Real-time content adaptive contrast enhancement for see-through fog and rain

In this paper we present a novel algorithm for improving the visibility of surveillance videos degraded by fog and/or rain. The proposed algorithm adaptively enhances the global and local contrast of a surveillance video. The algorithm is inspired on the human visual system, and accounts for the perceptual sensitivity to noise, compression artifacts, and the texture of image content. The model is combined with the classic Contrast Limited Adaptive Histogram Equalization (CLAHE) method to adaptively enhance surveillance videos. We have implemented a real-time video enhancement system and performed extensive experimental testing over a video database containing common surveillance videos recorded under fog and rain conditions. The proposed approach significantly improves the visual quality of surveillance videos by removing fog/rain effects, as well as reducing noise and artifacts.