Edge-Preserving Smoothing Using a Similarity Measure in Adaptive Geodesic Neighbourhoods

摘要

This paper introduces a novel image-dependent filtering approach derived from concepts known in mathematical morphology and aiming at edge-preserving smoothing natural images. Like other adaptive methods, it assumes that the neighbourhood of a pixel contains the essential information required for the estimation of local features. The proposed strategy does not require the definition of any spatial operator as it determines automatically, from the unfiltered input data, a weighting neighbourhood and a weighting kernel for each pixel location. It essentially consists in a weighted averaging combining both spatial and tonal information, for which a twofold similarity measure has to be calculated from local geodesic time functions. By designing relevant geodesic masks, two adaptive filtering algorithms are derived, that are particularly efficient at smoothing heterogeneous areas while preserving relevant structures in greyscale and multichannel images.