Lightweight image super-resolution reconstruction method based on progressive space
Applied Optics · 20.06.2026 · 65 (18) · 6030
DOI: 10.1364/ao.579497 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Optics, online erschienen am 11.06.2026. Verfasst von Pengnan Liu und 5 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
In image super-resolution reconstruction tasks, the size of the receptive field is crucial for capturing image information. Current mainstream reconstruction algorithms primarily rely on stacking convolutional layers or increasing the size of convolutional kernels to expand the receptive field and acquire more global image information. However, such approaches often lead to redundancy in image information and fail to fully utilize the richness of receptive field information across different scale spaces. To address this issue, this study proposes a lightweight image super-resolution reconstruction method based on progressive sampling. The core of this method lies in the design of a progressive multi-scale spatial module, which extracts multi-level feature information by progressively adjusting the image dimensions. Additionally, a progressive spatial attention mechanism is introduced to effectively explore the intrinsic correlations among multi-scale features within different receptive field spaces. Experimental results demonstrate that, compared to the existing popular reconstruction methods, the proposed approach achieves superior performance in both quantitative metrics and visual reconstruction quality. Specifically, our model achieves maximum quantitative evaluation scores of 38.18 dB for PSNR and 0.9611 for SSIM on the Set5 dataset (at ×2 scale) while maintaining a highly efficient and lightweight architecture with only 28G FLOPs and 670K parameters for ×4 scale reconstruction.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Liu, P., Su, Z., Cheng, Z., Niu, K., Cui, W., & Jiang, H. (2026). Lightweight image super-resolution reconstruction method based on progressive space. Applied Optics, 65(18), 6030. https://doi.org/10.1364/ao.579497
Vancouver
Liu P, Su Z, Cheng Z, Niu K, Cui W, Jiang H. Lightweight image super-resolution reconstruction method based on progressive space. Applied Optics. 2026;65(18):6030. doi:10.1364/ao.579497
BibTeX
@article{Liu2026,
author = {Liu, Pengnan and Su, Zhen and Cheng, Zhiwei and Niu, Kaikai and Cui, Wen and Jiang, He},
title = {Lightweight image super-resolution reconstruction method based on progressive space},
journal = {Applied Optics},
year = {2026},
volume = {65},
number = {18},
pages = {6030},
doi = {10.1364/ao.579497}
}
EUVIMED