Unsupervised nonlinear unmixing of hyperspectral images using Gaussian processes

Yoann Altmann, Nicolas Dobigeon, Steve McLaughlin, Jean-Yves Tourneret

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

Abstract

This paper describes a Gaussian process based method for nonlinear hyperspectral image unmixing. The proposed model assumes a nonlinear mapping from the abundance vectors to the pixel reflectances contaminated by an additive white Gaussian noise. The parameters involved in this model satisfy physical constraints that are naturally expressed within a Bayesian framework. The proposed abundance estimation procedure is applied simultaneously to all pixels of the image by maximizing an appropriate posterior distribution which does not depend on the endmembers. After determining the abundances of all image pixels, the endmembers contained in the image are estimated by using Gaussian process regression. The performance of the resulting unsupervised unmixing strategy is evaluated through simulations conducted on synthetic data.

Original languageEnglish
Title of host publication2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
PublisherIEEE
Pages1249-1252
Number of pages4
ISBN (Print)9781467300469
DOIs
Publication statusPublished - 31 Aug 2012
Event2012 IEEE International Conference on Acoustics, Speech, and Signal Processing - Kyoto, Japan
Duration: 25 Mar 201230 Mar 2012

Conference

Conference2012 IEEE International Conference on Acoustics, Speech, and Signal Processing
Abbreviated titleICASSP 2012
CountryJapan
CityKyoto
Period25/03/1230/03/12

Fingerprint

Pixels

Keywords

  • Gaussian Processes
  • hyperspectral images
  • Nonlinear unmixing

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Altmann, Y., Dobigeon, N., McLaughlin, S., & Tourneret, J-Y. (2012). Unsupervised nonlinear unmixing of hyperspectral images using Gaussian processes. In 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 1249-1252). IEEE. https://doi.org/10.1109/ICASSP.2012.6288115
Altmann, Yoann ; Dobigeon, Nicolas ; McLaughlin, Steve ; Tourneret, Jean-Yves. / Unsupervised nonlinear unmixing of hyperspectral images using Gaussian processes. 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2012. pp. 1249-1252
@inproceedings{696ef875c22a40d389fc345c1b51932c,
title = "Unsupervised nonlinear unmixing of hyperspectral images using Gaussian processes",
abstract = "This paper describes a Gaussian process based method for nonlinear hyperspectral image unmixing. The proposed model assumes a nonlinear mapping from the abundance vectors to the pixel reflectances contaminated by an additive white Gaussian noise. The parameters involved in this model satisfy physical constraints that are naturally expressed within a Bayesian framework. The proposed abundance estimation procedure is applied simultaneously to all pixels of the image by maximizing an appropriate posterior distribution which does not depend on the endmembers. After determining the abundances of all image pixels, the endmembers contained in the image are estimated by using Gaussian process regression. The performance of the resulting unsupervised unmixing strategy is evaluated through simulations conducted on synthetic data.",
keywords = "Gaussian Processes, hyperspectral images, Nonlinear unmixing",
author = "Yoann Altmann and Nicolas Dobigeon and Steve McLaughlin and Jean-Yves Tourneret",
year = "2012",
month = "8",
day = "31",
doi = "10.1109/ICASSP.2012.6288115",
language = "English",
isbn = "9781467300469",
pages = "1249--1252",
booktitle = "2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)",
publisher = "IEEE",
address = "United States",

}

Altmann, Y, Dobigeon, N, McLaughlin, S & Tourneret, J-Y 2012, Unsupervised nonlinear unmixing of hyperspectral images using Gaussian processes. in 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, pp. 1249-1252, 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, Kyoto, Japan, 25/03/12. https://doi.org/10.1109/ICASSP.2012.6288115

Unsupervised nonlinear unmixing of hyperspectral images using Gaussian processes. / Altmann, Yoann; Dobigeon, Nicolas; McLaughlin, Steve; Tourneret, Jean-Yves.

2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2012. p. 1249-1252.

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

TY - GEN

T1 - Unsupervised nonlinear unmixing of hyperspectral images using Gaussian processes

AU - Altmann, Yoann

AU - Dobigeon, Nicolas

AU - McLaughlin, Steve

AU - Tourneret, Jean-Yves

PY - 2012/8/31

Y1 - 2012/8/31

N2 - This paper describes a Gaussian process based method for nonlinear hyperspectral image unmixing. The proposed model assumes a nonlinear mapping from the abundance vectors to the pixel reflectances contaminated by an additive white Gaussian noise. The parameters involved in this model satisfy physical constraints that are naturally expressed within a Bayesian framework. The proposed abundance estimation procedure is applied simultaneously to all pixels of the image by maximizing an appropriate posterior distribution which does not depend on the endmembers. After determining the abundances of all image pixels, the endmembers contained in the image are estimated by using Gaussian process regression. The performance of the resulting unsupervised unmixing strategy is evaluated through simulations conducted on synthetic data.

AB - This paper describes a Gaussian process based method for nonlinear hyperspectral image unmixing. The proposed model assumes a nonlinear mapping from the abundance vectors to the pixel reflectances contaminated by an additive white Gaussian noise. The parameters involved in this model satisfy physical constraints that are naturally expressed within a Bayesian framework. The proposed abundance estimation procedure is applied simultaneously to all pixels of the image by maximizing an appropriate posterior distribution which does not depend on the endmembers. After determining the abundances of all image pixels, the endmembers contained in the image are estimated by using Gaussian process regression. The performance of the resulting unsupervised unmixing strategy is evaluated through simulations conducted on synthetic data.

KW - Gaussian Processes

KW - hyperspectral images

KW - Nonlinear unmixing

UR - http://www.scopus.com/inward/record.url?scp=84867604334&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2012.6288115

DO - 10.1109/ICASSP.2012.6288115

M3 - Conference contribution

SN - 9781467300469

SP - 1249

EP - 1252

BT - 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

PB - IEEE

ER -

Altmann Y, Dobigeon N, McLaughlin S, Tourneret J-Y. Unsupervised nonlinear unmixing of hyperspectral images using Gaussian processes. In 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE. 2012. p. 1249-1252 https://doi.org/10.1109/ICASSP.2012.6288115