Self-Guided Image Dehazing Using Progressive Feature Fusion

We propose an effective image dehazing algorithm which explores useful information from the input hazy image itself as the guidance for the haze removal. The proposed algorithm first uses a deep pre-dehazer to generate an intermediate result, and takes it as the reference image due to the clear structures it contains. To better explore the guidance information in the generated reference image, it then develops a progressive feature fusion module to fuse the features of the hazy image and the reference image. Finally, the image restoration module takes the fused features as input to use the guidance information for better clear image restoration. All the proposed modules are trained in an end-to-end fashion, and we show that the proposed deep pre-dehazer with progressive feature fusion module is able to help haze removal. Extensive experimental results show that the proposed algorithm performs favorably against state-of-the-art methods on the widely-used dehazing benchmark datasets as well as real-world hazy images.

I. Introduction

Images captured in hazy environments are always of low quality. The main reason of haze is the scattering effect caused by floating particles in the atmosphere. This process can be formulated by the atmospheric scattering model [1]–[3]: \begin{equation*} \mathbf {I}(\mathrm {x})=\mathbf {J}(\mathrm {x})t(\mathrm {x})+\mathbf {A}(1-t(\mathrm {x})), \tag{1}\end{equation*} where and represent the intensity of the observed hazy image and the clean image; is the transmission map; is the global atmospheric light. If the haze is homogeneous, we have , where is the atmospheric scattering parameter, and is the scene depth.