Using video analytics for early fire detection

Introduction. Alarming trends in the statistics of fires in buildings and premises in recent years with large-scale consequences require the search and development of new methods and approaches in the field of early fire detection. A promising approach to low-inertia identification of fires is considered to be the use of intelligent systems based on visual assessment of fire hazard in the premises, based on various types of video cameras. Unlike traditional point fire detectors, this method is not limited by the volume of the room and allows detecting fires even in large open spaces and unpredictable movement of the flame front.

The aim of the research is to substantiate the feasibility of reliably identifying the fire source in a room at an early stage using video recording. The main task is to develop an algorithm for training a neural network module that allows for the accurate determination of the coordinates of the fire source location in a room at an early stage using video recording.

Materials and methods. Experimental studies were carried out using a 3 × 3 × 2.3 m room model with a gas analysis system, fire alarms, video recording equipment, and a control and monitoring system for collecting and recording information installed in it.

Results and Discussion. As a result of the conducted research, an approach to the use of video analytics for identifying the source of fires at an early stage was proposed.

Conclusions. Based on experimental studies, the optimal size of the pre-trained neural network model for the task was selected, and the feasibility of using a video surveillance camera for low-inertia identification of fires in premises was substantiated.

1. Festag S. The Statistical Effectiveness of Fire Protection Measures: Learning from Real Fires in Germany. Fire Technology. 2021; 57:1589-1609. DOI: 10.1007/s10694-020-01073-y

2. Sheng D., Deng J., Zhang W., Cai J., Zhao W., Xiang J. A Statistical Image Feature-Based Deep Belief Network for Fire Detection. Complexity. 2021; (1):1-12. DOI: 10.1155/2021/5554316

3. Ma C., Van Coile R., Gernay T. Fire protection costs in composite buildings for cost-benefit analysis of fire designs. Journal of Constructional Steel Research. 2024; 215:108517. DOI: 10.1016/j.jcsr.2024.108517

4. Nan C., Xianmeng M., Wenhui D. Experimental Study on the Testing Environment Improvement of Fire Smoke Detectors. Procedia Engineering. 2012; 45:610-616. DOI: 10.1016/j.proeng.2012.08.211

5. Kalmykov S.P., Esin V.M. Fire detection time. Pozharovzryvobezopasnost/Fire and explosion safety. 2017; 26(11):52-63. (rus).

6. Chlenov A.N., Bucynskaya T.A., Zhuravlev S.Yu., Nikolaev V.A. On the efficiency of the multi-criteria fire alarm. Pozharovzryvobezopasnost/Fire and explosion safety. 2016; 25(12):55-60. DOI: 10.18322/PVB.2016.25.12.55-60.EDN XWVMGZ. (rus).

7. Kuznetsov G.V., Volkov R.S., Sviridenko A.S., Strizhak P.A. Fire detection and suppression in rooms with different geometries. Journal of Building Engineering. 2024; 90:109427. DOI: 10.1016/j.jobe.2024.109427

8. Kuznetsov G.V., Zhdanova A.O., Volkov R.S., Sviridenko A.S., Strizhak P.A. Smoke deposition and extraction in compartment fires with different ignition sources. Process Safety and Environmental Protection. 2024; 187:581-592. DOI: 10.1016/j.psep.2024.05.010

9. Artamonov V.S., Polyakov A.S., Ivanov A.N. Ultra-early and early detection of fires: concepts, application limits and unity. Pozharovzryvobezopasnost/Fire and explosion safety. 2016; 25(9):78-83. DOI: 10.18322/PVB.2016.25.09.78-83 (rus).

10. Song Z., Huang X., Jiang J., Pan X. A laboratory approach to CO2 and CO emission factors from underground coal fires. International Journal of Coal Geology. 2020; 219:103382. DOI: 10.1016/j.coal.2019.103382

11. Li Y., Yu L., Zheng C., Ma Z., Yang S., Song F. et al. Development and field deployment of a mid-infrared CO and CO2 dual-gas sensor system for early fire detection and location. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 2022; 270:120834. DOI: 10.1016/j.saa.2021.120834

12. Bu F., Gharajeh M.S. Intelligent and vision-based fire detection systems: A survey. Image and Vision Computing. 2019; 91:103803. DOI: 10.1016/j.imavis.2019.08.007

13. Cetin A.E., Dimitropoulos K., Gouverneur B., Grammalidis N., Gunay O., Habiboglu Y.H. et al. Video fire detection : review. Digital Signal Processing. 2013; 23:1827-1843. DOI: 10.1016/j.dsp.2013.07.003. EDN RIKHCF.

14. Xie D., Chen Q., Zhu Y. Optimal layout scheme of fire detectors and operation condition monitoring technology in urban integrated substation. Measurement. 2024; 236:115093. DOI: 10.1016/j.measurement.2024.115093

15. Hashemzadeh M., Zademehdi A. Fire detection for video surveillance applications using ICA K-medoids-based color model and efficient spatio-temporal visual features. Expert Systems with Applications. 2019; 130:60-78. DOI: 10.1016/j.eswa.2019.04.019

16. Ahn Y., Choi H., Kim B.S. Development of early fire detection model for buildings using computer vision-based CCTV. Journal of Building Engineering. 2023; 65:105647. DOI: 10.1016/j.jobe.2022.105647

17. Yazdi M., Bouwmans T. New trends on moving object detection in video images captured by a moving camera: A survey. Computer Science Review. 2018; 28:157-177. DOI: 10.1016/j.cosrev.2018.03.001

18. Hackner A., Oberpriller H., Ohnesorge A., Hechtenberg V., Müller G. Heterogeneous sensor arrays: Merging cameras and gas sensors into innovative fire detection systems. Sensors and Actuators B: Chemical. 2016; 231:497-505. DOI: 10.1016/j.snb.2016.02.081

19. Yar H., Khan Z.A., Rida I., Ullah W., Kim M.J., Baik S.W. An efficient deep learning architecture for effective fire detection in smart surveillance. Image and Vision Computing. 2024; 145:104989. DOI: 10.1016/j.imavis.2024.104989

20. Evsikov A.A., Samarin I.V. Detecting Fire Sources at Technological Facilities Using a Convolutional Neural Network. Pozharovzryvobezopasnost/Fire and explosion safety. 2023; 32(5):40-48. DOI: 10.22227/0869-7493.2023.32.­05.­­­­­40-48 (rus).