State Key Lab for Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, China.
Collaborative Innovation Center of Geospatial Technology, Wuhan University, Wuhan, China.
PLoS One. 2018 Oct 26;13(10):e0206230. doi: 10.1371/journal.pone.0206230. eCollection 2018.
As an informative proxy measure for a range of urbanisation and socioeconomic variables, satellite-derived night-time light data have been widely used to investigate the diverse anthropogenic activities and reveal social economy development disparities from the regional to the national scale. The new-generation night-time light data have been proven to potentially improve our understanding in the development and inequality of urban social economy due to its high spatial resolution, strong timeliness and minimal background noise. These night-time light data are derived from the visible infrared imaging radiometer suite (VIIRS) instrument with day/night band located on the Suomi National Polar-orbiting Partnership (NPP) satellite. This study proposed a hybrid model to estimate urban consumption potentiality based on the comprehensive information of human activities obtained from the VIIRS night-time light data. Our method established a flexible geographically weighted regression-based estimation model based on the residential consumption data and DN values of the VIIRS data to predict the possible consumption potentiality of other urban areas in dynamic time. The experiment conducted in Guiyang, a provincial capital in China, affirms that our model is proven to have higher accuracy compared with traditional regression models and can potentially provide guidance for improved business management and increased profit.
作为一系列城市化和社会经济变量的信息代理指标，卫星衍生的夜间灯光数据已被广泛用于研究各种人为活动，并揭示从区域到国家层面的社会经济发展差距。新一代夜间灯光数据由于具有高空间分辨率、强时效性和最小背景噪声，有望提高我们对城市社会经济发展和不平等的理解。这些夜间灯光数据是从搭载在苏美国家极地轨道伙伴关系（NPP）卫星上的可见红外成像辐射计套件（VIIRS）仪器的日夜带上获取的。本研究提出了一种基于 VIIRS 夜间灯光数据中获取的人类活动综合信息来估算城市消费潜力的混合模型。我们的方法建立了一个灵活的基于地理加权回归的估计模型，基于居住消费数据和 VIIRS 数据的 DN 值，以在动态时间内预测其他城市地区的可能消费潜力。在中国贵州省省会贵阳进行的实验证实，与传统回归模型相比，我们的模型具有更高的准确性，并且可能为改进业务管理和增加利润提供指导。
