Automatic fire identification, early warning and monitoring system based on computer vision

Abstract

The invention discloses an automatic fire identification, early warning and monitoring system based on computer vision. The automatic fire identification, early warning and monitoring system comprisesan early warning and monitoring system and an early warning and monitoring method, the early warning monitoring system comprises a data acquisition module, a data transmission module, a data processing module and an alarm system. The early warning monitoring method comprises the following steps: 1, the data acquisition device acquires an on-site dynamic image and transmits the on-site dynamic image to the computer; 2, a black image with the same size and channel number as the original image is created; 3, an object of which the color is in a flame characteristic interval is identified, and confirming is further carried out in combination with flame dynamic characteristics and 4, flame characteristics are identified, and an alarm system is started; the system can autonomously identify thefire and send out an early warning signal, so that the time for discovering the fire is shortened to the maximum extent; and the system can also be fused with a 5G technology, a scene picture under rescue is transmitted back in an almost zero-delay manner, fire scene information is provided in real time, and the fire behavior is controlled.

Description

technical field [0001] The invention relates to the technical field of fire prevention and control, in particular to a computer vision-based automatic fire identification, early warning and monitoring system. Background technique [0002] Monitoring is currently widely used in various places such as communities, warehouses, schools, hotels, etc. However, due to the limitations of the information processing capabilities of the monitoring room staff, they often cannot notice special situations in time, especially fires. The initial fire is small and difficult to detect, but The fire spreads rapidly, and once the best time to call the police is missed, it will cause heavy property damage and casualties. [0003] Most of the common fire detectors on the market are temperature sensors and smoke detectors. This kind of detection method can only be identified when the fire reaches a certain level, which is contrary to the idea of ​​real-time fire warning. [0004] Therefore, a com...

AI Technical Summary

Problems solved by technology

[0002] Monitoring is currently widely used in various places such as communities, warehouses, schools, hotels, etc. However, due to the limitations of the information processing capabilities of the monitoring room staff, they often cannot notice special situations in time, especially fires. The initial fire is small and difficult to detect, but The fire spread

Images