一种基于轨迹特征划分的交通轨迹数据分析方法

doi:10.13474/j.cnki.11-2246.2018.0353

测绘通报 ›› 2018, Vol. 0 ›› Issue (11): 73-77,82.doi: 10.13474/j.cnki.11-2246.2018.0353

一种基于轨迹特征划分的交通轨迹数据分析方法

赵庶旭, 屈睿涛, 刘昌荣

兰州交通大学电子与信息工程学院, 甘肃兰州 730070

收稿日期:2018-02-01 出版日期:2018-11-25 发布日期:2018-11-29
作者简介:赵庶旭(1977-),男,博士,教授,主要研究方向为交通信息工程与控制。E-mail:21319768@qq.com
基金资助:
甘肃省科技厅基金项目（1504GKCA018）

An Traffic Trajectory Data Analysis Method Based on Trajectory Feature Division

ZHAO Shuxu, QU Ruitao, LIU Changrong

School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China

Received:2018-02-01 Online:2018-11-25 Published:2018-11-29

摘要/Abstract

摘要：

当前大多交通轨迹数据的划分方法并未考虑空间分布的任意性，划分点较单一，从而导致分析效果并不理想。针对此问题，本文提出了多特征轨迹数据点结合数据空间划分方法，对提取采集到的海量交通轨迹的记录点，利用α-Shapes算法进行预处理并去噪，计算轨迹特征点，对这些特征点按空间接近度进行分组，之后按照位置关系进行Voronoi划分。该方法克服了轨迹数据划分时因空间分布任意性导致的划分效果不明显的缺点，有效地提高了轨迹数据分析效果。采用山东省淄博市出租车数据对本方法进行验证，结果证明该方法较传统数据划分方法在效果上优势明显，在轨迹数据去噪方面也做出了贡献。

关键词: 轨迹数据, α-Shapes算法, 轨迹特征点, Voronoi划分

Abstract:

Most of the current division methods of traffic trajectory data do not take into account the arbitrary nature of its spatial distribution,combined with the division of a single point,resulting in the analysis is not ideal.In order to solve this problem,a multi-feature trajectory data point combined with data space division method is proposed.The recorded points of the massively-trafficked traffic trajectory are extracted,pre-processed and denoised by α-Shapes method,and the trajectory feature points are calculated.Click the spatial proximity to group,and then according to the location of the Voronoi division.The method overcomes the shortcomings that the division effect is not obvious due to the randomness of spatial distribution when the trajectory data is divided,and effectively improves the trajectory data analysis effect.This method is validated by the taxi data of Zibo city,Shandong province.The results of the distribution heat map after the data are divided show that this method is more effective than the traditional data classification method and also contributes to the de-noising of the trajectory data.

Key words: trajectory data, α-Shapes algorithm, trajectory feature points, voronoi division

中图分类号:

P208

赵庶旭, 屈睿涛, 刘昌荣. 一种基于轨迹特征划分的交通轨迹数据分析方法[J]. 测绘通报, 2018, 0(11): 73-77,82.

ZHAO Shuxu, QU Ruitao, LIU Changrong. An Traffic Trajectory Data Analysis Method Based on Trajectory Feature Division[J]. 测绘通报, 2018, 0(11): 73-77,82.

