Fire and smoke detection in videos using background removal and Convolutional Neural Networks
Abstract
In open environments, when there is a low relative humidity and excessive heat, a small spark can be the trigger for large wildfires to happen. Facing the unpredictability of the fire, detecting it early is important to streamline the fire fighting and then minimize its consequences. Research with convolutional neural networks - CNN - has increased exponentially with the use of various technologies for the detection of fire. With that, this paper presents a fire and smoke detection method on videos based on removing background and then applying a CNN to detect it. As preliminary results, the proposed process was able to achieve an accuracy of 92,73% in the Fire Smoke Detection Network (FSDN) architecture and 94,88% in the XCeption architecture. These results are similar to the ones found in literature showing that apply CNN after background motion subtraction is a good way to solve the fire and smoke detection problem.References
C. F. de Castro, G. Serra, J. Parola, J. Reis, L. Lourenço, and S. Correia, "Combate a incêndios florestais," Escola Nacional de Bombeiros, vol. 13, 2003.
INPE. (2019) Monitoramento dos focos ativos por países. [On- line]. Available: http://queimadas.dgi.inpe.br/queimadas/portal-static/ estatisticas paises/
L. F. L. Fraga, "Verificação de adequação do projeto de uma tubulação de incêndio de uma refinaria de petróleo conforme as normas da petrobras," 2010.
B. M. N. de Souza, "Detecção e localização de fogo em imagens digitais usando técnicas de aprendizado de máquina," Ph.D. dissertation, Pontifícia Universidade Católica do Paraná, 2019. [Online]. Available: https://www.ppgia.pucpr.br/pt/arquivos/doutorado/teses/2019/ 066 Tese Bruno.pdf
T. Celik, H. Demirel, and H. Ozkaramanli, "Automatic fire detection in video sequences," in 2006 14th European Signal Processing Conference. IEEE, 2006, pp. 1–5.
T. Celik and H. Demirel, "Fire detection in video sequences using a generic color model," Fire safety journal, vol. 44, no. 2, pp. 147–158, 2009.
S. Rudz, K. Chetehouna, A. Hafiane, O. Sero-Guillaume, and H. Laurent, "On the evaluation of segmentation methods for wildland fire," in International Conference on Advanced Concepts for Intelligent Vision Systems. Springer, 2009, pp. 12–23.
W.-B. Horng, J.-W. Peng, and C.-Y. Chen, "A new image-based real- time flame detection method using color analysis," in Proceedings. 2005 IEEE Networking, Sensing and Control, 2005. IEEE, 2005, pp. 100– 105.
B. C. Ko, K.-H. Cheong, and J.-Y. Nam, "Fire detection based on vision sensor and support vector machines," Fire Safety Journal, vol. 44, no. 3, pp. 322–329, 2009.
R. S. priya and K. Vani, "Deep learning based forest fire classification and detection in satellite images," in 2019 11th International Conference on Advanced Computing (ICoAC), 2019, pp. 61–65.
Y. Lee, D. Im, and J. Shim, "Data labeling research for deep learning based fire detection system," in 2019 International Conference on Systems of Collaboration Big Data, Internet of Things Security (SysCoBIoTS), 2019, pp. 1–4.
M. Dua, M. Kumar, G. Singh Charan, and P. Sagar Ravi, "An improved approach for fire detection using deep learning models," in 2020 International Conference on Industry 4.0 Technology (I4Tech), 2020, pp. 171–175.
F. Chollet, "Xception: Deep learning with depthwise separable convolutions," in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 2017, pp. 1251–1258.
P. Foggia, A. Saggese, and M. Vento, "Real-time fire detection for video-surveillance applications using a combination of experts based on color, shape, and motion," IEEE TRANSACTIONS on circuits and systems for video technology, vol. 25, no. 9, pp. 1545–1556, 2015.
P. KaewTraKulPong and R. Bowden, "An improved adaptive background mixture model for real-time tracking with shadow detection," in Videobased surveillance systems. Springer, 2002, pp. 135–144.
INPE. (2019) Monitoramento dos focos ativos por países. [On- line]. Available: http://queimadas.dgi.inpe.br/queimadas/portal-static/ estatisticas paises/
L. F. L. Fraga, "Verificação de adequação do projeto de uma tubulação de incêndio de uma refinaria de petróleo conforme as normas da petrobras," 2010.
B. M. N. de Souza, "Detecção e localização de fogo em imagens digitais usando técnicas de aprendizado de máquina," Ph.D. dissertation, Pontifícia Universidade Católica do Paraná, 2019. [Online]. Available: https://www.ppgia.pucpr.br/pt/arquivos/doutorado/teses/2019/ 066 Tese Bruno.pdf
T. Celik, H. Demirel, and H. Ozkaramanli, "Automatic fire detection in video sequences," in 2006 14th European Signal Processing Conference. IEEE, 2006, pp. 1–5.
T. Celik and H. Demirel, "Fire detection in video sequences using a generic color model," Fire safety journal, vol. 44, no. 2, pp. 147–158, 2009.
S. Rudz, K. Chetehouna, A. Hafiane, O. Sero-Guillaume, and H. Laurent, "On the evaluation of segmentation methods for wildland fire," in International Conference on Advanced Concepts for Intelligent Vision Systems. Springer, 2009, pp. 12–23.
W.-B. Horng, J.-W. Peng, and C.-Y. Chen, "A new image-based real- time flame detection method using color analysis," in Proceedings. 2005 IEEE Networking, Sensing and Control, 2005. IEEE, 2005, pp. 100– 105.
B. C. Ko, K.-H. Cheong, and J.-Y. Nam, "Fire detection based on vision sensor and support vector machines," Fire Safety Journal, vol. 44, no. 3, pp. 322–329, 2009.
R. S. priya and K. Vani, "Deep learning based forest fire classification and detection in satellite images," in 2019 11th International Conference on Advanced Computing (ICoAC), 2019, pp. 61–65.
Y. Lee, D. Im, and J. Shim, "Data labeling research for deep learning based fire detection system," in 2019 International Conference on Systems of Collaboration Big Data, Internet of Things Security (SysCoBIoTS), 2019, pp. 1–4.
M. Dua, M. Kumar, G. Singh Charan, and P. Sagar Ravi, "An improved approach for fire detection using deep learning models," in 2020 International Conference on Industry 4.0 Technology (I4Tech), 2020, pp. 171–175.
F. Chollet, "Xception: Deep learning with depthwise separable convolutions," in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 2017, pp. 1251–1258.
P. Foggia, A. Saggese, and M. Vento, "Real-time fire detection for video-surveillance applications using a combination of experts based on color, shape, and motion," IEEE TRANSACTIONS on circuits and systems for video technology, vol. 25, no. 9, pp. 1545–1556, 2015.
P. KaewTraKulPong and R. Bowden, "An improved adaptive background mixture model for real-time tracking with shadow detection," in Videobased surveillance systems. Springer, 2002, pp. 135–144.
Published
2020-11-07
How to Cite
NAGAI, Mayla Toshimi; SOUZA, Bruno M N.
Fire and smoke detection in videos using background removal and Convolutional Neural Networks. In: WORKSHOP OF UNDERGRADUATE WORKS - CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI), 33. , 2020, Evento Online.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2020
.
p. 203-206.
DOI: https://doi.org/10.5753/sibgrapi.est.2020.13012.
