R package 'bayesImageS' implements algorithms for segmentation of 2D and 3D images, such as computed tomography (CT) and satellite remote sensing. It provides functions for Bayesian image analysis using the hidden Potts/Ising model with external field prior. Latent labels are updated using chequerboard Gibbs sampling or Swendsen-Wang. Algorithms for the smoothing parameter include: pseudolikelihood; path sampling (thermodynamic integration); the exchange algorithm; approximate Bayesian computation (ABC-MCMC and ABC-SMC); and Bayesian indirect likelihood (BIL).
Computational Statistics and Data Analysis (2018)
Storvik Geir and Hubin Aliaksandr