Efficient Aerial Fire Detection on Resource-Constrained Devices Using Cross-Architecture Knowledge Distillation
Jangirova, Sabina Jankovic, Branislava Ullah, Waseem Khan, Latif U. Guizani, Mohsen
Jangirova, Sabina
Jankovic, Branislava
Ullah, Waseem
Khan, Latif U.
Guizani, Mohsen
Supervisor
Department
Machine Learning
Embargo End Date
Type
Conference proceeding
Date
2025
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Fire poses significant ecological danger and can lead to human losses if not mitigated early. Fire detection systems on remote sensing devices have become critical to disaster prevention and management. In this paper, we propose a lightweight and effective fire detection model based on MobileViT-XS, compressed through the distillation of knowledge from a stronger teacher model. Our model combines the local feature extraction capabilities of convolutional networks with the global context capabilities of vision transforms. This allows our proposed model to achieve high detection accuracy and real-time inference, making it suitable for deployment on UAVs and other resource-constrained devices. Through rigorous experiments on the ADSF dataset, the proposed method achieves an accuracy, F1-score, and recall of 95.50%, along with a precision of 95.52%, thereby outperforming contemporary methods. Extensive evaluations on ADSF confirm that our method not only surpasses existing state-of-the-art techniques in accuracy but also provides significantly faster processing speeds.
Citation
S. Jangirova, B. Jankovic, W. Ullah, L. U. Khan and M. Guizani, "Efficient Aerial Fire Detection on Resource-Constrained Devices Using Cross-Architecture Knowledge Distillation," 2025 International Wireless Communications and Mobile Computing (IWCMC), Abu Dhabi, United Arab Emirates, 2025, pp. 1732-1737, doi: 10.1109/IWCMC65282.2025.11059492
Source
Proceedings of the International Wireless Communications and Mobile Computing
Conference
2025 International Wireless Communications and Mobile Computing (IWCMC), 2025
Keywords
Aerial images, Knowledge distillation, Vision Transformer, Convolution Neural Network, Fire detection
Subjects
Source
2025 International Wireless Communications and Mobile Computing (IWCMC), 2025
Publisher
IEEE