EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

FCMI-YOLO: An efficient deep learning-based algorithm for real-time fire detection on edge devices

Junjie Lu iD, Yuchen Zheng, Liwei Guan iD, Bing Lin, Wenzao Shi, Junyan Zhang, Yunping Wu

PLOS One · 07.08.2025 · 20 (8) · e0329555

DOI: 10.1371/journal.pone.0329555 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 07.08.2025. Verfasst von Junjie Lu und 6 weiteren Autor:innen. Beteiligte Einrichtung: College of Photonic and Electronic Engineering, Fujian Normal University, Fujian, China. Hauptthemen laut MEDLINE: Deep Learning, Algorithms, Fires. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Department of Education, Fujian Province, Science and Technology Bureau of Quanzhou, Natural Science Foundation of Fujian Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid development of Internet of Things (IoT) technology and deep learning has propelled the deployment of vision-based fire detection algorithms on edge devices, significantly exacerbating the trade-off between accuracy and inference speed under hardware resource constraints. To address this issue, this paper proposes FCMI-YOLO, a real-time fire detection algorithm optimized for edge devices. Firstly, the FasterNext module is proposed to reduce computational cost and enhance detection precision through lightweight design. Secondly, the Cross-Scale Feature Fusion Module (CCFM) and the Mixed Local Channel Attention (MLCA) mechanism are incorporated into the neck network to improve detection performance for small fire targets and reduce resource consumption. Finally, the Inner-DIoU loss function is proposed to optimize bounding box regression. Experimental results on a custom fire dataset demonstrate that FCMI-YOLO increases mAP@50 by 1.5%, reduces parameters by 40%, and lowers GFLOPs to 28.9% of YOLOv5s, demonstrating its practical value for real-time fire detection in edge scenarios with limited computational resources. The core code and dataset are available at https://github.com/ JunJieLu20230823/code.git .

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningAlgorithmsFiresInternet of Things

Autor:innen und Institutionen

  • College of Photonic and Electronic Engineering, Fujian Normal University, Fujian, China.
  • College of Physics and Energy, Fujian Normal University, Fujian, China.

ORCID-Profile: Junjie Lu, Liwei Guan

Förderung

  • Department of Education, Fujian Province
  • Science and Technology Bureau of Quanzhou
  • Natural Science Foundation of Fujian Province

Zitieren

APA

Lu, J., Zheng, Y., Guan, L., Lin, B., Shi, W., Zhang, J., & Wu, Y. (2025). FCMI-YOLO: An efficient deep learning-based algorithm for real-time fire detection on edge devices. PLOS One, 20(8), e0329555. https://doi.org/10.1371/journal.pone.0329555

Vancouver

Lu J, Zheng Y, Guan L, Lin B, Shi W, Zhang J, et al. FCMI-YOLO: An efficient deep learning-based algorithm for real-time fire detection on edge devices. PLOS One. 2025;20(8):e0329555. doi:10.1371/journal.pone.0329555

BibTeX
@article{Lu2025,
  author = {Lu, Junjie and Zheng, Yuchen and Guan, Liwei and Lin, Bing and Shi, Wenzao and Zhang, Junyan and Wu, Yunping},
  title = {FCMI-YOLO: An efficient deep learning-based algorithm for real-time fire detection on edge devices},
  journal = {PLOS One},
  year = {2025},
  volume = {20},
  number = {8},
  pages = {e0329555},
  doi = {10.1371/journal.pone.0329555}
}
Kostenlos starten →