Edge preserving probabilistic smoothing algorithm

In the presented paper a new probabilistic approach to the problem of noise reduction has been presented. It is based on the concept of a virtual particle performing a random walk on the image lattice, with transition probabilities derived from the median distribution. The probabilistic smoothing algorithm combined with a cooling procedure, known from the simulated annealing optimization method, constitutes a new powerful technique of noise suppression, capable of preserving edges and other image features.