基于核密度与融合数据的城市功能分区研究

A Study of Urban Functional Zoning Based on Kernel Density Estimation and Fusion Data

识别城市功能区并理解其分布特征,对于把握城市结构及科学制订城市规划具有重要作用。该文基于2015年上海市15.4万条有效POI数据与道路网数据,结合地理学第一定律,利用核密度分析技术,构建了一个城市功能区自动化分析模型。结果显示,在研究单元划分阶段,以路网形成的不规则格网为研究单元使城市功能区分割更合理;在定量分析阶段,相比于传统城市功能区识别模型,融合核密度分析的模型充分利用了POI点的空间自相关性,不仅实现了对无数据区域较为准确的识别,而且显著提高了对单一功能区与混合功能区的识别精度。

Identifying urban functional areas and understanding their distribution characteristics plays an important role in grasping urban structure and formulating the scientific and rational urban planning.In this paper,an automated analysis model on urban functional zones was constructed based on kernel density analysis method and the first law of geography.This model was verified with 0.154 million points of interest(POIs) data and road network data in Shanghai in 2015.In the generation phase of the research units,the result showed that the division of urban functional areas based on the irregular road network was effective.Compared with the traditional recognition model,the proposed model made full use of the spatial autocorrelation of POIs,which not only achieved more accurate identification of no data areas,but also significantly improved the recognition accuracy of single functional areas and mixed functional areas.