(3.235.191.87) 您好!臺灣時間:2021/05/13 04:34
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:郝維中
研究生(外文):Wae-Jone Hao
論文名稱:使用探測點與遮罩技巧辨識樂譜
論文名稱(外文):Music Score Recognition Using Masking and Probing Techniques
指導教授:陳淑媛陳淑媛引用關係
指導教授(外文):Shu-Yuan Chen
學位類別:碩士
校院名稱:元智大學
系所名稱:資訊工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2005
畢業學年度:94
語文別:中文
論文頁數:57
中文關鍵詞:樂譜辨識辨識樂譜
外文關鍵詞:Music Score Recognition
相關次數:
  • 被引用被引用:0
  • 點閱點閱:123
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
本論文提出一個以探測點與遮罩的技巧用以辨識樂譜符號。探測點的做法是於樂譜的影像上,對於每一次樂譜符號比對之前,並不直接做完整的符號樣本比對,而是先對該符號上較具代表性的幾處點先檢測是否初期符合,若是,才進一步做完整的符號樣本比對,這對於種類甚多的樂譜符號的辨識來說可以節省大量處理的時間;而遮罩的做法是將比對之後辨識出的符號以該符號的輪廓外形為範圍改變範圍內色點的值,舉例來說,於黑白樂譜的圖面上某一處找到一個升記號,則在該處將它改成紅色,其它的符號也各有賦予的專屬色,如此可以在圖面上搜尋時略過這些已被顏色標記的區域,只尋找代表未被辨識的區域色-黑色,節省大量運算時間。整個樂譜的辨識就是在圖面上尋找未被染色的黑點並以探測點的方式加以比對直到所有的點都被染色為止。除了本文表述的這兩種技巧之外,也利用投影技巧來尋找五線譜的位置,以及以幾何關聯位置判斷數量最多的音符符頭及沒有固定造形的符號如連音線等。
A method to recognize music score using masking and probing techniques is proposed in this thesis. On the score map, we firstly exam the pattern using check points before completed pattern matching. If all check points pass the pattern exam, in other words all check points do not touch white point on map, fine matching will be employed. In this way, time will be saved, especially for dozens patterns of musical note symbol are to be proceeded for recognition. As to the masking, once a symbol is recognized on the score map, the region of the symbol will be colored by a pattern image. For example, if a symbol sharp was located on the map, we use a bolder sharp pattern to cover this region. Only the overlap will be labeled with different color which is other than black. During scan-line session black points will be found and considered as unrecognized region, all the colored region will be skipped for time saving. In addition, techniques are adopted to locate staff lines, and heuristic of music theory could be used to aid note symbol recognition.
摘要 i
Abstract iii
索引 iv
圖索引 1
第一章 簡介 3
1.1. 動機 3
1.2. 相關文獻參閱 3
1.3. 本論文方法建構的依據 6
第二章 說明探測點與遮罩的要義與方法 9
2.1. 為何採用探測點 9
2.2. 為何採用遮罩 11
第三章 建立樣本與遮罩 15
第四章 具固定形狀與方向的符號之辨識程序 25
4.1. 探測點的比對順序 25
4.2. 符號樣本辨識流程 27
4.3. 樣本比對的評分 29
4.3. 符號比對的順序 31
第五章 無固定形狀與方向的符號之辨識程序 32
第六章 辨識程序整合與實驗結果 45
6.1. 辨識程序整合 45
6.2. 實驗結果 48
第七章 結論與未來研究 50
參考文獻 51
[1]Marinai, S. and Nesi, P., “Projection based segmentation of musical sheets,” Proceedings of the Fifth International Conference on Document Analysis and Recognition, 20-22 Sept. 1999, pp.515 – 518.
[2]Miyao, H. and Nakano, Y., “Head and stem extraction from printed music scores using a neural network approach,” Proceedings of the Third International Conference on Document Analysis and Recognition, 14-16 Aug. 1995, pp.1074 - 1079 vol.2.
[3]Ng, K.C., Boyle, R.D. and Cooper, D., “Low- and high-level approaches to optical music score recognition,” IEE Colloquium on Document Image Processing and Multimedia Environments, 2 Nov 1995, pp. 3/1 - 3/6.
[4]Randriamahefa, R., Cocquerez, J.P., Fluhr, C., Pepin, F. and Philipp, S., “Printed music recognition,” Proceedings of the Second International Conference on Document Analysis and Recognition, 20-22 Oct. 1993, pp. 898 – 901.
[5]Sicard, E., “An efficient method for the recognition of printed music,” Proceedings of 11th IAPR International Conference on Pattern Recognition vol. 3. Conference C: Image, Speech and Signal Analysis, 30 Aug.-3 Sept. 1992, pp. 573 – 576.
[6]Wijaya, K., Bainbridge, D., “Staff line restoration,” Seventh International Conference on Image Processing and Its Applications, (Conf. Publ. No. 465) , July 1999, pp.760 - 764 vol.2
[7]Leplumey, I., Camillerapp, J., Lorette, G., “A robust detector for music staves,” Proceedings of the Second International Conference on Document Analysis and Recognition, 20-22 Oct. 1993 pp.902 – 905
[8]Yip-San Wong, Choi, A., “A two-level model-based object recognition technique,” Proceedings of International Symposium on Speech, Image Processing and Neural Networks, 13-16 April 1994 pp.319 - 322 vol.1
[9]Stuckelberg, M.V., Doermann, D., “On musical score recognition using probabilistic reasoning,” Proceedings of the Fifth International Conference on Document Analysis and Recognition, 20-22 Sept. 1999, pp.115 – 118
[10]Stuckelberg, M.V., Pellegrini, C., Hilario, M., “An architecture for musical score recognition using high-level domain knowledge,” Proceedings of the Fourth International Conference on Document Analysis and Recognition, 18-20 Aug. 1997 pp.813 - 818 vol.2
[11]Baumann, S., “A simplified attributed graph grammar for high-level music recognition,” Proceedings of the Third International Conference on Document Analysis and Recognition, Volume: 2 , 14-16 Aug. 1995 pp.1080 - 1083 vol.2
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