An Image dehazing approach based on the airlight field estimation

TL;DR

This paper introduces a novel single image dehazing method that models atmospheric light as a field function, improving results especially in images with large sky regions by jointly estimating the airlight field, transmission, and haze-free image.

Contribution

It proposes modeling atmospheric light as a field function and a MAP-based joint estimation approach, addressing limitations of constant-airlight assumptions in large sky regions.

Findings

Outperforms existing dehazing methods on real images.

Effectively handles images with large sky regions.

Provides better brightness and color accuracy.

Abstract

This paper proposes a scheme for single image haze removal based on the airlight field (ALF) estimation. Conventional image dehazing methods which are based on a physical model generally take the global atmospheric light as a constant. However, the constant-airlight assumption may be unsuitable for images with large sky regions, which causes unacceptable brightness imbalance and color distortion in recovery images. This paper models the atmospheric light as a field function, and presents a maximum a-priori (MAP) method for jointly estimating the airlight field, the transmission rate and the haze free image. We also introduce a valid haze-level prior for effective estimate of transmission. Evaluation on real world images shows that the proposed approach outperforms existing methods in single image dehazing, especially when the large sky region is included.