在一場球類比賽的轉播中，通常會在場地上不同角度處架設多台攝影機，看轉播之觀眾所看到的播出畫面，通常是由導播依自己的意思，從這眾多的畫面來決定要選播那一畫面。隨著互動式需求的增加，這種強迫觀眾接受觀看統一畫面的方式將漸被唾棄，取而代之的是，播出畫面會依觀眾的需求來安排。因此，本研究嘗試著以觀眾的觀點為前題，考慮結合不同之因素來選用要播出畫面的選擇。兩個主要因素為(1)籃球所在之位置；(2)特定球員。為要達到知道所要之特定物件在球場上之位置變化，影像處理技術被用來對每一畫面做分割，並利用追蹤技術來決定該特定物件在各個連續畫面中的移動情況。評估模組考慮到對鏡頭畫面進行特定球員是否存在、是否存在籃球、籃球框與球員之間距離關係、區塊之間遮蔽情況以及感興趣球星與籃球之間的位置關係來評估並給予一個分數，播出鏡頭的選擇是根據這些評估出來的分數中選其最大者之鏡頭。若只考慮評估出來的分數，吾人發覺，畫面會因為不同鏡頭間之轉換頻繁，致使觀看畫面時會感不舒服，再加入畫面播出之平滑考量，讓畫面撥出不過於頻繁地切換鏡頭。本文提出評估因素使用不同的權重條件以及不同的平滑化方法來選擇要輸出的影像，實驗結果顯示，本論文所提出的方法可以符合觀眾所喜愛的觀看因素以及舒適程度。

There are many cameras installed in the basketball court for capturing the scene of the competition and broadcasting to the audience. The frames viewed by the audience are selected by the director. These frames may not meet the needs or interest of the audience. In this research, we consider the selecting of the presented frames based on the audience’s view. The two main factors to be considered are the basketball and the interesting player. Image processing techniques are used to preprocess each frame and find the objects on the frame. Then object tracking algorithm is applied to track the object moving trajectory over the contiguous frames. An evaluation module is created to estimate the score of each frame according to the factors: (1) if the basketball found or not; (2) the distance between the basketball and the score plate; (3) if the specific player existing; (4) the distance between the specific player and the score plate; and (5) the occluding condition among the players. The frame with the highest score is selected for presenting to the audience. Due to the frequently switching among different camera, it is felt not comfortable for audience’s eye. Smoothing factor is added for increasing the comfortable of the view’s eye. From the experimental results shown, the proposed method can match the results conducted from the subjective evaluation.

[1] B. Fan, Y. Du, L. Zhu, ”A robust template tracking algorithm with weighted active drift correction,” Pattern Recognition Letters, Vol. 32, Issue 9 , pp. 1317-1327, Jul. 2011.
[2] Z. Han, J. Jiao, B. Zhang, Q. Ye, J. Liu, ”Visual object tracking via sample-based Adaptive Sparse Representation (AdaSR),” Pattern Recognition, Vol. 44, Issue 9, pp. 2170-2183, Sep. 2011.
[3] S.M. Qi, X.W. Huang, H.F. Yi, ”Object tracking by Anisotropic Kernel Mean Shift,” Jonurnal of Electronics & Information Technology, Vol. 29, No. 3, pp. 686-689, May. 2007.
[4] V. Papadourakis, A. Argyros, ”Multiple objects tracking in presence of long-term occlusions,” Computer Vision and Image Understanding, Vol. 114, Issue 7, pp. 835-846, Jul. 2010.
[5] M. Wu, X. Peng, Q. Zhang, R. Zhao, ”Segmenting and tracking multiple objects under occlusion using multi-label graph cut,” Computers & Electrical Engineering, Vol. 36, Issue 5, pp. 927-934, Sep. 2010.
[6] 謝明逢, 范國清, ”Construction of a surveillance system for large monitoring spaces by a dual-camera module,” 國立中央大學資訊工程研究所碩士論文, Jun. 2005.