参考文献

[1] 王祖超,袁晓如.轨迹数据可视分析研究[J].计算机辅助设计与图形学学报,2015,27(1):9-25.
[2] LEE J G,HAN J W,WHANG K Y.Trajectory Clustering:A Partition-and-group Framework[C]//Proceedings of the 2007 ACM SIG MOD International Conference on Management of Data.Beijing:[s.n.],2007:593-604.
[3] CHANG C,ZHOU B Y.Multi-granularity Visualization of Trajectory Clusters Using Sub-trajectory Clustering[C]//Proceedings of the 7th IEEE International Conference on Data Mining Workshops.Miami,Florida:IEEE,2009:577-582.
[4] 袁冠,夏士雄,张磊,等.基于结构相似度的轨迹聚类算法[J].通信学报,2011,32(9):103-110.
[5] ZHAN X Y,HASAN S,UKKUSURI S V,et al.Urban Linktravel Time Estimation Using Large-scale Taxi Data with Partial Information[J].Transportation Research Part C:Emerging Technologies,2013,33:37-49
[6] SPRETKE D,BAK P,JANETZKO H,et al.Exploration Through Enrichment:A Visual Analytics Approach for Animal Movement[C]//Proceedings of ACM SIGSPA-TIAL International Conference on Advances in Geographic Information Systems.New York:ACM Press,2011:421-424.
[7] ZHANG W,LI S,PAN G.Mining the Semantics of Origin-destination Flows Using Taxi Traces[C]//ACM Conference on Ubiquitous Computing.New York:ACM,2012:943-949.
[8] PAN G,QI G,WU Z,et al.Land-use Classification Using Taxi GPS Traces[J].IEEE Transactions on Intelligent Transportation Systems,2013,14(1):113-123.
[9] 赵晓光.基于多尺度区域划分和运动模式的车辆轨迹预测[D].北京:北京邮电大学,2017.
[10] 王亮,胡琨元,库涛,等.基于多尺度空间划分与路网建模的城市移动轨迹模式挖掘[J].自动化学报,2015,41(1):47-58.
[11] SCHEEPENS R,WETERING H V D,WIJK J J V.Non-overlapping Aggregated Multivariate Glyphs for Moving Objects[C]//Proceedings of IEEE Pacific Visualization Symposium.Los Alamitos:IEEE Computer Society Press,2014:17-24.
[12] 刘子政,李默颖,黄长青,等.顾及速度约束的基于时间序列GPS漂移数据处理方法[J].测绘地理信息,2017,42(1):14-18.
[13] 王敏芬.基于速度约束的轨迹划分方法[J].科技创新与应用,2013(26):3-4.
[14] 赵东保,刘雪梅,郭黎.网格索引支持下的大规模浮动车实时地图匹配方法.计算机辅助设计与图形学学报,2014,26(9):1550-1556.
[15] 沈蔚,李京,陈云浩,等.基于LiDAR数据的建筑轮廓线提取及规则化算法研究[J].遥感学报,2008,12(5):692-698.
[16] 范启忠,陈是扦,刁燕,等.基于Alpha Shapes三维重建原理的机器人工作空间求解方法[J].机械传动,2014(9):27-29.
[17] 王宗跃,马洪超,徐宏根,等.海量点云的边缘快速提取算法[J].计算机工程与应用,2010,46(36):213-215.
[18] 张宏,温永宁,刘爱利,等.地理信息系统算法基础[M].北京:科学出版社,2006.
[19] AN Z H,ZHANG J G.Generation Algorithm and Application of Matlab-based Voronoi Diagram with Urban Obstacles[C]//International Conference on Artificial Intelligence,Management Science and Electronic Commerce.[S.l.]:IEEE,2011:2123-2126.
[20] 徐鹏飞,陈志刚,刘刚.一种Voronoi划分减量构造算法[J].计算机工程与应用,2011,47(11):7-10.
[21] LUPU M,FERON E,MAO Z H.Traffic Complexity of Intersecting Flows of Aircraft under Variations of Pilot Preferences in Maneuver Choice[C]//49th IEEE Conference on Decision and Control.Atlanta:IEEE,2010:1189-1194.
[22] 李圣权,胡鹏,闫卫阳.基于加权Voronoi图的城市影响范围划分[J].武汉大学学报(工学版),2004,37(1):94-97.
[23] 张金秋,赵庶旭,屈睿涛.基于点与热度的交通时空数据可视化[J].兰州交通大学学报,2017,36(3):63-69.

一种基于轨迹特征划分的交通轨迹数据分析方法

An Traffic Trajectory Data Analysis Method Based on Trajectory Feature Division

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 11

编辑推荐

Metrics

本文评价

[1]	邬群勇, 王祥健. 出租车轨迹数据的频繁轨迹识别[J]. 测绘通报, 2021, 0(11): 70-75.
[2]	张昊, 张健钦, 郭小刚, 卢剑, 陆浩. 轨迹大数据云存储与热力图生成方法[J]. 测绘通报, 2021, 0(10): 146-149.
[3]	黄亚锋, 向隆刚, 高萌. 轨迹数据的非关系管理及相似性分析[J]. 测绘通报, 2020, 0(6): 81-86.
[4]	武英豪, 李成名, 吴政, 武鹏达. 基于地理格网的复杂路线车辆通行时间估算方法[J]. 测绘通报, 2019, 0(6): 61-65.
[5]	梁明, 陈文静, 段平, 李佳. 轨迹压缩的典型方法评价[J]. 测绘通报, 2019, 0(4): 60-64,70.
[6]	李金磊, 慈谕瑶, 郑坤, 郭绍龙, 燕继红, 陈宇萍, 吴艳民, 李仕漪. 面向地理信息大数据的时空事件关系可视化分析框架[J]. 测绘通报, 2019, 0(12): 101-104.
[7]	王善辉, 张和生. 一种基于车辆轨迹数据的城市建成区边界识别方法[J]. 测绘通报, 2019, 0(1): 56-59,64.
[8]	胡瀚, 向隆刚, 王德浩. 出租车轨迹数据的道路提取[J]. 测绘通报, 2018, 0(7): 53-57.
[9]	刘旭, 陈云波, 施昆, 黄强. 结合Canopy-K-means算法和出租车轨迹数据的公交车站预测方法[J]. 测绘通报, 2018, 0(11): 63-68.
[10]	牟乃夏, 徐玉静, 张恒才, 陈洁, 张灵先, 刘希亮. 移动轨迹聚类方法研究综述[J]. 测绘通报, 2018, 0(1): 1-7.
[11]	段颖超, 张健钦, 李明轩, 杜明义. 一种公交到站时间预测方法[J]. 测绘通报, 2016, 0(5): 50-53.