Automatic change detection for the monitoring of cluttered underwater areas
Automating the surveillance of cluttered underwater areas (such as ports) poses several problems, the main one being the high number of contacts that any automatic target recognition (ATR) system will produce when surveying the area. In this paper an approach based on Change Detection which is capable of coping with the expected high number of potential threats is proposed for the surveillance of cluttered areas. To that end an automatic detector is coupled with a Data Association mechanism in order to determine the location of previously undetected contacts which may constitute newthreats. The proposed system is tested on a real scenario using synthetic aperture sonar (SAS) data, and the influence of conducting surveys in different orientations on the change detection results is studied.
Pinto, Marc A.