利用電腦輔助設計與製造作為齲齒陶瓷復形物、陶瓷牙冠或陶瓷美白貼片已成為臨床可接受的治療方式；其過程中所重建復形窩洞之三維模型，除了以傳統的翻模之外，利用三維光學重建也是目前常用的方法，其可改善傳統補綴物不美觀、強度不足等缺點，與改善牙冠製作流程耗時等缺點。本研究將採用三維光學量測方式，並探討重建結果與分析，以作為未來牙科臨床之應用。
　　本研究利用光學量測之主動式三角量測法做為物體的高度重建，並配合上相位移技術提高量測之精確度。研究包括了光學系統校正、光學系統架設、數位影像處理。光學系統架設利用電腦製作四步相位移條紋，透過數位投影機投射於待測物，經由電耦合元件將條紋影像擷取並儲存於電腦；再將擷取之條紋影像做影像增強、圖像雜訊濾除、帶通濾波…等前置處理。將處理完成之四張條紋圖像利用相位移法求得各像素點之相位，並透過相位展開技術求得連續相位，重建待測物表面輪廓。
　　目前重建項目有方形校正塊、方形校正塊相疊、斜面、蓋子曲面與酒瓶。結果顯示：高度為14.0 mm之方形校正塊重建後平均高度為12.476 mm；校正塊高度分別14.0 mm與8.0 mm相疊，重建結果分別為12.316 mm與8.054 mm；斜面重建後之平均絕對誤差為1.778 mm；蓋子曲面重建後之平均絕對誤差為1.189 mm；而酒瓶因瓶身邊緣高度大，因此所重建後之高度誤差超過一倍。
　　本實驗重建結果準確度不足，其造成之原因為：（1）物體高度所造成的陰影，使相位判別出現錯誤；（2）架設系統之精確度不足；（3）相位展開演算法為與展開路徑有關。未來研究方向建議：（1）發展與路徑無關之相位展開法；（2）提高系統之精確度；（3）將系統微小化，且可置於口內於牙齒之重建。

　　Computer aided design and manufacture technology has been accepted as one of the clinical approaches to fabricate restoration block for dental caries treatment. To fabricate the restoration block, three-dimensional computer model is required. Optical method is one of the available method to build the three-dimensional model.
This research uses phase-shifting active optical measurement method to reconstruct the height of three-dimensional surface as a preliminary step for future dental applications. The process included the setting of optics system, calibration the system, and digital image processing to reconstruct the height. The optics system included a digital projector and a CCD camera. Four fringe patterns, with different phases, were projected onto the sample object surface by the projector and the images were captured by the CCD. The images intensity was enhanced and filtered in both space and frequency domains. The phase angles of each pixel was established through these four images. The phase angles were then unwrapped to reconstruct in a continuous way and the height was obtained.
　　The reconstructed samples included a squared calibration block, stacking of two squared calibration block, an inclined wedge block, a curved lid and a bottle. The results showed that : the absolute height of the squared calibration is 14.0 mm while the reconstruct height was 12.476 mm; the absolute heights of the two stacking square squared calibration were : 14.0 mm and 8.0 mm respectively while the reconstructed heights were 12.316 mm and 8.054 mm respectively; The Mean Absolute Error (MAE) of the wedge block after reconstruction was 1.778 mm. As for the lid and bottle, the general surface patterns were obtained but the accuracy needs further refine.
　　In this research, the generated reconstruction results were not accurate enough. The reasons might due to: (1) the shadow induced by the sample edge height which would create error in the phase reconstruction; (2) setting of the optical system was not accurately calibrated; (3) the phase unwrapped algorithm dependent on the phase expansion’s route. Suggested improvements are: (1) develop a path independent phase unwrapped algorithm; (2) increase system accuracy by more detailed calibration; (3) decrease the system’s size to the size of being possible to reconstruct a teeth from the patient’s oral.

參考文獻
[1] P. E. Petersen, “World Map on Dental Caries, 35-44 Years,” W.H.O., 2003.
