Robust air-light estimation for a single hazy image using haze-lines in Plüker coordinates

Abstract

Image dehazing is a valid way to reduce the negative impacts of bad weather, such as fog and haze, on outdoor optical imaging. The effect of the image dehazing heavily relies on the air-light estimation. Most existing methods exploit the prior information, which is generally obtained by simply assessing the brightest pixels’ intensity in a hazy image, to estimate the air-light. However, bright objects, such as lamps, lead to the failure of these methods. So, a novel method for robust and optimal air-light estimation using haze-lines described in Plüker coordinates is proposed. First, the exact analytic solution of the air-light, i.e., the intersection of haze-lines, is derived via analyzing the geometric relationship between haze-lines in three-dimensional RGB space using their Plüker coordinates. Then, a novel air-light estimation algorithm using haze-lines in Plüker coordinates is elaborated. Finally, extensive experiments are performed to validate the proposed algorithm. Experimental results reveal that the proposed algorithm can more robustly and accurately estimate the air-light.