Radar target detection based on methods of image pattern matching

This paper studies the performance of pattern matching algorithms with the goal of the detection and tracking of vessel targets in maritime radar images. We compare different methods using a database which consists of radar images from two different manufactures. The data covers a timespan of roughly 4 hours with a one second time resolution. The performance of 3 template matching and 5 feature detector algorithms is tested and the most suitable algorithm is chosen for further optimizations. These optimizations are based on the properties of the radar images and the properties of the radar target.