Aiming at the measurement and positioning process of assembly and docking of aerospace and rail transit vehicle parts, in order to solve the problems of large measurement field range and complicated use of multiple measurement equipment combination, the calibration method of dual-vision measurement system facing assembly and docking was studied. By constructing the model of the dual-vision measurement system, analyzing the relationship between the camera parameters between the systems, introducing closed-loop constraints and LM algorithm to optimize the camera parameters in the single binocular system, and using the improved EPnP algorithm to calibrate the dual-vision system, and combining the self-designed double-sided one-dimensional target to complete the global calibration and coordinate conversion of the dual-vision system. The simulation docking experiment results show that the optimized calibration algorithm is superior to the traditional algorithm in accuracy, and simplifies the measurement steps and improves the measurement efficiency in the process of use.