Bayesian deconvolution of poissonian point sources
暂无分享,去创建一个
In this article, we address the problem of Bayesian de-convolution of point sources with Poisson statistics. A high level Bayesian approach is proposed to solve this problem. The original image is modeled list of an unknown number of points sources with unknow parameters. A prior distribution reflecting our degree of belief is introduced on all these unknown parameters, including the number of sources. All Bayesian inference relies on the posterior distribution. This latter admitting no analytical expression, we estimate it using an original Reversible Jump Markov Chain Monte Carlo method. The algorithm developed is tested over real data. It displays satisfactory results compared to traditional low level Bayesian approaches.
[1] D. J. Strauss. A model for clustering , 1975 .
[2] Guillaume Stawinski,et al. Reversible jump Markov chain Monte Carlo for Bayesian deconvolution of point sources , 1998, Optics & Photonics.
[3] J. Berger. Statistical Decision Theory and Bayesian Analysis , 1988 .
[4] A. Baddeley,et al. Stochastic geometry models in high-level vision , 1993 .
[5] J. Besag,et al. Bayesian Computation and Stochastic Systems , 1995 .