Abstract: Coal spontaneous combustion in open pit poses a great threat to coal resources, surrounding environment and safety production. Coal fire detection in open pit mine is of great significance. A method of day and night surface temperature fusion is proposed to detect coal fire in open pit by combining with other surface anoma-lies caused by spontaneous coal combustion. The method obtained the surface temperature during the day and at night respectively, and eliminated the influence of solar radiation and other factors through algebraic super-position to highlight the persistent thermal anomaly area. Based on the surface temperature anomaly, a two-stage band-pass filter of settlement anomaly and ecological anomaly is made to filter the surface temperature anomaly, and the distribution characteristics of coal fires in the study area are integrated. The research shows that the fusion of day and night surface temperature information is more effective than the single time-phase temperature detection method of coal fire. For the surrounding areas of open pit mine, the combination of thermal infrared remote sensing technology of day and night surface temperature fusion and radar interferom-etry technology and remote sensing ecological index method of coal fire detection has higher reliability and accuracy.