A Hybrid Deep Learning Model for Early Forest Fire Detection

Guardado en:

Detalles Bibliográficos
Publicado en:	Forests vol. 16, no. 5 (2025), p. 863
Autor principal:	Akhror, Mamadmurodov
Otros Autores:	Umirzakova Sabina, Mekhriddin, Rakhimov, Alpamis, Kutlimuratov, Zavqiddin, Temirov, Nasimov Rashid, Azizjon, Meliboev, Akmalbek, Abdusalomov, Im Cho Young
Publicado:	MDPI AG
Materias:	Forest fires Accuracy Public health Deep learning Forest fire detection Edge computing Biodiversity Architecture Visual stimuli Climate change Efficiency Wildfires Vegetation Preprocessing Environmental impact Sensors Effectiveness Natural environment Drones Forest & brush fires Surveillance Latency Object recognition Real time Semantics Edge detection Environmental
Acceso en línea:	Citation/Abstract Full Text + Graphics Full Text - PDF
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Descripción
Resumen:	Forest fires pose an escalating global threat, severely impacting ecosystems, public health, and economies. Timely detection, especially during early stages, is critical for effective intervention. In this study, we propose a novel deep learning-based framework that augments the YOLOv4 object detection architecture with a modified EfficientNetV2 backbone and Efficient Channel Attention (ECA) modules. The backbone substitution leverages compound scaling and Fused-MBConv/MBConv blocks to improve representational efficiency, while the lightweight ECA blocks enhance inter-channel dependency modeling without incurring significant computational overhead. Additionally, we introduce a domain-specific preprocessing pipeline employing Canny edge detection, CLAHE + Jet transformation, and pseudo-NDVI mapping to enhance fire-specific visual cues in complex natural environments. Experimental evaluation on a hybrid dataset of forest fire images and video frames demonstrates substantial performance gains over baseline YOLOv4 and contemporary YOLO variants (YOLOv5–YOLOv9), with the proposed model achieving 97.01% precision, 95.14% recall, 93.13% mAP, and 92.78% F1-score. Furthermore, our model outperforms fourteen state-of-the-art approaches across standard metrics, confirming its efficacy, generalizability, and suitability for real-time deployment in UAV-based and edge computing platforms. These findings highlight the synergy between architectural optimization and domain-aware preprocessing for high-accuracy, low-latency wildfire detection systems.
ISSN:	1999-4907
DOI:	10.3390/f16050863
Fuente:	Agriculture Science Database