電腦運算速度在科技日新月異的帶動下越來越快，而利用電腦的高速運算速度來實現影像處理已實現在許多研究中，譬如人臉特徵辨識及表情辨識。在表情辨識方面，最為廣泛的就是微笑辨識，因為微笑是人類的共同的面部特徵，而許多消費性電子產品例如數位相機都有微笑偵測功能。在此篇論文中，提出一套能有效辨識微笑的方法，該方法分為三大步驟：先基於膚色及橢圓模板的人臉偵測，再根據相對位置及運用小型模板進行的臉部器官偵測，最後監測嘴巴形狀變化度的微笑辨識。在人臉偵測及器官偵測方面，使用膚色網狀結構模板的橢圓區域來偵測是否存在眼睛及嘴巴等面部特徵。在微笑檢測中，使用適應演算法來檢測微笑嘴唇的變異程度。由實驗結果可以得知，本論文所研究之微笑偵測方式，可以在檢測面部時容忍眾多的使用者來利用系統，使得辨識系統更為強健，另一方面，在其他影像前處理時能將運算時間有效的降低，使得在即時判斷微笑系統中能更為流暢。

The speed of computation in computer is growing powerfully as time going through. Many researchers have used this chrematistic advantage to illustrate the image processing such as the face recognition. After all, smile is the common facial feature in human, and there are many electronic products like camera that are able to detect smile. This thesis contributes to the development of novel smile recognition. It includes 3 parts: the face detection based on skin color and ellipse template, facial feature detection by small template according to relative position and a method of smile detection by variation of mouth. In face detection, it uses the mesh ellipse filter to detect the skin region and existence of eyes and lip. In smile detection, it detects smile with variation of lip using adaptive theorem. Simulation results show that the smile detection of adaptive system in real-time can robustly accept different persons to use this application and recognize smile. On the other hand, based on proposed preprocesses, it can reduce the computation time to judge smile in smooth way.

[1]Chellappa, R., Wilson, C.L. and Sirohey, S., “Human and machine recognition of faces: a survey”, Proceedings of the IEEE, Vol. 83, pp.705-741, Dec 1995.
[2]Lin C.H., Wu J.L., “Automatic facial feature extraction by genetic algorithms” Proceeding of IEEE Transactions on Image Processing, Vol. 8, pp834-845, Jun 1999.
[3]Clippingdale, S. and Ito, T., “A unified approach to video face detection, tracking and recognition”, Proceeding of International Conference on Image Processing (ICIP), Vol. 1, pp.662-666, 1999.
[4]Cox, I.J., Ghosn, J. and Yianilos, P.N., “Feature-based face recognition using mixture-distance”, Proceeding of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), pp.209-216, Jun 1996.
[5]Heinzmann, J. and Zelinsky, A., “3-D facial pose and gaze point estimation using a robust real-time tracking paradigm”, Proceeding of Third IEEE International Conference on Automatic Face and Gesture Recognition, Apr 1998.
[6]Michael Chester, NEURAL NETWORKS A Tutorial, Englewood Cliffs, N.J. :PTR Prentice Hall, 1993.
[7]Jing M.C., “A simple image processing demo under .Net environment using C# ”, Feb 2007.
[8]Mohammad Maher Elrefaei, R, Detecting Smile and Forgery in Digital Color Image, National Taiwan University, electrical engineering. 2009
[9]Rafael C. Gonzalez, Richard E. Woods and Steven L. Eddins., “Digital Image processing using MATLAB”, Upper Saddle River, N. J. :Pearson Prentice Hall, 2004.
[10]Hsu R.L, Abdel-Mottaleb, M. and Jain, A.K., “Face detection in color images”, Proceeding of IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, pp.696-706, May 2002.
[11]Soriano, M., Martinkauppi, B. and Huovinen, S., “Laaksonen, M., Skin detection in video under changing illumination conditions,” Proceeding of 15th International Conference on Pattern Recognition, Vol. 1, pp839-842, July 2000.
[12]Wang D. and Zhou S., “Color Image Recognition Method Based on the Prewitt Operator”, International Conference on Computer Science and Software Engineering, Vol.6, pp.170-173, Dec 2008.
[13]Gao W.S., Zhang X.G., Yang L. and Liu H.Z., “An improved Sobel edge detection”, Proceeding of 2010 3rd IEEE International Conference on Computer Science and Information Technology (ICCSIT), Vol.5, pp.67-71, Jul 2010.
[14]Chen T.Y., Design and implementation of the face tracking and recognition system, National Cheng Kung University, Engineering Science, Jun 2007.