Fine-Scale Population Distributions Mapping Based on Remote Sensing and Social Sensing Data

Fine-scale population distribution data play an essential role in numerous fields, In this paper, we have developed an appropriate method that using a random forest algorithm to down-scale the town-level population distribution to the grid level. the method was supported by remote sensing data and social sensing data to obtain a population distribution map with a spatial resolution of
$5\mathrm{m}\times 5\mathrm{m}$
. Compared with WorldPop
$(100\mathrm{m}\times 100\mathrm{m})$
. our method has well accuracy (PearsonR = 0.7985, RMSE = 613.5037, p <0.0001). This method has reference significance for the micro-mapping of urban population distribution.