EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

EF-YOLO: Detecting Small Targets in Early-Stage Agricultural Fires via UAV-Based Remote Sensing

Jun Tao, Zhihan Wang, Jianqiu Wu, Yunqin Li, Tomohiro Fukuda, Jiaxin Zhang

Remote Sensing · 09.04.2026 · 18 (8) · 1119

DOI: 10.3390/rs18081119 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 09.04.2026. Verfasst von Jun Tao und 5 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation Youth Fund Project of Jiangxi Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Early detection of agricultural fires with Unmanned Aerial Vehicles (UAVs) is important for environmental safety, yet it remains difficult because ignition cues are extremely small, smoke patterns vary widely, and farmland scenes often contain strong background interference such as specular reflections. Model development is further constrained by the scarcity of data from the early ignition stage. To address these challenges, we propose a joint data and model optimization framework. We first build a hybrid dataset through an ROI-guided synthesis pipeline, in which latent diffusion models are used to insert high-fidelity, carefully screened fire samples into real farmland backgrounds. We then introduce EF-YOLO, a detector designed for high sensitivity to small targets. The network uses SPD-Conv to reduce feature loss during spatial downsampling and includes a high-resolution P2 head to improve the detection of minute objects. To reduce background clutter, a Dual-Path Frequency–Spatial Enhancement (DP-FSE) module serves as a lightweight statistical surrogate that extracts global contextual cues and local salient features in parallel, thereby suppressing high-frequency noise. Experimental results show that EF-YOLO achieves an APS of 40.2% on sub-pixel targets, exceeding the YOLOv8s baseline by 15.4 percentage points. With a recall of 88.7% and a real-time inference speed of 78 FPS, the proposed framework offers a strong balance between detection performance and efficiency, making it well suited for edge-deployed agricultural fire early-warning systems.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation Youth Fund Project of Jiangxi Province

Zitieren

APA

Tao, J., Wang, Z., Wu, J., Li, Y., Fukuda, T., & Zhang, J. (2026). EF-YOLO: Detecting Small Targets in Early-Stage Agricultural Fires via UAV-Based Remote Sensing. Remote Sensing, 18(8), 1119. https://doi.org/10.3390/rs18081119

Vancouver

Tao J, Wang Z, Wu J, Li Y, Fukuda T, Zhang J. EF-YOLO: Detecting Small Targets in Early-Stage Agricultural Fires via UAV-Based Remote Sensing. Remote Sensing. 2026;18(8):1119. doi:10.3390/rs18081119

BibTeX
@article{Tao2026,
  author = {Tao, Jun and Wang, Zhihan and Wu, Jianqiu and Li, Yunqin and Fukuda, Tomohiro and Zhang, Jiaxin},
  title = {EF-YOLO: Detecting Small Targets in Early-Stage Agricultural Fires via UAV-Based Remote Sensing},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {8},
  pages = {1119},
  doi = {10.3390/rs18081119}
}
Kostenlos starten →