由於人們可以透過音樂互相交流情感和生活體驗，所以音樂已在許多人的日常生活中成為不可或缺的部分。音樂可以依照樂譜形式記載，讓世界各地的人們可以了解、演奏及編寫。因此，樂譜可以說是音樂文字，常見的樂譜有簡譜與五線譜等等。然而，對於接觸音樂領域的初學者來說，識譜能力不足是個首要的問題。此外，隨著機器人的發展越來越多樣化及人性化，人機互動的功能對於機器人來說也顯的越來越重要。因此，機器人的發展將朝著具有表演及娛樂性質的方向，像是視譜唱歌或是按譜演奏音樂等等。在機器人劇場中，可作為表演視譜唱歌之項目，提供具有娛樂性質之表演。
有鑑於此，樂譜辨識的方法便可用來協助識譜能力較差的音樂初學者，以提供他們增進識譜的能力或與加快識譜的速度，使他們能夠在學習上達到事半功倍的效果。亦或者是將樂譜辨識的方法應用於機器人的看譜表演中。而已知的樂譜辨識方法，通常是利用掃瞄器來取得樂譜影像，且樂譜影像的內容為單純的樂譜符號。因此，習知的樂譜辨識方法所能應用的環境較為有限。例如:日本山葉(Yahama)股份有限公司的樂譜辨識裝置，為了避免樂譜
影像因人為拍攝或外在之環境因素而導致歪斜(skewing) 或扭曲(warping)，因此所辨識的樂譜影像為利用掃描器(scanner)掃描之影像，而樂譜影像的內容則為單純的樂譜符號(不包含歌詞)。在本篇論文中，我們將發展一套具有普遍性及強健性的演算法來解決上述的限
制。

Music plays an important role in human life. People can understand, perform, and write music through musical notations. Diverse forms of musical notations have been developed in various favor, such as the numbered musical notation and the five-line staff. However, it is a serious problem that people first touching on music cannot understand musical notations. Besides, it is more important for human—computer interaction because of diversification and humanity for robots. Robots will be developed to face performance and entertainment, e.g., the robot reading and singing. It can be regard as a performance with entertainment in the robot theater.
Recently, several musical notation recognition systems have been developed successfully under some constraints. For Yamaha’s musical notation recognition system, the constraint is that the musical notation should be captured by the scanner in order to avoid the warping and the skewing effects. Due to the constraint, the captured musical notation images usually have good quality. In this thesis, we relax the above constraint and design a more general musical notation recognition system which resolves these considerations, such as skew and warp. Under different kinds of real testing musical notation images, experimental results show that our proposed novel generalized musical notation recognition system is robust and encouraging.

[1] 鍾國亮，影像處理與電腦視覺，第三版，東華書局，台北，2006。
[2] 鍾國亮，資料壓縮的原理與應用，第二版，全華科技圖書，台北，2004。
[3] R. C. Gonzalez and R. E. Wood, Digital Image Processing. Second Edition,
PrenticeHall, New York 2002.
[4] N. Otsu, “A threshold selection method from gray level histogram,” IEEE
Trans. on Systems, Man, and Cybernetics, vol. 8, pp. 62 – 66, 1978.
[5] J. N. Kapur, P. K. Sahoo, and A. K. C. Wong, “ A new method for gray-level
picture thresholding using the entropy of the histogram, ” Computer Vision,
Graphics, and Image Processing, 29(3), 1985, pp. 273 – 285.
[6] K. L. Chung, T. C. Chen, and W. M. Yan, “ New memory-and
computation-efficient Hough transform for detecting lines, ” Pattern
Recognition, 37 (5), pp. 953 – 963, May 2004.
[7] T. C. Chen and K. L. Chung, “ A new randomized algorithm for
detecting lines, ” Real-Time Imaging, 7 (6), pp. 473 – 482, Jun 2001.
[8] R. Hartley and A. Zisserman, Multiple view geometry in computer
vision. Cambridge University Press, New York, 2000.
[9] D. Nehab, “Staff Line Detection by Skewed Projection,” May 2003.
[10] W. Niblack, An Introduction to Digital Image Processing. Prentice
Hall, New Jersey, 1986.
[11] Wei-chen Lee, Hung-Yan Gu, Kuo-Liang Chung, Chyi-Yeu Lin,
Chin-Shyurng Fahn, Yah-Syun Lai, Chih-Cheng Chang, Chia-Lun
Tsai, Kai-Jay Lu, Huang-Liang Liau, Mao-Kuo Hsu, " The
Realization of a Music Reading and Singing Two-Wheeled Robot "
IEEE Workshop on Advanced Robotics and its Social Impacts
(ARSO 2007), pp. 67 – 72, 2007.
[12] C.Y. Lin, C. K. Tseng, H.Y. Gu, K.L. Chung, C.S Fahn, K. J. Lu,
C.C. Chang, “ An autonomous singing face robot, ” accepted for
publication, 2008.
[13] R. J. Randriamahefa, J. P. Cocquerez, C. Fluhr, F. Pepin, and S.
Philipp, “ Printed music recognition, ” in Proc. of the International
Conference on Document Analysis and Recognition, pp. 898 – 901,
October 1993.
[14] K. T. Reed and J.R. Parker, “ Automatic Computer Recognition of
Printed Music, ” in Proc. of the International Conference on Pattern
Recognition, vol. 3, pp. 803 – 807, 1996.
[15] D. Deutsch, “ Music Recognition, ” Psychological Review, 76, pp.
300 – 307, 1969.