The trend extraction of the burning zone temperature is the basis for the burning state identification and stability control of the cement. In this paper, the trend extraction method of the burning zone temperature based on singular spectrum analysis is studied. Firstly, a visual inspection device is used to detect the temperature of the burning zone. Then, the singular spectrum analysis is carried out to obtain the singular spectrum eigenvalue of the temperature. After that, the appropriate contribution rate of the singular spectrum eigenvalue is selected, and the signal is reconstructed to extract the trend of the temperature signal. At last, the trend extraction method is validated on some burning zone temperature data, which are obtained from a cement company. The result shows that this paper provides a feasible method for extracting the trend of the burning zone temperature.