In order to realize 3D visualization velocimetry of microfluidic, an micro-particle tracking velocity measurement system based on optics coherence tomography was established and the principles of the system, algorithms such as image extraction of micro-particles, matching, and velocity computation were investigated. First, velocity measurement system based on optics coherence tomography and scanning and imaging method were presented. Then, micro-particles in fluid were detected using median filtering, OTSU binarization method and volume filtering to realize 3D visualization of flow characteristics. Finally, to find an optimal matching of micro-particles, the cost function was defined using the quadratic distance between particles, as well as the quadratic differences in the intensity moment of order 2, and velocity was computed using 3D coordinates of particles. Experimental results such as 3D particle images, and velocity vector with micrometer spatial resolution in convective flow were given. It is important for the velocity measurement of the complex microfluidic and study of microfluidic devices.