基于兴趣点数据的南昌城市功能区识别  

Nanchang City Functional Area Recognition Based on POI Data

作　　者：李岳 邓志杰 王毓乾[1,2,3] LI Yue;DENG Zhijie;WANG Yuqian(Key Laboratory for Digital Land and Resources of Jiangxi Province,East China University of Technology,330013,Nanchang,PRC;Faculty of Geomatics,East China University of Technology,330013,Nanchang,PRC;Key Laboratory of Mine Environmental Monitoring and Improving around Poyang Lake,Ministry of Natural Resources,330013,Nanchang,PRC)

机构地区：[1]东华理工大学江西省数字国土重点实验室,南昌330013 [2]东华理工大学测绘工程学院,南昌330013 [3]自然资源部环鄱阳湖区域矿山环境监测与治理重点实验室,南昌330013

出　　处：《江西科学》2022年第3期508-513,共6页Jiangxi Science

基　　金：东华理工大学江西省数字国土重点实验室开放研究基金资助项目(DLLJ201915);江西省自然科学基金项目(20202BABL202045)。

摘　　要：城市功能区对于城市规划和发展有着重要意义,随着中国经济在快速增长,城市功能区也在迅速变化,互联网大数据为城市的功能区识别和空间布局分析提供了新的方法。兴趣点(Point of interest,POI)数据是比较常见且容易获取的地理空间大数据,能够真实有效地反映社会经济活动,满足城市空间布局的要求。以南昌市为研究区域,基于高德地图POI,对所得数据进行筛选、清洗、重分类,得到城市功能用地数据,利用频数密度和类型比例识别城市功能区,通过构建混淆矩阵对单一功能区识别结果进行检验,然后利用高德地图对混合功能区识别结果进行目视解译验证。实验结果表明:1)单一功能区在南昌市分布最多且集中分布于城市郊区和城市核心之间,共计4187个网格,占比82.29%,混合功能区主要分布在城市核心地带,共901个网格,占比17.71%;2)核心区域多以交通、公共和商业混合用地为主;3)单一功能区总体精度为85.00%,Kappa系数为0.80,混合功能区目视解译验证结果与实际城市功能区一致,说明识别结果可信。Urban functional areas are significant to urban planning and development.With the rapid growth of China’s economy,urban functional areas are also changing rapidly.Internet big data provides new methods for urban functional area identification and spatial layout analysis.Point of interest(POI)is a relatively common and available geospatial big data,which can truly and effectively reflect social and economic activities and fulfil the requirements of urban spatial layout.This paper takes Nanchang city as the research area.The POI data obtained from Amap is filtered,cleaned,and reclassified to obtain urban functional land data.The frequency density and type ratio are used to identify urban functional areas.The accuracies of single functional areas identify are evaluated by a confusion matrix,while the accuracies of mixed functional area identify are evaluated by visually interpret on the Amap.The experimental results show that:1)The single functional areas are the most widely distributed in Nanchang and concentrated between the suburbs and the urban core,with a total of 4187 grids,accounting for 82.29%.The mixed functional areas are mainly distributed in the urban core with a total of 901 grids,accounting for 17.71%;2)The core urban areas are mostly mixed land for transportation,public and commercial use;3)The overall accuracy of a single functional area is 85.00%,the Kappa coefficient is 0.80,and the recognition result is credible.

关 键 词：城市功能区识别 兴趣点 频数密度 类型比例 南昌市 

分 类 号：P237[天文地球—摄影测量与遥感]