The monitoring of burning buildings with convolutional neural network

Florin Batog; Simona Moldovanu

doi:10.52846/stccj.2023.3.2.50

Authors

Florin Batog Dunarea de Jos University of Galati
Simona Moldovanu Dunarea de Jos University

DOI:

https://doi.org/10.52846/stccj.2023.3.2.50

Keywords:

color space, convolutional neural network

Abstract

Artificial intelligence is constantly expanding. This is used in many domains, from agriculture to the medical field. Lately, the focus has been on image-based learning (deep learning) because it has greater applicability in reality. An example is the identification of buildings that burn in real-time, the purpose being the saving as many lives as possible by the firefighters. This paper aims to identify the optimal color space for the identification of fire in images. To achieve this the images were selected from two databases taken from the Kaggle platform, and the images were processed with six color spaces: RGB, YCbCr, HSV, HLS, L*a*b, and L*u*v. In this form the images fed convolutional neural networks (CNNs). After that, the models were trained on six datasets, a dataset for each color space, and after compiling of CNNs, a testing set for the prediction model was proposed. The results of the model were analyzed and interpreted according to accuracy and loss function and the space YCbCr identified the fire from the images with an accuracy of 100% and loss function of 4.02*10^-05.

References

K. Muhammad, J. Ahmad, and S. W. Baik, “Early fire detection using convolutional neural networks during surveillance for effective disaster management,” Neurocomputing, 2018, vol. 288, pp. 30–42.

K Y. J. Kaufman, D. D. Herring, K. J. Ranson and G. J. Collatz, "Earth Observing System AM1 mission to Earth," in IEEE Transactions on Geoscience and Remote Sensing, 1998, vol. 36, no. 4, pp. 1045-1055.

L. Giglio, J. D. Kendall, and C. O. Justice, “Evaluation of global fire detection algorithms using simulated AVHRR infrared data,” International Journal of Remote Sensing, 1999, 20, pp. 1947 – 1985.

L. Giglio, J. D. Kendall and R. Mack, “A multi-year fire data set for the tropics derived from the TRMM VIRS,” International Journal of Remote Sensing, 2003, pp.1-10.

Li and Zhao, P. Li, W. Zhao, Image fire detection algorithms based on convolutional neural networks, Case Stud. Thermal Eng., 2020, vol. 19, pp. 100625.

Majid S., Alenezi F., Masood S., Ahmad M., Gündüz E.S., Polat K., Attention based CNN model for fire detection and localization in real-world imagesExpert Syst. Appl., 2022, vol. 189, pp. 116114.

Celik T., Demirel H., Fire detection in video sequences using a generic color model. Fire Saf. J., 2009, vol. 44, pp. 147–158.

K. Muhammad, J. Ahmad, I. Mehmood, S. Rho, and S. W. Baik, "Convolutional Neural Networks Based Fire Detection in Surveillance Videos," in IEEE Access, 2018 vol. 6, pp. 18174-18183.

R. Szeliski, "Computer Vision: Algorithms and Applications", Springer Nature, 2022, pp. 72-79.

S. Moldovanu, C.D. Obreja, K. Biswas and L. Moraru, "Towards accurate diagnosis of skin lesions using feedforward back propagation neural networks," Diagnostics, 2021, vol. 11, pp. 936.

L.Chen, H. Qu, J. Zhao, B. Chen and J. C. Principe,”Efficient and robust deep learning with correntropy-induced loss function. Neural Comput. & Appl., 2016, vol. 27, pp. 1019–1031.

M. Mihaela, A.L. Culea-Florescu and S. Moldovanu, “A Convolutional Neural Network for skin lesion classification in Dermoscopic Images using discrete wavelet transform,” IEEE European Control Conference (ECC), Bucharest, Romania, pp. 1-5, 13-16, June 2023.