Remote Sensing Image Super-Resolution via Residual Aggregation and Split Attentional Fusion Network
Remote sensing images contain various land surface scenes and different scales of ground objects, which greatly increases the difficulty of super-resolution tasks. The existing deep learning-based methods cannot solve this problem well. To achieve high-quality super-resolution of remote sensing imag...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2021-01-01
|
Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9541020/ |
id |
doaj-41d468347f144618a2f951ed66d168f0 |
---|---|
record_format |
Article |
spelling |
doaj-41d468347f144618a2f951ed66d168f02021-10-06T23:00:09ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01149546955610.1109/JSTARS.2021.31136589541020Remote Sensing Image Super-Resolution via Residual Aggregation and Split Attentional Fusion NetworkLong Chen0https://orcid.org/0000-0002-7586-0780Hui Liu1Minhang Yang2Yurong Qian3Zhengqing Xiao4Xiwu Zhong5https://orcid.org/0000-0002-3515-4566College of Software, Xinjiang University, Ürümqi, ChinaCollege of Information Science and Engineering, Xinjiang University, Ürümqi, ChinaCollege of Software, Xinjiang University, Ürümqi, ChinaCollege of Software, Xinjiang University, Ürümqi, ChinaCollege of Mathematics and System Sciences, Xinjiang University, Ürümqi, ChinaCollege of Software, Xinjiang University, Ürümqi, ChinaRemote sensing images contain various land surface scenes and different scales of ground objects, which greatly increases the difficulty of super-resolution tasks. The existing deep learning-based methods cannot solve this problem well. To achieve high-quality super-resolution of remote sensing images, a residual aggregation and split attentional fusion network (RASAF) is proposed in this article. It is mainly divided into the following three parts. First, a split attentional fusion block is proposed. It uses a basic split–fusion mechanism to achieve cross-channel feature group interaction, allowing the method to adapt to various land surface scene reconstructions. Second, to fully exploit multiscale image information, a hierarchical loss function is used. Third, residual learning is used to reduce the difficulty of training in super-resolution tasks. However, the respective residual branch features are used quite locally and fail to represent the real value. A residual aggregation mechanism is used to aggregate the local residual branch features to generate higher quality local residual branch features. The comparison of RASAF with some classical super-resolution methods using two widely used remote sensing datasets showed that the RASAF achieved better performance. And it achieves a good balance between performance and model parameter number. Meanwhile, the RASAF’s ability to support multilabel remote sensing image classification tasks demonstrates its usefulness.https://ieeexplore.ieee.org/document/9541020/Remote sensing imageresidual aggregationsplit attentional fusionsuper-resolution (SR) |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Long Chen Hui Liu Minhang Yang Yurong Qian Zhengqing Xiao Xiwu Zhong |
spellingShingle |
Long Chen Hui Liu Minhang Yang Yurong Qian Zhengqing Xiao Xiwu Zhong Remote Sensing Image Super-Resolution via Residual Aggregation and Split Attentional Fusion Network IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Remote sensing image residual aggregation split attentional fusion super-resolution (SR) |
author_facet |
Long Chen Hui Liu Minhang Yang Yurong Qian Zhengqing Xiao Xiwu Zhong |
author_sort |
Long Chen |
title |
Remote Sensing Image Super-Resolution via Residual Aggregation and Split Attentional Fusion Network |
title_short |
Remote Sensing Image Super-Resolution via Residual Aggregation and Split Attentional Fusion Network |
title_full |
Remote Sensing Image Super-Resolution via Residual Aggregation and Split Attentional Fusion Network |
title_fullStr |
Remote Sensing Image Super-Resolution via Residual Aggregation and Split Attentional Fusion Network |
title_full_unstemmed |
Remote Sensing Image Super-Resolution via Residual Aggregation and Split Attentional Fusion Network |
title_sort |
remote sensing image super-resolution via residual aggregation and split attentional fusion network |
publisher |
IEEE |
series |
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
issn |
2151-1535 |
publishDate |
2021-01-01 |
description |
Remote sensing images contain various land surface scenes and different scales of ground objects, which greatly increases the difficulty of super-resolution tasks. The existing deep learning-based methods cannot solve this problem well. To achieve high-quality super-resolution of remote sensing images, a residual aggregation and split attentional fusion network (RASAF) is proposed in this article. It is mainly divided into the following three parts. First, a split attentional fusion block is proposed. It uses a basic split–fusion mechanism to achieve cross-channel feature group interaction, allowing the method to adapt to various land surface scene reconstructions. Second, to fully exploit multiscale image information, a hierarchical loss function is used. Third, residual learning is used to reduce the difficulty of training in super-resolution tasks. However, the respective residual branch features are used quite locally and fail to represent the real value. A residual aggregation mechanism is used to aggregate the local residual branch features to generate higher quality local residual branch features. The comparison of RASAF with some classical super-resolution methods using two widely used remote sensing datasets showed that the RASAF achieved better performance. And it achieves a good balance between performance and model parameter number. Meanwhile, the RASAF’s ability to support multilabel remote sensing image classification tasks demonstrates its usefulness. |
topic |
Remote sensing image residual aggregation split attentional fusion super-resolution (SR) |
url |
https://ieeexplore.ieee.org/document/9541020/ |
work_keys_str_mv |
AT longchen remotesensingimagesuperresolutionviaresidualaggregationandsplitattentionalfusionnetwork AT huiliu remotesensingimagesuperresolutionviaresidualaggregationandsplitattentionalfusionnetwork AT minhangyang remotesensingimagesuperresolutionviaresidualaggregationandsplitattentionalfusionnetwork AT yurongqian remotesensingimagesuperresolutionviaresidualaggregationandsplitattentionalfusionnetwork AT zhengqingxiao remotesensingimagesuperresolutionviaresidualaggregationandsplitattentionalfusionnetwork AT xiwuzhong remotesensingimagesuperresolutionviaresidualaggregationandsplitattentionalfusionnetwork |
_version_ |
1716840440544624640 |