A Plug-and-Play Priors Framework for Hyperspectral Unmixing

Min Zhao,Xiuheng Wang,Jie Chen,Wei Chen

DOI: https://doi.org/10.1109/tgrs.2020.3047479

IF: 8.2

2022-01-01

IEEE Transactions on Geoscience and Remote Sensing

Abstract:Spectral unmixing is a widely used technique in hyperspectral image processing and analysis. It aims to separate mixed pixels into the component materials and their corresponding abundances. Early solutions to spectral unmixing are performed independently on each pixel. Nowadays, investigating proper priors into the unmixing problem has been popular as it can significantly enhance the unmixing performance. However, it is nontrivial to handcraft a powerful regularizer, and complex regularizers may introduce extra difficulties in solving optimization problems in which they are involved. To address this issue, we present a plug-and-play (PnP) priors framework for hyperspectral unmixing. More specifically, we use the alternating direction method of multipliers (ADMM) to decompose the optimization problem into two iterative subproblems. One is a regular optimization problem depending on the forward model, and the other is a proximity operator related to the prior model and can be regarded as an image denoising problem. Our framework is flexible and extendable which allows a wide range of denoisers to replace prior models and avoids handcrafting regularizers. Experiments conducted on both synthetic data and real airborne data illustrate the superiority of the proposed strategy compared with other state-of-the-art hyperspectral unmixing methods.

imaging science & photographic technology,remote sensing,engineering, electrical & electronic,geochemistry & geophysics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to effectively use prior information to improve the performance of spectral unmixing in hyperspectral image processing. Traditional spectral unmixing methods usually process each pixel independently, ignoring the spatial and spectral correlations existing in hyperspectral images. Although in recent years, by introducing appropriate priors into the unmixing problem, the unmixing effect can be significantly enhanced, it is not easy to manually design a powerful regularizer, and a complex regularizer may increase the difficulty of solving the optimization problem.

For this reason, the paper proposes a plug - and - play priors framework based on the alternating direction multiplier method (ADMM) for hyperspectral unmixing. This framework solves the above problems by decomposing the optimization problem into two iterative sub - problems: one is a conventional optimization problem depending on the forward model, and the other is a proximity operator problem related to the prior model, which can be regarded as an image denoising problem. This framework is flexible and extensible, allowing a wide range of denoisers to be used to replace the prior model, avoiding the difficulty of manually designing regularizers.

The main contributions of the paper include:
1. Proposing a robust hyperspectral unmixing method based on the ADMM algorithm. This method is achieved by decomposing the associated optimization problem into two sub - problems, one of which is a proximity operator related to the prior model and can be replaced by a denoising operator.
2. Integrating a variety of denoisers (linear or nonlinear, band - like or 3D, traditional or deep - learning - based) into the proposed unmixing framework to utilize various prior information of the image and bypass the problem of designing regularizers.
3. Designing two strategies to combine denoisers: one is to extract spatial and spectral prior information from the reconstructed hyperspectral image; the other is to directly capture spectral and spatial correlations from the fractional abundance map, thereby reducing the computational burden.

Experiments on synthetic data and actual aerial data prove the superiority of the proposed strategy compared to other advanced hyperspectral unmixing methods.