Title :
Linear-quadratic and polynomial Non-Negative Matrix Factorization; application to spectral unmixing
Author :
Meganem, Ines ; Deville, Yannick ; Hosseini, Shahram ; Deliot, Philippe ; Briottet, Xavier ; Duarte, Leonardo T.
Author_Institution :
IRAP, Univ. de Toulouse, Toulouse, France
fDate :
Aug. 29 2011-Sept. 2 2011
Abstract :
In this article, we present a source separation method for linear-quadratic models. This class of mixing models is encountered in various real applications, such as hyperspectral unmixing for urban environments. Linear-quadratic mixing models are less studied in the literature than linear ones but there exist some methods for handling them, essentially Bayesian or based on Independent Component Analysis.
Keywords :
Bayes methods; independent component analysis; matrix decomposition; polynomial matrices; source separation; spectral analysis; Bayesian analysis; NMF; artificial mixtures; artificial signals; hyperspectral unmixing; independent component analysis; linear-quadratic models; mixing models; polynomial nonnegative matrix factorization; reflectance spectra; source separation method; spectral unmixing; urban environments; Adaptation models; Estimation error; Hyperspectral imaging; Mathematical model; Polynomials;
Conference_Titel :
Signal Processing Conference, 2011 19th European
Conference_Location :
Barcelona
