Application of the Dempster-Shafer Algorithm to the Detection of Sar Ship Wakes

An Automatic Detection Algorithm (ADA) for ship wakes based upon the Dempster-Shafer method of data fusion was developed. The technique was tested on an extensive set of SEASAT SAR ocean imagery. The SEASAT files analyzed in this study were chosen specifically to test the limitations of the ADA. Files exhibiting multiple ships, short or faint wake features, striated ocean backgrounds or the presence of other naturally-occuring linear ocean features constituted the majority of the image set. Even under such circumstances, the ADA correctly classified the great majority of the image files.