Pseudo-likelihood equations for Potts model on higher-order neighborhood systems: A quantitative approach for parameter estimation in image analysis
Levada, Alexandre L. M. ; Mascarenhas, Nelson D. A. ; Tannús, Alberto
Braz. J. Probab. Stat., Tome 23 (2009) no. 1, p. 120-140 / Harvested from Project Euclid
This paper presents analytical pseudo-likelihood (PL) equations for Potts Markov random field (MRF) model parameter estimation on higher-order neighborhood systems by expanding the derivative of the log-PL function based on the enumeration of all possible contextual configuration patterns given a neighborhood system. The proposed equations allow the modeling of less restrictive neighborhood systems in a large number of MRF applications in a computationally feasible way. To evaluate the proposed estimation method we propose a hypothesis testing approach, derived by approximating the asymptotic variance of MPL parameter estimators using the observed Fisher information. The definition of the asymptotic variance, together with the test size α and p-values, provide a complete framework for quantitative analysis. Experiments with synthetic images generated by Markov chain Monte Carlo simulation methods assess the accuracy of the proposed estimation method, indicating that higher-order neighborhood systems reduce the MPL estimator asymptotic variance and improve estimation performance.
Publié le : 2009-12-15
Classification:  Markov random fields,  Potts model,  maximum pseudo-likelihood estimation,  Markov chain Monte Carlo simulation
@article{1256562754,
     author = {Levada, Alexandre L. M. and Mascarenhas, Nelson D. A. and Tann\'us, Alberto},
     title = {Pseudo-likelihood equations for Potts model on higher-order neighborhood systems: A quantitative approach for parameter estimation in image analysis},
     journal = {Braz. J. Probab. Stat.},
     volume = {23},
     number = {1},
     year = {2009},
     pages = { 120-140},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1256562754}
}
Levada, Alexandre L. M.; Mascarenhas, Nelson D. A.; Tannús, Alberto. Pseudo-likelihood equations for Potts model on higher-order neighborhood systems: A quantitative approach for parameter estimation in image analysis. Braz. J. Probab. Stat., Tome 23 (2009) no. 1, pp.  120-140. http://gdmltest.u-ga.fr/item/1256562754/