Implementation strategies for hyperspectral unmixing using Bayesian source separation

TL;DR

This paper introduces an efficient implementation strategy for Bayesian Positive Source Separation in hyperspectral unmixing, enabling its application to full images by optimizing pixel sampling and computational resources.

Contribution

It presents a novel implementation approach that reduces computation time and memory usage for BPSS algorithms in hyperspectral image analysis.

Findings

Effective pixel selection improves unmixing accuracy.

Sampling strategies impact the quality of estimated spectra and abundance maps.

The method is validated on synthetic and real Mars hyperspectral data.

Abstract

Bayesian Positive Source Separation (BPSS) is a useful unsupervised approach for hyperspectral data unmixing, where numerical non-negativity of spectra and abundances has to be ensured, such in remote sensing. Moreover, it is sensible to impose a sum-to-one (full additivity) constraint to the estimated source abundances in each pixel. Even though non-negativity and full additivity are two necessary properties to get physically interpretable results, the use of BPSS algorithms has been so far limited by high computation time and large memory requirements due to the Markov chain Monte Carlo calculations. An implementation strategy which allows one to apply these algorithms on a full hyperspectral image, as typical in Earth and Planetary Science, is introduced. Effects of pixel selection, the impact of such sampling on the relevance of the estimated component spectra and abundance maps, as well as on the computation times, are discussed. For that purpose, two different dataset have been used: a synthetic one and a real hyperspectral image from Mars.