A new context-based procedure for the detection and removal of cloud shadow from moderate-and-high resolution satellite data over land

This paper has focused on a new automated detection and removal of cloud shadows of remotely sensed data to service quantitative interval of atmospheric parameters, using the FengYun-3 (FY-3), moderate spatial resolution and HuanJing-1 (HJ-1), high spatial resolution satellite data. Taking into account interaction between neighboring pixels in remote sensing images, this research aims at the special context relationships that are used to detect cloud shadows and then being carried out on related pixel match and cloud shadow removal. The tests show that the algorithm of detection and removal of cloud shadow is very simple and valid with the high discrimination accuracy and low leaking discrimination rate and so on, as well as yet associated with scale and scaling effect.