Ship abnormal behavior detection based on KD-Tree and clustering algorithm

This paper aims to assist the maritime department to monitor waters and identify more meaningful outliers. In order to assist the managers to identify the anomaly behavior of ships, which improve the efficiency of regulation. In order to better mine information from the data of the Automatic Ship Identification System (AIS), scientifically perceive the water traffic situation, and identify abnormal ships, this paper proposes an improved density spatial clustering algorithm (DBSCAN), combined with the KD-tree nearest neighbor point search algorithm to detect abnormal points of ship trajectory points Finding abnormalities and analyzing them can simplify the way of abnormal identification and monitoring, provide a good ship monitoring environment for maritime inspection departments, reduce the work intensity of monitoring personnel, and provide reasonable suggestions for waterway construction departments, emergency response departments and maritime supervision departments.