Multi-stations integral measurement field is currently the most effective method for large-scale 3D measurement in manufacturing plants. The accuracy evaluation of the method is of great significance for ensuring the reliability of measurement results and the rationality of measurement task planning. To fuse the multi-source observation data in the integral measurement field accuracy evaluation process, a model based on geometric constraints was established, and an accuracy evaluation method for integral measurement fields was proposed. Indicators such as instrument observation uncertainty, instrument pose parameter uncertainty, and point coordinate uncertainty were included in the accuracy evaluation method. Firstly, a rapid and accurate assessment of the instrument observation on-site uncertainty was implemented based on the statistical information of the measurement results. Then the uncertainty propagation process was analyzed using the method recommended in guide to the expression of uncertainty in measurement (GUM), and evaluation methods for the uncertainty of instrument pose parameters and target point coordinates were established. Finally, the rationality of the integral measurement field accuracy evaluation method was verified through simulation and on-site experiments.