Hyperspectral Unmixing via Turbo Bilinear Approximate Message Passing

TL;DR

This paper introduces a turbo belief propagation method for hyperspectral unmixing that leverages spectral and spatial coherence, utilizing BiG-AMP within a factor graph framework, and demonstrates improved performance on synthetic and real data.

Contribution

It presents a novel turbo BP approach combining BiG-AMP for bilinear factorization and EM for parameter tuning in hyperspectral unmixing, enhancing accuracy over existing methods.

Findings

Favorable unmixing performance on synthetic data

Effective exploitation of spectral and spatial coherence

Robustness demonstrated on real-world datasets

Abstract

The goal of hyperspectral unmixing is to decompose an electromagnetic spectral dataset measured over M spectral bands and T pixels into N constituent material spectra (or "end-members") with corresponding spatial abundances. In this paper, we propose a novel approach to hyperspectral unmixing based on loopy belief propagation (BP) that enables the exploitation of spectral coherence in the endmembers and spatial coherence in the abundances. In particular, we partition the factor graph into spectral coherence, spatial coherence, and bilinear subgraphs, and pass messages between them using a "turbo" approach. To perform message passing within the bilinear subgraph, we employ the bilinear generalized approximate message passing algorithm (BiG-AMP), a recently proposed belief-propagation-based approach to matrix factorization. Furthermore, we propose an expectation-maximization (EM) strategy to tune the prior parameters and a model-order selection strategy to select the number of materials N. Numerical experiments conducted with both synthetic and real-world data show favorable unmixing performance relative to existing methods.