對於聽障或對口述語言表達有困難的人來說，手語是最常見到的語言表達方式。對於一般正常人而言，手語並不是一個普遍的語言，以致於一般正常人與聽障者做資訊傳達時，常會有溝通不良或資訊傳達錯誤的問題產生。本論文的主要目的是為了增進正常人與聽障人士溝通的便利性而研究。在此，我們以手語手型的辨識為主要的研究方向。

　　在圖形識別中，以像素為手型的表達方式，對於影像的旋轉，有很高的敏感度影響分類的結果好壞。相對於以輪廓為手型的表達方式，對於影像的旋轉，有旋轉不變性的特性，其分類結果並不會有所影響。在過去的研究中可發現，以輪廓為手型的表達方式的分類結果，大多都高於以像素為手型的表達方式。本篇論文提出的像素為基底的表示方式擁有影像大小不變性與影像旋轉不變性這兩個特性。根據我們分析，每個手語中的手型都有其主要手指的指向，手指的指向大部分會與其手型的最大軸平行。我們利用這個特徵，提出角度正規化，解決原本的影像旋轉不變性的問題。經由分析各個類別的共變異矩陣，可以看出角度正規化的確大幅的降低各個手型中每個樣本間的差異性，這也就是說，降低各個類別間重疊性。

　　在本論文最後的實驗中，比較以像素為基底的傳統表達方式跟我們所提出的方法，而實驗結果證實了我們所提出的方法能提高手型辨識的準確率。並且，在相同解析度的條件下，也比較了以輪廓為手型的表達方式的準確率。實驗結果證實以像素為手型的表達方式的準確率能夠提升到相近於以輪廓為手型的表達方式的準確率。

　The pixel-based shape representation has been sensitive to rotation. In this paper, we propose a pixel-based descriptor that is invariant with rotation and scale for the hand shape recognition in Taiwanese Sign Language (TSL). Based on the property that a hand shape is characteristic of a unique pointing direction, angle normalization is used to meet the rotation-invariant requirement. With angle normalization, the traces of class covariance matrices have been reduced almost all over the classes of hand shapes, implying a less overlap between classes. This leads to an increase in recognition accuracy.

[1]Y. Cui and J. Weng, “A learning-based prediction-and-verification segmentation scheme for hand sign image sequence,” IEEE Trans. Pattern Anal. Machine Intell., vol. 21, no. 8, Aug. 1999.
[2]N. H. Habili, and C.C. Lim, and A. Moini, “Segmentation of the face and hands in sign language video sequences using color and motion cues,” IEEE Trans.Circuit Syst. Video Technol., vol.14, no. 8, Aug. 2004.
[3]J. Cai and Z.Q. Liu, “Hidden markov models with spectral feature for 2D shape recognition,” IEEE Trans. Pattern Anal. Machine Intell., vol. 23, no. 12, Dec. 2001.
[4]H. Kauppinen, T. Seppanen and M. Pietikaainen, “An experimental comparison of autoregressive and fourier-based descriptors in 2D shape classification,” IEEE Trans. Pattern Anal. Machine Intell., vol. 17, no. 2, Feb. 1995.
[5]K. Arbter, W. E. Snyder, H. Burkhardt and G. Hirzinger, “Application of affine-invariant fourier descriptors to recognition of 3-D objects,” IEEE Trans. Pattern Anal. Machine Intell., vol. 12, no.7, July 1990.
[6]史文漢、丁立芬，手能生橋，第一冊~第二冊，中華民國聾人協會發行，2004.
[7]D. Chai and K. N. Ngan, “Face segmentation using skin-color map in videophone applications,” IEEE Trans. Circuits Syst. Video Technol., vol. 9, pp. 551–564, June 1999.
[8]H. Wang and S.-F. Chang, “A highly efficient system for automatic faceregion detection in MPEG video,” IEEE Trans. Circuits Syst. VideoTechnol., vol. 7, pp. 615–628, Aug. 1997.
[9]K. Sobottka and I. Pitas, “A novel method for automatic face segmentation, facial feature extraction and tracking,” Signal Process. Image Commun., vol. 12, pp. 263–281, June 1998.
[10]X. Zhu, J. Yang, and A. Waibel, “Segmenting hands of arbitrary color,” in Proc. IEEE 4th Int. Conf. FG, Grenoble, France, pp. 446–453, Mar. 2000.
[11]M. J. Jones and J. M. Rehg, “Statistical color models with application to skin detection,” in Proc. CVPR, Ft. Collins, Colorado, pp. 274–280, June 1999.
[12]R. C. Gonzalez and R. E. Woods, Digital Image Processing, Addison-Wesley, 2002.
[13]J. D. Foley et al., Computer Graphics, Addition Wesley, 1992
[14]K. Fukunaga, Introduction to Statistical Pattern Recognition, New York: Academic Press, 1990.
[15]R. M. Haralick and L. G. Shapiro, Computer and Robot Vision, Volume I, Addison-Wesley, pp. 28–48, 1992.
[16]Q. Jackson and D. A. Landgrebe, “An adaptive method for combined covariance estimation and Classification,” IEEE Trans. Geosci. Remote Sensing, vol. 40, pp. 1082–1087, May 2002.
[17]S. P. Lin and M. D. Perlman, “A monte carlo comparison of four estimators of a covariance matrix,” in Multivariate Analysis–VI: Proceeding of the Sixth International Symposium on Multivariate Analysis, P.R. Krishnaiah, Ed. Amsterdan, The Netherlands: Elsevier, 1985.
[18]P. W. Wahl and R. A. Kronmall, “Discriminant functions when covariances are equal and sample size are moderate,” Biometrics, vol. 33, pp. 479–484, 1977.
[19]S. Marks and O.J. Dunn, “Discriminant functions when the covariance matrices are unequal,” J. Amer. Stat. Assoc., vol. 69, pp.555–559, 1974.
[20]J. P. Hoffbeck and D. A. Landgrebe, “Covariance matrix estimation and classification with limited training data,” IEEE Trans. Pattern Anal. Machine Intell., vol. 18, pp. 763–767, July 1996.