“…The use of Markov Random Fields in imaging has a long tradition [2], and has been used in a variety of applications, including segmentation [3], low level vision [4], object identification [5], as well as pattern recognition in dermatoscopic images [6]. The underlying graph is a regular lattice corresponding to the pixels of the image, which has a Markov property, i.e., the value of any pixel is a random variable conditionally independent of all except the neighboring pixels defined in a neighborhood structure.…”