Embedded forest fire monitoring and positioning system based on machine vision

According to the bulky, high cost and interfered seriously of present forest fire monitoring system, the embedded forest fire monitoring and positioning system is designed based on machine vision. The system consists of four parts: embedded forest fire detection, 3G wireless communication, positioning system and held terminal. Forest fire recognition and positioning algorithm are studied based on FPGA and DSP, and 3G wireless communication is achieved. The results show that the system can overcome a variety of interference and has low false alarm rate, high real time, low power consumption. This system has good operation effect through the test in Inner Mongolia and the positioning error is less than ten meters.