Markov random field and fuzzy logic modeling in sonar imagery: Applicationto the classification of underwater floor

This paper proposes an original method For the classification of seafloors from high resolution sidescan sonar images. We aim at classifying the sonar images into five kinds of regions: sand, pebbles, rocks, ripples, and dune s. The proposed method adopts a pattern recognition approach based on the e xtraction and the analysis of the cast shadows exhibited by each seabottom type. This method consists of three stages of processing. First, the origin al image is segmented into two kinds of regions: shadow (corresponding to a lack of acoustic reverberation behind each "object" lying on the seabed) a nd seabottom reverberation. Second, based on the extracted shadows. shape p arameter vectors are computed on subimages and classified with a fuzzy clas sifier This preliminary classification is finally refined thanks to a Marko v random field model which allows to incorporate spatial homogeneity proper ties one would expect for the final classification map. Experiments on a va riety of real high-resolution sonar images are reported. (C) 2000 Academic Press.