Abstract: Soil infiltrability is an important issue in water resources transformation and utilization as well as agricultural water management. In this study, an automatic system with machine vision and image processing technology was developed for measuring soil infiltrability based on the linear run-on method. Hardware was integrated and the operational software was developed. The system uses a digital camera to take images of the run-on water advances at real-time from arbitrary orientation before they are processed and analyzed to determine the wetted areas. According to the wetted areas determined the system computed the soil infiltration rate with the mathematic models outlined in the linear run-on method and displayed infiltration curve on the real-time basis. Laboratory experiments were carried out to verify the system with desired functions such as real-time measurements, infiltration computation, and display. The measurement error was 5.27%, which demonstrates the feasibility of the system. This system provides an automated yet high efficient tool for the measurement of soil infiltrability.