Ship positioning by matching radar images and map data

Abstract

Satellite network based GPS system is the most accurate, accountable and practical method for determining geographical ship locations. However, GPS systems are not preferable since they are prone to external interferences, (ie, jammers). In this study, a method for determining ship positions by matching navigation or surveillance radar images and satellite images is introduced. For this purpose, an estimate for radar image is generated by a novel method using satellite or map image. Then, the estimated radar image is compared with the real radar image using cross correlation to generate a success rate. Then, this success rate is improved iteratively by updating estimated ship position. The algorithm was run with both simulated radar images and real radar images to evaluate the correctness of the estimated ship positions. The study results are promising in the sense that introduced algorithm can be industrialized after improvements.