计算机科学 ›› 2019, Vol. 46 ›› Issue (4): 14-21.doi: 10.11896/j.issn.1002-137X.2019.04.003

• 大数据与数据科学 • 上一篇    下一篇

交通事故时空模式可视分析方法

饶永明, 张延孔, 谢文军, 刘璐, 刘新月, 罗月童   

  1. 合肥工业大学计算机与信息学院 合肥230000
  • 收稿日期:2018-10-31 出版日期:2019-04-15 发布日期:2019-04-23
  • 通讯作者: 罗月童(1978-),男,博士,教授,硕士生导师,主要研究方向为可视分析、科学可视化、信息可视化,E-mail:ytluo@hfut.edu.cn(通信作者)。
  • 作者简介:饶永明(1993-),男,硕士生,主要研究方向为可视分析、图像处理;张延孔(1990-),男,讲师,主要研究方向为信息可视化;谢文军(1984-),男,博士,教授,硕士生导师,主要研究方向为计算机图形学;刘 璐(1996-),女,硕士生,主要研究方向为可视分析、信息可视化;刘新月(1995-),男,硕士生,主要研究方向为可视分析、信息可视化;
  • 基金资助:
    本文受国家重点研发项目(2017YFB1402200),安徽省科技强警计划项目(1604d0802009),浙江大学CAD&CG国家重点实验室开放课题(A1814),中央高校基本科研业务费专项资金(JZ2017HGBH0915),安徽省高等学校省级质量工程项目(2017jyxm0045)资助。

Visual Analysis Method of Traffic Accident Spatial-Temporal Pattern

RAO Yong-ming, ZHANG Yan-kong, XIE Wen-jun, LIU Lu, LIU Xin-yue, LUO Yue-tong   

  1. School of Computer Science and Information Engineering,Hefei University of Technology,Hefei 230000,China
  • Received:2018-10-31 Online:2019-04-15 Published:2019-04-23

摘要: 随着城市化进程的推进,城市人口和车辆迅速增长,城市交通事故日益频发,成为社会关注的热点。以合肥市近十年的交通事故记录数据为研究对象,运用可视分析方法分析交通事故记录数据中事故发生的时间和地点信息,探究交通事故的时空模式,构建交通事故可视分析系统,以辅助相关部门改善交通事故频发问题。文中首次提出了道路事故危险度的概念,并以之为判定依据,结合多尺度时间统计折线图和周期性时间统计环形图等可视化方法,构建了一种新的事故多发路段的识别方法。与传统事故多发路段识别方法相比,本方法无需对道路进行分段处理,从而避免了分段优劣对识别结果的影响。在此基础上,将交通事故数据与城市路网数据相结合,运用可视分析技术构建交通事故可视分析系统。本系统可以帮助相关部门了解总体城市交通事故和单条道路的时间模式及事故多发路段,并探究连续时间限定或周期时间限定下的事故多发路段。除时间条件外,本系统还能识别不同天气等其他限定条件下的事故多发路段,从而使得交警部门能根据不同情况下的道路事故危险度来进行决策管理,并合理部署救援警力,降低事故危害。所提系统对缓解和遏制交通事故增长势头、减少和预防道路交通事故具有重要的现实意义,并且也有利于道路交通的科学有效管理。

关键词: 道路分段, 道路事故危险度, 交通事故, 可视分析, 事故多发路段

Abstract: With the advancement of urbanization and the rapid growth of urban population and vehicles,urban traffic accidents are increasingly frequent,which becomes a hot social concern.This paper took the traffic accident record data of Hefei City in the past ten years as the research object,used the visual analysis method to analyze the accident time and the place information in the traffic accident record data,explored the time and space pattern of the traffic accident,and constructed the traffic accident analysis system,so as to assist relevant departments to improve frequent traffic accidents.In this paper,the concept of road accident risk degree was put forward for the first time,and a new identification method of accident-prone road section was constructed based on multi-scale time statistical zigzag chart and periodic time statistical circle map.Compared with the traditional method,this method does not need to deal with the road segmentation,thus avoiding the impact of the quality of the segmentation on the recognition results.On this basis,this paper combined traffic accident data with urban road network data,and used visual analysis technology to build a visual analysis system of traffic accidents.This system can help relevant departments to understand the time pattern of the overall urban traffic accidents and single road and the accident-prone sections,and explore the accident-prone sections under the continuous time limit or cycle time limit.In addition to time conditions,the system can also identify accident-prone sections under different weather and other limited conditions,so that traffic police departments can make decision management through road accident risk under different circumstances,deploy rescue police forces reasonably and reduce accident hazards.The proposed system has important practical significance for mitigating and curbing the growth of traffic accidents,reducing and preventing road traffic accidents.It is also conducive to the scientific and effectivemana-gement of road traffic.

