基於電視擷取影像之依打者識別好球帶位置之研究｜國立臺灣科技大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	李崇甫 Chung-Fu Lee
論文名稱：	基於電視擷取影像之依打者識別好球帶位置之研究 Study on Batter-dependent Strike Zone Identification from TV broadcast Images
指導教授：	蘇順豐 Shun-Feng Su
口試委員:	王偉彥 none 莊鎮嘉 none 許孟超 none
學位類別：	碩士 Master
系所名稱：	電資學院 - 電機工程系 Department of Electrical Engineering
論文出版年：	2008
畢業學年度：	97
語文別：	英文
論文頁數：	91
中文關鍵詞：	線性預測、追蹤技術
外文關鍵詞：	linear prediction, tracking
相關次數：	點閱：345 下載：2
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

上一筆

一場精彩的棒球比賽，除了球員精湛的技術之外，比賽的公平性也是相當重要的一環。往往在比賽中，一個錯誤的判決，將會對整場比賽，甚至整個系列戰造成很大的影響，其中主審裁判在好壞球的判定上，更是佔了比賽中很關鍵的一部分。本論文將提出一種方式，重建棒球的軌跡與不同打者的好球帶範圍，進而判斷投手是否有將球投入好球帶的範圍。在本篇論文當中，我們僅使用電視轉播畫面，並假設攝影機角度不變之情形下，利用移動物體追蹤的技術，如此可以排除一些不必要考慮的像素，節省運算時間，之後再利用路徑預測技術來解決接近本壘板時可能產生的雜訊與誤差的問題，以達到重建棒球軌跡與進壘位置的目的，並使用傳統影像處理所用的色彩學與形態學，來進行打者好球帶的建構，進而判斷出好壞球。本系統的建立有助於訓練裁判在好壞球過於主觀的缺失，並且在重大比賽中輔助裁判做出關鍵性的判決。

Besides players’ wonderful acting, fair judgments for all plays in the field is also a very important essential for a good baseball game. The judgment of a strike or a ball made by the head umpire sometimes may have a great influence on deciding who is the winner or the loser in a game or even a series. In this study, we intend to investigate ways of rebuilding the ball trajectories from television images to decide whether the trajectory passes the strike zone or not. It should be noted that the strike zone will depend upon the batters in the field. Thus, algorithms of detecting whether the pitched ball is a strike are proposed and discussed. In our study, the broadcast television images are used to rebuild the baseball trajectories and to find the position when the baseball passes over the home plate. It is noted that in this study we assume the shooting angle of the camera is fixed but unknown. By considering tracking effects, some unnecessary pixels are abandoned in our algorithm to avoid unnecessary pattern matching and to speed up the recognition process. In our algorithms, noise problems caused by various objects around the home plate are resolved by using linear prediction when the ball approaches to the home plate. Afterward, the strike zone for the current batter is built by using chromatological and morphological techniques. Various experiments are conducted to demonstrate the effectiveness of the proposed algorithms. This system can be employed to help training umpires to avoid making mistakes and can be used to resolve debates in some key situations.

Contents
摘要	I
Abstract	II
誌謝	III
Contents	V
List of Figures	VII
List of tables	IX
Chapter 1	1
Introduction	1
1.1 Research motivation	1
1.2 Research objective	2
1.3 Research scope and limit	2
1.4 Thesis Organization	2
Chapter 2	3
The Foundation of Theories	3
2.1 Color Space Model	3
2.1.1 RGB Color Space Model	3
2.1.2 Normalize RGB Color Space Model	4
2.1.3 YUV Color Space Model	5
2.1.4 HSI Color Space Model	6
2.1.5 YCbCr Color Space Model	7
2.2 Digital Images Process	8
2.2.1 Gray Image	8
2.2.2 Binary Image	9
2.2.3 Histogram and Histogram equalization	10
2.2.4 Difference	12
2.3 The Structure of System	13
Chapter 3	15
Baseball Detection System	15
3.1 Image Pre-processing	15
3.1.1 Image Subtraction	16
3.1.2 Image Binarization	21
3.1.3 Morphological Technique	22
3.1.4 Region Selection	26
3.2 Component Labeling	27
3.3 Searching Rectangle	32
3.4 Linear Prediction	36
3.5 Z-axis Orientation	39
Charter 4	42
Strike Zone Establishment	42
4.1 Area Selection	43
4.2 Face Detection	44
4.2.1 Skin Color Detection	45
4.2.2 Morphological Technique	47
4.2.3 Size Filter and Ratio Processing	48
4.2.4 Adding Clothes Color Selection	51
4.3 Knees Detection	51
4.4 Strike Zone Establishment	54
4.5 Expand Strike Zone and Judgment	56
Chapter 5	59
Simulation Results	59
5.1 Simulation of Baseball Detection and Tracking	59
5.2 Face and Knees Detection	66
5.3 Successful Results Display	70
Chapter 6	84
Conclusions and Future Work	84
6.1 Conclusions	84
Reference	87
















