The measurement of dynamic surface velocities in rivers holds significant importance for hydrological studies, environmental monitoring, and water resource management. With the rapid evolution of image processing technologies, methods based on imagery for velocity measurement have garnered widespread attention due to their non-intrusive nature and operational simplicity. Addressing the invasive nature, limited measurement capabilities, and poor adaptability to complex environments inherent in traditional velocity measurement techniques, a novel technical scheme is proposed. Initially, an enhanced Mean shift algorithm is employed for effective tracking of river surface targets, overcoming stability and accuracy issues faced by conventional algorithms in complex aquatic environments. Subsequently, a new velocity measurement method, integrating optical flow techniques with calibration technology, is introduced to augment accuracy and mitigate environmental interferences. This research not only enhances the reliability of non-contact velocity measurement technologies but also offers new perspectives and tools for river monitoring, contributing significantly to the sustainable management and utilization of water resources.