A Study On The Method Of Mining Accompany Vehicle Group In Urban Road Network

With the improvement of residents’ living standards,the number of domestic motor vehicles continues to rise,and vehicle travel has become one of the main choices of the public.At the same time,there has been a surge in vehicle-related violations,violations and other crimes.How to better manage road vehicles and the corresponding illegal and criminal acts has become an important problem to be solved by management and law enforcement departments.With the improvement of the new infrastructure represented by 5G,artificial intelligence and the Internet of things,it provides an indispensable basis for solving this problem.Nowadays,the automatic detection bayonet recognition system in the road network can produce a large amount of traffic track big data,that is,license plate recognition data,every day,which provides sufficient "nutrients" for the study of road vehicles.The identification of accompany vehicle group is one of the research directions of traffic track big data,and its application scene is usually car-related crime cases.For this kind of illegal behavior,the traditional practice is to manually investigate the vehicles involved,which is time-consuming and laborious and the accuracy is not high.In this paper,under the framework of Spark computing,a distributed parallel FP-Growth association algorithm for mining peer vehicles is proposed to improve the recognition efficiency.In this process,the data structure of peer segment is proposed,which is mainly used to store all the motor vehicle information that the monitoring bayonet passes within a certain time threshold,and to abandon the position information of the monitoring bayonet and the time information of the motor vehicle passing through the monitoring bayonet,so as to reduce the storage space of the data.Then the results of peer vehicles are processed,mainly by using the principle of confidence in association rules to eliminate random peer vehicles.Finally,the trajectory of the same vehicle is visually displayed to further verify the feasibility of this method.