Nonlinear unmixing of hyperspectral images using a generalized bilinear model

Abderrahim Halimi, Yoann Altmann, Nicolas Dobigeon, Jean-Yves Tourneret

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper studies a generalized bilinear model and a hierarchical Bayesian algorithm for unmixing hyperspectral images. The proposed model is a generalization of the accepted linear mixing model but also of a bilinear model recently introduced in the literature. Appropriate priors are chosen for its parameters in particular to satisfy the positivity and sum-to-one constraints for the abundances. The joint posterior distribution of the unknown parameter vector is then derived. A Metropolis-within-Gibbs algorithm is proposed which allows samples distributed according to the posterior of interest to be generated and to estimate the unknown model parameters. The performance of the resulting unmixing strategy is evaluated via simulations conducted on synthetic and real data.

Original languageEnglish
Title of host publication2011 IEEE Statistical Signal Processing Workshop (SSP)
PublisherIEEE
Pages413-416
Number of pages4
ISBN (Electronic)9781457705700
DOIs
Publication statusPublished - 29 Jul 2011
Event2011 IEEE Statistical Signal Processing Workshop - Nice, France
Duration: 28 Jun 201130 Jun 2011

Publication series

NameIEEE Statistical Signal Processing Workshop (SSP)
ISSN (Print)2373-0803

Conference

Conference2011 IEEE Statistical Signal Processing Workshop
Abbreviated titleSSP 2011
CountryFrance
CityNice
Period28/06/1130/06/11

Fingerprint

Bilinear Model
Hyperspectral Image
Posterior distribution
Joint Distribution
Positivity
Unknown Parameters
Model
Unknown
Estimate
Simulation

Keywords

  • Bayesian algorithm
  • bilinear model
  • Gibbs sampler
  • Hyperspectral imagery
  • MCMC methods
  • spectral unmixing

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Applied Mathematics
  • Signal Processing
  • Computer Science Applications

Cite this

Halimi, A., Altmann, Y., Dobigeon, N., & Tourneret, J-Y. (2011). Nonlinear unmixing of hyperspectral images using a generalized bilinear model. In 2011 IEEE Statistical Signal Processing Workshop (SSP) (pp. 413-416). (IEEE Statistical Signal Processing Workshop (SSP)). IEEE. https://doi.org/10.1109/SSP.2011.5967718
Halimi, Abderrahim ; Altmann, Yoann ; Dobigeon, Nicolas ; Tourneret, Jean-Yves. / Nonlinear unmixing of hyperspectral images using a generalized bilinear model. 2011 IEEE Statistical Signal Processing Workshop (SSP). IEEE, 2011. pp. 413-416 (IEEE Statistical Signal Processing Workshop (SSP)).
@inproceedings{5289141761134cbb9937d7a5846004db,
title = "Nonlinear unmixing of hyperspectral images using a generalized bilinear model",
abstract = "This paper studies a generalized bilinear model and a hierarchical Bayesian algorithm for unmixing hyperspectral images. The proposed model is a generalization of the accepted linear mixing model but also of a bilinear model recently introduced in the literature. Appropriate priors are chosen for its parameters in particular to satisfy the positivity and sum-to-one constraints for the abundances. The joint posterior distribution of the unknown parameter vector is then derived. A Metropolis-within-Gibbs algorithm is proposed which allows samples distributed according to the posterior of interest to be generated and to estimate the unknown model parameters. The performance of the resulting unmixing strategy is evaluated via simulations conducted on synthetic and real data.",
keywords = "Bayesian algorithm, bilinear model, Gibbs sampler, Hyperspectral imagery, MCMC methods, spectral unmixing",
author = "Abderrahim Halimi and Yoann Altmann and Nicolas Dobigeon and Jean-Yves Tourneret",
year = "2011",
month = "7",
day = "29",
doi = "10.1109/SSP.2011.5967718",
language = "English",
series = "IEEE Statistical Signal Processing Workshop (SSP)",
publisher = "IEEE",
pages = "413--416",
booktitle = "2011 IEEE Statistical Signal Processing Workshop (SSP)",
address = "United States",

}

Halimi, A, Altmann, Y, Dobigeon, N & Tourneret, J-Y 2011, Nonlinear unmixing of hyperspectral images using a generalized bilinear model. in 2011 IEEE Statistical Signal Processing Workshop (SSP). IEEE Statistical Signal Processing Workshop (SSP), IEEE, pp. 413-416, 2011 IEEE Statistical Signal Processing Workshop, Nice, France, 28/06/11. https://doi.org/10.1109/SSP.2011.5967718

Nonlinear unmixing of hyperspectral images using a generalized bilinear model. / Halimi, Abderrahim; Altmann, Yoann; Dobigeon, Nicolas; Tourneret, Jean-Yves.

2011 IEEE Statistical Signal Processing Workshop (SSP). IEEE, 2011. p. 413-416 (IEEE Statistical Signal Processing Workshop (SSP)).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Nonlinear unmixing of hyperspectral images using a generalized bilinear model

AU - Halimi, Abderrahim

AU - Altmann, Yoann

AU - Dobigeon, Nicolas

AU - Tourneret, Jean-Yves

PY - 2011/7/29

Y1 - 2011/7/29

N2 - This paper studies a generalized bilinear model and a hierarchical Bayesian algorithm for unmixing hyperspectral images. The proposed model is a generalization of the accepted linear mixing model but also of a bilinear model recently introduced in the literature. Appropriate priors are chosen for its parameters in particular to satisfy the positivity and sum-to-one constraints for the abundances. The joint posterior distribution of the unknown parameter vector is then derived. A Metropolis-within-Gibbs algorithm is proposed which allows samples distributed according to the posterior of interest to be generated and to estimate the unknown model parameters. The performance of the resulting unmixing strategy is evaluated via simulations conducted on synthetic and real data.

AB - This paper studies a generalized bilinear model and a hierarchical Bayesian algorithm for unmixing hyperspectral images. The proposed model is a generalization of the accepted linear mixing model but also of a bilinear model recently introduced in the literature. Appropriate priors are chosen for its parameters in particular to satisfy the positivity and sum-to-one constraints for the abundances. The joint posterior distribution of the unknown parameter vector is then derived. A Metropolis-within-Gibbs algorithm is proposed which allows samples distributed according to the posterior of interest to be generated and to estimate the unknown model parameters. The performance of the resulting unmixing strategy is evaluated via simulations conducted on synthetic and real data.

KW - Bayesian algorithm

KW - bilinear model

KW - Gibbs sampler

KW - Hyperspectral imagery

KW - MCMC methods

KW - spectral unmixing

UR - http://www.scopus.com/inward/record.url?scp=80052205709&partnerID=8YFLogxK

U2 - 10.1109/SSP.2011.5967718

DO - 10.1109/SSP.2011.5967718

M3 - Conference contribution

T3 - IEEE Statistical Signal Processing Workshop (SSP)

SP - 413

EP - 416

BT - 2011 IEEE Statistical Signal Processing Workshop (SSP)

PB - IEEE

ER -

Halimi A, Altmann Y, Dobigeon N, Tourneret J-Y. Nonlinear unmixing of hyperspectral images using a generalized bilinear model. In 2011 IEEE Statistical Signal Processing Workshop (SSP). IEEE. 2011. p. 413-416. (IEEE Statistical Signal Processing Workshop (SSP)). https://doi.org/10.1109/SSP.2011.5967718