EUVIMEDEUVIMED entdecken
Crossref

Integrated Visual Sensing and Efficient Deep Learning for UAV-Based Track Foreign Object Detection

Haifeng Song, Yu Mei, Shaoqing Liu, Zixuan Zhang, Hairong Dong

IEEE Sensors Journal · 01.04.2026 · 26 (7) · 10316-10326

DOI: 10.1109/jsen.2026.3667586 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 01.04.2026. Verfasst von Haifeng Song und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 30 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deploying deep learning models on uncrewed aerial vehicle (UAV)-based edge platforms for railway track foreign object detection is often limited by large model size and heavy computation. To address these issues, this article presents an efficient foreign object detection model, termed RVGC-YOLO. The model introduces three key improvements: 1) designing a new C2f-RVG module to enhance multiscale feature fusion; 2) replacing standard convolutions with GSConv to reduce parameters while maintaining feature extraction capability; and 3) embedding a coordinate attention (CA) mechanism to strengthen focus on critical regions. Furthermore, for efficient deployment on edge devices, this article proposes an efficient model pruning and distillation framework, based on which a lightweight version of RVGC-YOLO, named RVGC-YOLO-Tiny, is introduced. Experiments on the UAV-RSOD dataset demonstrate that RVGC-YOLO achieves 89.60% mAP@0.5 and 63.72% mAP@0.5:0.95. Building upon this performance, the optimized RVGC-YOLO-Tiny significantly reduces parameters and computation by 61.86% and 55.06%, respectively, while maintaining a detection performance nearly consistent with the baseline. These results demonstrate that the proposed model optimization framework provides a practical solution for real-time railway safety monitoring.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Song, H., Mei, Y., Liu, S., Zhang, Z., & Dong, H. (2026). Integrated Visual Sensing and Efficient Deep Learning for UAV-Based Track Foreign Object Detection. IEEE Sensors Journal, 26(7), 10316-10326. https://doi.org/10.1109/jsen.2026.3667586

Vancouver

Song H, Mei Y, Liu S, Zhang Z, Dong H. Integrated Visual Sensing and Efficient Deep Learning for UAV-Based Track Foreign Object Detection. IEEE Sensors Journal. 2026;26(7):10316-10326. doi:10.1109/jsen.2026.3667586

BibTeX
@article{Song2026,
  author = {Song, Haifeng and Mei, Yu and Liu, Shaoqing and Zhang, Zixuan and Dong, Hairong},
  title = {Integrated Visual Sensing and Efficient Deep Learning for UAV-Based Track Foreign Object Detection},
  journal = {IEEE Sensors Journal},
  year = {2026},
  volume = {26},
  number = {7},
  pages = {10316--10326},
  doi = {10.1109/jsen.2026.3667586}
}
Kostenlos starten →