The combustion environment of pulverized coal particles is provided by the flat flame entrained flow combustor system based on McKenna burner. The moving coal particles are photographed by a color camera imaging system, which is calibrated by a black body furnace. The temperature of the combustion coal particles is measured based on the two-color method using the responses of the r and g channels. The calibration of the temperature measurement system is carried out based on the BP neural network training method. By photographing the burning flame of the candle, the results are compared with the thermocouple data to verify the reliability of the temperature measuring system. The experimental study obtained the temperature information of the coal particles at different distances from the burner exit. The results show that the temperature of the coal particles rises first and then decreases along the height of the nozzle outlet. The acquisition of temperature information provides a data reference for studying the combustion process and ignition mechanism of pulverized coal. The system can be combined with the backlight imaging system to form a dual optical path system for simultaneous experimental research.