List of Figures
Figure 2.1 RGB color space model	4
Figure 2.3 HSI color space model	7
Figure 2.4 YCbCr color space model	8
Figure 3.5 The reults of various thresholds.	22
Figure 3.6 The dilation result	24
Figure 3.7 The erosion result	24
Figure 3.8 An illustration of the opening and closing operations	25
Figure 3.9 The result of opening operation	26
Figure 3.10 Region of selection	27
Figure 3.11 Selected region	27
Figure 3.12 Flow chart of component labeling for clustering.	29
Figure 3.13 The result of component labeling	29
Figure 3.14 Component labeling in real example.	30
Figure 3.15 The process of detecting baseball	32
Figure 3.16 The tracking method	33
Figure 3.17 The results of tracking operation.	34
Figure 3.18 The processing of the system	35
Figure 3.19 The result of baseball detection	36
Figure 3.20 The chart representing the method of least square	38
Figure 3.21 Linear prediction	38
Figure 3.22 The chart of Z-axis locating	41
Figure 5.1 Successful baseball detecting and tracking example 1	61
Figure 5.2 Successful baseball detecting and tracking example 2	62
Figure 5.3 Successful baseball detecting and tracking example 3	63
Figure 5.4 Fail detection caused by white background example 1	64
Figure 5.5 Fail detection caused by too small angle example 2	65
Figure 5.6 The result of face detection example 1	66
Figure 5.7 The result of face detection example 2	67
Figure 5.8 The result of face detection example 3	67
Figure 5.9 The fail result of face detection	68
Figure 5.10 The knees detection step by step example 1	69
Figure 5.11 The knees detection step by step example 2	69
Figure 5.12 The same judgment made by a umpire and the system example 1	70
Figure 5.13 Different judgments made by a umpire and the system example 1	71
Figure 5.14 The same judgment made by a umpire and the system example 2	72
Figure 5.15 Different judgments made by a umpire and the system example 2	73
Figure 5.16 The same judgment made by a umpire and the system example 3	74
Figure 5.17 Different judgments made by a umpire and the system example 3	75
Figure 5.18 Different judgments made by a umpire and the system example 4	76
Figure 5.19 Different judgments made by a umpire and the system example 5	77
Figure 5.20 The same judgment made by a umpire and the system example 4	78
Figure 5.21 The same judgment made by a umpire and the system example 5	79
Figure 5.22 The same judgment made by a umpire and the system example 6	80
Figure 5.23 The same judgment made by a umpire and the system example 7	81
Figure 5.24 The same judgment made by a umpire and the system example 8	82
Figure 5.25 The same judgments made by a umpire and the system example 9	83
Figure 5.26 The selection by users ……………...………………………...84











List of tables
Table 4.1 The average pixels of a baseball	58
Table 5.1 Elapsed time for experiments conducted in the simulation	83

                                

