面向高速公路逃费车辆甄别的实时检测方法
|
摘要
高速公路逃费车辆的在线甄别对高速公路运行管理和运行效率的提升具有重要意义。高速公路逃费车辆甄别具有时效性约束。针对现有高速公路收费数据稽查低效率、高延迟以及精度低等问题,提出了Storm环境下在线逃费车辆实时检测甄别技术。针对假冒军警牌车辆、倒卡逃费、套牌车逃费三种逃费方式,利用某省高速公路真实收费数据模拟实时收费收据流,设计了基于真实军警牌数据、站点阈值表和滑动时间窗口的三种在线计算模型。实验结果表明,三种逃费方式计算的准确率分别达到99%、98%、98.5%,单条收费数据的处理时间达到毫秒级,实现了在线检测可疑逃费车辆。最后,在三种计算模型的基础上,完成了偷逃费车辆稽查系统,能实时甄别并显示高速公路路网出现的所有可疑逃费车辆收费信息。
The screening of highway escaping vehicles online with timeliness is of great significance to the operation and management of highway and the improvement of highway operation efficiency. Aiming at the existing problems of low efficiency,high delay and low accuracy of fee evasion,we propose a real-time approach to find suspected vehicles online based on Apache Storm. According to different behavior modes of fee evasion,three online calculation models based on real military police card data,the threshold table of station and the sliding time window are designed by simulating the real-time charging receipt flow by using the real highway toll data of a province. The experiment shows that the calculation accuracy of the three types of evasion methods is 99%,98%,98.5%. Also,a piece of toll data can be dealt with in milliseconds,realization of the online detection of suspicious fee evasion vehicles. Finally,on the basis of three kinds of calculation models,the vehicle inspection system for fee evasion is completed. The system can identify and display the toll information of all suspected fee evasion vehicles in the highway network in real time.
The screening of highway escaping vehicles online with timeliness is of great significance to the operation and management of highway and the improvement of highway operation efficiency. Aiming at the existing problems of low efficiency,high delay and low accuracy of fee evasion,we propose a real-time approach to find suspected vehicles online based on Apache Storm. According to different behavior modes of fee evasion,three online calculation models based on real military police card data,the threshold table of station and the sliding time window are designed by simulating the real-time charging receipt flow by using the real highway toll data of a province. The experiment shows that the calculation accuracy of the three types of evasion methods is 99%,98%,98.5%. Also,a piece of toll data can be dealt with in milliseconds,realization of the online detection of suspicious fee evasion vehicles. Finally,on the basis of three kinds of calculation models,the vehicle inspection system for fee evasion is completed. The system can identify and display the toll information of all suspected fee evasion vehicles in the highway network in real time.
引文
[1] ANDERSON Q.Storm real-time processing cookbook[M].[s.l.]:Packt Publishing,2013.
[2] KARUNARATNE P,KARUNASEKERA S,HARWOOD A.Distributed stream clustering using micro-clusters on Apache Storm[J].Journal of Parallel & Distributed Computing,2017,108:74-84.
[3] 朱美玲,王雄斌,张守利,等.基于大规模流式车牌识别数据的即时伴随车辆发现[J].中国科学技术大学学报,2016,46(1):47-55.
[4] 亓开元,赵卓峰,房俊,等.针对高速数据流的大规模数据实时处理方法[J].计算机学报,2012,35(3):477-490.
[5] 吴淑玲.基于数据挖掘的高速公路联网收费系统防逃费研究[D].广州:华南理工大学,2016.
[6] 陈海华,刘小峰.广东省高速公路联网收费背景下用户信用体系及防逃费技术的研究与应用[J].中国公共安全:学术版,2016(12):110-115.
[7] 李悦,刘晨.基于历史车牌识别数据的套牌车并行检测方法[J].计算机应用,2016,36(3):864-870.
[8] 龙伟,陈志,万亚平.基于Hadoop的高速公路联网收费稽查系统[J].西部交通科技,2014(8):73-78.
[9] KNIGHT B,VEERMAN E,DICKINSON G,et al.Professional microsoft SQL server 2008 integration services[M].Birmingham,UK:Wrox Press Ltd.,2012.
[10] 陈波.数据挖掘在高速公路防逃费中的应用研究[D].广州:华南理工大学,2014.
[11] 邓如丰.数据挖掘在联网收费系统中的应用[D].广州:华南理工大学,2013.
[12] 李小运.数据挖掘技术在高速公路联网收费稽查上的研究与应用[D].广州:华南理工大学,2014.
[13] 王涛,王顺,沈益民.交通流大数据中的套牌车并行检测算法[J].湖北工程学院学报,2014,34(6):29-32.
[14] CHEN Cheng,LIU Zhong,LIN Weihua,et al.Distributed modeling in a MapReduce framework for data-driven traffic flow forecasting[J].IEEE Transactions on Intelligent Transportation Systems,2013,14(1):22-33.
[2] KARUNARATNE P,KARUNASEKERA S,HARWOOD A.Distributed stream clustering using micro-clusters on Apache Storm[J].Journal of Parallel & Distributed Computing,2017,108:74-84.
[3] 朱美玲,王雄斌,张守利,等.基于大规模流式车牌识别数据的即时伴随车辆发现[J].中国科学技术大学学报,2016,46(1):47-55.
[4] 亓开元,赵卓峰,房俊,等.针对高速数据流的大规模数据实时处理方法[J].计算机学报,2012,35(3):477-490.
[5] 吴淑玲.基于数据挖掘的高速公路联网收费系统防逃费研究[D].广州:华南理工大学,2016.
[6] 陈海华,刘小峰.广东省高速公路联网收费背景下用户信用体系及防逃费技术的研究与应用[J].中国公共安全:学术版,2016(12):110-115.
[7] 李悦,刘晨.基于历史车牌识别数据的套牌车并行检测方法[J].计算机应用,2016,36(3):864-870.
[8] 龙伟,陈志,万亚平.基于Hadoop的高速公路联网收费稽查系统[J].西部交通科技,2014(8):73-78.
[9] KNIGHT B,VEERMAN E,DICKINSON G,et al.Professional microsoft SQL server 2008 integration services[M].Birmingham,UK:Wrox Press Ltd.,2012.
[10] 陈波.数据挖掘在高速公路防逃费中的应用研究[D].广州:华南理工大学,2014.
[11] 邓如丰.数据挖掘在联网收费系统中的应用[D].广州:华南理工大学,2013.
[12] 李小运.数据挖掘技术在高速公路联网收费稽查上的研究与应用[D].广州:华南理工大学,2014.
[13] 王涛,王顺,沈益民.交通流大数据中的套牌车并行检测算法[J].湖北工程学院学报,2014,34(6):29-32.
[14] CHEN Cheng,LIU Zhong,LIN Weihua,et al.Distributed modeling in a MapReduce framework for data-driven traffic flow forecasting[J].IEEE Transactions on Intelligent Transportation Systems,2013,14(1):22-33.