A BAYESIAN-APPROACH FOR THE MEDIAN FILTER IN IMAGE-PROCESSING

In this paper we propose the use of the median filter (MF) within the Bayesian framework. This allows us to develop global methods for both image smoothing and image approximation by the MF 'roots'. A new metho d for solving the approximation problem is proposed, which is based on stochastic optimization with constraints. Results of the proposed met hod for both simulated and real binary images are illustrated and comp ared to results from a known deterministic method. (C) 1997 Elsevier S cience B.V.