A Lightweight Forest Fire Detection Algorithm Based on an Improved YOLOv11
International Journal of Advanced Computer Science and Applications · 2026 · 17 (10)
DOI: 10.14569/ijacsa.2026.0171007 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Computer Science and Applications, 2026. Verfasst von Jingzhou Tao und 4 weiteren Autor:innen.
Abstract (Originalsprache)
This study presents IWL-YOLOv11 (Improved Wildfire Lightweight YOLOv11), a detector that balances forest fire detection accuracy and parameter count. The model introduces three components. A Dual-Branch Gated Depthwise Convolution Block (DGCB) modifies the C3K2 backbone through depthwise separable convolutions and element-wise interaction between two feature branches. A Hybrid Convolution-Attention Network (HCAN) combines features at multiple scales with self-attention at the junction between the backbone and neck. The regression objective uses WDIoU, a combination of Wise Intersection-over-Union loss and Distribution Focal Loss. Experiments use a modified Flame dataset containing 8,625 images and a separately constructed ForestFire dataset containing 2,096 images. IWL-YOLOv11 achieves mAP@0.50 of 88.9 percent and 90.2 percent, respectively, exceeding the YOLOv11 baseline by 9.6 and 6.0 percentage points. Its mAP averaged over IoU thresholds from 0.50 to 0.95 is 62.4 percent on Flame and 78.2 percent on ForestFire. The model has 2,738,750 parameters, approximately 71 percent fewer than the baseline. Cumulative ablations show progressively higher detection metrics across the tested configurations. The reported results indicate a favorable accuracy–parameter trade-off on these image splits. Inference latency, edge-device performance, and uncertainty across repeated training runs are not reported, so the parameter reduction alone does not establish real-time deployment performance.
Zitieren
APA
Tao, J., Liu, Z., Lyu, R., Li, J., & Cao, X. (2026). A Lightweight Forest Fire Detection Algorithm Based on an Improved YOLOv11. International Journal of Advanced Computer Science and Applications, 17(10). https://doi.org/10.14569/ijacsa.2026.0171007
Vancouver
Tao J, Liu Z, Lyu R, Li J, Cao X. A Lightweight Forest Fire Detection Algorithm Based on an Improved YOLOv11. International Journal of Advanced Computer Science and Applications. 2026;17(10). doi:10.14569/ijacsa.2026.0171007
BibTeX
@article{Tao2026,
author = {Tao, Jingzhou and Liu, Zijing and Lyu, Ruoxi and Li, Jie and Cao, Xu},
title = {A Lightweight Forest Fire Detection Algorithm Based on an Improved YOLOv11},
journal = {International Journal of Advanced Computer Science and Applications},
year = {2026},
volume = {17},
number = {10},
doi = {10.14569/ijacsa.2026.0171007}
}
EUVIMED