Contemporary sonar systems are used mainly to search for and detect underwater objects which might affect the safety of navigation or the safety of moored watercraft. The task of the sonar operator is to display the optimal settings of the sonar image and to select the proper operational parameters that enable detection and identification of hidden targets. The operator performs corrections of the recorded sonar data, thereby improving the image quality and the ability to interpret the survey results. The paper deals with the research problem of changing display images, conducting geometric corrections, and extracting objects from the sonar imagery background. The presented method for the automatic searching of objects utilizes the morphologic features of sonograms and geometric and arithmetic methods of digital signal processing. In searching for the features of a sonar target, the minimal-distance methods used in pattern recognition were applied. The developed application makes it possible to perform research experiments that follow the stages of sonar detection of objects. The obtained results can indicate the validity of the research assumptions and the appropriateness of the methods used.

29-42