Kalman Filtering and Expectation Maximization for Multitemporal Spectral Unmixing

TL;DR

This paper introduces a novel multitemporal spectral unmixing method that combines Kalman filtering and EM algorithm to effectively model and estimate spectral variability and abundances in hyperspectral image sequences.

Contribution

It proposes a physically motivated parametric model for spectral variability and employs a Bayesian filtering approach with EM for efficient parameter estimation in multitemporal data.

Findings

Outperforms existing multitemporal spectral unmixing algorithms in simulations.

Effectively models spectral variability using a state-space formulation.

Provides accurate estimates of endmember variability and abundances.

Abstract

The recent evolution of hyperspectral imaging technology and the proliferation of new emerging applications presses for the processing of multiple temporal hyperspectral images. In this work, we propose a novel spectral unmixing (SU) strategy using physically motivated parametric endmember representations to account for temporal spectral variability. By representing the multitemporal mixing process using a state-space formulation, we are able to exploit the Bayesian filtering machinery to estimate the endmember variability coefficients. Moreover, by assuming that the temporal variability of the abundances is small over short intervals, an efficient implementation of the expectation maximization (EM) algorithm is employed to estimate the abundances and the other model parameters. Simulation results indicate that the proposed strategy outperforms state-of-the-art multitemporal SU algorithms.