1 April 2004 Fast joint separation and segmentation of mixed images
Author Affiliations +
J. of Electronic Imaging, 13(2), (2004). doi:10.1117/1.1666873
Abstract
We consider the problem of the blind separation of noisy instantaneously mixed images. The images are modeled by hidden Markov fields with unknown parameters. Given the observed images, we give a Bayesian formulation and we propose a fast version of the MCMC (Monte Carlo Markov Chain) algorithm based on the Bartlett decomposition for the resulting data augmentation problem. We separate the unknown variables into two categories: 1. The parameters of interest which are the mixing matrix, the noise covariance and the parameters of the sources distributions. 2. The hidden variables which are the unobserved sources and the unobserved pixel segmentation labels. The proposed algorithm provides, in the stationary regime, samples drawn from the posterior distributions of all the variables involved in the problem leading to great flexibility in the cost function choice. Finally, we show the results for both synthetic and real data to illustrate the feasibility of the proposed solution.
Hichem Snoussi, Ali Mohammad-Djafari, "Fast joint separation and segmentation of mixed images," Journal of Electronic Imaging 13(2), (1 April 2004). http://dx.doi.org/10.1117/1.1666873
JOURNAL ARTICLE
13 PAGES


SHARE
KEYWORDS
Image segmentation

Signal to noise ratio

Matrices

Monte Carlo methods

Computer simulations

Sensors

Data modeling

RELATED CONTENT

Image processing and low-discrepancy sequences
Proceedings of SPIE (November 02 1999)
Holography and the Wigner function
Proceedings of SPIE (July 09 2002)
Graphic-oriented environmental monitoring system
Proceedings of SPIE (May 07 1999)
Depolarizing collisions of 174Yb with rare gas atoms
Proceedings of SPIE (April 17 2008)

Back to Top