Background: In intensity modulated radiation therapy, cone beam computed tomography (CT) has been used to evaluate patients prior to treatment. This study conducted a comparative evaluation of the image reconstruction ability of the clinically used half-fan bowtie filter and the full-fan bowtie filter. Materals and Methods: A CT simulation marker was inserted inside a human phantom, and the pelvic region, a large field-of-view region, was scanned by moving the isocenter along the x-axis ±1–5 cm with the full-fan mode. Furthermore, image verification was conducted based on the planning CT image and bone to confirm the setup correction value. The obtained value was then compared with that from the clinically used half-fan scan. Results: The evaluation of the reconstructed image (from the isocenter to the marker) after setting the median line did not show a significant difference with respect to the image obtained using the half-fan scan. Planning CT images and setup errors were compared in three directions, and the results showed that each mean value was within the margin of error (±3 mm). The 3D vector value was determined to be within 0–2.45 mm, and the comparison of the value obtained from the half-fan scan showed no statistically significant result. Conclusion: The application of a phantom study to actual patients in the future will reduce the error caused by movement during the treatment due to the short scan time and will reduce the imaging dose for patients during setup error confirmation and correction.