Single image haze removal based on the improved atmospheric scattering model

In this paper, we propose an improved atmospheric scattering model (IASM) to overcome the inherent limitation of the traditional atmospheric scattering model. Based on the IASM, a fast single image dehazing algorithm is also presented. In this algorithm, by constructing a linear model between the transmission and the haze aware density feature, the transmission map can be directly estimated through a linear operation on three components: luminance, saturation and gradient. Combining the sky-relevant feature and the proposed guided energy model (GEM), we can accurately estimate the atmospheric light and scene incident light, and can further restore the scene albedo via the IASM. Finally, an accelerating framework (AF) based on the Gaussian-Laplacian pyramid is proposed to increase the computational speed. Experimental results demonstrate that the proposed algorithm outperforms most of the prevalent algorithms in terms of visual effect and computational efficiency. Besides, it is also capable of processing various types of degraded images in addition to hazy images.