NPP-VIIRS年度夜间灯光数据合成方法

Synthesis Method of NPP-VIIRS Annual Night Light Data

夜间灯光数据能有效反映人类夜间活动强度,但由于年度NPP-VIIRS夜间灯光数据的缺乏,很大程度上限制了其应用发展。传统的年度数据合成是直接基于年度均值影像进行数据校正合成,但这会使误差累积。因此,文章提出结合建设用地数据利用阈值法和中值法对NPP-VIIRS月度影像进行逐月校正,剔除背景噪声和极高值后再合成年度数据,并利用GDP数据对其进行精度评定。结果表明,合成年度夜间灯光数据与GDP相关性达到0.996,其数据的可靠性得到了很大提高,同时也说明该合成方法更具科学性。

The night light data can effectively reflect the intensity of human nocturnal activity,but due to the lack of annual NPP-VIIRS night light data,its application development is greatly limited.Traditional annual data synthesis is directly based on the annual mean image to correct the synthesis,but this will cause errors to accumulate.Therefore,this paper proposes a synthesis method of NPP-VIIRS annual night light data.Firstly,it combines the threshold method and the median method of construction land data to correct the monthly images of NPP-VIIRS month by month,and eliminate background noise and extremely high values,and then synthesizes the annual data,and uses the GDP data to evaluate their accuracy.The results show that the correlation between the annual night light data and GDP of the synthesis reaches 0.996,and the reliability of the data is greatly improved,and it also shows that the synthesis method is more scientific.