A Novel Similarity Measure for Clustering Vessel Trajectories Based on Dynamic Time Warping

Abstract

Clustering methods that use a similarity measurement for evaluating vessel trajectories are important for mining spatial distribution information in water transportation. To better measure the similarity of vessel trajectories, a novel similarity measure is proposed based on the dynamic time warping distance, which considers the course change of track points and the meaning at the route level. Parallel experiments were conducted based on a month of Automatic Identification System (AIS) data collected from the Zhoushan Islands area, China. After evaluation of the accuracy and the cluster degree, the novel measure demonstrated its capabilities for distinguishing different vessel trajectories and detecting similar vessel trajectories with high accuracy and has a better performance compared to some existing methods.