Research On Visualization Of Inland Waterway Transportation Information Based On Massive AIS Data

With the rapid development of inland river shipping industry in China,the water traffic environment has become increasingly complex,putting maritime safety supervision under huge pressure.Visualization of shipping information is an effective means of maritime supervision.It presents valuable ship traffic information in a graphic and understandable way,which is of profound significance to scientifically guide maritime management and ensure the safety of maritime traffic.AIS(Automatic Identification System)data is the major information in inland river shipping.It is also the basic data for research on water transportation at home and abroad.AIS data has the same characteristics as big data,including enormous data volume,high processing speed(a record is generated at the fastest 2 seconds per ship),many data types(vessel type,speed,location etc.),and low value density(each ship voyage information contains hundreds of AIS records).Such characteristics have made the traditional centralized calculation and processing system based on AIS data difficult to meet the application needs of big data.It is urgent to adopt the latest distributed processing architecture.However,the navigational patterns and models of ships are implicated in AIS big data,which is not easy for humans to discover.Therefore,it is necessary to dig and analyze AIS data and make full use of it in smart maritime services.In addition,different ship traffic nature has their own characteristics,so the visualization method under different characteristics needs to be studied.Based on the characteristics of inland river shipping,this paper takes the visualization of inland navigation information as the starting point and carries out researches on the analysis and storage of massive AIS data,mining and analysis based on AIS data,and the visualization of ship shipping information.Firstly,a distributed computing and application platform based on Hadoop's inland navigation information visualization is designed.Then,aiming at inland river ship traffic,ship density,ship speed,and ship trajectory,the data mining method is studied;besides,an intuitive,easy-to-understand visualization method is studied in both single and multiple dimensions targeting different objects.In order to test the effectiveness of the system design,in this paper,it is taken the Wuhan section of the Yangtze River as the experimental water area,to study the methods on the collection,analysis,distributed storage and reading of AIS data,and conduct statistical analysis on the ship traffic volume,ship density and ship speed,digged ship tracks in a clustering patterns,and demonstrated the single-and multi-dimension information received in visualization methods such as histograms,pie charts,parallelograms,space-time cube maps and geographical information maps.The designed system has achieved ideal results in the visualization of inland ship information and verified the effectiveness and practicality of the system.The research in this paper may help promote the development of visualization of inland navigation based on big data.