Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze

The purpose of image dehazing is the reduction of the image degradation caused by suspended particles for supporting high-level visual tasks. Besides the atmospheric scattering model, convolutional neural network (CNN) has been used for image dehazing. However, the existing image dehazing algorithms...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Wenxuan Zhao, Yaqin Zhao, Liqi Feng, Jiaxi Tang
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	serial Unet++ module image dehazing dep learning unevenly distributed haze Electronics TK7800-8360
Acceso en línea:	https://doaj.org/article/8654ae3123a940a482b029005f4f856d
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:8654ae3123a940a482b029005f4f856d
record_format	dspace
spelling	oai:doaj.org-article:8654ae3123a940a482b029005f4f856d2021-11-25T17:25:29ZAttention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze10.3390/electronics102228682079-9292https://doaj.org/article/8654ae3123a940a482b029005f4f856d2021-11-01T00:00:00Zhttps://www.mdpi.com/2079-9292/10/22/2868https://doaj.org/toc/2079-9292The purpose of image dehazing is the reduction of the image degradation caused by suspended particles for supporting high-level visual tasks. Besides the atmospheric scattering model, convolutional neural network (CNN) has been used for image dehazing. However, the existing image dehazing algorithms are limited in face of unevenly distributed haze and dense haze in real-world scenes. In this paper, we propose a novel end-to-end convolutional neural network called attention enhanced serial Unet++ dehazing network (AESUnet) for single image dehazing. We attempt to build a serial Unet++ structure that adopts a serial strategy of two pruned Unet++ blocks based on residual connection. Compared with the simple Encoder–Decoder structure, the serial Unet++ module can better use the features extracted by encoders and promote contextual information fusion in different resolutions. In addition, we take some improvement measures to the Unet++ module, such as pruning, introducing the convolutional module with ResNet structure, and a residual learning strategy. Thus, the serial Unet++ module can generate more realistic images with less color distortion. Furthermore, following the serial Unet++ blocks, an attention mechanism is introduced to pay different attention to haze regions with different concentrations by learning weights in the spatial domain and channel domain. Experiments are conducted on two representative datasets: the large-scale synthetic dataset RESIDE and the small-scale real-world datasets I-HAZY and O-HAZY. The experimental results show that the proposed dehazing network is not only comparable to state-of-the-art methods for the RESIDE synthetic datasets, but also surpasses them by a very large margin for the I-HAZY and O-HAZY real-world dataset.Wenxuan ZhaoYaqin ZhaoLiqi FengJiaxi TangMDPI AGarticleserial Unet++ moduleimage dehazingdep learningunevenly distributed hazeElectronicsTK7800-8360ENElectronics, Vol 10, Iss 2868, p 2868 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	serial Unet++ module image dehazing dep learning unevenly distributed haze Electronics TK7800-8360
spellingShingle	serial Unet++ module image dehazing dep learning unevenly distributed haze Electronics TK7800-8360 Wenxuan Zhao Yaqin Zhao Liqi Feng Jiaxi Tang Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze
description	The purpose of image dehazing is the reduction of the image degradation caused by suspended particles for supporting high-level visual tasks. Besides the atmospheric scattering model, convolutional neural network (CNN) has been used for image dehazing. However, the existing image dehazing algorithms are limited in face of unevenly distributed haze and dense haze in real-world scenes. In this paper, we propose a novel end-to-end convolutional neural network called attention enhanced serial Unet++ dehazing network (AESUnet) for single image dehazing. We attempt to build a serial Unet++ structure that adopts a serial strategy of two pruned Unet++ blocks based on residual connection. Compared with the simple Encoder–Decoder structure, the serial Unet++ module can better use the features extracted by encoders and promote contextual information fusion in different resolutions. In addition, we take some improvement measures to the Unet++ module, such as pruning, introducing the convolutional module with ResNet structure, and a residual learning strategy. Thus, the serial Unet++ module can generate more realistic images with less color distortion. Furthermore, following the serial Unet++ blocks, an attention mechanism is introduced to pay different attention to haze regions with different concentrations by learning weights in the spatial domain and channel domain. Experiments are conducted on two representative datasets: the large-scale synthetic dataset RESIDE and the small-scale real-world datasets I-HAZY and O-HAZY. The experimental results show that the proposed dehazing network is not only comparable to state-of-the-art methods for the RESIDE synthetic datasets, but also surpasses them by a very large margin for the I-HAZY and O-HAZY real-world dataset.
format	article
author	Wenxuan Zhao Yaqin Zhao Liqi Feng Jiaxi Tang
author_facet	Wenxuan Zhao Yaqin Zhao Liqi Feng Jiaxi Tang
author_sort	Wenxuan Zhao
title	Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze
title_short	Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze
title_full	Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze
title_fullStr	Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze
title_full_unstemmed	Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze
title_sort	attention enhanced serial unet++ network for removing unevenly distributed haze
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/8654ae3123a940a482b029005f4f856d
work_keys_str_mv	AT wenxuanzhao attentionenhancedserialunetnetworkforremovingunevenlydistributedhaze AT yaqinzhao attentionenhancedserialunetnetworkforremovingunevenlydistributedhaze AT liqifeng attentionenhancedserialunetnetworkforremovingunevenlydistributedhaze AT jiaxitang attentionenhancedserialunetnetworkforremovingunevenlydistributedhaze
_version_	1718412359875166208

Attention Enhanced Serial Unet++ Network for Removing Unevenly Distributed Haze

Ejemplares similares