Approximate Bayes model selection procedures for Gibbs-Markov random fields |
| |
| Institution: | 1. Department of Statistics, The University of Georgia, GA 30602-1952, Athens, Greece;2. Department of Statistics, The University of North Carolina, NC, USA;1. Paranaense Company of Energy, Londrina, Brazil;2. Federal University of Goiás, Institute of Computer Science, Brazil;3. University of São Paulo, Institute of Mathematical Science and Computation, Brazil;4. University of São Paulo, Sao Carlos School of Engineering, Brazil |
| |
| Abstract: | For applications in texture synthesis, we derive two approximate Bayes criteria for selecting a model from a collection of Markov random fields. The first criterion is based on a penalized maximum likelihood. The second criterion, a Markov chain Monte Carlo approximation to the first, has distinct computational advantages. Some simulation results are also presented. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
