Motivated by applications in texture synthesis, we propose a model
selection procedure for Markov random fields based on penalized
pseudolikelihood. The procedure is shown to be consistent for choosing the
true model, even for Gibbs random fields with phase transitions. As a
by-product, rates for the restricted mean-square error and moderate deviation
probabilities are derived for the maximum pseudolikelihood estimator. Some
simulation results are presented for the selection procedure.
Publié le : 1996-05-14
Classification:
Markov random fields,
Gibbs random fields,
model selection,
pseudolikelihood,
texture synthesis,
image analysis,
62M40,
62F12,
68U10
@article{1034968138,
author = {Ji, Chuanshu and Seymour, Lynne},
title = {A consistent model selection procedure for Markov random fields
based on penalized pseudolikelihood},
journal = {Ann. Appl. Probab.},
volume = {6},
number = {1},
year = {1996},
pages = { 423-443},
language = {en},
url = {http://dml.mathdoc.fr/item/1034968138}
}
Ji, Chuanshu; Seymour, Lynne. A consistent model selection procedure for Markov random fields
based on penalized pseudolikelihood. Ann. Appl. Probab., Tome 6 (1996) no. 1, pp. 423-443. http://gdmltest.u-ga.fr/item/1034968138/