RECTILINEAR PATH DETECTION OF A TARGET IN THE WATER COLUMN BYUSING A THREE-DIMENSIONAL HOUGH TRANSFORM OF FORWARDLOOKING SONAR DATA

International audienceThis paper presents a new approach to the problem of tracking objects insequences of forward looking sonar images. Unlike previous work, we propose a “trackbefore detect” procedure based on a specific Hough Transform. The Hough transform is afamous method for detecting lines in binary images but in this study, FLS sonar data are ingrayscale and plotted in three dimensions. In typical FLS sonar images indeed, pixel intensityrepresents the backscattering at a given distance and a given azimuth but for a large range ofelevations. The standard Hough transform has been adapted to this particular data set, andpreliminary results have been obtained on real data