CNN based Haze Removal for Aerospace and UAV Applications

Aerial imaging has profound applications in Military surveillance, Earthquake assessment, Aerospace, etc. Often due to sudden weather fluctuations, there are variations in visibility. In this paper a CNN based architecture is proposed which removes the haze without much degradation in colour and contrast of the image. The Proposed model is trained on RESIDE (OTS) dataset and provided a 66%, 65%, and 9.15% improvement in PSNR, BRISQUE and Entropy respectively when compared with the best of the available haze removal techniques. Also the model has shown 25% reduction in the computational time making it ideal for Real time applications.