Objective As two important methods for studying flow fields in blood vessels, comparison of phase contrast magnetic resonance imaging (PC MRI) and computational fluid dynamics (CFD) is of great significance. Previous studies have focused on the relative comparison of the velocity vector size in the results obtained from the two methods, being short of the distribution comparisons of the two. In this paper, a regional correlation method is proposed to quantitatively compare the distribution of velocity vectors in PC MRI and CFD results. Methods The blood flow velocity field of left and right internal carotid arteries in 9 healthy volunteers was measured by PC MRI and compared with the velocity field calculated by CFD. The relative error and regionalcorrelation were calculated to quantitatively compare the velocity vector size and velocity distribution of the two methods respectively. Results Relative errors were larger near the vessel wall and the siphon bend, while smaller near the center of the blood vessel. The distribution of regional correlation and relative error was overall opposite, but in some positions where flow rate was small and relative errors were large, the correlation was not low. Conclusions The comparison method based on regional correlation can quantitatively compare the velocity distribution of PC MRI and CFD results, which reflects the similarity or difference of velocity distribution and can be used as an important indicator in comparative study of PC MRI and CFD.