Single Image Dehazing via Relativity-of-Gaussian

To avoid halo artifacts and brightness distortion and achieve ideal visual effects, in this paper a relativity-of-Gaussian method (ROG) for single image dehazing is proposed. Firstly, we estimate the global atmospheric light only in the sky regions to minimize interference in other regions. Then, we use an adaptive boundary constraint ROG filtering to optimize the transmission properly which can reduce the halo artifacts. Finally, we remove the haze and produce a more natural visual recovery effect with an improved dark channel method. Experimental results show that this method outperforms state-of-the-art haze removal methods in terms of both efficiency and the dehazing visual effect.