Dual domain multi-scale feature enhancement network for light field denoising
Optics Express · 15.06.2026 · 34 (12) · 21499
DOI: 10.1364/oe.596627 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optics Express, online erschienen am 04.06.2026. Verfasst von Xing Liu und 5 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Noise significantly degrades light field (LF) imaging quality, particularly in handheld LF cameras due to the light-scattering effects of microlens arrays. However, current LF denoising methods suffer from several limitations: (1) predominant focus on the pixel domain restricts sufficient exploration of LF frequency characteristics; (2) excessive dependence on local information or straightforward addition/concatenation for integrating global and local features impairs global structural consistency and introduces artifacts. Therefore, we propose a dual domain multi-scale network (DDMSN) that integrates frequency-domain noise priors with pixel-domain texture and epipolar geometric structure awareness. Specifically, in the frequency domain module, we propose a dual frequency collaborative enhancement (DFCE) strategy for differentiated treatment of high frequency noise characteristics and valid structural information, and adaptively enhance low frequency and high frequency band components to achieve a fine-grained restoration of the frequency domain structure. In the pixel domain module, to achieve texture detail restoration and structural consistency preservation, we introduce a global-local feature interaction (GLFI) mechanism embedded within the encoder-decoder architecture, which fuses spatial-angular local features with EPI global features through channel attention weighting, optimizing local receptive fields and global structural consistency dependencies jointly. Extensive experiments on synthetic and real-world LF noise datasets demonstrate that our method achieves significant improvements over existing methods in both qualitative visual results and quantitative metrics.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Liu, X., Zhang, W., An, P., Huang, X., Yang, C., & Zhu, C. (2026). Dual domain multi-scale feature enhancement network for light field denoising. Optics Express, 34(12), 21499. https://doi.org/10.1364/oe.596627
Vancouver
Liu X, Zhang W, An P, Huang X, Yang C, Zhu C. Dual domain multi-scale feature enhancement network for light field denoising. Optics Express. 2026;34(12):21499. doi:10.1364/oe.596627
BibTeX
@article{Liu2026,
author = {Liu, Xing and Zhang, Wenbo and An, Ping and Huang, Xinpeng and Yang, Chao and Zhu, Ce},
title = {Dual domain multi-scale feature enhancement network for light field denoising},
journal = {Optics Express},
year = {2026},
volume = {34},
number = {12},
pages = {21499},
doi = {10.1364/oe.596627}
}
EUVIMED