UMFNet: Frequency-Guided Multi-Scale Fusion with Dynamic Noise Suppression for Robust Low-Light Object Detection
Applied Sciences · 11.05.2025 · 15 (10) · 5362
DOI: 10.3390/app15105362 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 11.05.2025. Verfasst von Shihao Gong und 2 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The dominant low-light object detectors face the following spectral trilemma: (1) the loss of high-frequency structural details during denoising, (2) the amplification of low-frequency illumination distortion, and (3) cross-band interference in multi-scale features. To resolve these intertwined challenges, we present UMFNet—a frequency-guided detection framework that unifies adaptive frequency distillation with inter-band attention coordination. Our technical breakthroughs manifest through three key innovations: (1) a frequency-adaptive fusion (FAF) module employing learnable wavelet kernels (16–64 decomposition basis) with dynamic SNR-gated thresholding, achieving an 89.7% photon utilization rate in ≤1 lux conditions—2.4× higher than fixed-basis approaches; (2) a spatial-channel coordinated attention (SCCA) mechanism with dual-domain nonlinear gating that reduces high-frequency hallucination by 37% through parametric phase alignment (verified via gradient direction alignment coefficient ρG = 0.93); (3) a spectral perception loss combining the frequency-weighted structural similarity index measure (SSIM) with gradient-aware focal modulation, enforcing physics-constrained feature recovery. Extensive validation demonstrates UMFNet’s leadership: 73.1% mAP@50 on EXDark (+6.4% over YOLOv8 baseline), 58.7% on DarkFace (+3.1% over GLARE), and 40.2% on thermal FLIR ADAS (+9.7% improvement). This work pioneers a new paradigm for precision-critical vision systems in photon-starved environments.
Zitieren
APA
Gong, S., Ma, Z., & Li, X. (2025). UMFNet: Frequency-Guided Multi-Scale Fusion with Dynamic Noise Suppression for Robust Low-Light Object Detection. Applied Sciences, 15(10), 5362. https://doi.org/10.3390/app15105362
Vancouver
Gong S, Ma Z, Li X. UMFNet: Frequency-Guided Multi-Scale Fusion with Dynamic Noise Suppression for Robust Low-Light Object Detection. Applied Sciences. 2025;15(10):5362. doi:10.3390/app15105362
BibTeX
@article{Gong2025,
author = {Gong, Shihao and Ma, Zheng and Li, Xiang},
title = {UMFNet: Frequency-Guided Multi-Scale Fusion with Dynamic Noise Suppression for Robust Low-Light Object Detection},
journal = {Applied Sciences},
year = {2025},
volume = {15},
number = {10},
pages = {5362},
doi = {10.3390/app15105362}
}
EUVIMED