Hyperspectral unmixing accounting for spatial correlations and endmember variability

Abderrahim Halimi; Nicolas Dobigeon; Jean-Yves Tourneret; Paul Honeine

doi:10.1109/WHISPERS.2015.8075442

Hyperspectral unmixing accounting for spatial correlations and endmember variability

Abderrahim Halimi, Nicolas Dobigeon, Jean-Yves Tourneret, Paul Honeine

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper presents an unsupervised Bayesian algorithm for hyper-spectral 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.

Original language	English
Title of host publication	2015 7th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS)
Publisher	IEEE
ISBN (Electronic)	9781467390156
DOIs	https://doi.org/10.1109/WHISPERS.2015.8075442
Publication status	Published - 23 Oct 2017

Keywords

endmember variability
Hyperspectral imagery
image classification
Markov chain Monte-Carlo

ASJC Scopus subject areas

Computer Vision and Pattern Recognition
Signal Processing

Access to Document

10.1109/WHISPERS.2015.8075442

Cite this

@inproceedings{f0ab77adc7a44a5080d01f25cbe45978,

title = "Hyperspectral unmixing accounting for spatial correlations and endmember variability",

abstract = "This paper presents an unsupervised Bayesian algorithm for hyper-spectral 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.",

keywords = "endmember variability, Hyperspectral imagery, image classification, Markov chain Monte-Carlo",

author = "Abderrahim Halimi and Nicolas Dobigeon and Jean-Yves Tourneret and Paul Honeine",

year = "2017",

month = oct,

day = "23",

doi = "10.1109/WHISPERS.2015.8075442",

language = "English",

booktitle = "2015 7th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS)",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - Hyperspectral unmixing accounting for spatial correlations and endmember variability

AU - Halimi, Abderrahim

AU - Dobigeon, Nicolas

AU - Tourneret, Jean-Yves

AU - Honeine, Paul

PY - 2017/10/23

Y1 - 2017/10/23

N2 - This paper presents an unsupervised Bayesian algorithm for hyper-spectral 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.

AB - This paper presents an unsupervised Bayesian algorithm for hyper-spectral 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.

KW - endmember variability

KW - Hyperspectral imagery

KW - image classification

KW - Markov chain Monte-Carlo

UR - http://www.scopus.com/inward/record.url?scp=85039151806&partnerID=8YFLogxK

U2 - 10.1109/WHISPERS.2015.8075442

DO - 10.1109/WHISPERS.2015.8075442

M3 - Conference contribution

AN - SCOPUS:85039151806

BT - 2015 7th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS)

PB - IEEE

ER -

Hyperspectral unmixing accounting for spatial correlations and endmember variability

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this