数据挖掘技术及其在车辆监控系统中的应用

发布时间：2018-09-02 07:47

【摘要】：随着信息化技术在道路交通领域的广泛应用,以数据挖掘技术为核心的道路车辆监控管理和辅助决策已经成为现代智能交通的重要发展方向之一。数据挖掘可以很好地解决道路交通领域中信息海量但缺乏深入研究的问题,然而,我国目前对于数据挖掘技术及其在车辆监控系统中的应用研究还处于起步阶段。基于以上背景,本文讨论了数据挖掘技术和其中的聚类分析技术,包括五种常用聚类分析技术的定义、基本思想和主要算法步骤；根据数据挖掘技术的应用场景,提出了车辆监控系统的总体设计方案,包括系统的技术指标、总体架构和技术路线；分析了车辆监控系统数据的特点,完成了车辆数据的预处理过程；将数据挖掘技术与车辆应用系统的场景相结合,提出了一种具有噪声的基于密度聚类算法的优化方法,该方法具有聚类速度快、抗噪性能良好以及可以发现任意形状的空间聚类的优点,并将改进后的聚类算法应用于车辆监控系统之中,实现了车辆监控系统中超速多发路段的发现和重点车辆信息监控的功能；本文最后展示了车辆监控系统的相关界面,分析了超速多发路段的聚类分析和重点车辆信息监控的结果。实验结果表明,改进后的聚类算法应用于车辆监控系统之中,能够有效地实现车辆监控系统中发现超速多发路段和监控重点车辆信息的功能,并显著提升了发现超速多发路段的准确性,从而为道路交通监管机构的决策提供了有力的支持。
[Abstract]:With the wide application of information technology in the field of road traffic, data mining technology as the core of the monitoring and management of road vehicles and auxiliary decision-making has become one of the important development directions of modern intelligent transportation. Data mining can solve the problem of huge amount of information but lack of in-depth research in the field of road traffic. However, the research on data mining technology and its application in vehicle monitoring system is still in its infancy in our country. Based on the above background, this paper discusses the data mining technology and clustering analysis technology, including five commonly used clustering analysis technology definition, basic ideas and main algorithm steps. The overall design scheme of the vehicle monitoring system is put forward, including the technical index, the overall structure and the technical route of the system, the characteristics of the vehicle monitoring system data are analyzed, and the preprocessing process of the vehicle data is completed. Combining the data mining technology with the scene of vehicle application system, an optimization method based on density clustering algorithm with noise is proposed, which has fast clustering speed. The improved clustering algorithm is applied to vehicle monitoring system, which has good anti-noise performance and can find the advantages of arbitrary shape spatial clustering. Realized the detection of overspeed section in the vehicle monitoring system and the function of key vehicle information monitoring. Finally, the paper showed the related interface of the vehicle monitoring system. The results of clustering analysis and monitoring of key vehicles are analyzed. The experimental results show that the improved clustering algorithm is applied to the vehicle monitoring system, which can effectively realize the function of finding overspeed sections and monitoring key vehicle information in the vehicle monitoring system. It also improves the accuracy of finding overspeed section, and provides strong support for the decision of road traffic supervision organization.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495;TP311.13

【参考文献】