Non-Locally up-Down Convolutional Attention Network for Remote Sensing Image Super-Resolution
Recently, single image super-resolution (SISR) has been widely applied in the field of remote sensing image processing and obtained remarkable performance. However, existing CNN-based remote sensing image super-resolution methods are unable to exploit shallow visual characteristics at global recepti...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9189886/ |
| _version_ | 1818917749017018368 |
|---|---|
| author | Huan Wang Qian Hu Chengdong Wu Jianning Chi Xiaosheng Yu |
| author_facet | Huan Wang Qian Hu Chengdong Wu Jianning Chi Xiaosheng Yu |
| author_sort | Huan Wang |
| collection | DOAJ |
| description | Recently, single image super-resolution (SISR) has been widely applied in the field of remote sensing image processing and obtained remarkable performance. However, existing CNN-based remote sensing image super-resolution methods are unable to exploit shallow visual characteristics at global receptive fields, which results in the limited perceptual capability of these models. Furthermore, the low-resolution inputs and features contain abundant low-frequency information, which are weighed in channels and space equally, hence limiting the representational ability of CNNs. To solve these problems, we propose a non-locally up-down convolutional attention network (NLASR) for remote sensing image super-resolution. First, a non-local features enhancement module (NLEB) is constructed to obtain the spatial context information of high-dimensional feature maps, which allows our network to utilize global information to enhance low-level similar structured texture information with effect, overcoming the defects of deficiency perceptual ability of shallow convolutional layers. Second, an enhanced up-sampling channel-wise attention (EUCA) module and enhanced down-sampling spatial-wise attention (EDSA) module are proposed to weight the features at multiple scales. By integrating the channel-wise and multi-scale spatial information, the attention modules are able to compute the response values from the multi-scale regions of each neuron and then establish the accurate mapping from low to high resolution space. Extensive experiments on NWPU-RESISC45 and UCMerced-LandUse datasets show that the proposed method can provide state-of-the-art or even better performance in both quantitative and qualitative measurements. |
| first_indexed | 2024-12-20T00:39:00Z |
| format | Article |
| id | doaj.art-776304a5b9eb47d589d4db9e13eabe01 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T00:39:00Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-776304a5b9eb47d589d4db9e13eabe012022-12-21T19:59:41ZengIEEEIEEE Access2169-35362020-01-01816630416631910.1109/ACCESS.2020.30228829189886Non-Locally up-Down Convolutional Attention Network for Remote Sensing Image Super-ResolutionHuan Wang0https://orcid.org/0000-0002-1404-8529Qian Hu1https://orcid.org/0000-0003-3150-962XChengdong Wu2https://orcid.org/0000-0001-5842-2458Jianning Chi3https://orcid.org/0000-0002-2118-0931Xiaosheng Yu4https://orcid.org/0000-0001-7012-0612Faculty of Robot Science and Engineering, Northeastern University, Shenyang, ChinaFaculty of Information Science and Engineering, Northeastern University, Shenyang, ChinaFaculty of Robot Science and Engineering, Northeastern University, Shenyang, ChinaFaculty of Robot Science and Engineering, Northeastern University, Shenyang, ChinaFaculty of Robot Science and Engineering, Northeastern University, Shenyang, ChinaRecently, single image super-resolution (SISR) has been widely applied in the field of remote sensing image processing and obtained remarkable performance. However, existing CNN-based remote sensing image super-resolution methods are unable to exploit shallow visual characteristics at global receptive fields, which results in the limited perceptual capability of these models. Furthermore, the low-resolution inputs and features contain abundant low-frequency information, which are weighed in channels and space equally, hence limiting the representational ability of CNNs. To solve these problems, we propose a non-locally up-down convolutional attention network (NLASR) for remote sensing image super-resolution. First, a non-local features enhancement module (NLEB) is constructed to obtain the spatial context information of high-dimensional feature maps, which allows our network to utilize global information to enhance low-level similar structured texture information with effect, overcoming the defects of deficiency perceptual ability of shallow convolutional layers. Second, an enhanced up-sampling channel-wise attention (EUCA) module and enhanced down-sampling spatial-wise attention (EDSA) module are proposed to weight the features at multiple scales. By integrating the channel-wise and multi-scale spatial information, the attention modules are able to compute the response values from the multi-scale regions of each neuron and then establish the accurate mapping from low to high resolution space. Extensive experiments on NWPU-RESISC45 and UCMerced-LandUse datasets show that the proposed method can provide state-of-the-art or even better performance in both quantitative and qualitative measurements.https://ieeexplore.ieee.org/document/9189886/Single image super-resolution (SISR)channel-wise and space-wise attention mechanismsdeep learningremote sensing image processing |
| spellingShingle | Huan Wang Qian Hu Chengdong Wu Jianning Chi Xiaosheng Yu Non-Locally up-Down Convolutional Attention Network for Remote Sensing Image Super-Resolution IEEE Access Single image super-resolution (SISR) channel-wise and space-wise attention mechanisms deep learning remote sensing image processing |
| title | Non-Locally up-Down Convolutional Attention Network for Remote Sensing Image Super-Resolution |
| title_full | Non-Locally up-Down Convolutional Attention Network for Remote Sensing Image Super-Resolution |
| title_fullStr | Non-Locally up-Down Convolutional Attention Network for Remote Sensing Image Super-Resolution |
| title_full_unstemmed | Non-Locally up-Down Convolutional Attention Network for Remote Sensing Image Super-Resolution |
| title_short | Non-Locally up-Down Convolutional Attention Network for Remote Sensing Image Super-Resolution |
| title_sort | non locally up down convolutional attention network for remote sensing image super resolution |
| topic | Single image super-resolution (SISR) channel-wise and space-wise attention mechanisms deep learning remote sensing image processing |
| url | https://ieeexplore.ieee.org/document/9189886/ |
| work_keys_str_mv | AT huanwang nonlocallyupdownconvolutionalattentionnetworkforremotesensingimagesuperresolution AT qianhu nonlocallyupdownconvolutionalattentionnetworkforremotesensingimagesuperresolution AT chengdongwu nonlocallyupdownconvolutionalattentionnetworkforremotesensingimagesuperresolution AT jianningchi nonlocallyupdownconvolutionalattentionnetworkforremotesensingimagesuperresolution AT xiaoshengyu nonlocallyupdownconvolutionalattentionnetworkforremotesensingimagesuperresolution |