From heavy rain removal to detail restoration: A faster and better network

The profound accumulation of precipitation during intense rainfall events can markedly degrade the quality of images, leading to the erosion of textural details. Despite the improvements observed in existing learning-based methods specialized for heavy rain removal, it is discerned that a significan...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Pattern recognition Ročník 148; s. 110205
Hlavní autoři: Wen, Yuanbo, Gao, Tao, Zhang, Jing, Zhang, Kaihao, Chen, Ting
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 01.04.2024
Témata:
Detail reconstruction
Dilated convolution
Dual-stage network
Pixel-wise attention
Single image deraining
Dual-stage network
Dilated convolution
Single image deraining
Pixel-wise attention
Detail reconstruction
ISSN:0031-3203
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The profound accumulation of precipitation during intense rainfall events can markedly degrade the quality of images, leading to the erosion of textural details. Despite the improvements observed in existing learning-based methods specialized for heavy rain removal, it is discerned that a significant proportion of these methods tend to overlook the precise reconstruction of the intricate details. In this work, we introduce a simple dual-stage progressive enhancement network, denoted as DPENet, aiming to achieve effective deraining while preserving the structural accuracy of rain-free images. This approach comprises two key modules, a rain streaks removal network (R2Net) focusing on accurate rain removal, and a details reconstruction network (DRNet) designed to recover the textural details of rain-free images. Firstly, we introduce a dilated dense residual block (DDRB) within R2Net, enabling the aggregation of high-level and low-level features. Secondly, an enhanced residual pixel-wise attention block (ERPAB) is integrated into DRNet to facilitate the incorporation of contextual information. To further enhance the fidelity of our approach, we employ a comprehensive loss function that accentuates both the marginal and regional accuracy of rain-free images. Extensive experiments conducted on publicly available benchmarks demonstrates the noteworthy efficiency and effectiveness of our proposed DPENet. The source code and pre-trained models are currently available at https://github.com/chdwyb/DPENet. •Dual-stage deraining network performs better than the single-stage network.•Multi-level feature fusion is significant for rain streaks removal.•Contextual information aggregation facilitates textural details reconstruction.
ISSN:0031-3203
DOI:10.1016/j.patcog.2023.110205