Hyperspectral Unmixing via Nonnegative Matrix Factorization with Handcrafted and Learnt Priors

TL;DR

This paper introduces a flexible hyperspectral unmixing framework combining handcrafted and learned priors within NMF, utilizing image denoisers and sparsity regularization, validated on synthetic and real data.

Contribution

It proposes a novel NMF-based unmixing method that jointly incorporates handcrafted and learned priors, enhancing spectral unmixing performance.

Findings

Effective on synthetic data

Validated on real airborne data

Improves unmixing accuracy

Abstract

Nowadays, nonnegative matrix factorization (NMF) based methods have been widely applied to blind spectral unmixing. Introducing proper regularizers to NMF is crucial for mathematically constraining the solutions and physically exploiting spectral and spatial properties of images. Generally, properly handcrafting regularizers and solving the associated complex optimization problem are non-trivial tasks. In our work, we propose an NMF based unmixing framework which jointly uses a handcrafting regularizer and a learnt regularizer from data. we plug learnt priors of abundances where the associated subproblem can be addressed using various image denoisers, and we consider an l_2,1-norm regularizer to the abundance matrix to promote sparse unmixing results. The proposed framework is flexible and extendable. Both synthetic data and real airborne data are conducted to confirm the effectiveness of our method.