Abstarct: Precision measurements of rainfall with high spatial and temporal resolution are important for modeling surface currents, proper location of dams, flood and drought prediction, and water resource management and agricultural projects. Over the past few decades, far-reaching data from remote sensing and satellite technologies have been used to estimate precipitation values in order to increase the accuracy of world-class precipitation recordings and overcome the problems caused by traditional rainfall measurement methods. In this research, extracted data from the analysis and processing of MODIS sensor images, also Shahrood, Damghan and Semnan stations 6 hourly precipitations observed data in the period 2005-2015 be used. In order to simulate rainfall, ANN method was used for modeling. the obtained data from processing of MODIS sensor images include the reflection values of visible bands and infrared bands, the emission of thermal infrared bands, brightness temperature of thermal infrared bands and normalized difference vegetation index Separately or in combination to each other as input and observed rainfall data were fed into the model as output. Then simulated data were compared with observation data. The result shows that, the simulated precipitations by combining visible reflection data of band 3 and Infrared reflection data of bands 5, 18 and 19 are more similar to observational precipitations and the performance of the model is acceptable.