Unsupervised bayesian convex deconvolution based on a field with an explicit partition function - Statistics > MethodologyReportar como inadecuado




Unsupervised bayesian convex deconvolution based on a field with an explicit partition function - Statistics > Methodology - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: This paper proposes a non-Gaussian Markov field with a special feature: anexplicit partition function. To the best of our knowledge, this is an originalcontribution. Moreover, the explicit expression of the partition functionenables the development of an unsupervised edge-preserving convex deconvolutionmethod. The method is fully Bayesian, and produces an estimate in the sense ofthe posterior mean, numerically calculated by means of a Monte-Carlo MarkovChain technique. The approach is particularly effective and the computationalpracticability of the method is shown on a simple simulated example.



Autor: Jean-Francois Giovannelli

Fuente: https://arxiv.org/







Documentos relacionados