基于POI数据的城市功能区识别方法研究  

Research on identifying urban regions of different functions based on POI data

作　　者：蒋云良 董墨萱[1] 范婧 高少文 刘勇 马新强[4] 

机构地区：[1]浙江师范大学数理与信息工程学院,浙江金华321004 [2]湖州师范学院信息工程学院,浙江湖州313000 [3]浙江大学控制科学与工程学院,浙江杭州310027 [4]重庆文理学院机器视觉与智能信息系统重点实验室,重庆402160

出　　处：《浙江师范大学学报（自然科学版）》2017年第4期398-405,共8页Journal of Zhejiang Normal University:Natural Sciences

基　　金：国家自然科学基金资助项目(61370173)

摘　　要：提出了基于兴趣点数据对城市不同区域的功能进行识别的思想,根据手机基站位置将城市划分为基本单元,对基本单元中的兴趣点统计数据提出3种处理方案,并进行了模糊聚类分析,通过带有明显类别特征的兴趣点分布与聚类结果进行重叠率计算,从而确定了区域功能.为验证方法的有效性,选取浙江省杭州市一定范围内的城市区域进行实验.实验结果表明:根据兴趣点数据对城市功能区进行识别,能较好地实现城市区域的功能划分与特征分析,有助于对城市发展作出建设性规划.Due to the rapid development and evolution of cities,the functional area of city became different from early planning. Decision makers often did not know the current spatial structure of the city quickly and accurately. However,the data gathered from city brought some new thoughts of understanding the city life to people,people could identify urban regions of different functions from POI（ Point of Interest） data. First,one could divide urban space with the locations of mobile base stations. Second,fuzzy clustering could be used to analyze the POI data which could deal with three different plans. At last,one could identify the function of the result from fuzzy clustering by the distribution of POI data with noticeable features. The method was evaluated in a certain area of Hangzhou,Zhejiang Province. The results justified that identifying urban regions of different functions from POI data succeded in dividing urban regions and feature analysis,and provided technical support for urban structure layout and land use,and provided realistic basis for effective use of urban space.

关 键 词：城市计算 城市功能区 兴趣点 归一化 模糊聚类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]