Analysis And System Realization Of Marine Ship Trajectory Anomalies Based On Spatiotemporal Data

Ship trajectory data contains a large amount of spatiotemporal information,and its potential information can be effectively mined from these data.The mined information is of great significance for the detection of illegal crimes,early warning of collision accidents,and maritime search and rescue.Thus,Analysis and research of marine ship trajectory data have become hot issues.The main purpose of this paper is to study the anomalous trajectories that appear in the operation of the ships and analyze the abnormal behavior of the ships by using fishing boat data in Shandong Province.This paper designs a similarity measurement method to associate and fuse the trajectories of multiple data sources,analyze the fused trajectories based on the law of abnormal behavior,mine abnormal behaviors in abnormal trajectories,and realize real-time detection and early warning of different abnormal behaviors.The specific design work is as follows:（1）This paper firstly preprocesses the fishing vessel radar,AIS,and Beidou data to realize data fusion,perform trajectory correlation on the fused data,delete duplicate data,and make up for the inaccuracy of a single data source and various relationships that may be lost,enhance Trajectory calculation accuracy and analysis reliability.（2）To study the similarity of fishing boat trajectories,and to reduce the workload of trajectory comparison caused by massive data,an improved Euclidean metric method of trajectory similarity based on spatiotemporal data is designed.By increasing the distance threshold and time threshold,the traditional Euclidean metric method is improved and a grid filtering model is established,and the trajectories beyond the set distance are discarded,thereby reducing the computational cost of the system.（3）According to the designed algorithm,establish a real-time detection model,screen suspicious data,classify the filtered trajectory data according to the type and completeness according to the relationship between the trajectories,and classify the detected and classified trajectory characteristics and abnormal behavior trajectory characteristics.Match,identify,and analyze the unusual behavior of fishing vessels at sea.（4）Design and implement an abnormal analysis system of marine ship trajectory based on the proposed algorithm.The system uses Arc Map for map mapping,Arc GIS server for map publishing,MVC architecture,and Java language for development based on the IDEA development environment.SQL Sever database is used for data storage,and Tomcat is selected as the web server.The system includes tracking modules and abnormal tracks.The detection module and the abnormal trajectory repair module are divided into three modules,and the practicability of the system is verified through functional tests and performance tests.