基于出租车热点骨架分析的城市结构提取

Urban Structure Extraction Based on the Skeleton Analysis of Taxi Hotspot

出租车作为城市公共交通的重要组成部分,其活跃程度已经成为了能够反映城市结构与交通状况的因素之一。出租车车载GPS系统累积了海量的出租车轨迹数据,这些数据中包括了乘客乘车行为的起始地点与乘车时间,同时也隐含了城市空间功能结构信息。本文在统计出租车运营特征指标的基础上,结合上海市空间地理信息,使用基于核密度的空间聚类方法,分析了上海市域内出租车乘客出行行为的时空分布规律。根据聚类结果生成出租车热点图,并在Matlab中采用骨架分析的方法生成热点图主轴线,将主轴线分布与上海市新的城市规划结合分析,发现出租车的热点分布、核心车流向与上海市新规划中的城市副中心分布相吻合。

As an important part of urban public transport, taxis have become one of the factors which can reflect the urban structure and traffic condition. Taxi GPS system has accumulated massive taxi trajectory data. These data not only include the behavior of passenger, but also reveal the function of urban spatial structure. Based on the statistics of the taxi operation characteristic index, this paper uses spatial clustering method of kernel density combined with the spatial geographic information of Shanghai, and analyzes taxi passenger travel behaviors and spatial-time distribution in Shanghai. According to the clustering results, taxi hotspot map was generated, and by the method of skeleton analyses on matlab, the main transport axis of Shanghai were calculated. Compared with the new city planning of Shanghai,the results of taxi hotspot and the core taxi flow distribution match well the new city center of Shanghai new city planning.