Fusion of local statistical detectors in SAS imagery classification

Detection of underwater mines is a present crucial strategic task. The images provided by synthetic aperture sonar (SAS) are then of great interest for the detection and classification of objects lying on the sea floor or buried in the sea bed. This paper proposes a detection method based on data fusion, using local statistical characteristics extracted from the SAS data. These values come from first, second, third, and fourth order statistical properties of the sonar images.