Visibility in bad weather from a single image

TLDR

Bad weather such as fog and haze degrades scene visibility by absorbing and scattering light, a process commonly modeled as a linear combination of direct attenuation and airlight, and existing dehazing methods typically require multiple images with varying polarization or atmospheric conditions, which is often impractical. The authors propose an automated single‑image method to restore visibility in bad weather. The method exploits that clear images have higher contrast and that airlight varies smoothly with distance, formulating a Markov random field cost function that can be optimized efficiently with graph cuts or belief propagation. It operates without needing geometric information and is applicable to both color and grayscale images.

Abstract

Bad weather, such as fog and haze, can significantly degrade the visibility of a scene. Optically, this is due to the substantial presence of particles in the atmosphere that absorb and scatter light. In computer vision, the absorption and scattering processes are commonly modeled by a linear combination of the direct attenuation and the airlight. Based on this model, a few methods have been proposed, and most of them require multiple input images of a scene, which have either different degrees of polarization or different atmospheric conditions. This requirement is the main drawback of these methods, since in many situations, it is difficult to be fulfilled. To resolve the problem, we introduce an automated method that only requires a single input image. This method is based on two basic observations: first, images with enhanced visibility (or clear-day images) have more contrast than images plagued by bad weather; second, airlight whose variation mainly depends on the distance of objects to the viewer, tends to be smooth. Relying on these two observations, we develop a cost function in the framework of Markov random fields, which can be efficiently optimized by various techniques, such as graph-cuts or belief propagation. The method does not require the geometrical information of the input image, and is applicable for both color and gray images.