Title :
Estimation of the Ising field parameter thanks to the exact partition function
Author :
Giovannelli, J.-F.
Author_Institution :
Lab. d´´Integration du Materiau au Syst., Equipe Signal-Image, Univ. de Bordeaux 1, Talence, France
Abstract :
We propose new estimation methods for the Ising field parameter based on an explicit expression for the partition function. This expression is known for a long time, but, to the best of our knowledge, it has never been used for parameter estimation. The paper develops standard Bayesian estimates and proposes a numerical comparative evaluation.
Keywords :
belief networks; image reconstruction; parameter estimation; Bayesian estimation; Ising field parameter estimation; image processing; partition function; Approximation methods; Bayesian methods; Computational modeling; Estimation; Image processing; Markov processes; Parameter estimation; Bayesian estimation; Ising field; parameter estimation; partition function;
Conference_Titel :
Image Processing (ICIP), 2010 17th IEEE International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-7992-4
Electronic_ISBN :
1522-4880
DOI :
10.1109/ICIP.2010.5650185
