MFFDet: Enhancing Multi-Scale Forest Fire Detection in UAV Imagery
Fire · 04.07.2026 · 9 (7) · 278
DOI: 10.3390/fire9070278 ↗Kurz gefasst
Fachartikel in der Zeitschrift Fire, online erschienen am 04.07.2026. Verfasst von Zhengshen Huang und 7 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Yunnan International Joint Laboratory of Natural Rubber Intelligent Monitor and Digital Applications, Yunnan International Joint R&D Center (China-Malaysia) for Digital Monitoring, Management and Applications of Nature Reserves.
Abstract (Originalsprache)
In Unmanned aerial vehicle (UAV) forest fire detection, flames and smoke exhibit dramatic scale variations. Existing methods often struggle with multi-scale feature extraction, fusion quality, and localization reliability, resulting in limited accuracy improvements. To address this issue, this study optimizes the backbone, neck, and head of YOLOv11n to propose a novel multi-scale forest fire detector (MFFDet), which consists of three key modules: (1) the Multi-Scale Feature Calibration Module (MFCM) is designed to improve multi-scale feature representation by context aggregation and detail calibration; (2) the Cross-Scale Semantic Alignment Module (CSAM) is proposed to enhance fusion quality by applying channel reorganization and local spatial refinement; and (3) the Location Quality Estimator Head (LQEH) is presented for reliable localization by mapping the statistical information of regression distributions into a localization quality score, which systematically boosts the accuracy and stability of multi-scale object detection. In addition, to alleviate the scarcity of UAV forest fire detection data, this study constructs a UAV Forest Fire Dataset (UF2D), providing important data support for UAV-based fire detection. Experiments on UF2D show that MFFDet achieves an mAP@0.5 of 70.1%, the best among all compared models, representing a 4.4% improvement over the baseline. Moreover, it attains the top performance on small, medium, and large objects, with APs of 20.3%, APm of 31.5%, and APl of 44.8%, highlighting MFFDet’s robustness and accuracy for multi-scale flame and smoke detection in a complex forest fire environment, which bears important practical significance for the intelligent upgrade of forest fire prevention and control.
Förderung
- National Natural Science Foundation of China
- Yunnan International Joint Laboratory of Natural Rubber Intelligent Monitor and Digital Applications
- Yunnan International Joint R&D Center (China-Malaysia) for Digital Monitoring, Management and Applications of Nature Reserves
- Yunnan Fundamental Research Projects
- Xingdian Talent Support Program
- Academician Li Wei Workstation of Yunnan Province
Zitieren
APA
Huang, Z., Wang, R., Li, X., Kou, W., Gu, Q., Li, Z., Ye, J., & Wang, Q. (2026). MFFDet: Enhancing Multi-Scale Forest Fire Detection in UAV Imagery. Fire, 9(7), 278. https://doi.org/10.3390/fire9070278
Vancouver
Huang Z, Wang R, Li X, Kou W, Gu Q, Li Z, et al. MFFDet: Enhancing Multi-Scale Forest Fire Detection in UAV Imagery. Fire. 2026;9(7):278. doi:10.3390/fire9070278
BibTeX
@article{Huang2026,
author = {Huang, Zhengshen and Wang, Rui and Li, Xin and Kou, Weili and Gu, Qinyan and Li, Zengxing and Ye, Jiangxia and Wang, Qiuhua},
title = {MFFDet: Enhancing Multi-Scale Forest Fire Detection in UAV Imagery},
journal = {Fire},
year = {2026},
volume = {9},
number = {7},
pages = {278},
doi = {10.3390/fire9070278}
}
EUVIMED