Thermal imaging fire alarm system based on artificial intelligence and image processing method thereof

Abstract

The invention discloses a thermal imaging fire alarm system based on artificial intelligence and an image processing method thereof, and relates to the technical field of fire alarms. The system comprises a CAN controller, a CAN transceiver, a CAN bus, a data concentrator, an infrared acquisition image preprocessing module, a monitored target recognition module, a fire detection and diagnosis unit, a result output module, a fire monitoring center, and fire detection units arranged at different fire detection monitoring area positions. The thermal imaging fire alarm system based on the artificial intelligence provided by the invention solves the problems that the judgment accuracy of single smoke particle judgment and early warning on the fire condition is low, the system is easily interfered by external cigarette butts and matches and other factors, the single infrared image judging time is long, the position of a monitoring target is difficult to be located, the timely positioning ofan ignition point of the monitoring target position is influenced, and the time of optimally determining the position of the ignition point and the fire is missed.

Description

technical field [0001] The invention belongs to the technical field of fire alarms, in particular to an artificial intelligence-based thermal imaging fire alarm system and an image processing method thereof. Background technique [0002] Most of the fires will release a lot of smoke. Smoke particles are one of the important characteristics of the early fire, so smoke detection can detect the fire earlier, but the smoke emitted by some cigarette butts, matches or lighters will alarm the smoke fire. Fire alarms can cause false alarms; especially in aircraft cargo holds, factories and other places, smoke from factories or smoke from cigarette butts and matches can cause false alarms from fire alarms, resulting in forced landings or delays of flights and factory malfunctions. Operation happens. [0003] Infrared imaging uses photoelectric technology to detect the infrared specific band signal of the thermal radiation of the object, converts the signal into a figure that can be ...

AI Technical Summary

Problems solved by technology

[0004] The object of the present invention is to provide a thermal imaging fire alarm system based on artificial intelligence and its image processing method, by providing a fire detection unit that is set at different locations for monitoring target points and connected to the CAN bus, the fire detection unit includes several detection The smoke sensor at the monitoring target point and the infrared imager connected to the smoke sensor through the rotating positioning tracking module are collected by the data concentrator to collect the infrared image data of the monitoring target collected by the fire detection unit, and then passed through the infrared acquisition image preprocessing module and the Monitoring target recognition unit, fire detection, diagnosis unit, result output module, fire monitoring center, comprehensive smoke particle judgment and infrared image judgment to make accurate judgment, solves the problem of low precision of single smoke particle judgment and early warning on fire situation Interferenced by external cigarette butts, matches and other factors, a single infrared image takes a long time to judge, and it is difficult to locate the monitoring target position, which affects the real-time positioning of the ignition point at the monitoring target position, and misses the problem of the best time to confirm the ignition point position and confirm the fire.

Images