NGSTGAN: N-Gram Swin Transformer and Multi-Attention U-Net Discriminator for Efficient Multi-Spectral Remote Sensing Image Super-Resolution
Remote Sensing · 17.06.2025 · 17 (12) · 2079
DOI: 10.3390/rs17122079 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 17.06.2025. Verfasst von Chao Zhan und 5 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 77 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Chunhui Program Cooperative Research Project of the Chinese Ministry of Educatio, Henan Provincial Science and Technology Research Project, Key Research Project Fund of Institution of Higher Education in Henan Province.
Abstract (Originalsprache)
The reconstruction of high-resolution (HR) remote sensing images (RSIs) from low-resolution (LR) counterparts is a critical task in remote sensing image super-resolution (RSISR). Recent advancements in convolutional neural networks (CNNs) and Transformers have significantly improved RSISR performance due to their capabilities in local feature extraction and global modeling. However, several limitations remain, including the underutilization of multi-scale features in RSIs, the limited receptive field of Swin Transformer’s window self-attention (WSA), and the computational complexity of existing methods. To address these issues, this paper introduces the NGSTGAN model, which employs an N-Gram Swin Transformer as the generator and a multi-attention U-Net as the discriminator. The discriminator enhances attention to multi-scale key features through the addition of channel, spatial, and pixel attention (CSPA) modules, while the generator utilizes an improved shallow feature extraction (ISFE) module to extract multi-scale and multi-directional features, enhancing the capture of complex textures and details. The N-Gram concept is introduced to expand the receptive field of Swin Transformer, and sliding window self-attention (S-WSA) is employed to facilitate interaction between neighboring windows. Additionally, channel-reducing group convolution (CRGC) is used to reduce the number of parameters and computational complexity. A cross-sensor multispectral dataset combining Landsat-8 (L8) and Sentinel-2 (S2) is constructed for the resolution enhancement of L8’s blue (B), green (G), red (R), and near-infrared (NIR) bands from 30 m to 10 m. Experiments show that NGSTGAN outperforms the state-of-the-art (SOTA) method, achieving improvements of 0.5180 dB in the peak signal-to-noise ratio (PSNR) and 0.0153 in the structural similarity index measure (SSIM) over the second best method, offering a more effective solution to the task.
Förderung
- Chunhui Program Cooperative Research Project of the Chinese Ministry of Educatio
- Henan Provincial Science and Technology Research Project
- Key Research Project Fund of Institution of Higher Education in Henan Province
- Henan Polytechnic University for the Double First-Class Project of Surveying and Mapping Disciplines
- Japan Society for the Promotion of Science (JSPS) KAKENHI
Zitieren
APA
Zhan, C., Wang, C., Lu, B., Yang, W., Zhang, X., & Wang, G. (2025). NGSTGAN: N-Gram Swin Transformer and Multi-Attention U-Net Discriminator for Efficient Multi-Spectral Remote Sensing Image Super-Resolution. Remote Sensing, 17(12), 2079. https://doi.org/10.3390/rs17122079
Vancouver
Zhan C, Wang C, Lu B, Yang W, Zhang X, Wang G. NGSTGAN: N-Gram Swin Transformer and Multi-Attention U-Net Discriminator for Efficient Multi-Spectral Remote Sensing Image Super-Resolution. Remote Sensing. 2025;17(12):2079. doi:10.3390/rs17122079
BibTeX
@article{Zhan2025,
author = {Zhan, Chao and Wang, Chunyang and Lu, Bibo and Yang, Wei and Zhang, Xian and Wang, Gaige},
title = {NGSTGAN: N-Gram Swin Transformer and Multi-Attention U-Net Discriminator for Efficient Multi-Spectral Remote Sensing Image Super-Resolution},
journal = {Remote Sensing},
year = {2025},
volume = {17},
number = {12},
pages = {2079},
doi = {10.3390/rs17122079}
}
EUVIMED