Objective: The aim of the study was to compare the manual tracing of lateral cephalometric radiographs with the use of cephalometric software obtained by 2 methods: use of a scanner and use of cameras to test the accuracy of measurement for clinical orthodontic diagnosis. Material and Methods: The original method of manual tracing was compared to 2 radiographic digitization methods: A) captured from a scanner, B) captured from digital cameras applied to 40 randomly selected cephalograms of orthodontic patients at Mahidol University. The digital images from cameras and scanner were analyzed with image processing software (CephSmile). The repeatability test was obtained for all methods, and the paired t-test determined the statistical significance between digitized images and original manual tracings for thirty-four measurements. Results: The paired t-test showed no significant differences among all 34 variables. Most mean differences were less than 0.25 mm and 0.25 degrees. Conclusion: The results demonstrated that there were no significant differences on measurement parameters for cephalometric analysis among manual tracing and computerized digital images from scanner and cameras.