Cargando…
High-Resolution Representations Network for Single Image Dehazing
Deep learning-based image dehazing methods have made great progress, but there are still many problems such as inaccurate model parameter estimation and preserving spatial information in the U-Net-based architecture. To address these problems, we propose an image dehazing network based on the high-r...
Autores principales: | , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8949864/ https://www.ncbi.nlm.nih.gov/pubmed/35336428 http://dx.doi.org/10.3390/s22062257 |
_version_ | 1784675005357883392 |
---|---|
author | Han, Wensheng Zhu, Hong Qi, Chenghui Li, Jingsi Zhang, Dengyin |
author_facet | Han, Wensheng Zhu, Hong Qi, Chenghui Li, Jingsi Zhang, Dengyin |
author_sort | Han, Wensheng |
collection | PubMed |
description | Deep learning-based image dehazing methods have made great progress, but there are still many problems such as inaccurate model parameter estimation and preserving spatial information in the U-Net-based architecture. To address these problems, we propose an image dehazing network based on the high-resolution network, called DeHRNet. The high-resolution network originally used for human pose estimation. In this paper, we make a simple yet effective modification to the network and apply it to image dehazing. We add a new stage to the original network to make it better for image dehazing. The newly added stage collects the feature map representations of all branches of the network by up-sampling to enhance the high-resolution representations instead of only taking the feature maps of the high-resolution branches, which makes the restored clean images more natural. The final experimental results show that DeHRNet achieves superior performance over existing dehazing methods in synthesized and natural hazy images. |
format | Online Article Text |
id | pubmed-8949864 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-89498642022-03-26 High-Resolution Representations Network for Single Image Dehazing Han, Wensheng Zhu, Hong Qi, Chenghui Li, Jingsi Zhang, Dengyin Sensors (Basel) Article Deep learning-based image dehazing methods have made great progress, but there are still many problems such as inaccurate model parameter estimation and preserving spatial information in the U-Net-based architecture. To address these problems, we propose an image dehazing network based on the high-resolution network, called DeHRNet. The high-resolution network originally used for human pose estimation. In this paper, we make a simple yet effective modification to the network and apply it to image dehazing. We add a new stage to the original network to make it better for image dehazing. The newly added stage collects the feature map representations of all branches of the network by up-sampling to enhance the high-resolution representations instead of only taking the feature maps of the high-resolution branches, which makes the restored clean images more natural. The final experimental results show that DeHRNet achieves superior performance over existing dehazing methods in synthesized and natural hazy images. MDPI 2022-03-15 /pmc/articles/PMC8949864/ /pubmed/35336428 http://dx.doi.org/10.3390/s22062257 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Han, Wensheng Zhu, Hong Qi, Chenghui Li, Jingsi Zhang, Dengyin High-Resolution Representations Network for Single Image Dehazing |
title | High-Resolution Representations Network for Single Image Dehazing |
title_full | High-Resolution Representations Network for Single Image Dehazing |
title_fullStr | High-Resolution Representations Network for Single Image Dehazing |
title_full_unstemmed | High-Resolution Representations Network for Single Image Dehazing |
title_short | High-Resolution Representations Network for Single Image Dehazing |
title_sort | high-resolution representations network for single image dehazing |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8949864/ https://www.ncbi.nlm.nih.gov/pubmed/35336428 http://dx.doi.org/10.3390/s22062257 |
work_keys_str_mv | AT hanwensheng highresolutionrepresentationsnetworkforsingleimagedehazing AT zhuhong highresolutionrepresentationsnetworkforsingleimagedehazing AT qichenghui highresolutionrepresentationsnetworkforsingleimagedehazing AT lijingsi highresolutionrepresentationsnetworkforsingleimagedehazing AT zhangdengyin highresolutionrepresentationsnetworkforsingleimagedehazing |