Hyperspectral Unmixing Hierarchies

TL;DR

This paper introduces Hierarchical Unmixing using Deep Nonnegative Matrix Factorization and BLUTHs, improving spectral variability handling and endmember detection in hyperspectral images.

Contribution

It proposes a hierarchical unmixing framework with a novel network architecture, enhancing endmember detection and abundance estimation.

Findings

BLUTHs outperform state-of-the-art algorithms on laboratory scenes.

BLUTHs are competitive on remote sensing scenes.

Demonstrated ocean color unmixing on satellite hyperspectral data.

Abstract

Unmixing reveals the spatial distribution and spectral details of different constituents, called endmembers, in a hyperspectral image. Because unmixing has limited ground truth requirements, can accommodate mixed pixels, and is closely tied to light propagation, it is a uniquely powerful tool for analyzing hyperspectral images. However, spectral variability inhibits unmixing performance, the proper way to determine the number of endmembers is ambiguous, and the clarity of the endmembers degrades as more are included. Hierarchical structure is a possible solution to all three problems. Here, hierarchical unmixing is defined by imposing a hierarchical abundance sum constraint on Deep Nonnegative Matrix Factorization. Binary Linear Unmixing Tactile Hierarchies (BLUTHs) solve the hierarchical unmixing problem with a simple network architecture. Sparsity modulation unmixing growth tailors the topology of a BLUTH to each scene. The structure imposed by BLUTHs allows endmembers with varying levels of spectral contrast to be revealed, mitigating the challenge of spectral variability. The performance of BLUTHs exceeds state-of-the-art unmixing algorithms on laboratory scenes, particularly with regard to abundance estimation, while their performance remains competitive on remote sensing scenes. In addition, ocean color unmixing by BLUTHs is demonstrated on hyperspectral scenes from the HYPSO and PACE satellites.