Hierarchical Bayesian image analysis: from low-level modeling to robust supervised learning

TL;DR

This paper introduces a hierarchical Bayesian framework that jointly performs low-level latent variable modeling, clustering, and supervised classification, specifically applied to hyperspectral image analysis, unifying unmixing and classification tasks.

Contribution

It presents a novel unified hierarchical Bayesian model that integrates latent variable estimation, clustering, and supervised classification in a single framework.

Findings

Effective latent variable extraction improves classification accuracy.

Unifies unmixing and classification in hyperspectral imaging.

Demonstrates robustness to poorly labeled data.

Abstract

Within a supervised classification framework, labeled data are used to learn classifier parameters. Prior to that, it is generally required to perform dimensionality reduction via feature extraction. These preprocessing steps have motivated numerous research works aiming at recovering latent variables in an unsupervised context. This paper proposes a unified framework to perform classification and low-level modeling jointly. The main objective is to use the estimated latent variables as features for classification and to incorporate simultaneously supervised information to help latent variable extraction. The proposed hierarchical Bayesian model is divided into three stages: a first low-level modeling stage to estimate latent variables, a second stage clustering these features into statistically homogeneous groups and a last classification stage exploiting the (possibly badly) labeled data. Performance of the model is assessed in the specific context of hyperspectral image interpretation, unifying two standard analysis techniques, namely unmixing and classification.