Nonlinearity detection in hyperspectral images using a polynomial post-nonlinear mixing model

Yoann Altmann, Nicolas Dobigeon, Jean-Yves Tourneret

Research output: Contribution to journalArticle

Abstract

This paper studies a nonlinear mixing model for hyperspectral image unmixing and nonlinearity detection. The proposed model assumes that the pixel reflectances are nonlinear functions of pure spectral components contaminated by an additive white Gaussian noise. These nonlinear functions are approximated by polynomials leading to a polynomial post-nonlinear mixing model. We have shown in a previous paper that the parameters involved in the resulting model can be estimated using least squares methods. A generalized likelihood ratio test based on the estimator of the nonlinearity parameter is proposed to decide whether a pixel of the image results from the commonly used linear mixing model or from a more general nonlinear mixing model. To compute the test statistic associated with the nonlinearity detection, we propose to approximate the variance of the estimated nonlinearity parameter by its constrained Cramér-Rao bound. The performance of the detection strategy is evaluated via simulations conducted on synthetic and real data. More precisely, synthetic data have been generated according to the standard linear mixing model and three nonlinear models from the literature. The real data investigated in this study are extracted from the Cuprite image, which shows that some minerals seem to be nonlinearly mixed in this image. Finally, it is interesting to note that the estimated abundance maps obtained with the post-nonlinear mixing model are in good agreement with results obtained in previous studies.

Original languageEnglish
Pages (from-to)1267-1276
Number of pages10
JournalIEEE Transactions on Image Processing
Volume22
Issue number4
DOIs
Publication statusPublished - Apr 2013

Fingerprint

Polynomials
Pixels
Minerals
Statistics

Keywords

  • Constrained Cramér-Rao bound
  • nonlinearity detection
  • post-nonlinear mixing model (PPNMM)
  • spectral unmixing (SU)

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

@article{7d30bcc517a94eec8f0636291c5c6664,
title = "Nonlinearity detection in hyperspectral images using a polynomial post-nonlinear mixing model",
abstract = "This paper studies a nonlinear mixing model for hyperspectral image unmixing and nonlinearity detection. The proposed model assumes that the pixel reflectances are nonlinear functions of pure spectral components contaminated by an additive white Gaussian noise. These nonlinear functions are approximated by polynomials leading to a polynomial post-nonlinear mixing model. We have shown in a previous paper that the parameters involved in the resulting model can be estimated using least squares methods. A generalized likelihood ratio test based on the estimator of the nonlinearity parameter is proposed to decide whether a pixel of the image results from the commonly used linear mixing model or from a more general nonlinear mixing model. To compute the test statistic associated with the nonlinearity detection, we propose to approximate the variance of the estimated nonlinearity parameter by its constrained Cram{\'e}r-Rao bound. The performance of the detection strategy is evaluated via simulations conducted on synthetic and real data. More precisely, synthetic data have been generated according to the standard linear mixing model and three nonlinear models from the literature. The real data investigated in this study are extracted from the Cuprite image, which shows that some minerals seem to be nonlinearly mixed in this image. Finally, it is interesting to note that the estimated abundance maps obtained with the post-nonlinear mixing model are in good agreement with results obtained in previous studies.",
keywords = "Constrained Cram{\'e}r-Rao bound, nonlinearity detection, post-nonlinear mixing model (PPNMM), spectral unmixing (SU)",
author = "Yoann Altmann and Nicolas Dobigeon and Jean-Yves Tourneret",
year = "2013",
month = "4",
doi = "10.1109/TIP.2012.2210235",
language = "English",
volume = "22",
pages = "1267--1276",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "IEEE",
number = "4",

}

Nonlinearity detection in hyperspectral images using a polynomial post-nonlinear mixing model. / Altmann, Yoann; Dobigeon, Nicolas; Tourneret, Jean-Yves.

In: IEEE Transactions on Image Processing, Vol. 22, No. 4, 04.2013, p. 1267-1276.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Nonlinearity detection in hyperspectral images using a polynomial post-nonlinear mixing model

AU - Altmann, Yoann

AU - Dobigeon, Nicolas

AU - Tourneret, Jean-Yves

PY - 2013/4

Y1 - 2013/4

N2 - This paper studies a nonlinear mixing model for hyperspectral image unmixing and nonlinearity detection. The proposed model assumes that the pixel reflectances are nonlinear functions of pure spectral components contaminated by an additive white Gaussian noise. These nonlinear functions are approximated by polynomials leading to a polynomial post-nonlinear mixing model. We have shown in a previous paper that the parameters involved in the resulting model can be estimated using least squares methods. A generalized likelihood ratio test based on the estimator of the nonlinearity parameter is proposed to decide whether a pixel of the image results from the commonly used linear mixing model or from a more general nonlinear mixing model. To compute the test statistic associated with the nonlinearity detection, we propose to approximate the variance of the estimated nonlinearity parameter by its constrained Cramér-Rao bound. The performance of the detection strategy is evaluated via simulations conducted on synthetic and real data. More precisely, synthetic data have been generated according to the standard linear mixing model and three nonlinear models from the literature. The real data investigated in this study are extracted from the Cuprite image, which shows that some minerals seem to be nonlinearly mixed in this image. Finally, it is interesting to note that the estimated abundance maps obtained with the post-nonlinear mixing model are in good agreement with results obtained in previous studies.

AB - This paper studies a nonlinear mixing model for hyperspectral image unmixing and nonlinearity detection. The proposed model assumes that the pixel reflectances are nonlinear functions of pure spectral components contaminated by an additive white Gaussian noise. These nonlinear functions are approximated by polynomials leading to a polynomial post-nonlinear mixing model. We have shown in a previous paper that the parameters involved in the resulting model can be estimated using least squares methods. A generalized likelihood ratio test based on the estimator of the nonlinearity parameter is proposed to decide whether a pixel of the image results from the commonly used linear mixing model or from a more general nonlinear mixing model. To compute the test statistic associated with the nonlinearity detection, we propose to approximate the variance of the estimated nonlinearity parameter by its constrained Cramér-Rao bound. The performance of the detection strategy is evaluated via simulations conducted on synthetic and real data. More precisely, synthetic data have been generated according to the standard linear mixing model and three nonlinear models from the literature. The real data investigated in this study are extracted from the Cuprite image, which shows that some minerals seem to be nonlinearly mixed in this image. Finally, it is interesting to note that the estimated abundance maps obtained with the post-nonlinear mixing model are in good agreement with results obtained in previous studies.

KW - Constrained Cramér-Rao bound

KW - nonlinearity detection

KW - post-nonlinear mixing model (PPNMM)

KW - spectral unmixing (SU)

UR - http://www.scopus.com/inward/record.url?scp=84873647070&partnerID=8YFLogxK

U2 - 10.1109/TIP.2012.2210235

DO - 10.1109/TIP.2012.2210235

M3 - Article

AN - SCOPUS:84873647070

VL - 22

SP - 1267

EP - 1276

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 4

ER -