The remote sensing data at 1.1km×1.1km resolution from the polar-orbiting weather satellite are retrieved by a range of split window algorithms to obtain ground surface temperatures,which together with minimum surface temperatures(MST) measured at a number of meteorological stations are then used to get gridded MSTs by use of regression and variational techniques.After comparison of errors from these algorithms the best one is selected for retrieving.By use of the retrieved variational corrected MSTs,injury indices and conditions for wheat’s growth stages,the regional distribution of various degree freezing injuries is plotted,thereby realizing the remote sensing monitoring of wheat’s injury and the accurate calculation of the injury areas.