The Research Of Mine Fire Recognition System Based On Infrared Thermal Imaging Technology

In recent years, it has shown that coal mine fire is become one of main disasters in coal mine. It is not easy to find when and where to cause malignant accidents of gas explosion or dust explosion. Sometimes will bring huge production pressure to coal enterprise, but also seriously threat the life of the underground workers. If we can develop a monitoring equipment which can real-time monitoring of coal mine goaf or other key areas, then may discover potential threat of coal spontaneous combustion earlier and reduce coal spontaneous combustion danger level, so that we can take some effective measures in time, avoiding coal mine fire and downhole safety accidents with the maximum extent.The thesis analysis after coal fire early warning system and image recognition technology at home and abroad, for the prevention of Coal mine fire to theat life safety, the thesis puts forward a new kind of mine fire recognition system based on infrared thermal imaging technology. The system can effectively remote monitoring coal goaf, response quickly to abnormal situation. The system’s algorithm is based on gaussian mixture background and combined with 3 times standard deviation principle for judging whether the smoke and realizing the function of risk rating. On the hardware, Using TI company’s DSP TMS320DM6437 to implement fire recognition algorithm, when found abnormal monitoring image will immediately take pictures upload to the controller and PC to real-time display. Controller using the stmicroelectronics company’s STM32F407ZET6 to take real-time infrared image display on the server and communication with the client. On the software, this paper take the Visual Studio 2012 platform as the foundation, Using C# language exploit PC monitoring terminal, combined the Ethernet communication technology to realize long distance data transmission and remote monitoring. Due to the Coal mine goaf fire recognition real-time monitoring and effective management, simplify operation and greatly improve the work efficiency of the system.Finally, the thesis carried on the Matlab7.1 for simulation algorithm, verifying the feasibility. At the same time, the experiments of software and hardware to prove the reliability of the system which about testing speed fast, high precision, well reliability, PC display is fluent, human-machine effect is good. For the subsequent development of coal mine fire recognition system research provides certain experience.