作者: Xavier Descombes , Robin Morris , Josiane Zerubia , Marc Berthod
关键词:
摘要: Recent developments in statistics now allow maximum likelihood estimators for the parameters of Markov Random Fields to be constructed. We detail theory required, and present an algorithm which is easily implemented practical terms computation time. demonstrate this on three MRF models, standard Potts model, inhomogeneous variation model a long-range interaction better adapted modelling real-world images. estimate from synthetic real image, then resynthesise models features image have been captured by model.