English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Haze usually leads to deteriorated images with low contrast, color shift andstructural distortion. We observe that many deep learning based models exhibitexceptional performance on removing homogeneous haze, but they usually fail toaddress the challenge of non-homogeneous dehazing. Two main factors account forthis situation. Firstly, due to the intricate and non uniform distribution ofdense haze, the recovery of structural and chromatic features with highfidelity is challenging, particularly in regions with heavy haze. Secondly, theexisting small scale datasets for non-homogeneous dehazing are inadequate tosupport reliable learning of feature mappings between hazy images and theircorresponding haze-free counterparts by convolutional neural network(CNN)-based models. To tackle these two challenges, we propose a novel twobranch network that leverages 2D discrete wavelete transform (DWT), fastFourier convolution (FFC) residual block and a pretrained ConvNeXt model.Specifically, in the DWT-FFC frequency branch, our model exploits DWT tocapture more high-frequency features. Moreover, by taking advantage of thelarge receptive field provided by FFC residual blocks, our model is able toeffectively explore global contextual information and produce images withbetter perceptual quality. In the prior knowledge branch, an ImageNetpretrained ConvNeXt as opposed to Res2Net is adopted. This enables our model tolearn more supplementary information and acquire a stronger generalizationability. The feasibility and effectiveness of the proposed method isdemonstrated via extensive experiments and ablation studies. The code isavailable at https://github.com/zhouh115/DWT-FFC.

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