Key words: Black spot, Road accident risk degree, Road segmentation, Traffic accident, Visual analysis

中图分类号: 

  • TP391.9
[1]COOK K A,THOMAS J J.Illuminating the Path:The Research and Development Agenda for Visual Analytics[M].National Visualization and Analytics Ctr,2005.
[2]ZHU Q,FU X.An overview of multimodal spatial and temporal visual analysis methods [J].Journal of Surveying and Mapping,2017,46(10):1672-1677.(in Chinese)
朱庆,付萧.多模态时空大数据可视分析方法综述[J].测绘学报,2017,46(10):1672-1677.
[3]REN L,DU Y,MA S,et al.Overview of visual analysis of big data [J].Journal of Software,2014,25(9):1909-1936.(in Chinese)
任磊,杜一,马帅,等.大数据可视分析综述[J].软件学报,2014,25(9):1909-1936.
[4]ZHANG J,YANLI E,MA J,et al.Visual Analysis of Public
Utility Service Problems in a Metropolis[J].IEEE Transactions on Visualization & Computer Graphics,2014,20(12):1843-1852.
[5]CAO N,LIN Y R,SUN X,et al.Whisper:Tracing the Spatiotemporal Process of Information Diffusion in Real Time[J].IEEE Transactions on Visualization & Computer Graphics,2012,18(12):2649-2658.
[6]ZHOU Z G,HU D X,LIU Y N,et al.Visual analysis of spatial and temporal multidimensional properties of air quality monitoring data [J].Journal of computer-aided design and graphics,2017,29(8):1477-1487.(in Chinese)
周志光,胡迪欣,刘亚楠,等.面向空气质量监测数据时空多维属性的可视分析方法[J].计算机辅助设计与图形学学报,2017,29(8):1477-1487.
[7]WANG Z,YE T,LU M,et al.Visual Exploration of Sparse Traffic Trajectory Data[J].IEEE Transactions on Visualization &Computer Graphics,2014,20(12):1813-1822.
[8]GUO H,WANG Z,YU B,et al.TripVista:Triple Perspective
Visual Trajectory Analytics and its application on microscopic traffic data at a road intersection[C]∥IEEE Pacific Visualization Symposium.IEEE Computer Society,2011:163-170.
[9]WANG Z,LU M,YUAN X,et al.Visual Traffic Jam Analysis Based on Trajectory Data[J].IEEE Transactions on Visualization & Computer Graphics,2013,19(12):2159.
[10]LU M,WANG Z,LIANG J,et al.OD-Wheel:Visual design to explore OD patterns of a central region[C]∥Visualization Symposium.IEEE,2015:87-91.
[11]LU M,WANG Z,YUAN X.TrajRank:Exploring travel behaviour on a route by trajectory ranking[C]∥Proceedings of IEEE Pacific Visualization Symposium (Pacific Vis’15).IEEE,2015:311-318.
[12]ZHANG J Q,ZHAO S X,QU R T.Spatial and temporal data visualization based on point and heat [J].Journal of Lanzhou Jiaotong University,2017,36(3):63-69.(in Chinese)
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.
[13]PACK M L,WONGSUPHASAWAT K,VANDANIKER M,et al.
ICE--visual analytics for transportation incident datasets[C]∥IEEE International Conference on Information Reuse & Integration.IEEE,2009:200-205.
[14]PIRINGER H,BUCHETICS M,BENEDIK R.AlVis:Situation awareness in the surveillance of road tunnels[C]∥Visual Analytics Science and Technology.IEEE,2013:153-162.
[15]FAN X,HE B,PATRICK B.Context-Aware Big Data Analytics and Visualization for City-Wide Traffic Accidents[C]∥ International & Interdisciplinary Conference on Modeling & Using Context.Cham:Springer,2017.
[16]SHAFABAKHSH G A,FAMILI A,BAHADORI M S.GIS-
based spatial analysis of urban traffic accidents:Case study in Mashhad,Iran[J].Journal of Traffic and Transportation Engineering (English Edition),2017(3):82-91.
[17]CHEN Q,SONG X,YAMADA H,et al.Learning deep representation from big and heterogeneous data for traffic accident inference[C]∥Thirtieth AAAI Conference on Artificial Intelligence.2016:338-344.
[18]LU H S,MAO Z J,ZHONG T Y,et al.Study on the linear intelligent screening system based on GIS-T [C]∥International Urban Transportation Academic Conference.2011.(in Chinese)
卢辉恕,毛志坚,钟天宇,等.基于GIS-T的事故多发点段线性智能排查系统研究[C]∥多国城市交通学术会议.2011.
[1] 温啸林, 李长林, 张馨艺, 刘尚松, 朱敏.
基于DPoS共识机制的区块链社区演化的可视分析方法
Visual Analysis Method of Blockchain Community Evolution Based on DPoS Consensus Mechanism
计算机科学, 2022, 49(1): 328-335. https://doi.org/10.11896/jsjkx.201200118
[2] 罗月童, 汪涛, 杨梦男, 张延孔.
基于历史行车轨迹集的车辆行为可视分析方法
Historical Driving Track Set Based Visual Vehicle Behavior Analytic Method
计算机科学, 2021, 48(9): 86-94. https://doi.org/10.11896/jsjkx.200900040
[3] 姚林, 王翔坤, 贾钰沛, 耿仕洪, 朱敏.
基于气象因子的气候区划可视分析系统
Visual Analysis System of Climatic Regionalization Based on Meteorological Factors
计算机科学, 2021, 48(6A): 552-557. https://doi.org/10.11896/jsjkx.200900127
[4] 白雪, 努尔布力, 王亚东.
网络安全态势感知研究现状与发展趋势的图谱分析
Map Analysis for Research Status and Development Trend on Network Security Situational Awareness
计算机科学, 2020, 47(6A): 340-343. https://doi.org/10.11896/JsJkx.190500169
[5] 贾熹滨,叶颖婕,陈军成.
基于关联规则的交通事故影响因素的挖掘
Influence Factors Mining of Traffic Accidents Based on Association Rules
计算机科学, 2018, 45(6A): 447-452.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!