EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep learning-based forest fire detection using an improved SSD algorithm with CBAM

Diansheng Zhang iD, Yueyuan Zhang iD, Leilei Dong, Shifeng Ruan, Zhiwei Liu

PLOS One · 18.11.2025 · 20 (11) · e0333574

DOI: 10.1371/journal.pone.0333574 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 18.11.2025. Verfasst von Diansheng Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Information and Software Engineering, East China Jiaotong University, Nanchang, China. Hauptthemen laut MEDLINE: Deep Learning, Forests, Algorithms, Wildfires. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Graduate Innovative Special Fund Projects of Jiangxi Province, Anyang City Science and Technology Research Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Fires are characterized by their sudden onset, rapid spread, and destructive nature, often causing irreversible damage to ecosystems. To address the challenges in forest fire detection, including the varying scales and complex features of flame and smoke, as well as false positives and missed detections caused by environmental interference, we propose a novel object detection model named CBAM-SSD. Firstly, data augmentation techniques involving geometric and color transformations are employed to enrich the dataset, effectively mitigating issues of insufficient and incomplete data collected in real-world scenarios. This significantly enhances the SSD model’s ability to detect flames, which exhibit highly variable morphological characteristics. Furthermore, the CBAM module is integrated into the SSD backbone network to reconstruct its feature extraction structure. This module adaptively weights flame color and smoke texture along the channel dimension and highlights critical fire regions in the spatial dimension, substantially improving the model’s perception of key fire features. Experimental results demonstrate that the CBAM-SSD model is lightweight and suitable for real-time detection, achieving a mAP@0.5 of 97.55% for flames and smoke, a 1.53% improvement over the baseline SSD. Specifically, the AP50 for flame detection reaches 96.61%, a 3.01% increase compared to the baseline, with a recall of 96.40%; while the AP50 for smoke detection reaches 98.49%, with a recall of 98.80%. These results indicate that the improved model delivers higher detection accuracy and lower false and missed detection rates, offering an efficient, convenient, and accurate solution for forest fire detection.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningForestsAlgorithmsWildfiresFires

Autor:innen und Institutionen

  • School of Information and Software Engineering, East China Jiaotong University, Nanchang, China.
  • School of Communications and Information Engineering, Xi'an University of Posts and Telecommunications, Xian, China.
  • School of Flight, Anyang Institute of Technology, Anyang, China.

ORCID-Profile: Diansheng Zhang, Yueyuan Zhang

Förderung

  • Graduate Innovative Special Fund Projects of Jiangxi Province
  • Anyang City Science and Technology Research Project

Zitieren

APA

Zhang, D., Zhang, Y., Dong, L., Ruan, S., & Liu, Z. (2025). Deep learning-based forest fire detection using an improved SSD algorithm with CBAM. PLOS One, 20(11), e0333574. https://doi.org/10.1371/journal.pone.0333574

Vancouver

Zhang D, Zhang Y, Dong L, Ruan S, Liu Z. Deep learning-based forest fire detection using an improved SSD algorithm with CBAM. PLOS One. 2025;20(11):e0333574. doi:10.1371/journal.pone.0333574

BibTeX
@article{Zhang2025,
  author = {Zhang, Diansheng and Zhang, Yueyuan and Dong, Leilei and Ruan, Shifeng and Liu, Zhiwei},
  title = {Deep learning-based forest fire detection using an improved SSD algorithm with CBAM},
  journal = {PLOS One},
  year = {2025},
  volume = {20},
  number = {11},
  pages = {e0333574},
  doi = {10.1371/journal.pone.0333574}
}
Kostenlos starten →