Integration of Physics-Based and Data-Driven Models for Hyperspectral Image Unmixing: A summary of current methods

Spectral unmixing is one of the most important quantitative analysis tasks in hyperspectral data processing. Conventional physics-based models are characterized by clear interpretation. However they may not be suitable for analyzing scenes with unknown complex physical characteristics. Data-driven methods have developed rapidly recent years, particular deep learning because possess superior capability modeling and nonlinear systems. Simply transferring these as black-boxes to conduct lead low interpretability generalization ability. This article reviews works that integrate advantages both data-driven means neural network structures design, prior design loss design. Most derive from a common mathematical optimization framework, combine good high accuracy.