Abstract: Objective To compare the effects of different image-guided radiotherapy （IGRT） registration methods on the positioning error of patients with lung cancer. Methods 64 patients diagnosed with lung cancer admitted to Deyang People's Hospital from January 2019 to October 2020 were selected for the study. The patients were divided into two groups by the double-color ball method, each with 32 cases. After simulating positioning and formulating the radiotherapy plan, group A adopted gray-scale registration, and group B adopted bone registration for CT image registration. Compare the translational deviation and rotation deviation of the two groups of patients after automatic registration, and calculate and compare the translational error and rotation error according to the radiotherapist's registration result after manual fine-tuning. Results There was no significant difference in the rotation error Rx, Ry, and Rz between the two groups of patients （P>0.05）. The translation error （G） of patient A was smaller than that of group B, and the difference was statistically significant （P<0.05）. There was no statistically significant difference between the results of the two groups'automatic registration of lower lung lesions and manual fine-tuning. There was no statistically significant difference between the two groups' conversion errors Rx, Ry, and Rz of upper lung lesions （P>0.05）. However, the translation error （G） of patients with upper lung lesions in group A was smaller than that of group B, and the difference was statistically significant （P<0.05）. Conclusions When lung cancer patients undergo IGRT, gray-scale registration had apparent advantages in translation error, especially in patients with upper lung lesions. Manual fine-tuning based on automatic registration can obtain registration results that meet the needs of clinical radiotherapy.