Title :
A Bayesian Framework for Abundance Estimation in Hyperspectral Data using Markov Random Fields
Author :
Stites, Matthew R. ; Moon, Todd K. ; Gunther, Jacob H. ; Williams, Gustavious P.
Author_Institution :
Utah State Univ., Logan
Abstract :
A model is proposed which uses neighborhoods of pixels as priors in a Bayesian setting to extract abundance information from a hyperspectral image. It is assumed that elements of the abundance vector for a pixel are independent, but that corresponding elements of abundance vectors for neighboring pixels are correlated. A posterior density encourages estimated abundances in neighboring pixels to be similar. Minimum mean- square error estimates are obtained by averaging samples from this density, where the samples are obtained by Gibbs sampling.
Keywords :
Bayes methods; Markov processes; estimation theory; image sampling; mean square error methods; spectral analysis; Bayesian framework; Gibbs sampling; Markov random fields; abundance vectors; hyperspectral data abundance estimation; hyperspectral image; minimum mean-square error estimates; neighboring pixels; posterior density; Atmospheric modeling; Bayesian methods; Data mining; Gaussian noise; Hyperspectral imaging; Image sampling; Jacobian matrices; Markov random fields; Moon; Pixel;
Conference_Titel :
Signals, Systems and Computers, 2007. ACSSC 2007. Conference Record of the Forty-First Asilomar Conference on
Conference_Location :
Pacific Grove, CA
Print_ISBN :
978-1-4244-2109-1
Electronic_ISBN :
1058-6393
DOI :
10.1109/ACSSC.2007.4487310