基于DMSP/OLS修正的福建省人口数据网格化  

Data Griding of Population in Fujian Province Based on Modified DMSP/OLS

作　　者：黄耀裔[1] 陈文成[1] HUANG Yaoyi;CHEN Wencheng(School of Resource and Environmental Science,Quanzhou Normal University,Quanzhou Fujian 362000,China)

机构地区：[1]泉州师范学院资源与环境科学学院,福建泉州362000

出　　处：《泉州师范学院学报》2020年第6期86-93,共8页Journal of Quanzhou Normal University

基　　金：福建省自然科学项目(2016J01195);泉州市科技项目(2018Z027)。

摘　　要：基于福建省人口普查的县域常住人口、DMSP/OLS、NDVI等数据,据研究的植被指数能有效降低夜间灯光数据过饱和问题,结合植被指数改进优化HSI指数模型.在改进的HSI指数下,基于地理探测器分区后分别直接以常住人口的网格个数为权进行加权个案与加权最小二乘法进行人口网格化建模,再进行人口零误差调整得到1 km×1 km尺度的福建省人口统计数据网格化.结果表明:结合地理探测器的改进HSI指数模型能优化夜间灯光过饱和,有效回避可变面元效应等问题,提高模型拟合精度,对人口网格化效果更好,更符合实际的人口空间分布.According to the data of resident population,DMSP/OLS,NDVI in Fujian Province,the vegetation index studied can effectively reduce the oversaturation of the night light data,and improve the HSI index model with the vegetation index,under the improved HSI index,based on the number of grids of the resident population,the weighted cases and the weighted least squares were modeled by the grid number of the resident population,and then the population zero-error adjustment was carried out to get the grid of the population statistics data of Fujian Province at the scale of 1 km×1 km.The results showed that the improved HSI index model combined with the geographical detector can optimize the night light oversaturation,effectively avoid the variable panel effect,improve the fitting accuracy of the model,to obtain better effect of population gridding,and to be more in line with the actual spatial distribution of population.

关 键 词：DMSP/OLS NDVI 改进HSI 人口网格化 福建省 

分 类 号：K910[历史地理—人文地理学]