）
Deep Deterministic Independent Component Analysis for Hyperspectral Unmixing
University of Florida（佛罗里达大学）；UiT - The Arctic University of Norway（UiT-挪威北极大学）
我们开发了一种新的基于神经网络的独立分量分析（ICA）方法，通过直接最小化所有提取的组件之间的依赖。使用基于矩阵的R{é}nyi的$α$ -阶熵泛函，我们的网络可以直接通过随机梯度下降（SGD）进行优化，而无需任何变分近似或对抗训练。作为一个坚实的应用程序，我们评估我们的ICA在高光谱解混（HU）的问题，并驳斥了一个说法，即“\emph{ICA does not play a role in unmixing hyperspectral data}“，这是最初提出的\cite{nascimento2005does}。我们的DDICA代码和附加说明可在https://github.com/hongmingli1995/DDICA上找到。
Endmember-Guided Unmixing Network (EGU-Net): A General Deep Learning Framework for Self-Supervised Hyperspectral Unmixing
German Aerospace Center (DLR)（德国航空航天中心）；Univ. Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；Aerospace Information Research Institute, Chinese Academy of Sciences（中国科学院空天信息创新研究院）；The University of Tokyo（东京大学）；RIKEN Center for Advanced Intelligence Project (AIP)（理化学研究所先进智能项目中心）
Department of Engineering Cybernetics, Norwegian University of Science and Technology（工程 cybernetics 部，挪威科学与技术大学）；Department of Biology, Norwegian University of Science and Technology（生物学部，挪威科学与技术大学）；Department of Automatic Control and Systems Engineering, University Politehnica Bucharest（自动控制与系统工程部，布加勒斯特理工大学）
Deep Hyperspectral Unmixing using Transformer Network
Jalpaiguri Engineering College（贾尔派古里工程学院）；University of Antwerp（安特卫普大学）；Helmholtz-Zentrum Dresden-Rossendorf（亥姆霍兹德累斯顿罗森多夫研究中心）；Helmholtz Institute Freiberg for Resource Technology（亥姆霍兹弗赖贝格资源技术研究所）
目前，这篇论文正在IEEE进行评审。Transformers以其在自然语言处理方面的最先进性能引起了视觉研究界的兴趣。Transformers凭借其优越的性能，在高光谱图像分类领域找到了自己的道路，并取得了可喜的成果。在这篇文章中，我们利用Transformers的力量来克服高光谱解混的任务，并提出了一种新的深度解混模型与Transformers。我们的目标是利用Transformers的能力，以更好地捕捉全球功能的依赖关系，以提高端元光谱和丰度图的质量。所提出的模型是卷积自动编码器和Transformer的组合。高光谱数据由卷积编码器编码。Transformer捕获从编码器导出的表示之间的长范围依赖性。使用卷积解码器重建数据。我们将所提出的解混模型应用于三个广泛使用的解混数据集，即，Samson、Apex和Washington DC mall的数据，并在均方根误差和光谱角距离方面将其与最新技术进行了比较。拟议模型的源代码将在\url{https://github.com/preetam22n/DeepTrans-HSU}上公开提供。
Hyperspectral Image Generation with Unmixing Guided Diffusion Model
School of Statistics and Data Science, KLMDASR, LEBPS, and LPMC, Nankai University（统计与数据科学学院，KLMDASR，LEBPS和LPMC，南开大学）；Image Processing Center, School of Astronautics, Beihang University（图像处理中心，航天学院，北航大学）
A Dual Symmetric Gauss-Seidel Alternating Direction Method of Multipliers for Hyperspectral Sparse Unmixing
School of Information Science and technology, and the Provincial Key Lab of Information Coding and Trans- mission, Southwest Jiaotong University（信息科学与技术学院，信息编码与传输省重点实验室，西南交通大学）；School of Mathematics, Southwest Jiaotong University（数学学院，西南交通大学）；School of Computer and Computing Science, Zhejiang University City College（计算机与计算科学学院，浙江大学城市学院）
Improving Autoencoder Training Performance for Hyperspectral Unmixing with Network Reinitialisation
Institute of Theoretical and Applied Informatics, Polish Academy of Sciences（波兰科学院理论与应用信息学研究所）；Silesian University of Technology（西里西亚工业大学）；Sapientia Hungarian University of Transylvania（特兰西瓦尼亚萨皮恩扎匈牙利大学）
Hyperspectral Unmixing via Nonnegative Matrix Factorization with Handcrafted and Learnt Priors
School of Marine Science and Technology, Northwestern Polytechnical University, China（西北工业大学海洋科学与技术学院）；State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, China（北京交通大学轨道交通控制与安全国家重点实验室）
Extending the Unmixing methods to Multispectral Images
Universite Bourgogne Franche-Comté（勃艮第-弗朗什-孔泰大学）；Le Centre de Recherche et de Restauration des Musées de France（法国博物馆研究与修复中心）；Centre National de la Recherche Scientifique（法国国家科学研究中心）
Using Low-rank Representation of Abundance Maps and Nonnegative Tensor Factorization for Hyperspectral Nonlinear Unmixing
Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences（中国科学院空天信息创新研究院数字地球科学重点实验室）；University of Chinese Academy of Sciences（中国科学院大学）；Hong Kong Baptist University（香港浸会大学）；Dalian Maritime University（大连海事大学）；Université Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；INRIA（法国国家信息与自动化研究所）；CNRS（法国国家科学研究中心）；Grenoble INP（格勒诺布尔理工学院）；LJK（让·昆兹曼实验室）
Preprocessing Algorithm Leveraging Geometric Modeling for Scale Correction in Hyperspectral Images for Improved Unmixing Performance
Department of Electrical and Electronic Engineering, University of Peradeniya（珀斯德尼亚大学电子与电气工程系）；School of Electrical and Computer Engineering, Purdue University（普渡大学电子与计算机工程学院）
