基于夜间灯光遥感数据多尺度城市聚类分析
发布时间：2018-11-27 19:34
【摘要】：先前的研究已经证明从DMSP/OLS(Defense Meteorological Satellite Program/Operational Linescan System)稳定夜间灯光数据中可以提取城市建成区,并且以此估算了人口、GDP等其他各项指标,但对提取出的城市建成区空间分布模式研究较少。本文提出了一种改进的基于密度的聚类(Density-based Spatial Clustering of Qpplications with Boise,DBSCAN)算法,将对于点要素的聚类扩展到面要素,以面状对象的边缘作为界定包含关系的准则;确定了聚类参数,以要素间距离的突变点作为距离参数。按此方法对夜间灯光数据提取的中国城市对象按照密度分布的特点进行聚类,通过改变距离参数得到在不同尺度下中国城市的集聚形态。通过与实证资料的对比验证了该算法的有效性,为研究中国城市的空间分布及其演变提供了有力的研究方法。
[Abstract]:Previous studies have shown that urban built-up areas can be extracted from DMSP/OLS (Defense Meteorological Satellite Program/Operational Linescan System) stable night light data, and other indicators such as population, GDP and so on have been estimated. But there is little research on the spatial distribution model of urban built-up area. In this paper, an improved density-based clustering (Density-based Spatial Clustering of Qpplications with Boise,DBSCAN) algorithm is proposed, which extends the clustering of point elements to surface elements and uses the edge of the plane object as the criterion to define the inclusion relationship. The clustering parameters are determined, and the abrupt point of the distance between the elements is taken as the distance parameter. According to this method, the Chinese urban objects extracted by night lighting data are clustered according to the characteristics of density distribution, and the gathering patterns of Chinese cities at different scales are obtained by changing the distance parameters. The validity of the algorithm is verified by comparing with the empirical data, which provides a powerful research method for studying the spatial distribution and evolution of Chinese cities.
【作者单位】： 华东师范大学地理科学学院;华东师范大学地理信息科学教育部重点实验室;
【基金】：国家自然科学基金人才培养项目(J1310028) 上海市教育委员会科研创新项目(15ZZ026) 中央高校基本科研业务费专项资金项目
【分类号】：TP311.13;TP751
本文编号：2361824
[Abstract]:Previous studies have shown that urban built-up areas can be extracted from DMSP/OLS (Defense Meteorological Satellite Program/Operational Linescan System) stable night light data, and other indicators such as population, GDP and so on have been estimated. But there is little research on the spatial distribution model of urban built-up area. In this paper, an improved density-based clustering (Density-based Spatial Clustering of Qpplications with Boise,DBSCAN) algorithm is proposed, which extends the clustering of point elements to surface elements and uses the edge of the plane object as the criterion to define the inclusion relationship. The clustering parameters are determined, and the abrupt point of the distance between the elements is taken as the distance parameter. According to this method, the Chinese urban objects extracted by night lighting data are clustered according to the characteristics of density distribution, and the gathering patterns of Chinese cities at different scales are obtained by changing the distance parameters. The validity of the algorithm is verified by comparing with the empirical data, which provides a powerful research method for studying the spatial distribution and evolution of Chinese cities.
【作者单位】： 华东师范大学地理科学学院;华东师范大学地理信息科学教育部重点实验室;
【基金】：国家自然科学基金人才培养项目(J1310028) 上海市教育委员会科研创新项目(15ZZ026) 中央高校基本科研业务费专项资金项目
【分类号】：TP311.13;TP751
【相似文献】
相关期刊论文 前1条
1 周纪;陈云浩;李京;马伟;占文凤;;城市区域热辐射方向性研究进展[J];地球科学进展;2009年05期
，本文编号：2361824
本文链接：https://www.wllwen.com/guanlilunwen/gongchengguanli/2361824.html
