We propose a novel framework for large-scale maritime ship group surveillance using spaceborne optical imaging satellite data and Electronic Intelligence (ELINT) satellite data. Considering that the size of a ship is usually less than the distance between different ships for large-scale maritime surveillance, we treat each ship as a mass point and ship groups are modelled as point sets. Motivated by the observation that ship groups performing tactical or strategic operations often have a stable topology and their attributes remain unchanged, we combine both topological features and attributive features within the framework of Dempster-Shafer (D-S) theory for coherent ship group analysis. Our method has been tested using different sets of simulated data and recorded data. Experimental results demonstrate our method is robust and efficient for large-scale maritime surveillance.
