OATAO - Open Archive Toulouse Archive Ouverte Open Access Week

A hierarchical Bayesian model accounting for endmember variability and abrupt spectral changes to unmix multitemporal hyperspectral images

Thouvenin, Pierre-Antoine and Dobigeon, Nicolas and Tourneret, Jean-Yves A hierarchical Bayesian model accounting for endmember variability and abrupt spectral changes to unmix multitemporal hyperspectral images. (2018) IEEE Transactions on Computational Imaging, 4 (1). 32-45. ISSN 2333-9403

[img]
Preview
(Document in English)

PDF (Author's version) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
1MB

Official URL: https://doi.org/10.1109/TCI.2017.2777484

Abstract

Hyperspectral unmixing is a blind source separation problem that consists in estimating the reference spectral signa- tures contained in a hyperspectral image, as well as their relative contribution to each pixel according to a given mixture model. In practice, the process is further complexified by the inherent spec- tral variability of the observed scene and the possible presence of outliers. More specifically, multitemporal hyperspectral images, i.e., sequences of hyperspectral images acquired over the same area at different time instants, are likely to simultaneously exhibit mod- erate endmember variability and abrupt spectral changes either due to outliers or to significant time intervals between consecutive acquisitions. Unless properly accounted for, these two perturba- tions can significantly affect the unmixing process. In this context, we propose a new unmixing model for multitemporal hyperspec- tral images accounting for smooth temporal variations, construed as spectral variability, and abrupt spectral changes interpreted as outliers. The proposed hierarchical Bayesian model is inferred us- ing a Markov chain Monte Carlo method allowing the posterior of interest to be sampled and Bayesian estimators to be approxi- mated. A comparison with unmixing techniques from the literature on synthetic and real data allows the interest of the proposed ap- proach to be appreciated.

Item Type:Article
HAL Id:hal-02002120
Audience (journal):International peer-reviewed journal
Uncontrolled Keywords:
Institution:French research institutions > Centre National de la Recherche Scientifique - CNRS (FRANCE)
Université de Toulouse > Institut National Polytechnique de Toulouse - Toulouse INP (FRANCE)
Université de Toulouse > Université Toulouse III - Paul Sabatier - UT3 (FRANCE)
Université de Toulouse > Université Toulouse - Jean Jaurès - UT2J (FRANCE)
Université de Toulouse > Université Toulouse 1 Capitole - UT1 (FRANCE)
Laboratory name:
Funders:
Hypanema - MapInvPlnt - CIMI thematic trimester on Image Processing - Direction générale de l'Armement (DGA)
Statistics:download
Deposited On:20 Nov 2018 16:36

Repository Staff Only: item control page