Research on extraction and reconstruction of power equipment image data based on dual-feature fusion
International Journal of Low-Carbon Technologies · 27.01.2026 · 21
DOI: 10.1093/ijlct/ctag081 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Low-Carbon Technologies, online erschienen am 15.07.2026. Verfasst von Junjie Zha und 5 weiteren Autor:innen. Die Arbeit nennt 383 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Dual-channel fusion module leverages a channel attention mechanism to achieve cross-channel fusion and enhancement between depth and intensity features, effectively improving the expression of critical information. Experimental results on a representative power equipment image dataset demonstrate that the proposed method outperforms existing approaches such as Reciprocal Rank Fusion (RRF), joint bilateral upsampling, and Depth-Net in terms of peak signal-to-noise ratio (PSNR) and root mean square error (RMSE). In particular, for 4× super-resolution reconstruction tasks, the method achieves an average PSNR improvement of 6.79 dB and an average RMSE reduction of 0.94, validating the effectiveness and superiority of the proposed approach in power equipment image data extraction and reconstruction.
Zitieren
APA
Zha, J., Shan, X., Teng, A., Li, M., Cheng, X., & Liu, Z. (2026). Research on extraction and reconstruction of power equipment image data based on dual-feature fusion. International Journal of Low-Carbon Technologies, 21. https://doi.org/10.1093/ijlct/ctag081
Vancouver
Zha J, Shan X, Teng A, Li M, Cheng X, Liu Z. Research on extraction and reconstruction of power equipment image data based on dual-feature fusion. International Journal of Low-Carbon Technologies. 2026;21. doi:10.1093/ijlct/ctag081
BibTeX
@article{Zha2026,
author = {Zha, Junjie and Shan, Xinwen and Teng, Aiguo and Li, Meng and Cheng, Xinyun and Liu, Zihan},
title = {Research on extraction and reconstruction of power equipment image data based on dual-feature fusion},
journal = {International Journal of Low-Carbon Technologies},
year = {2026},
volume = {21},
doi = {10.1093/ijlct/ctag081}
}
EUVIMED