Clustering Research Based on Ship Trajectory Density and DBSCAN Algorithm

Abstract

Automatic Identification System (AIS) is composed of shore base station facilities and shipboard equipment. It is a new type of digital navigation aid system and equipment that integrates network technology, modern communication technology, computer technology and electronic information display technology. Automatic ship identification system can provide communication and management between ships and between ships and shore base stations, so in the process of navigation informatization, AIS has received more and more attention. The message data of the automatic ship identification system contains a large amount of information about ships at sea. In order to find the movement rules of ships in a specific area in the AIS data, Improved DBSCAN algorithm based on ship trajectory segment density was proposed. The improved DBSCAN clustering algorithm is used to perform clustering analysis on the real AIS track segment data instead of the whole trajectory. The experimental results show that the improved DBSCAN algorithm can get the common track within the region, and can improve the accuracy of clustering to a certain extent, which lays a foundation for the further analysis of abnormal ship behavior.