基于出租车轨迹数据的交通异常识别算法

详细信息    查看全文 | 推荐本文 |

英文篇名：Traffic Anomaly Recognition Algorithm Based on Taxi Track Date 作者：王雷 ; 安实 ; 杨海强 ; 马晓龙 英文作者：WANG Lei;AN Shi;YANG Hai-qiang;MA Xiao-long;School of Transportation Science and Engineering,Harbin Institute of Technology;Qingdao Hisense Trans Tech Co.,Ltd.; 关键词：交通异常识别算法 ; 多光谱分隔算法 ; 路径选择模式 ; 区域轨迹模式 ; 交通异常图 英文关键词：traffic anomaly recognition algorithm;;multispectral separation algorithm;;path selection model;;regional trajectory model;;traffic anomaly map 中文刊名：KXJS 英文刊名：Science Technology and Engineering 机构：哈尔滨工业大学交通科学与工程学院;青岛海信网络科技股份有限公司; 出版日期：2018-11-18 出版单位：科学技术与工程 年：2018 期：v.18;No.465 基金：国家自然科学基金(51478151)资助 语种：中文; 页：KXJS201832037 页数：9 CN：32 ISSN：11-4688/T 分类号：244-252

摘要

为了实现基于出租车轨迹数据的交通异常识别,首先以城市栅格地图模型为框架,提出了一种针对城市路网的多光谱分隔算法;并根据城市路网分别从区域增长与区域融合两种角度,实现了多光谱地图的分割。其次在分割的城市路网基础上,设计了交通异常的识别算法。算法依据单元区域内道路网络拓扑结构构建交通异常图;然后根据出租车路径选择模式的历史规律,计算每个单元区域内不同路径上的出租车轨迹流量的变化;最后根据三倍均方差指标识别单元区域内的交通异常。最后以哈尔滨为例,进行了算例分析。算例结果表明,提出的异常识别算法取得了良好的效果,验证了算法的有效性及准确性。
        In order to realize the traffic anomaly recognition based on the taxi track data,the urban grid map model was taken as the framework firstly,a multi spectral separation algorithm for urban road network was proposed,and the multi spectral map segmentation was realized according to the urban road network from two angles of regional growth and regional integration. Secondly,based on the segmentation of urban road network,a traffic anomaly recognition algorithm was designed. According to the road network topology in the unit area,the algorithm constructed the traffic anomaly map,and then calculated the change of the taxi path flow on the different paths in each unit area according to the history law of the taxi path selection model. The traffic anomalies in the unit area were identified based on the three times mean square error index. Finally,a complete urban road network traffic anomaly map was constructed. At the end,the example of Harbin was taken as an example. The results show that the proposed anomaly recognition algorithm has achieved good results and verifies the effectiveness and accuracy of the algorithm.

引文

1韦旭棉.基于固定型检测器的高速公路自动事件检测算法研究.济南:山东大学,2011Wei Xumian.Automatic incident detection algorithm for freeway based on fixed detector.Jinan:Shandong University,2011
    2 Persaud B N,Hall F L.Catastrophe theory and patterns in 30 second freeway traffic data,implications for incident detection.Transportation Research Part A:General,1989;23(2):103-113
    3 Chew R L,Ritchie S G.Automatic detection of lane-blocking freeway incidents using artificial neural networks.Transportation Research:Part C,1995;(6):371-388
    4 Abdulhai B,Ritchie S G.Enhancing the university and transferability of freeway incident detection using a Bayesian-based neural network.Transportation Research:Part C,1996;7:261-280
    5 Lee J G,Han J,Li X.Trajectory outlier detection:A partition-anddetect framework.IEEE International Conference on Data Engineering.New York:IEEE Computer Society,2008:140-149
    6 Ge Y,Xiong H,Zhou Z H,et al.Top-Eye:Top-k evolving trajectory outlier detection.Proceedings of the 19th ACM International Conference on Information and Knowledge Management.New York:ACM,2010:1733-1736
    7叶敏.基于轨迹数据挖掘的异常检测方法研究.西安:长安大学,2017Ye Min.An anomaly detection method based on trajectory data mining.Xi'an:Chang'an University,2017
    8朱绍辉.基于强相关子网的交通瘫痪预警系统研究与设计.武汉:华中科技大学,2016Zhu Shaohui.Research and design of traffic paralysis early warning system based on strongly correlated subnet.Wuhan:Huazhong University of Science and Technology,2016
    9蒋文娟,张亚娟,刘经天,等.混合动态纹理与李群论相结合的交通异常事件监测技术.科学技术与工程,2017;17(26):86-90Jiang Wenjuan,Zhang Yajuan,Liu Jingtian,et al.Traffic anomaly monitoring techniques combined with mixed dynamic texture and Lee group theory.Science Technology and Engineering,2017;17(26):86-90
    10帅丹.道路交通异常事件检测及关键帧提取算法研究.重庆:重庆交通大学,2016Shuai Dan.Road traffic incident detection and key frame extraction algorithm.Chongqing:Chongqing Jiaotong University,2016