主管单位:中国科学技术协会
主办单位:中国地理学会
承办单位:华东师范大学

世界地理研究 ›› 2020, Vol. 29 ›› Issue (4): 804-813.DOI: 10.3969/j.issn.1004-9479.2020.04.2019186

• 产业与布局 • 上一篇    下一篇

基于POI数据的城市用地功能识别与评价研究

窦旺胜(), 王成新(), 薛明月, 王召汉   

  1. 山东师范大学地理与环境学院/“人地协调与绿色发展”山东省高校协同创新中心,济南 250358
  • 收稿日期:2019-05-05 修回日期:2019-08-02 出版日期:2020-07-30 发布日期:2020-08-15
  • 通讯作者: 王成新
  • 作者简介:窦旺胜(1996-),男,硕士研究生,主要研究方向为区域发展与城镇规划,E-mail:1137198068@qq.com
  • 基金资助:
    山东省重点研发计划项目(GG201703150142)

Identification and evaluation of urban functional land based on POI data——A case study of five districts in Jinan

Wangsheng DOU(), Chengxin WANG(), Mingyue XUE, Zhaohan WANG   

  1. College of Geography and Environment, Human-earth Coordination and Green Development, Shandong University Collaborative Innovation Center,Shandong Normal University, Jinan 250358,China
  • Received:2019-05-05 Revised:2019-08-02 Online:2020-07-30 Published:2020-08-15
  • Contact: Chengxin WANG

摘要:

充分利用大数据开展城市用地功能识别,有助于把握城市空间结构,推动城市内部空间合理布局。POI数据是大数据时代一种较易获得且极具代表性的空间点状数据,能够有效地确定城市用地的实际功能。以济南市内五区的185126条POI数据为基础,对所得数据进行去重、纠偏、重分类,构建城市用地功能分类体系,运用频数密度、类型比例及核密度估计,识别济南市内五区城市用地功能并利用误差矩阵对识别结果进行检验。结果表明:①混合功能用地与单一功能用地呈现圈层化地域分布特征,“核心-外围”分异明显;②由内向外单一功能用地集聚趋势减弱,混合功能用地多样性降低,不同用地表现出不同的空间分布模式;③通过误差矩阵及与用地规划图中规划用地及电子地图的实际用地对比,识别总体精度为75.67%,识别结果较为准确。

关键词: 城市用地功能, POI数据, 定量识别, 济南市

Abstract:

Making full use of big data to identify urban land use function can help to grasp the urban spatial structure and promote the rational layout of urban interior space. POI data is a kind of easily available and representative spatial point-like data in the era of big data, which can effectively determine the actual function of urban land. Based on 185,126 POI data in five districts of jinan, this paper deleted duplicates, corrected deviations and reclassified the obtained data, constructed a functional classification system of urban land, and used frequency density, type ratio and kernel density estimation to identify and evaluate the urban land functions in five districts of jinan, and use the error matrix to test the recognition results.The results show that: (1) mixed functional land and single functional land presents the regional distribution characteristics of the circle, the "core-periphery" distinction is obvious; (2) from the inward to the outward, the concentration trend of single functional land is weakened, the diversity of mixed functional land is reduced, and different land shows different spatial distribution patterns; (3) Through the error matrix and the actual land use of the planned land and electronic map in the land use plan, the overall accuracy is 75.67%, and the recognition result is more accurate.

Key words: urban functional land, POI, quantitative identification, Jinan