基于DMSP-OLS灯光数据的广东省城市人口估算
发布时间：2019-04-16 10:57
【摘要】：使用DMSP/OLS近20年数据对广东省城市的人口规模进行估算.首先,对广东省的DMSP/OLS数据进行校正.以2000、2010年人口普查数据与灯光数据为样本,构建"灰度密度-人口密度"模型,估算广东21个地级市常住人口.该模型既充分考虑灰度阀值的设定对人口估算结果的影响,又考虑城市人口数量变化在灯光灰度总值和灯光覆盖面积这2个变量上的体现.由于使用灯光数据指标与人口数据指标的比值作为估算基础,各城市的估算函数参数均不同,符合各城市经济发展状况和人口分布特点.通过2005年全国1%人口抽样数据检验,21个地级市平均估算相对误差为3.69%,全省总常住人口估算相对误差为5.02%,证明该模型能有效地估算广东省各地级市的人口规模.与已有的人口统计数据获取途径相比,该文构建的模型比传统的人口统计方法增强了时效性,又比其他遥感估算方法节省了地物判读和解译的步骤,且估算误差更小.基于DMSP/OLS数据的城市人口估算对研究中国城市人口规模具有重要的应用价值.
[Abstract]:The population size of cities in Guangdong Province was estimated using DMSP/OLS data for nearly 20 years. Firstly, the DMSP/OLS data of Guangdong Province are corrected. Taking 2000 and 2010 census data and lighting data as samples, the "gray density-population density" model is constructed to estimate the resident population of 21 prefectural cities in Guangdong province. The model not only considers the influence of the setting of gray threshold on the result of population estimation, but also takes into account the change of urban population on the two variables of light gross value and light coverage area. Because the ratio of light data index to population data index is used as the basis of estimation, the estimation function parameters of each city are different, which accord with the economic development situation and population distribution characteristics of each city. Through the national sampling data of 1% of the population in 2005, the average relative error of estimation in 21 prefectural cities was 3.69%, and the relative error in the estimation of the total resident population in the whole province was 5.02%. It is proved that the model can effectively estimate the population size of different prefecture-level cities in Guangdong Province. Compared with the existing methods of obtaining demographic data, the model proposed in this paper improves the timeliness of the traditional demographic method, and saves the steps of interpretation and interpretation of ground objects compared with other remote sensing estimation methods, and the estimation error is less than that of other remote sensing estimation methods. The estimation of urban population based on DMSP/OLS data has important application value in studying the size of urban population in China.
【作者单位】： 华南师范大学地理科学学院;
【基金】：国家973计划前期研究专项项目(2014CB460614) 国家自然科学基金项目(41201141)
【分类号】：C921
[Abstract]:The population size of cities in Guangdong Province was estimated using DMSP/OLS data for nearly 20 years. Firstly, the DMSP/OLS data of Guangdong Province are corrected. Taking 2000 and 2010 census data and lighting data as samples, the "gray density-population density" model is constructed to estimate the resident population of 21 prefectural cities in Guangdong province. The model not only considers the influence of the setting of gray threshold on the result of population estimation, but also takes into account the change of urban population on the two variables of light gross value and light coverage area. Because the ratio of light data index to population data index is used as the basis of estimation, the estimation function parameters of each city are different, which accord with the economic development situation and population distribution characteristics of each city. Through the national sampling data of 1% of the population in 2005, the average relative error of estimation in 21 prefectural cities was 3.69%, and the relative error in the estimation of the total resident population in the whole province was 5.02%. It is proved that the model can effectively estimate the population size of different prefecture-level cities in Guangdong Province. Compared with the existing methods of obtaining demographic data, the model proposed in this paper improves the timeliness of the traditional demographic method, and saves the steps of interpretation and interpretation of ground objects compared with other remote sensing estimation methods, and the estimation error is less than that of other remote sensing estimation methods. The estimation of urban population based on DMSP/OLS data has important application value in studying the size of urban population in China.
【作者单位】： 华南师范大学地理科学学院;
【基金】：国家973计划前期研究专项项目(2014CB460614) 国家自然科学基金项目(41201141)
【分类号】：C921
【参考文献】
相关期刊论文 前2条
1 卓莉,陈晋,史培军,辜智慧,范一大,一之p,俊明;基于夜间灯光数据的中国人口密度模拟[J];地理学报;2005年02期
2 曹丽琴;李平湘;张良培;;基于DMSP/OLS夜间灯光数据的城市人口估算——以湖北省各县市为例[J];遥感信息;2009年01期
【共引文献】
相关期刊论文 前10条
1 赵虎;李霖;张志军;朱海红;;基于扩展空间句法的城区人口密度估算方法[J];测绘科学;2009年04期
2 杜国明;张树文;;面向防洪救灾的人口统计数据空间化研究——以扶余县为例[J];长江流域资源与环境;2007年02期
3 唐华秀;马劲松;战金艳;邓祥征;;江苏省人口数据空间离散化研究及其精度分析[J];长江流域资源与环境;2008年04期
4 曹晓晗;;辽宁省人口密度分布模拟研究[J];测绘与空间地理信息;2010年02期
5 张桂芳;单新建;尹京苑;;基于TM影像的上海地区人口密度数字模拟[J];地震;2007年01期
6 王鹤饶;郑新奇;袁涛;;DMSP/OLS数据应用研究综述[J];地理科学进展;2012年01期
7 康停军;张新长;赵元;王海鹰;张维;;基于多智能体的城市人口分布模型[J];地理科学;2012年07期
8 卓莉;李强;史培军;陈晋;郑t，
本文编号：2458717
本文链接：https://www.wllwen.com/shekelunwen/renkou/2458717.html
