Smoothing images by a probability filter

Image smoothing is a useful and necessary part of image processing. Both the median and mean filters are popular smoothing methods, but there can be problems when these two filters are applied to an image corrupted by unknown noise. We present a spatial domain probability filter which achieves the advantages of median and mean filters. The aim of our probability filter is to provide a compromise between the two more conventional filters. We present experimental results and performance evaluation of the median, mean filters and this probability filter operating on an image corrupted by impulse-like or random noise.