在本論文中對運動影片提出了一種新的廣告偵測流程。首先，自電視台所錄影的運動影片被分為不同的分鏡(shot) ，之後對每一個獨立的分鏡使用記分板偵測的方式來決定在分鏡中是否存有計分板。如分鏡的內容中含有計分板，則此分鏡就一定不屬於廣告片段。其次，對沒有包含記分板畫面的分鏡，取出其關鍵影格(key-frame)並與包含有記分板畫面之分鏡的關鍵影格做比較，如其相似程度較高，則系統進一步以此分鏡的特徵，利用向量支援機器(Support Vector Machine, SVM)分類器來判斷此分鏡是否為廣告節目。實驗結果顯示本論文所提出的方法能夠有效的區分出節目與廣告的分鏡。最後的實驗數據顯示此方法在實際應用上是可行的。

In this thesis, a novel TV commercial detection scheme for sports videos is proposed. Firstly a sport video is partitioned into shots and then the scoreboard detection method is applies to each individual shot to determine if any scoreboard existed in the shot. If the shot contains scoreboard, then the shot belongs to a program shot. Next, the key-frame of a shot without scoreboard would be compared with that with scoreboard. If high similarity is achieved between them, then a SVM classifier based on the features of the shot is used to determine if the shot belongs to a commercial one. Experimental results demonstrate that the proposed scheme can effectively differentiate commercial shots from programs ones. The high detection rate shows the feasibility of the proposed scheme for practical applications.

[1] ISO 11172, "Coding of Moving Pictures and Associated Audio–for Digital Storage Media at up to about 1.5 Mbit/s."

[2] MPEG ORG , http://www.mpeg.org

[3] LIBSVM: A Library for Support Vector Machines, Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm, 2001.

[4] A. Albiol, M. Ch, F. Albiol, and L. Torres. "Detection of TV commercials, " in Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, vol. 3, pp.541-544, 2004.

[5] M. Aizerman, E. Braverman, and L. Rozonoer, " Theoretical Foundations of the Potential Function Method in Pattern Recognition Learning, " Automation and Remote Control, Vol. 25, 1964.

[6] B. E., Boser, I. M. Guyon, and V. N. Vapnik, "A Training Algorithm for Optimal Margin Classifiers, " in Proceedings of the Fifth Annual Workshop on Computational Learning Theory 5, pp. 144-152, 1992.

[7] L. Y. Duan, M. Xu, Q. Tian, C. S. Xu, and J. S. Jin, "A unified framework for semantic shot classification in sports video, " IEEE Transactions Multimedia, vol. 7, no. 6, pp. 1066-1083, 2005

[8] R. Fletcher, "Practical methods of optimization," Wiley, 2nd edition, 1987

[9] X. Hua, L.Lu, and H. Zhang, "Robust Learning-Based TV Commercial Detection", in Proceedings of the IEEE International Conference on Multimedia and Expo (ICME’05), vol. 4, pp. 6-8, July 2005.

[10] Y. Huang, L. Hsu, Sandnes, F.-E., "An Intelligent Subtitle Detection Model for Locating Television Commercials," IEEE Transactions on Systems, Man, and Cybernetics, Part B, Volume 37, Issue 2, pp. 485-492, 2007

[11] J. Jian, M. Hung, C. Hsieh, and Y. Chang, "Real-Time Scene Classification for Baseball Videos," in Proceedings of the CVGIP International Conference on Vision, Graphics and Image Processing, pp. 115-122, 2005

[12] D.Klein, "Lagrange Multipliers without Permanent Scarring," berkeley, 2004

[13] S. Li, H. Li, and Z. Wang, "A Novel Approach to Commercial Detection in News Video," in Proceedings of the IEEE International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing, Vol. 2, pp. 86 – 90, 2007

[14] Y. Li and C.-C. J. Kuo, "Detecting commercial breaks in real TV programs based on audiovisual information," in Proceedings of the Internet Multimedia Management Systems. vol. 421, pp. 225-236, 2000

[15] C. Liang, J. Kuo, W. Chu, and J. Wu, "Semantic units detection and summarization of baseball videos," in Proceedings of the 47th Midwest Symposium on Circuits and Systems, Vol. 1, pp. 297-300, 2004

[16] R. Lienhart, C. Kuhmunch, and W. Effelsberg, "On the detection and recognition of television commercials, " in Proceedings of the IEEE Conference on Multimedia Computing and Systems, pp. 509-516. 1997

[17] Y. Rubner, C. Tomasi and J. Guibas, "The Earth Mover’s Distance as a Metric for Image Retrieval," International Journal of Computer Vision vol. 40(2), pp.99-121, 2000

[18] D. Sadlier, S. Marlow, N. O'Connor, and N. Murphy, "Automatic TV advertisement detection from mpeg bitstream, " Pattern Recogition Society, vol. 35, no. 12, pp. 2-15, 2002.

[19] V . Vapnik, "The Nature of Statistical Learning Theory." Springer - Verlag, 1999

[20] L. Wu, X. Han, H. Cui, and W. Shen, "Shot Segmentation and Classification in Basketball Videos", in Proceedings of the IEEE International Conference on Intelligent Information Hiding and Multimedia Signal Processing, Vol. 2, pp. 539-542, 2007

[21] J. Yeh, J. Chen, and J. Kuo et al, "TV commercial detection in news program videos, " in Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS’05), Vol. 5, pp. 23-26, 2005.

[22] B.L. Yeo and B. Liu, "A Unified Approach to Temporal Segmentation of Motion JPEG and MPEG Compressed Video," in Proceedings of the IEEE International Conference on Multimedia Computing and System, pp. 81–83, 1995.

[23] 吳上立, 林宏墩, C語言數位影像處理, 全華, 2006

[24] 黃堯彬, 黃詰琳, 詹寶珠, "使用溫感影像做靜態手勢辨識," 數位生活研討會, 2008

[25] 戴顯權, 資料壓縮, 紳藍, 2001