Heavy rainfall from typhoons or tropical cyclones often causes inland flooding and mudslides that threaten lives and property. In this study, Special Sensor Microwave Imager (SSM/I) satellite data observed from 2000 to 2004 were used to calculate the rainfall rates of different typhoons in the northwestern Pacific. Geostationary weather satellite infrared images were also applied to estimate the typhoon rotation speed via the maximum cross-correlation technique. By including such information in the tropical rainfall potential (TRaP) technique, an improved typhoon rainfall potential technique can be constructed. Considering the fact that a typhoon's spiral rainbands move constantly, half-hourly or hourly infrared data observed from geostationary weather satellites were used to calculate the revolving speed, which was subsequently used to predict the rainband movement over the next hour. After comparing the predicted rainfall potential with the rain gauge data of Taiwan's small offshore islands, it was found that this new method can improve the typhoon's accumulated rainfall by approximately 40% over the original TRaP method. Therefore, to produce a more accurate short-term typhoon rainfall forecast, it is very important to factor in the satellite-estimated storm rotation speed.