Dynamical spectral unmixing of multitemporal hyperspectral images

TL;DR

This paper introduces a dynamic spectral unmixing method for multitemporal hyperspectral images, modeling spectral signatures and abundances as evolving latent variables, and demonstrates its effectiveness on synthetic and real datasets.

Contribution

It proposes a novel dynamical model for unmixing multitemporal hyperspectral images and an efficient algorithm based on alternating minimization.

Findings

Effective unmixing on synthetic data

Successful application to real hyperspectral images

Improved estimation of latent variables

Abstract

In this paper, we consider the problem of unmixing a time series of hyperspectral images. We propose a dynamical model based on linear mixing processes at each time instant. The spectral signatures and fractional abundances of the pure materials in the scene are seen as latent variables, and assumed to follow a general dynamical structure. Based on a simplified version of this model, we derive an efficient spectral unmixing algorithm to estimate the latent variables by performing alternating minimizations. The performance of the proposed approach is demonstrated on synthetic and real multitemporal hyperspectral images.