PCA-based hierarchical clustering of AIS trajectories with automatic extraction of clusters

With the rapid development of vessel industry, the density of traffic flow is gradually increasing, which increases the risk of water accidents and threatens the safety of human life and the economy of enterprises. Thus, the deep excavation of the ship's trajectory is very important. Clustering analysis is one of the hot issues in data mining. This paper conducts a cluster analysis of vessel's trajectories based on the AIS datasets of Wuhan Erqi Yangtze River Bridge area. The main contents of this paper are as follows: Firstly, the Frechet distance is introduced to measure the similarity between AIS trajectories; Secondly, Principal component analysis(PCA) is used to decompose the distance matrix generated in the first step to determine the number of clusters in the final level of clustering. Finally, the above distance matrix and clustering number are fused with the traditional hierarchical clustering algorithm. In order to verify the practicability and efficiency of this improved Hierarchical Clustering algorithm, a large number of experiments based on the AIS datasets are carried out. The experimental results verify the superior performance of the proposed algorithm in the study of vessel trajectory clustering.