Forest fire monitoring system based on aerial image

Since natural disaster annually leads to casualties and property damages, developments for ICT-based disaster management techniques are fostering to minimize economic and social losses. For this reason, it is essential to develop a customized response technology for a natural disaster. In this paper, we introduce a smart-eye platform which is developed for disaster recognition and response. In addition, we propose a deep-learning based forest fire monitoring technique, which utilizes images acquired from an unmanned aerial vehicle with an optical sensor. Via training for image set of past forest fires, the proposed deep-learning based forest fire monitoring technique is designed to be able to make human-like judgement for a new input image automatically whether forest fire exists or not. Through simulation results, the algorithm architecture and detection accuracy of the proposed scheme is verified. By applying the proposed automatic disaster recognition technique to decision support system for disaster management, we expect to reduce losses caused by disasters and costs required for disaster monitoring and response.

[1]  I. Colomina,et al.  Unmanned aerial systems for photogrammetry and remote sensing: A review , 2014 .

[2]  ByoungChul Ko,et al.  Modeling and Formalization of Fuzzy Finite Automata for Detection of Irregular Fire Flames , 2011, IEEE Transactions on Circuits and Systems for Video Technology.

[3]  Aníbal Ollero,et al.  Journal of Intelligent & Robotic Systems manuscript No. (will be inserted by the editor) An Unmanned Aircraft System for Automatic Forest Fire Monitoring and Measurement , 2022 .

[4]  ByoungChul Ko,et al.  Vision based forest smoke detection using analyzing of temporal patterns of smoke and their probability models , 2011, Electronic Imaging.

[5]  Eitan Altman,et al.  Towards efficient disaster management: 5G and Device to Device communication , 2015, 2015 2nd International Conference on Information and Communication Technologies for Disaster Management (ICT-DM).

[6]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[7]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[8]  Findra Kartika Sari Dewi,et al.  Natural disaster detection using wavelet and artificial neural network , 2015, 2015 Science and Information Conference (SAI).

[9]  Lawrence D. Jackel,et al.  Backpropagation Applied to Handwritten Zip Code Recognition , 1989, Neural Computation.

[10]  Nitish Srivastava,et al.  Improving neural networks by preventing co-adaptation of feature detectors , 2012, ArXiv.

[11]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[12]  Marimuthu Palaniswami,et al.  Smoke detection in video using wavelets and support vector machines , 2009 .

[13]  Ole-Christoffer Granmo,et al.  A methodology for fire data analysis based on pattern recognition towards the disaster management , 2015, 2015 2nd International Conference on Information and Communication Technologies for Disaster Management (ICT-DM).