Title :
Unmixing hyperspectral images using the generalized bilinear model
Author :
Halimi, Abderrahim ; Altmann, Yoann ; Dobigeon, Nicolas ; Tourneret, Jean-Yves
Author_Institution :
IRIT/INP, Univ. of Toulouse, Toulouse, France
Abstract :
Nonlinear models have recently shown interesting properties for spectral unmixing. This paper considers a generalized bilinear model recently introduced for unmixing hyperspectral images. Different algorithms are studied to estimate the parameters of this bilinear model. The positivity and sum-to-one constraints for the abundances are ensured by the proposed algorithms. The performance of the resulting unmixing strategy is evaluated via simulations conducted on synthetic and real data.
Keywords :
Bayes methods; Markov processes; Monte Carlo methods; geophysical image processing; geophysical techniques; gradient methods; mean square error methods; Bayesian model; Markov chain Monte Carlo method; Taylor series expansion; constrained gradient descent method; generalized bilinear model; hyperspectral image unmixing; joint posterior distribution; minimum mean square error estimator; nonlinear model; parameter estimation; positivity constraint; spectral unmixing; sum-to-one constraint; Bayesian methods; Computational modeling; Estimation; Hyperspectral imaging; Optimization; Bayesian inference; MCMC methods; bilinear model; gradient descent algorithm; hyperspectral imagery; least square algorithm; spectral unmixing;
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2011 IEEE International
Conference_Location :
Vancouver, BC
Print_ISBN :
978-1-4577-1003-2
DOI :
10.1109/IGARSS.2011.6049492
