In this paper, a new image-based disdrometer is developed to survey the movement of free falling raindrops. The proposed high-speed image velocimetry (HSIV) system comprises a high-speed camera mounted on a lens with long depth of field, a backlight source using an array of blue LEDs, and a computation/control unit for image acquisition and analysis. Further image processing algorithms are produced for raindrop detection and raindrop tracking. The performance of the HSIV system is compared with that of a commercial PARSIVEL2 disdrometer during a local rainfall event within 12 h, while a tippingbucket rain gauge is also used to estimate the amount rainfall rate during that period. The data obtained from the proposed HSIV system are in good agreement with those of PARSIVEL2. Furthermore, the results of the HSIV system are closer than those of PARSIVEL2 to the actual rainfall, which is determined by the rain gauge. To validate the raindrop tracking algorithm, another experiment is conducted to compare the tracking results with the frame-by-frame visual observation from 2880 consecutive frames within 5 min. Experimental results suggest that the proposed HSIV system facilitates the measurement of the precipitation properties, fall velocity, equivolume diameter, raindrop concentration, and rainfall rate, to real-time monitoring of the rainfall with stable and reliable analysis.