A Hybrid Deep Learning Model for Early Forest Fire Detection

Guardat en:
Dades bibliogràfiques
Publicat a:Forests vol. 16, no. 5 (2025), p. 863
Autor principal: Akhror, Mamadmurodov
Altres autors: Umirzakova Sabina, Mekhriddin, Rakhimov, Alpamis, Kutlimuratov, Zavqiddin, Temirov, Nasimov Rashid, Azizjon, Meliboev, Akmalbek, Abdusalomov, Im Cho Young
Publicat:
MDPI AG
Matèries:
Accés en línia:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Etiquetes: Afegir etiqueta
Sense etiquetes, Sigues el primer a etiquetar aquest registre!

MARC

LEADER 00000nab a2200000uu 4500
001 3211971391
003 UK-CbPIL
022 |a 1999-4907 
024 7 |a 10.3390/f16050863  |2 doi 
035 |a 3211971391 
045 2 |b d20250501  |b d20250531 
084 |a 231463  |2 nlm 
100 1 |a Akhror, Mamadmurodov  |u Department of Computer Engineering, Gachon University, Seongnam-si 13120, Republic of Korea; akhror1999@gachon.ac.kr (A.M.); sabinatuit@gachon.ac.kr (S.U.); akmaljon@gachon.ac.kr (A.A.) 
245 1 |a A Hybrid Deep Learning Model for Early Forest Fire Detection 
260 |b MDPI AG  |c 2025 
513 |a Journal Article 
520 3 |a 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. 
653 |a Forest fires 
653 |a Accuracy 
653 |a Public health 
653 |a Deep learning 
653 |a Forest fire detection 
653 |a Edge computing 
653 |a Biodiversity 
653 |a Architecture 
653 |a Visual stimuli 
653 |a Climate change 
653 |a Efficiency 
653 |a Wildfires 
653 |a Vegetation 
653 |a Preprocessing 
653 |a Environmental impact 
653 |a Sensors 
653 |a Effectiveness 
653 |a Natural environment 
653 |a Drones 
653 |a Forest & brush fires 
653 |a Surveillance 
653 |a Latency 
653 |a Object recognition 
653 |a Real time 
653 |a Semantics 
653 |a Edge detection 
653 |a Environmental 
700 1 |a Umirzakova Sabina  |u Department of Computer Engineering, Gachon University, Seongnam-si 13120, Republic of Korea; akhror1999@gachon.ac.kr (A.M.); sabinatuit@gachon.ac.kr (S.U.); akmaljon@gachon.ac.kr (A.A.) 
700 1 |a Mekhriddin, Rakhimov  |u Department of Computer Systems, Tashkent University of Information Technologies Named After Muhammad Al-Khwarizmi, Tashkent 100200, Uzbekistan; raximov022@gmail.com 
700 1 |a Alpamis, Kutlimuratov  |u Department of Applied Informatics, Kimyo International University in Tashkent, Toshkent 100121, Uzbekistan; kutlimuratov.aj@kiut.uz 
700 1 |a Zavqiddin, Temirov  |u Department of Digital Technologies, Alfraganus University, Yukori Karakamish Street 2a, Tashkent 100190, Uzbekistan; temirov@afu.uz 
700 1 |a Nasimov Rashid  |u Department of Artificial Intelligence, Tashkent State University of Economics, Tashkent 100066, Uzbekistan; rashid.nasimov@tsue.uz 
700 1 |a Azizjon, Meliboev  |u Department of Digital Technologies and Mathematics, Kokand University, Kokand 150700, Uzbekistan; a.meliboyev@kokanduni.uz 
700 1 |a Akmalbek, Abdusalomov  |u Department of Computer Engineering, Gachon University, Seongnam-si 13120, Republic of Korea; akhror1999@gachon.ac.kr (A.M.); sabinatuit@gachon.ac.kr (S.U.); akmaljon@gachon.ac.kr (A.A.) 
700 1 |a Im Cho Young  |u Department of Computer Engineering, Gachon University, Seongnam-si 13120, Republic of Korea; akhror1999@gachon.ac.kr (A.M.); sabinatuit@gachon.ac.kr (S.U.); akmaljon@gachon.ac.kr (A.A.) 
773 0 |t Forests  |g vol. 16, no. 5 (2025), p. 863 
786 0 |d ProQuest  |t Agriculture Science Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3211971391/abstract/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full Text + Graphics  |u https://www.proquest.com/docview/3211971391/fulltextwithgraphics/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3211971391/fulltextPDF/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch