基于DBSCAN密度聚类算法的高速公路交通流异常数据检测

详细信息    查看全文 | 推荐本文 |

英文篇名：Highway Traffic Flow Anomaly Data Detection Based on DBSCAN Density Clustering Algorithm 作者：阮嘉琨 ; 蔡延光 ; 乐冰 关键词：DBSCAN ; 密度聚类 ; 高速公路交通流 ; 异常数据检测 英文关键词：DBSCAN;;density clustering;;highway traffic flow;;abnormal data detection 中文刊名：GYKJ 英文刊名：Industrial Control Computer 机构：广东工业大学自动化学院; 出版日期：2019-07-25 出版单位：工业控制计算机 年：2019 期：v.32 基金：国家自然科学基金(61074147);; 广东省自然科学基金(S2011010005059);; 广东省教育部产学研结合项目(2012B091000171,2011B090400460);; 广东省科技计划项目(2012B050600028,2014B010118004,2016A050502060);; 广州市花都区科技计划项目(HD14ZD001);; 广州市科技计划项目(201604016055);; 广州市天河区科技计划项目(2018CX005) 语种：中文; 页：GYKJ201907039 页数：3 CN：07 ISSN：32-1764/TP 分类号：96-98

摘要

由于高速公路交通流数据的复杂多变性以及随机波动性强,而导致传统的异常数据识别方法很难准确检测出其交通流异常数据,提出了采用DBSCAN密度聚类算法来检测高速公路交通流异常数据。DBSCAN密度聚类算法能够有效地对高速路交通流数据进行准确地分类而分离出异常样本,从而检测出其异常交通流数据。结合实验表明,该高速路交通流异常数据检测方法达到了较好的效果,能够满足实际路况的检测需求。
        This paper proposes to use DBSCAN density clustering algorithm to detect high speed.Abnormal data on highway traffic flow.The DBSCAN density clustering algorithm can effectively classify the highway traffic flow data and separate the abnormal samples to detect the abnormal traffic flow data.Experiments show that the traffic flow anomaly data detection method of the expressway has achieved good results.

引文

[1]Yildirimoglu M,Geroliminis N.Experienced travel time prediction for congested freeways[J].Transportation Research,Part B:Methodological,2013,53(4):45-63
    [2]Habtemichael F G,Cetin M. Short-term traffic flow rate forecasting based on identifying similar traffic patterns[J].Transportation Research,Part C:Emerging Technologies,2016,66(5):61-78
    [3]刘国华,吕晓柯,石晨,等.初速数据判别方法研究[J].火炮发射与控制学报,2013(1):8-10,24
    [4]徐晨.异常数据均匀分布检验法的研究[D].沈阳:东北大学,2006
    [5]宋震龙,蒋刚毅,黄晁,等.基于偏度-峰度检验的无线局域网室内定位算法[J].通信学报,2012,33(5):99-105
    [6]Nath R P D,Lee H J,Chowdhury N K,et al.Modified Kmeans clustering for travel time prediction based on historical traffic data[C]∥Proc of the 14th IEEE International Conference on KnowledgeBased and Intelligent Information and Engineering Systems. Berlin:Springer,2010:511-521
    [7]Zhang Ting,Xia Yingjie,Zhu Qianqian,et al.Mining related information of traffic flows on lanes by K-medoids[C]∥Proc of the 11th IEEE International Conference on Fuzzy Systems and Knowledge Discovery. Piscataway,NJ:IEEE Press,2014:390-396
    [8]Cleverdon C W.On the inverse relationship of recall and precision[J]. Journal of Documentaion, 1972,28(3):195-201
    [9]蒋华,季丰,王鑫,等.基于DCNDA算法的数据异常检测[J].计算机工程与设计,2018,39(11):3425-3430,3485
    [10]杨先圣,姜磊,彭雄,等.基于大数据的异常检测方法研究[J].计算机工程与科学,2018,40(7):1180-1186
    [11]伍育红.聚类算法综述[J].计算机科学,2015,42(S1):491-499,524
    [12]Salamanis A,Margaritis G,Kehagias D D,et al.Identifying patterns under both normal and abnormal traffic conditions for short-term traffic prediction[J]. Transportation Research Procedia, 2017, 22:665-674