Research of New Method for Removal Thin Cloud and Fog of the Remote Sensing Images

For the questions of obfuscation in Google earth remote sensing images, the new method had been pointed out. The cause of obfuscation was the interference of thin cloud and fog. The new method was combination of the advanced multi-scale Retinex (AMSR) and the method of Homomorphic Filter. The results were shown the new method was effectual for removing the thin cloud and fog from the remote sensing images in Google earth. The AMSR included three steps. The first step was to get the complementary color image for the remote sensing images in Google Earth with thin cloud and fog. Next, the enhanced images were stretched with MSR in different k Multiples standard deviation near the average brightness. Finally, the new complementary color image of the enhanced images had been got. The criterion of information entropy was as the image quality criterion. The conclusion shows that the new images were high quality. The new images handled by the method of Homomorphic Filter after enhanced by AMSR. The parameter of Homomorphic Filter was best in 0.9±0.1 and the remote sensing images handled by new method have largest information entropy.