1. Introduction
Remote sensing imagery has been used in many geoscience observation fields such as land cover classification [1], [2], environment monitoring [3], change detection [4], [5], forest canopy closure estimation [6], and poverty prediction [7], [8]. However, remote sensing imagery is inevitably affected by many factors, such as cloud occlusions, weather, and climate effects. Thick cloud occlusions will lose much of the information. Therefore, cloud removal is an indispensable preprocessing step before using remote sensing imagery in various applications.