基于Django的交通事故数据可视分析系统设计与实现  

Design and Implementation of Traffic Accident Data Visual Analysis System Based on Django

作　　者：张同硕 廖明军 张荣华 黄伟东 邵明宇 ZHANG Tongshuo;LIAO Mingjun;ZHANG Ronghua;HUANG Weidong;SHAO Mingyu(School of Mechanical Engineering,Yancheng Institute of Technology,Yancheng 224051,China;Yancheng City Planning Municipal Design Institute Co.Ltd.,Yancheng 224055,China)

机构地区：[1]盐城工学院机械工程学院,江苏盐城224051 [2]盐城市规划市政设计院有限公司,江苏盐城224055

出　　处：《软件导刊》2023年第7期112-117,共6页Software Guide

基　　金：江苏省产学研合作项目(BY2022458);盐城市创新创业领军人才项目(yccx2019219);盐城工学院校级科研项目(xjr2019029)。

摘　　要：目前,交通管理部门已经积累了大量的交通事故数据,但由于数据分析、可视技术等手段的限制,尚未完全实现对大量事故数据应用及事故产生原因和事故预防决策的进一步研究。因此,采用一种新型的分析挖掘方式,搭建Web交通事故可视分析网站,网站后端采用Django架构开发,将Bootstrap作为系统前端框架。利用Echarts结合百度地图API进行数据多维可视化,实现对时间分析、空间分析、动态分析、事故预测等多维可视分析功能;使用关系型数据库MySQL存储数据,Redis数据库作为系统缓存;基于Logistic算法对历史交通事故数据的碰撞类型、季度、天气、是否周末、路段等字段进行训练建模,实现未来某天、某路段可能发生的交通事故次数预测。结果表明,使用该系统不需要掌握复杂的数据分析方法,只需通过界面上的功能按钮,进行相关选项组合即可实现对数据的可视分析,极大减少交通领域人员在可视分析技术学习上的时间、精力投入。At present,traffic management departments have accumulated a large number of traffic accident data,but due to the limitations of data analysis,visual technology and other means,has not fully realized the application of a large number of accident data and the cause of the accident and accident prevention decision further research.So this paper adopts a new analysis mining way-web traffic accident visual analysis website.The back end of the website is developed using Django architecture,and the front end uses Echarts combined with Baidu Map API to carry out multi-dimensional data visualization,realizing multi-dimensional visual analysis functions such as time analysis,spatial analysis,dynamic analysis and accident prediction.Relational database MySQL is used to store data,and Redis database is used as system cache.Based on the logistic algorithm,training modeling is carried out on the date,month,quarter,weather,road section and other fields of histori⁃cal traffic accident data,and the number of traffic accidents that may occur on a certain day and road section in the future is predicted.The re⁃sults show that the system does not need to master complex data analysis methods,but only needs to combine related options through the func⁃tion buttons on the interface to realize visual analysis of data,which greatly reduces the time and energy investment of transportation personnel in visual analysis technology learning.

关 键 词：DJANGO 交通事故 WEB网站 可视分析 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]