Inhomogeneous cloud removing from remote sensing image
Liya Zhou, Qin Zhiyuan, Qing Lin, Wei Shang · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2011
Cloud is the common phenomenon in the obtaining and application of the visible-light remote sensing image, and these images are usually degraded by the turbid medium (e.g., fog, cloud) in the atmosphere. This article researched the defogging techniques of the single satellite remote sensing image on the basis of the dark channel prior and the imaging model of the cloud images, and proposed a new approach to calculate the airlight in view of problems of asymmetric thickness of cloud in the satellite remote sensing images and unreasonable estimation of the airlight in the existing defogging algorithms. In my approach, cloud thickness information was added to effectively avoid the phenomenon of 'over-treatment' and 'under-treatment'. Finally, the feature of scientific and effective were validated by the successful testament of respectively using medium-resolution and high-resolution satellite remote sensing data.