Hyperspectral unmixing accounting for spatial correlations and endmember variability

Abstract : This paper presents an unsupervised Bayesian algorithm for hyperspectral image unmixing accounting for endmember variability. This variability is obtained by assuming that each pixel is a linear combination of random endmembers weighted by their corresponding abundances. An additive noise is also considered in the proposed model generalizing the normal compositional model. The proposed model is unsupervised since it estimates the abundances and both the mean and the covariance matrix of each endmember. A classification map indicating the class of each pixel is also obtained based on the estimated abundances. Simulations conducted on a real dataset show the potential of the proposed model in terms of unmixing performance for the analysis of hyperspectral images.
Complete list of metadatas

Cited literature [19 references]  Display  Hide  Download

Contributor : Open Archive Toulouse Archive Ouverte (oatao) <>
Submitted on : Thursday, October 6, 2016 - 5:56:32 PM
Last modification on : Thursday, July 25, 2019 - 4:34:15 PM
Long-term archiving on : Friday, February 3, 2017 - 7:01:28 PM


Files produced by the author(s)


  • HAL Id : hal-01377331, version 1
  • OATAO : 15278



Abderrahim Halimi, Nicolas Dobigeon, Jean-Yves Tourneret, Paul Honeine. Hyperspectral unmixing accounting for spatial correlations and endmember variability. 7th IEEE Workshop on Hyperspectral Image and SIgnal Processing: Evolution in Remote Sensing (WHISPERS 2015), Jun 2015, Tokyo, Japan. pp. 1-4. ⟨hal-01377331⟩



Record views


Files downloads