| For the past few years,with the continuous improvement of people’s living standards,more and more people have become car owners.Correspondingly,traffic-related criminal cases are also increasing every year.Thus,the vehicle information captured by checkpoints have played a very important role in catching the hit-and-run vehicle and detecting illegal vehicles.However,with the continued construction and investment in urban intelligent traffic information engineering,the vehicle data collected by capture system become increasingly large which increased from one million to hundreds of millions in just a few years.Storage capacity limit is reached,query speed becomes much more slower,system functions are unable to meet the new requirements,more and more problems exposed that the old version of the vehicle management and control platform can not meet the current needs of public security work.In this regard,this paper carried out an in-depth analysis on the shortage and problems which the original vehicle management and control platform exists,designed and implemented a new version of vehicle management system based on Hadoop.The followings are the main contents of this article:1.Make a deep deinvestigation and study of the architecture and main features of HDFS and HBase based on Hadoop.In particular,focusing on the HBase storage features.2.Detailed analyse the needs of vehicle management system,give the specific system design,and highlight the related design of the business logic and data storage layers.3.Give solutions to solving the problems that HBase database does not support multi-condition combination inquires based on Solr + HBase;The storage problems of capturing image are given for individual storage solutions.Finally,the system was actually built,and some of these functional modules were given specific implementation processes and operating results.Meanwhile,it is proved that the system has met the needs of the design by functional testing and performance testing. |
PDF Full Text Request