[2] http://pharma.kolon.co
[3] P. E. Petersen, “Changing Levels of Dental Caries Experience (DMFT) Among 12-year-olds in Developed and Developing Countries,” W.H.O.
[4] 張智毫, “高雄市五至十八歲學生口腔流行病學調查之研究,” 高雄醫學大學口腔衛生科學研究所碩士論文, 2002.
[5] http://www.bhp.doh.gov.tw
[6] 孫安迪,「牙齒與口腔疾病」, 健康世界發行出版社, 1992.
[7] http://www.dentalkennan.jp
[8] 高益鳴, 「專家解答：口腔疾病」, 史丹佛預防醫學, 2006.
[9] http://www.odontoclinica.cl/Fluor.htm
[10]http://www.edwardbyrne.com/decay.htm
[11]http://www.toothclub.gov.hk
[12]鍾國雄, 「牙科材料學」, 合記圖書, 1993.
[13]http://mysuperdentist.com/
[14]張家豪, “電腦輔助工程應用於牙科復形物及數位影像腳形、腳壓量測系統之開發,” 崑山科技大學機械工程系碩士論文, 2003.
[15]B. Bidanda, S. Motavalli, K. Harding, “Reverse Engineering: An Evaluation of Prospective Non-contact Technologies and Application in Manufacturing System, ” Int. J. Comput. Inter. Manuf., Vol.4, pp.145-156, 1991.
[16]P. F. Luo, Y. J. Chao, M. A. Sutton, “Application of Stereo Vision to Three-dimensional Deformation Analyses in Fracture Experiments,” Opt. Eng., Vol.33, pp.981-990, 1994.
[17]M. Magee, R. Weniger, “Isolation of Three Dimensional Features of Known Height Using a Light-striped Stereopic System,” SPIE, Vol.1822, pp.175-187, 1992.
[18]M. Halioua, R. S. Krishnamurthy, H. C. Liu, F. P. Chiang, “Automated 360∘Profilometry of 3D Diffuse Object,” Appl. Opt., Vol.24, pp.2193-2196, 1985.
[19]J. S. Massa, G. S. Buller, A. C. Walker, S. Cova, M. Umasuthan, A. M. Wallace, ”Time-of –flight Optical Ranging System Based on Time-correlated Single-photon Counting,” Appl. Opt., Vol.37, pp.7298-7304, 1998.
[20]H. Takasaki, “Moiré Topography,” Appl. Opt., Vol.9, pp.1467-1472, 1970.
[21]A. Asundi, W. Zhou, “Unified Calibration Technique and its Application in Optical Triangular Profilometry,” Appl. Opt., Vol.38, pp.3556-3561, 1999
[22]Chen F, Brown G M., Song M., "Overview of three-dimensional shape measurement using optical methods “J. Opt Eng. , pp. 10-29, 2000.
[23]M. Takeda, K. Mutoh,”Fourier transform profilometry for the automatic measurement of 3-D object shapes,” Appl. Opt., Vol. 22, 1983.
[24]P. Huang, S. Zhang, F.P. Chiang, “Trapezoidal phase-shifting method for 3-d shape measurement," in Proc. SPIE, pp. 142-152, 2004.
[25]V. Srinivasan, H. C. Liu, M. Halioua, “Automated Phase-measuring Profilometry of 3-D Diffuse Objects,” Opt. Soc. Am., Vol.23, 1984.
[26]F.Chen, G.M. Brown, M. Song, “Overview of Three-dimensional Shape Measurement Using Optical Methods.” Opt. Eng., pp. 10-22, 2000.
[27]張柏毅, “應用條紋投影法及條紋反射法量測量測物體表面形貌,” 國立成功大學機械工程系碩士論文, 2005.
[28]郭哲佑, “相位移條紋投影輪廓儀之特性與分析,” 國立中山大學材料科學研究所碩士論文, 2005.
[29]張家壽, “應用數位投影疊紋法於微小尺寸表面之量測,” 國立台灣大學應用力學研究所碩士論文, 2001.
[30]繆紹綱, 「數位影像處理」,灣培生教育出版社, 2004.