Markov Random Eld and Fuzzy Logic Modeling in Sonar Imagery: Application to the Classiication of Underwater Oor

1 This paper proposes an original method for the classiication of sea-oors from high resolution sidescan sonar images. We aim at classifying the sonar images into ve kinds of regions: sand, pebbles, rocks, ripples and dunes. The proposed method adopts a pattern recognition approach based on the extraction and the analysis of the cast shadows exhibited by each sea-bottom type. This method consists of three stages of processing. Firstly, the original image is segmented into two kinds of regions: shadow (corresponding to a lack of acoustic reverberation behind each \object" lying on the sea-bed) and sea-bottom reverberation. Secondly, based on the extracted shadows, shape parameter vectors are computed on sub-images and classiied with a fuzzy classiier. This preliminary classiication is nally reened thanks to a Markov random eld model which allows to incorporate spatial homogeneity properties one would expect for the nal classiication map. Experiments on a variety of real high resolution sonar images are reported.