Objectives: To evaluate the color accuracy of various calibration models on three commercial SLR digital cameras (CCD, CMOS and FoveonX3) for use in dental shade matching. Methods: CIE LAB colors of 264 Kodak Q60 color patches and 65 shade tabs were measured with a spectroradiometer, and digital images were taken with the three cameras. The CIE LAB and the image RGB values of the Kodak Q60 patches were compared to formulate four color regression calibration models: 2nd order, 2nd order 11 terms, 3rd order and tetrahedral interpolation. The RGB values of the shade tab images, obtained from the three cameras, were applied into these models to predict CIE LAB colors of the shade tabs. The color differences (Delta E) among the predicted and measured CIE LAB values of the shade tabs were compared to evaluate calibration accuracy. Repeated measures ANOVA was used to evaluate the between and within group differences among digital images taken with the three cameras and using the four calibration models. The Tukey- Kramer was used as a post-hoc test. Results: An interaction effect between cameras and calibration models was found (P=0.0002). The Delta Es ranged from 1.79 (CCD with 3rd order polynomial regression) to 5.25 after calibration protocols. No statistically significant difference was found in terms of calibration accuracy between CCD or CMOS cameras when calibrated with either the 3rd order regression or the tetrahedral interpolation model. Conclusion: The results show promising use of calibrated commercial digital cameras for shade matching if light source stability can be improved. Supported by: NIH R15 EY013527 and an ACP Research Fellowship in Geriatric Prosthodontics.

Back to the Prosthodontics Research Program
Back to the IADR/AADR/CADR 82nd General Session (March 10-13, 2004)

Top Level Search