基于多源数据的城市功能区识别方法  

A method for identifying urban functional areas based on multi-source data

作　　者：王雪琴 刘岳峰[1] WANG Xueqin;LIU Yuefeng(Institute of Remote Sensing and Geographic Information System,Peking University,Beijing 100871,China)

机构地区：[1]北京大学遥感与地理信息系统研究所,北京100871

出　　处：《时空信息学报》2023年第1期18-24,共7页JOURNAL OF SPATIO-TEMPORAL INFORMATION

摘　　要：本文提出一种基于多源数据的城市功能区识别方法,并以成都为例进行了研究。首先,利用路网数据将研究区域划分为街区单元;然后,利用文本表示模型对出租车轨迹数据和兴趣点数据进行挖掘,提取街区的社会经济活动特征;再针对建筑轮廓矢量数据提取建筑景观指数,作为街区的自然特征;最后,基于街区的自然特征和社会经济活动特征采用高斯混合模型对街区进行聚类分析,识别得到居住区、商业区、工业区、科教区、文娱区和生态区六类功能区。结合百度地图和谷歌影像地图,对识别结果进行精度评价。结果表明,本方法在城市功能区识别上有较高的准确率,能为城市的发展规划提供优化建议。In this paper,we propose a method for identifying urban functional areas based on multi-source data and conduct a case study in Chengdu.Firstly,we divide the study area into block units using road network data.Then,we extract the socio-economic activity features of each block by utilizing a text representation model to analyze taxi trajectory data and point of interest data.We also extract the building landscape metrics as a natural feature of each block from the building outline vector data.Finally,we employ a Gaussian mixture model to perform clustering analysis on the blocks based on their natural and socio-economic activity features,and identify six types of functional areas,including residential,commercial,industrial,scientific and educational,cultural and recreational,and ecological areas.We evaluate the accuracy of the identification results using Baidu map and Google satellite imagery.The results indicate that our method achieves high accuracy in identifying urban functional areas and can provide optimization suggestions for urban development planning.

关 键 词：城市功能区 兴趣点 出租车轨迹 景观指数 

分 类 号：P208[天文地球—地图制图学与地理信息工程]