FY-3D MERSI-II Cloud Detection and Clear Sky Product Production

Abstract

At the present stage, satellite remote sensing images often have the problem of cloud obscuration, which causes difficulties in the subsequent image processing and further research. In this paper, using the Moderate Resolution Spectral Imager II (MERSI-II) data from the domestic Fengyun-3D satellite as the data source, we study how to use the threshold method to detect clouds on the obtained data, and use the multi-temporal image superposition method to produce clear-sky products according to the results of the cloud detection, to eliminate the problem of cloud occlusion, and restore the real information of the features. Comparing the obtained cloud detection products with the actual true-color images and the official cloud detection products of Fengyun Satellite, it is found that the effect is still good. At the same time, the production results of the clear sky products can also clearly see the real features and better realize the de-cloud operation.