Fire Detection System with Utilization of Industrial Video Surveillance System

Fire occurrence is the most destructive hazard for production, as it causes irreparable damage to the enterprise, equipment, health and life of people, as well as the environment. Therefore, recognizing fires and preventing them in a timely manner is very important. Modern production facilities are equipped with widely distributed video surveillance systems. By analyzing the transmitted images, the algorithm is able to recognize the source of fire and activate an alarm, thereby ensuring the safety of the enterprise. The algorithm deploys various characteristics of flame images during data processing. The mixed Gaussian model is used to distinguish the dynamic area from the static background, and the color characteristics of pixels in the RGB model are analyzed to detect fire. In order to increase the accuracy of the results obtained, the frequency of changes in the contour of the dynamic area (flickering) in the target area of fire is checked.