Background <p>This study compared the diagnostic accuracy of cone beam computed tomography (CBCT) and intraoral radiographs (IOR) for detecting periodontal defects.</p> Materials and methods <p>A total of 67 periodontal defects, including 39 dehiscence, 13 fenestration, and 15 Class III furcation defects, were experimentally created were artificially created on eight dry skull mandibles. The images of these defects were taken with IOR and CBCT. The resulting images were evaluated by two observers in a double-blind manner. Kappa statistics, McNemar and McNemar-Browner tests, Chi-square test, and Bonferroni-corrected Z test were used for statistical analysis.</p> Results <p>Excellent intra-observer and inter-observer agreement was observed in our study. In detecting periodontal defects (dehiscence, fenestration, and Class III furcation defect), sensitivity, specificity, and accuracy were found to be 96%, 64%, and 72%, respectively, for IOR, and 100%, 96%, and 97%, respectively, for CBCT. While the rate of detecting periodontal defects was 49.5% for IOR, it was 70% for CBCT. Statistically significant difference was found between IOR and CBCT in detecting periodontal defects (<i>p</i> = 0.004). Statistically significant difference was found between IOR and CBCT in detecting dehiscence (<i>p</i> = 0.007). No statistically significant difference was found between IOR and CBCT in detecting fenestration and Class III furcation defects.</p> Conclusions <p>A statistically significant difference was found between CBCT and IOR in terms of the detection of dehiscence, one of the periodontal defects, and CBCT was concluded to be more successful than IOR in the diagnosis of dehiscence defects. No statistically significant difference was found between IOR and CBCT in detecting fenestration and Class III furcation type defects. CBCT was found to be more successful than IOR in detecting defects in the anterior region. However, no difference was found between the two methods in the posterior region.</p>