[7] J. Yang, O. Arif, P.A. Vela, J. Teizer, Z. Shi, ”Tracking multiple workers on construction sites using video cameras,” Advanced Engineering Informatics, Vol. 24, Issue 4, pp. 428-434, Nov. 2010.
[8] T.T. Santos, C.H. Morimoto, ”Multiple camera people detection and tracking using support integration,” Pattern Recognition Letters, Vol. 32, Issue 1, pp. 47-55, Jan. 2011.
[9] Y. Fu, Y. Guo, Y. Zhu, F. Liu, C. Song, Z. Zhou, ”Multi-View Video Summarization,” IEEE Trans. On Multimedia, Vol. 12, Issue 7, pp. 717-729, Nov. 2010.
[10] F. Daniyal, A. Cavallaro, ”Multi-camera scheduling for video production,” 2011 Conference for Visual Media Production(CVMP), London, UK, pp. 11-20, Nov. 2011.
[11] H. P. Moravec, “Toward Automatic Visual Obstacle Avoidance,” Proc. of 5th International Joint Conference on Artifitial Intelligence, pp. 584, Aug. 1977.
[12] E. Maggio, A. Cavallaro, VIDEO TRACKING Theory and Practice, 1st Ed. A John Wiley and Sons, Ltd., Publication, 2011.
[13] H. Aghajan, A. Cavallaro, Multi-Camera Networks: Principle and Applications, Elsevier/Academic Press, Boston, 2009.
[14] K.S. Kaawaase, F. Chi, J. Shuhong, Q.B. Ji, “A Review on Selected Target Tracking Algorithms,” Information Technology Journal, Vol. 10, Issue 4, pp. 691-702, 2011.
[15] Y. Freund, R.E. Schapire, “A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting,” Journal of Computer and System Sciences, Vol. 55, pp. 119-139, 1997.
[16] P. Viola, M. Jones, “Robust Real-time Object Detection,” ICCV 2nd Internaltion Workshop on Statistical and Computational Theories of Vision: Modeling, Learning, Computing and Sampling, Jul. 2001.
[17] L. Li, W. Huang, I.Y.H. Gu, Q. Tina, “Foreground object detection from videos containing complex background,” ACM Multimedia, Berkeley, CA, USA, pp.2-10, Nov. 2003.
[18] A. Senior, A. Hampapur, Y.L. Tian, L. Brown, S. Pankanti, R. Bolle, ”Appearance Models for Occlusion Handling,” in proceedings of Second International workshop on Performance Evaluation of Tracking and Surveillance systems in conjunction with CVPR’01, Kauai, Hawaii, USA, Dec. 2001.
[19] K. Nummiaro, E.K. Meier, L.V. Gool, “A color base particle filter,” in First International Workshop on Generative-Model-Based Vision, pp. 53-60, 2002.
[20] T.P. Chen, H. Haussecker, A. Bovyrin, R. Belenov, K. Rodyushkin, A. Kuranov, V. Eruhimov, ”Computer Vison Workload Analysis:Case Study of Video Surveillance Systems,” Intel Technology Journal, Vol. 9, Issue 2, pp. 109-118, May 2005.
[21] Mini-basketball rules, Apr. 10, 2005.
Available: http://www.fiba.com/downloads/Rules/2005_mini_bask_rule.pdf
[22] Official Basketball Rules 2012, Jul. 5, 2012 Available:
http://www.fiba.com/downloads/Rules/2012/OfficialBasketballRules2012.pdf
[23] 楊棠鈞, ”結合Adaboost分類器和支援向量機的路標辨識系統之實現,” 國立成功大學工程科學系碩士論文, Jul. 2009.
[24] G. Woods 原著, 繆紹綱譯, 數位影像處理 Digital image processing, 3rd Ed. 普林斯頓國際有限公司, May 2009.