Ship track clustering analysis method based on improved dbscan algorithm

Abstract

The invention discloses a ship trajectory clustering analysis method based on the improved DBSCAN algorithm, comprising the following steps: S1, extracting effective ship trajectory data from the AIS database; S2, using the fusion distance MD to measure the trajectory similarity, and obtaining the trajectory The shortest hypertrajectory after fusion between; S3, adopt the length of the shortest supertrajectory of secondary time O (mn); S4, obtain fusion distance MD from the length of trajectory; S5, determine the global parameter of the improved DBSCAN algorithm; S6 1. Scan the entire trajectory segment data set through the improved DBSCAN algorithm to obtain a collection of clusters; S7. For the acquired spatial motion mode, if a new trajectory conforms to one of the ship operating modes, then the trajectory is considered to be a normal state track. This method improves the measurement accuracy, reduces the time consumption compared with the traditional DBSCAN algorithm, and can handle multi-density data sets well, has better robustness and adaptability, and correctly classifies the trajectory .

Description

technical field [0001] The invention relates to the technical field of ship navigation safety, in particular to a ship track cluster analysis method based on an improved DBSCAN algorithm. Background technique [0002] With the development of the economy, the enhancement of waterway transportation and international trade has led to the rise of maritime traffic, and at the same time, a large number of ship motion trajectories have been generated, which is of great significance to the analysis and supervision of global maritime transportation. Nowadays, with the development of technology, the Automatic Identification System (AIS) is introduced to track ships and monitor maritime affairs. It can cooperate with the Global Positioning System (GPS) to exchange a series of ship dynamic and static information such as position, course, and speed with other ships and AIS base stations. Facilitates tracking and monitoring of vessel movements at sea. Through these AIS data, mining the s...

AI Technical Summary

Problems solved by technology

[0003] The DBSCAN algorithm is usually used for cluster analysis of ship trajectories, but the accuracy of the existing DBSCAN algorithm measurement is low, the obtained trajectory classification is not particularly accurate, and the clustering of ship trajectories takes a long time, and its robustness and adaptability sex is also poor

Images