Bayesian segmentation of hyperspectral images

TL;DR

This paper presents a Bayesian method for segmenting hyperspectral images using Hidden Markov Models and Potts Markov Random Fields, implemented with an MCMC algorithm, demonstrated through simulations.

Contribution

It introduces a novel Bayesian segmentation approach for hyperspectral images combining HMM and Potts MRF with an MCMC implementation.

Findings

Effective segmentation demonstrated in simulations

Bayesian framework effectively models spectral and spatial data

MCMC algorithm successfully infers hidden labels

Abstract

In this paper we consider the problem of joint segmentation of hyperspectral images in the Bayesian framework. The proposed approach is based on a Hidden Markov Modeling (HMM) of the images with common segmentation, or equivalently with common hidden classification label variables which is modeled by a Potts Markov Random Field. We introduce an appropriate Markov Chain Monte Carlo (MCMC) algorithm to implement the method and show some simulation results.