ACT-SR: Aggregation Connection Transformer for Remote Sensing Image Super-Resolution
Recently, transformer-based methods have shown impressive performances in remote sensing image super-resolution (RSISR). However, the application of transformer in RSISR frequently results in artifacts and the loss of image detail due to limited information transmission pathways and the constraints...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10767430/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849764194726969344 |
|---|---|
| author | Yingdong Kang Xinyu Wang Xuemin Zhang Shaoju Wang Guang Jin |
| author_facet | Yingdong Kang Xinyu Wang Xuemin Zhang Shaoju Wang Guang Jin |
| author_sort | Yingdong Kang |
| collection | DOAJ |
| description | Recently, transformer-based methods have shown impressive performances in remote sensing image super-resolution (RSISR). However, the application of transformer in RSISR frequently results in artifacts and the loss of image detail due to limited information transmission pathways and the constraints of unidimensional self-attention mechanisms. To solve these problems, an aggregation connection transformer (ACT-SR) is proposed for RSISR. ACT-SR employs an advanced attention mechanism designed to enrich information transmission across spatial and channel dimensions, thereby enlarging the receptive fields for more accurate feature extraction. A core component of ACT-SR is the novel aggregation connection attention block, which effectively captures spatial similarities and channel importance, aggregating this information through a combination of series and parallel connections for enhanced feature representation. Furthermore, a new gated feed-forward network is introduced to enhance the nonlinear mapping capabilities of the transformer and control the information flow through the network. In addition, ACT-SR integrates a shifted windows scheme alongside interpolation residual calculation to facilitate efficient detail recovery and artifact elimination. Experimental results confirm the effectiveness of the proposed modules, with ACT-SR outperforming several state-of-the-art RSISR methods in both objective metrics and visual quality. |
| format | Article |
| id | doaj-art-5bcf8b275c5c41fe8441ff55ba9a16c8 |
| institution | DOAJ |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-5bcf8b275c5c41fe8441ff55ba9a16c82025-08-20T03:05:11ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01188953896410.1109/JSTARS.2024.350671710767430ACT-SR: Aggregation Connection Transformer for Remote Sensing Image Super-ResolutionYingdong Kang0https://orcid.org/0009-0001-3465-4366Xinyu Wang1https://orcid.org/0000-0002-0493-3954Xuemin Zhang2https://orcid.org/0009-0000-5583-0982Shaoju Wang3https://orcid.org/0009-0005-5730-1597Guang Jin4School of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaRecently, transformer-based methods have shown impressive performances in remote sensing image super-resolution (RSISR). However, the application of transformer in RSISR frequently results in artifacts and the loss of image detail due to limited information transmission pathways and the constraints of unidimensional self-attention mechanisms. To solve these problems, an aggregation connection transformer (ACT-SR) is proposed for RSISR. ACT-SR employs an advanced attention mechanism designed to enrich information transmission across spatial and channel dimensions, thereby enlarging the receptive fields for more accurate feature extraction. A core component of ACT-SR is the novel aggregation connection attention block, which effectively captures spatial similarities and channel importance, aggregating this information through a combination of series and parallel connections for enhanced feature representation. Furthermore, a new gated feed-forward network is introduced to enhance the nonlinear mapping capabilities of the transformer and control the information flow through the network. In addition, ACT-SR integrates a shifted windows scheme alongside interpolation residual calculation to facilitate efficient detail recovery and artifact elimination. Experimental results confirm the effectiveness of the proposed modules, with ACT-SR outperforming several state-of-the-art RSISR methods in both objective metrics and visual quality.https://ieeexplore.ieee.org/document/10767430/Deep learningoptical remote sensingsuper-resolutiontransformer |
| spellingShingle | Yingdong Kang Xinyu Wang Xuemin Zhang Shaoju Wang Guang Jin ACT-SR: Aggregation Connection Transformer for Remote Sensing Image Super-Resolution IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning optical remote sensing super-resolution transformer |
| title | ACT-SR: Aggregation Connection Transformer for Remote Sensing Image Super-Resolution |
| title_full | ACT-SR: Aggregation Connection Transformer for Remote Sensing Image Super-Resolution |
| title_fullStr | ACT-SR: Aggregation Connection Transformer for Remote Sensing Image Super-Resolution |
| title_full_unstemmed | ACT-SR: Aggregation Connection Transformer for Remote Sensing Image Super-Resolution |
| title_short | ACT-SR: Aggregation Connection Transformer for Remote Sensing Image Super-Resolution |
| title_sort | act sr aggregation connection transformer for remote sensing image super resolution |
| topic | Deep learning optical remote sensing super-resolution transformer |
| url | https://ieeexplore.ieee.org/document/10767430/ |
| work_keys_str_mv | AT yingdongkang actsraggregationconnectiontransformerforremotesensingimagesuperresolution AT xinyuwang actsraggregationconnectiontransformerforremotesensingimagesuperresolution AT xueminzhang actsraggregationconnectiontransformerforremotesensingimagesuperresolution AT shaojuwang actsraggregationconnectiontransformerforremotesensingimagesuperresolution AT guangjin actsraggregationconnectiontransformerforremotesensingimagesuperresolution |