流式计算在交通管理中应用研究  

Study on the Application of Steam Computing in Traffic Management

作　　者：周建宁 徐晓东 蔡岗 

机构地区：[1]公安部交通管理科学研究所,无锡214151

出　　处：《中国公共安全（学术版）》2016年第1期70-75,共6页China Public Security(Academy Edition)

基　　金："十二五"国家科技支撑计划(2014BAG01B04-04)

摘　　要：近年来,为强化路面动态监控,各级公安交通管理部门开始大规模建设车辆智能监测记录系统(以下简称卡口系统),随着卡口系统接入的卡口数量日益增多,当卡口的过车数据量大到一定时候,基于传统关系型数据库的比对预警方式,会出现预警时间延迟,无法满足实战的实时性需求。本文针对卡口过车的实时预警要求,对大数据流式计算组件进行了分析和研究,提出了过车信息实时接入、实时比对预警的大数据流式计算技术解决方案,为基层路面民警的及时拦截查处争取了时间。In recent years, in order to strengthen the monitoring of pavement, police departments of traffic management at all levels start a massive construction of intelligent vehicle monitoring and recording system（referred to as the monitoring system）.With the increasing number of monitors accessing to monitoring system, and the quantity of the monitoring car information increases to certain amount, the traditional way of comparing early warning management based on database, will appear warning time delay, and can not meet the actual requirement.This paper introduces solutions of the vehicle real-time information access and real-time comparison warning of big data stream computing technology based on vehicle real-time warning requirements of data stream computing component analysis and research.

关 键 词：大数据 流式计算 SPARK STREAMING 智能交通 卡口 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]