Remote sensing using thermal infrared radiation and the use of physical models is considered a suitable tool for calculating the surface temperature in large areas. In this research, using bands 4 and 5 of AVHRR sensor as well as bands 31 and 32 of MODIS sensor, the surface temperature of the earth in Tehran and Qazvin provinces was retrieved. After checking the number of 114 images belonging to the two mentioned sensors, 16 images were selected for the final temperature calculation. Among the 18 split-window algorithms presented by the researchers, Price's algorithm was chosen to calculate surface temperature from NOAA/AVHRR images and Keen's algorithm was chosen to estimate temperature from MODIS images. In order to validate the obtained values, 5 cm soil temperature information was used in 11 meteorological stations. High correlation coefficients in most days of the study period, especially in the case of the AVHRR meter, indicate the reliability of the method used in estimating the surface temperature. Also, according to the temperature maps obtained from the surface temperature data and taking into account the topography of the region, the general conditions governing the temperature distribution in the region are justified.