《Single Image Dehazing》.pdf

Raanan Fattal Hebrew University of Jerusalem, Israel Input Output Depth Figure 1: Dehazing based on a single input image and the corresponding depth estimate. is due to the presence of aerosols such as dust, mist, and fumes which deflect light from its original course of propagation. In long In this paper we present a new method for estimating the optical distance photography or foggy scenes, this process has a substan- transmission in hazy scenes given a single input image. Based on tial effect on the image in which contrasts are reduced and surface this estimation, the scattered light is eliminated to increase scene colors become faint. Such degraded photographs often lack visual visibility and recover haze-free scene contrasts. In this new ap- vividness and appeal, and moreover, they offer a poor visibility of proach we formulate a refined image formation model that accounts the scene contents. This effect may be an annoyance to amateur, for surface shading in addition to the transmission function. This commercial, and artistic photographers as well as undermine the allows us to resolve ambiguities in the data by searching for a solu- quality of underwater and aerial photography. This may also be the tion in which the resulting shading and transmission functions are case for satellite imaging which is used for many purposes includ- locally statistically uncorrelated.