基于交通轨迹数据的城市区域划分可视化研究

发布时间：2018-03-28 14:09

本文选题：交通轨迹数据　切入点：可视化　出处：《兰州交通大学》2017年硕士论文

【摘要】：随着城市道路监控设备的完善、无线通信技术的不断改进以及车载定位终端设备的使用,产生了大量包含时间、位置等信息的交通轨迹数据。交通轨迹数据记载了车辆移动、居民出行的相关信息,并且具有数据量多、密度大、快速增长等特点,已成为重要的数据资源,不仅能够从中分析出一个城市道路交通运行状况(如道路拥堵水平),而且还可以分析出对象移动的特点(如在什么地方停留、经常去什么地方等)以及城市交通区域变化情况(如热点位置),因此,针对交通轨迹数据的研究,可以应用到许多重要领域,如气象、城市规划、旅游等部门。现阶段亟需有效的方法对数据隐藏的规律性信息进行挖掘,而可视化技术的应运而生,使得大数据更加易于处理与分析。基于此,本文以山东省淄博市张店区与临淄区的浮动车GPS轨迹数据作为研究样例,结合张店区与临淄区的路网信息展开研究。论文主要研究内容如下:(1)根据GPS轨迹数据与路网数据的特点,规定了GPS轨迹数据清洗条件、坐标转换机制以及路网数据二次优化遵循的原则。修正后的数据为后续的可视化界面仿真设计以及得到可视化研究结果提供了数据支撑;(2)提出相关的计算分析算法,主要包括地图匹配与聚类分析。叙述了地图匹配基本思想与过程,对复杂路段进行了重点匹配分析;聚类分析是选取不同的实验参数将匹配后的数据在路网中进行聚类,得出最优的结果;(3)提出可视化平台设计采取的方法,设计相关的评价标准得出城市区域划分模型;并与百度地图显示的城市实际布局作对比分析。本文的可视化研究成果可以为城市规划、政策制定以及指导居民出行等各个领域提供参考性意见,以此解决现在城市道路规划存在的客观问题以及引导如何健康地规划未来城市的发展。
[Abstract]:With the improvement of urban road monitoring equipment, the continuous improvement of wireless communication technology and the use of vehicle positioning terminal equipment, a large number of traffic trajectory data, including time, location and other information, have been generated. Residents travel related information, and with a large amount of data, density, rapid growth and other characteristics, has become an important data resource, Not only can we analyze the traffic status of a city road (such as road congestion level), but we can also analyze the characteristics of object movement (such as where to stay, As well as changes in urban traffic regions (such as hot spots, etc.), research on traffic trajectory data can be applied to many important areas, such as meteorology, urban planning, and so on. Tourism and other departments. At this stage, there is an urgent need for effective methods to mine the regular information of data hiding. With the emergence of visualization technology, big data is made easier to process and analyze. In this paper, the GPS track data of floating vehicle in Zhangdian District and Linzi District, Zibo City, Shandong Province, are taken as example. According to the characteristics of GPS track data and road network data, the cleaning conditions of GPS track data are defined. The mechanism of coordinate transformation and the principle of quadratic optimization of road network data are followed. The revised data provide data support for the design of visual interface simulation and the result of visualization research. It mainly includes map matching and clustering analysis. The basic idea and process of map matching are described, and the key matching analysis is carried out on complex road sections. Cluster analysis is to select different experimental parameters to cluster the matched data in the road network. The method of visual platform design is put forward, and the relevant evaluation criteria are designed to get the model of urban area division. And compared with the actual layout of the city displayed by Baidu map, the visual research results of this paper can provide reference advice for urban planning, policy formulation and guiding residents to travel, etc. In order to solve the existing objective problems of urban road planning and how to guide the healthy planning of future urban development.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U495;TP311.13

【参考文献】