Breaking Through the Haze: An Advanced Non-Homogeneous Dehazing Method based on Fast Fourier Convolution and ConvNeXt

TL;DR

This paper introduces a novel non-homogeneous image dehazing method combining frequency domain features, large receptive field convolutional blocks, and a pretrained ConvNeXt to effectively restore hazy images.

Contribution

It proposes a two-branch network utilizing DWT, FFC residual blocks, and a pretrained ConvNeXt for improved non-homogeneous dehazing performance.

Findings

Enhanced dehazing quality on complex hazy images

Effective global contextual information exploration

Improved generalization with pretrained ConvNeXt

Abstract

Haze usually leads to deteriorated images with low contrast, color shift and structural distortion. We observe that many deep learning based models exhibit exceptional performance on removing homogeneous haze, but they usually fail to address the challenge of non-homogeneous dehazing. Two main factors account for this situation. Firstly, due to the intricate and non uniform distribution of dense haze, the recovery of structural and chromatic features with high fidelity is challenging, particularly in regions with heavy haze. Secondly, the existing small scale datasets for non-homogeneous dehazing are inadequate to support reliable learning of feature mappings between hazy images and their corresponding haze-free counterparts by convolutional neural network (CNN)-based models. To tackle these two challenges, we propose a novel two branch network that leverages 2D discrete wavelete transform (DWT), fast Fourier convolution (FFC) residual block and a pretrained ConvNeXt model. Specifically, in the DWT-FFC frequency branch, our model exploits DWT to capture more high-frequency features. Moreover, by taking advantage of the large receptive field provided by FFC residual blocks, our model is able to effectively explore global contextual information and produce images with better perceptual quality. In the prior knowledge branch, an ImageNet pretrained ConvNeXt as opposed to Res2Net is adopted. This enables our model to learn more supplementary information and acquire a stronger generalization ability. The feasibility and effectiveness of the proposed method is demonstrated via extensive experiments and ablation studies. The code is available at https://github.com/zhouh115/DWT-FFC.