EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Multi-Scale Recursive Attention Feature Fusion Network for Image Super-Resolution Reconstruction Algorithm

Xiaowei Han iD, Lei Wang, Xiaopeng Wang iD, Pengchao Zhang, Haoran Xu

Sensors · 28.11.2023 · 23 (23) · 9458

DOI: 10.3390/s23239458 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 28.11.2023. Verfasst von Xiaowei Han und 4 weiteren Autor:innen. Beteiligte Einrichtung: The Key Laboratory of Industrial Automation of Shaanxi Province, Shaanxi University of Technology, Hanzhong 723000, China. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the National Natural Science Foundation of China, the National Social Science Foundation of China, the National Education Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In recent years, deep convolutional neural networks (CNNs) have made significant progress in single-image super-resolution (SISR) tasks. Despite their good performance, the single-image super-resolution task remains a challenging one due to problems with underutilization of feature information and loss of feature details. In this paper, a multi-scale recursive attention feature fusion network (MSRAFFN) is proposed for this purpose. The network consists of three parts: a shallow feature extraction module, a multi-scale recursive attention feature fusion module, and a reconstruction module. The shallow features of the image are first extracted by the shallow feature extraction module. Then, the feature information at different scales is extracted by the multi-scale recursive attention feature fusion network block (MSRAFFB) to enhance the channel features of the network through the attention mechanism and fully fuse the feature information at different scales in order to improve the network’s performance. In addition, the image features at different levels are integrated through cross-layer connections using residual connections. Finally, in the reconstruction module, the upsampling capability of the deconvolution module is used to enlarge the image while extracting its high-frequency information in order to obtain a sharper high-resolution image and achieve a better visual effect. Through extensive experiments on a benchmark dataset, the proposed network model is shown to have better performance than other models in terms of both subjective visual effects and objective evaluation metrics.

Schlagwörter

Stichwörter der Autor:innen

attention feature fusionmulti-scale featurerecursive networkssuper-resolution reconstruction

Autor:innen und Institutionen

  • The Key Laboratory of Industrial Automation of Shaanxi Province, Shaanxi University of Technology, Hanzhong 723000, China.

ORCID-Profile: Xiaowei Han, Xiaopeng Wang

Förderung

  • the National Natural Science Foundation of China
  • the National Social Science Foundation of China
  • the National Education Science Foundation of China
  • Key Project of Shaanxi Provincial Natural Science Basic Research Program
  • National Natural Science Foundation of China
  • National Social Science Foundation of China
  • National Education Science Foundation of China

Zitieren

APA

Han, X., Wang, L., Wang, X., Zhang, P., & Xu, H. (2023). A Multi-Scale Recursive Attention Feature Fusion Network for Image Super-Resolution Reconstruction Algorithm. Sensors, 23(23), 9458. https://doi.org/10.3390/s23239458

Vancouver

Han X, Wang L, Wang X, Zhang P, Xu H. A Multi-Scale Recursive Attention Feature Fusion Network for Image Super-Resolution Reconstruction Algorithm. Sensors. 2023;23(23):9458. doi:10.3390/s23239458

BibTeX
@article{Han2023,
  author = {Han, Xiaowei and Wang, Lei and Wang, Xiaopeng and Zhang, Pengchao and Xu, Haoran},
  title = {A Multi-Scale Recursive Attention Feature Fusion Network for Image Super-Resolution Reconstruction Algorithm},
  journal = {Sensors},
  year = {2023},
  volume = {23},
  number = {23},
  pages = {9458},
  doi = {10.3390/s23239458}
}
Kostenlos starten →