EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Glare-Aware Resi-YOLO: Tiny-Vessel Detection with Dual-Brain Edge Deployment for Maritime UAVs

Shang-En Tsai, Chia-Han Hsieh

Drones · 23.03.2026 · 10 (3) · 226

DOI: 10.3390/drones10030226 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Drones, online erschienen am 23.03.2026. Verfasst von Shang-En Tsai und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: En-Shou Investment Co., Ltd., Taiwan, Ching Lung Agricultural Technology Co., Ltd., Taiwan.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Maritime UAV perception must reliably detect and track tiny vessels under harsh specular glare. In practice, detection failures are dominated by two coupled factors: (i) vessels often occupy only a few pixels, causing small-object recall collapse and (ii) sun glint and sea-surface reflections generate over-exposed regions that trigger false positives and unstable associations. This paper presents Resi-YOLO, a system-level pipeline that improves tiny-vessel sensitivity while preserving embedded throughput on a Jetson Orin Nano. At the model level, Resi-YOLO combines a P2-enhanced feature path with CBAM-based glare suppression to strengthen high-resolution semantics and suppress glare-induced artifacts; optional SAHI-style slicing is supported for ultra-high-resolution scenes. At the system level, we adopt a heterogeneous dual-brain deployment, where the Orin Nano performs primary inference and an MCU-based safety-island tracker mitigates delay/jitter via time-stamped measurement replay and IMM-UKF updates. We further define a Glare Severity Score (GSS) to stratify robustness by illumination intensity. Experiments show that Resi-YOLO improves APsmall by 13.1 percentage points over YOLOv8n (18.4% to 31.5%), raises high-glare mAP@0.5 from 41.2% to 53.7%, and runs at 12.8 FPS end-to-end (~100 ms latency) on Jetson Orin Nano, while TensorRT inference-only throughput exceeds 30 FPS.

Förderung

  • En-Shou Investment Co., Ltd., Taiwan
  • Ching Lung Agricultural Technology Co., Ltd., Taiwan

Zitieren

APA

Tsai, S., & Hsieh, C. (2026). Glare-Aware Resi-YOLO: Tiny-Vessel Detection with Dual-Brain Edge Deployment for Maritime UAVs. Drones, 10(3), 226. https://doi.org/10.3390/drones10030226

Vancouver

Tsai S, Hsieh C. Glare-Aware Resi-YOLO: Tiny-Vessel Detection with Dual-Brain Edge Deployment for Maritime UAVs. Drones. 2026;10(3):226. doi:10.3390/drones10030226

BibTeX
@article{Tsai2026,
  author = {Tsai, Shang-En and Hsieh, Chia-Han},
  title = {Glare-Aware Resi-YOLO: Tiny-Vessel Detection with Dual-Brain Edge Deployment for Maritime UAVs},
  journal = {Drones},
  year = {2026},
  volume = {10},
  number = {3},
  pages = {226},
  doi = {10.3390/drones10030226}
}
Kostenlos starten →