DA-Net: Dual Attention Network for Haze Removal in Remote Sensing Image

Haze removal in remote sensing images is essential for practical applications in various fields such as weather forecasting, monitoring, mineral exploration and disaster management. The previous deep learning models make use of large convolutional kernel and attention mechanisms for efficient dehazi...

Full description

Saved in:
Bibliographic Details
Main Authors: Namwon Kim, Il-Seok Choi, Seong-Soo Han, Chang-Sung Jeong
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10679105/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850066205564469248
author Namwon Kim
Il-Seok Choi
Seong-Soo Han
Chang-Sung Jeong
author_facet Namwon Kim
Il-Seok Choi
Seong-Soo Han
Chang-Sung Jeong
author_sort Namwon Kim
collection DOAJ
description Haze removal in remote sensing images is essential for practical applications in various fields such as weather forecasting, monitoring, mineral exploration and disaster management. The previous deep learning models make use of large convolutional kernel and attention mechanisms for efficient dehazing. However, it has drawbacks such as the loss of image details and low performance. In this paper, we shall present a new dual attention network, called DA-Net, for dehazing remote sensing images which achieves better dehazing performance while reducing model complexity sharply by exploiting a novel dual attention block where two modules, channel-spatial attention and parallel attention are serially connected. We propose a new architecture for parallel attention which achieves better dehazing performance by concatenating three different attention mechanisms in parallel: global channel attention, local channel attention and spatial attention. Moreover, we shall show that the concatenation of channel-spatial attention to parallel attention module enables detecting haze component information more accurately while reducing the model complexity proportional to the number of parameters by combining the channel and spatial information generated respectively from two different channel and spatial branches. Our experimental results show that DA-Net achieves much better performance for both synthetic and real image data sets compared to the other dehazing models in terms of quantitative and qualitative evaluations.
format Article
id doaj-art-5a6dd40ec69c45fa8a28fc565a5d01a5
institution DOAJ
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-5a6dd40ec69c45fa8a28fc565a5d01a52025-08-20T02:48:49ZengIEEEIEEE Access2169-35362024-01-011213629713631210.1109/ACCESS.2024.345958810679105DA-Net: Dual Attention Network for Haze Removal in Remote Sensing ImageNamwon Kim0https://orcid.org/0000-0002-1865-5537Il-Seok Choi1Seong-Soo Han2Chang-Sung Jeong3Department of Electrical Engineering, Korea University, Seoul, South KoreaCTO SW Engineering Research and Development Laboratory, LG Electronics, Seoul, South KoreaDivision of Liberal Studies, Kangwon National University, Samcheok-si, South KoreaDepartment of Electrical Engineering, Korea University, Seoul, South KoreaHaze removal in remote sensing images is essential for practical applications in various fields such as weather forecasting, monitoring, mineral exploration and disaster management. The previous deep learning models make use of large convolutional kernel and attention mechanisms for efficient dehazing. However, it has drawbacks such as the loss of image details and low performance. In this paper, we shall present a new dual attention network, called DA-Net, for dehazing remote sensing images which achieves better dehazing performance while reducing model complexity sharply by exploiting a novel dual attention block where two modules, channel-spatial attention and parallel attention are serially connected. We propose a new architecture for parallel attention which achieves better dehazing performance by concatenating three different attention mechanisms in parallel: global channel attention, local channel attention and spatial attention. Moreover, we shall show that the concatenation of channel-spatial attention to parallel attention module enables detecting haze component information more accurately while reducing the model complexity proportional to the number of parameters by combining the channel and spatial information generated respectively from two different channel and spatial branches. Our experimental results show that DA-Net achieves much better performance for both synthetic and real image data sets compared to the other dehazing models in terms of quantitative and qualitative evaluations.https://ieeexplore.ieee.org/document/10679105/Dehazingremote sensing imagedeep learningattention
spellingShingle Namwon Kim
Il-Seok Choi
Seong-Soo Han
Chang-Sung Jeong
DA-Net: Dual Attention Network for Haze Removal in Remote Sensing Image
IEEE Access
Dehazing
remote sensing image
deep learning
attention
title DA-Net: Dual Attention Network for Haze Removal in Remote Sensing Image
title_full DA-Net: Dual Attention Network for Haze Removal in Remote Sensing Image
title_fullStr DA-Net: Dual Attention Network for Haze Removal in Remote Sensing Image
title_full_unstemmed DA-Net: Dual Attention Network for Haze Removal in Remote Sensing Image
title_short DA-Net: Dual Attention Network for Haze Removal in Remote Sensing Image
title_sort da net dual attention network for haze removal in remote sensing image
topic Dehazing
remote sensing image
deep learning
attention
url https://ieeexplore.ieee.org/document/10679105/
work_keys_str_mv AT namwonkim danetdualattentionnetworkforhazeremovalinremotesensingimage
AT ilseokchoi danetdualattentionnetworkforhazeremovalinremotesensingimage
AT seongsoohan danetdualattentionnetworkforhazeremovalinremotesensingimage
AT changsungjeong danetdualattentionnetworkforhazeremovalinremotesensingimage