Abstract: In order to explore the recognition of coal gangue hills by surface temperature retrieved by thermal infrared remote sensing，based on Landsat TM satellite remote sensing image and meteorological data and vector data of the study area，the surface temperature of the study area was retrieved by single window algorithm. The inversion of the surface temperature anomaly area is compared with the actual distribution of coal gangue piles in Yangquan City to verify the accuracy of using the inversion results to identify coal gangue piles. The results show that the high temperature anomaly area in Yangquan mining area is not only related to the distribution of coal gangue，but also to the distribution of coal and factories in the mining area. The high temperature anomaly area extracted by the inversion results can be used as a reference for the accurate detection of the temperature of the coal gangue pile，thereby improving the efficiency of judging the spontaneous combustion position of the coal gangue pile.

Key words: thermal infrared remote sensing, surface temperature, coal gangue pile