Delving Deeper Into Image Dehazing: A Survey
Images captured under foggy or hazy weather conditions are affected by the scattering of atmospheric particles, resulting in decreased contrast and color variation, thereby limiting their practical applications. In recent years, deep learning methods showcase significant advancements in image dehazi...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2023-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10325493/ |
_version_ | 1797321111147380736 |
---|---|
author | Guohou Li Jia Li Gongchao Chen Zhibin Wang Songlin Jin Chang Ding Weidong Zhang |
author_facet | Guohou Li Jia Li Gongchao Chen Zhibin Wang Songlin Jin Chang Ding Weidong Zhang |
author_sort | Guohou Li |
collection | DOAJ |
description | Images captured under foggy or hazy weather conditions are affected by the scattering of atmospheric particles, resulting in decreased contrast and color variation, thereby limiting their practical applications. In recent years, deep learning methods showcase significant advancements in image dehazing. However, the complexity and degradation factors in hazy images challenge the generalization capacity of dehazing methods. This paper comprehensively reviews the recent developments in single-image dehazing techniques based on deep learning. From the perspectives of Convolutional Neural Networks (CNN) and Generative Adversarial Networks (GAN), different models are introduced and classified into four categories: Encoder-Decoder, Multi-Module, Multi-Branch, and Dual-Generative Adversarial Networks. The robustness and effectiveness of deep learning models are analyzed by comparing their performance and model complexity on public datasets. Additionally, limitations of current benchmark datasets and evaluation metrics are identified, and unresolved issues and future research directions are discussed. Our efforts in this paper will serve as a comprehensive reference for future research and call for further development in deep learning-based image dehazing. |
first_indexed | 2024-03-08T04:52:49Z |
format | Article |
id | doaj.art-ad57431a48af480c8289dcf5815ad693 |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-08T04:52:49Z |
publishDate | 2023-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-ad57431a48af480c8289dcf5815ad6932024-02-08T00:00:55ZengIEEEIEEE Access2169-35362023-01-011113175913177410.1109/ACCESS.2023.333561810325493Delving Deeper Into Image Dehazing: A SurveyGuohou Li0Jia Li1https://orcid.org/0009-0006-8862-947XGongchao Chen2Zhibin Wang3Songlin Jin4Chang Ding5https://orcid.org/0000-0001-9978-9301Weidong Zhang6School of Information Engineering, Henan Institute of Science and Technology, Xinxiang, ChinaSchool of Information Engineering, Henan Institute of Science and Technology, Xinxiang, ChinaSchool of Information Engineering, Henan Institute of Science and Technology, Xinxiang, ChinaSchool of Information Engineering, Henan Institute of Science and Technology, Xinxiang, ChinaSchool of Information Engineering, Henan Institute of Science and Technology, Xinxiang, ChinaSchool of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, ChinaSchool of Information Engineering, Henan Institute of Science and Technology, Xinxiang, ChinaImages captured under foggy or hazy weather conditions are affected by the scattering of atmospheric particles, resulting in decreased contrast and color variation, thereby limiting their practical applications. In recent years, deep learning methods showcase significant advancements in image dehazing. However, the complexity and degradation factors in hazy images challenge the generalization capacity of dehazing methods. This paper comprehensively reviews the recent developments in single-image dehazing techniques based on deep learning. From the perspectives of Convolutional Neural Networks (CNN) and Generative Adversarial Networks (GAN), different models are introduced and classified into four categories: Encoder-Decoder, Multi-Module, Multi-Branch, and Dual-Generative Adversarial Networks. The robustness and effectiveness of deep learning models are analyzed by comparing their performance and model complexity on public datasets. Additionally, limitations of current benchmark datasets and evaluation metrics are identified, and unresolved issues and future research directions are discussed. Our efforts in this paper will serve as a comprehensive reference for future research and call for further development in deep learning-based image dehazing.https://ieeexplore.ieee.org/document/10325493/Image dehazingdeep learningconvolutional neural networks (CNNs)generative adversarial networks (GANs) |
spellingShingle | Guohou Li Jia Li Gongchao Chen Zhibin Wang Songlin Jin Chang Ding Weidong Zhang Delving Deeper Into Image Dehazing: A Survey IEEE Access Image dehazing deep learning convolutional neural networks (CNNs) generative adversarial networks (GANs) |
title | Delving Deeper Into Image Dehazing: A Survey |
title_full | Delving Deeper Into Image Dehazing: A Survey |
title_fullStr | Delving Deeper Into Image Dehazing: A Survey |
title_full_unstemmed | Delving Deeper Into Image Dehazing: A Survey |
title_short | Delving Deeper Into Image Dehazing: A Survey |
title_sort | delving deeper into image dehazing a survey |
topic | Image dehazing deep learning convolutional neural networks (CNNs) generative adversarial networks (GANs) |
url | https://ieeexplore.ieee.org/document/10325493/ |
work_keys_str_mv | AT guohouli delvingdeeperintoimagedehazingasurvey AT jiali delvingdeeperintoimagedehazingasurvey AT gongchaochen delvingdeeperintoimagedehazingasurvey AT zhibinwang delvingdeeperintoimagedehazingasurvey AT songlinjin delvingdeeperintoimagedehazingasurvey AT changding delvingdeeperintoimagedehazingasurvey AT weidongzhang delvingdeeperintoimagedehazingasurvey |