[1] A. Gueziec, et al., "Tracking Pitches for Broadcast Television," Proc. of IEEE Int'l Conference on Computer, Vol. 35, No. 3, pp.38-43, Mar. 2002
[2] G. Pingali, A. Opalach, and Y. Jean, “Ball Tracking and Virtual Replays for Innovative Tennis Broadcasts,” IEEE Trans. Pattern Recognition, Proceedings. 15th International Conference on Vol. 4, pp.4152, 2000
[3] “Hawk-Eye, ” Wikipedia, http://en.wikipedia.org/wiki/Hawk_Eye
[4] R. C. Gonzalez and R. E. Wood, Digital Image Processing, 2Ed. New Jersey: Prentice Hall, 2002
[5] 何耀宗, 色彩基礎, 三版, 台北巿, 東大, 民國76年
[6] D. Matthys, “Digital Image Processing,” http://academic.mu.edu/phys/matthysd/web226/L0221.htm
[7] G. Sharms and H. J. Trussell, “Digital Color Imaging,” IEEE Trans. Image Processing, Vol. 6, pp. 901–932, July 1997
[8] N. Otsu, “A Threshold Selection Method from Gray-level Histograms,” IEEE Trans. Systems, Man, and Cybernetics, Vol. 9, No. 1, pp. 62-66, 1979
[9] I. D. Trier and T. Taxt, “Evaluation of Binarization Methods for Document Images,” IEEE Trans. On Pattern Analysis and Machine Intelligence, Vol. 17, pp. 312-315, March 1995
[10] “Strike Zone,” Wikipedia, http://en.wikipedia.org/wiki/Strike_zone
[11] “Batter Box,” Wikipedia, http://en.wikipedia.org/wiki/Batter%27s_box
[12] M. Soriano, S. Huovinen, B. Martinkauppi, and M. Laaksonen, “Using the Skin Locus to Cope with Changing Illumination Conditions in Color-Based Face Tracking,” Proc. of IEEE Nordic Signal Processing Symposium, pp. 383-386, 2000
[13] Y. Arake, N. Shimada, and Y. Shirai, “Detection of Faces of Various Directions in Complex Backgrounds,” Proceedings of the 16th IEEE International Conference on Pattern Recognition, Vol. 1, pp 409-412, 2002
[14] D. S. Bloomberg, “Multiresolution Morphological Analysis of Document Images,” Proc. SPIE Visual Communication Image Processing, Vol. 1818, pp. 648–662, 1992
[15] K. K. Chin and J. Saniie, “Morphological Processing for Feature Extraction,” Proc. SPIE, vol. 2030, pp. 288–302, 1993
[16] Y. M. Y. Hasan and L. J. Karam, “Morphological Reversible Contour Eepresentation,” IEEE Trans. Pattern Anal. Machine Intelligence, Vol. 22, pp. 227–240, Mar. 2000
[17] 黃敏峰, The Study of Monitoring and Control System Using Human Face Tracking Method, 國立成功大學電機工程學系碩士論文, 2003
[18] 王貞元, Fire Detection and Region Location in Surveillance Image, 國立台灣科技大學電機工程學系碩士論文, 2008
[19] K. Agrawala and A. V. Kulkarni, “A Sequential Approach to the Extraction of Shape Features,” Comput. Graphics Image Processing, Vol. 6, pp. 538-5.57, 1977
[20] 羅彥雄, 視訊棒球偵測與軌跡重建, 國立台灣科技大學電子工程學系碩士論文, 2006
[21] 馬維章, 以多部攝影機實現棒球偵測與追蹤系統, 國立台灣科技大學電子工程學系碩士論文, 2008
[22] “Batter, Wikipedia,” http://en.wikipedia.org/wiki/Batter_(baseball)
[23] 林丕靜, 數值分析, 四版, 台北市, 儒林, 2005
[24] R. G. Kuehni, Color Space And Its Divisions: Color Order From Antiquity To The Present, New Jersey, Hoboken, 2003
[25] K. Sobottka and I. Pitas, “Extraction of Facial Regions and Features Using Color and Shape Information,” Proc.13th International Conference on Pattern Recognition, pp.421-425, 1996
[26] C. C. Lien, and C. L. Chiang, "The Scene-Based Event Detection for Baseball Sport Video," IPPR Conf. On CVGIP, Vol. 18, pp. 1-14, 2005
[27] W. N. Lie and W. C. Hsiao, "Content-Based Video Retrieval Based on Object Motion Trajectory," IEEE Workshop on Multimedia Signal Processing, pp.237-240, Dec. 2002
[28] B. Satterwhite and O. Marques, “Automatic Detection of TV Commercials,” IEEE Potentials, Vol. 23, pp. 1-12, 2004
[29] Z. Liu and Y. Wang, “Face Detection and Tracking in Video Using Dynamic Programming, “IEEE ICIP, Vol. 1, pp. 53-56, 2000
[30] H. Pan, B. Li, and M. I. Sezan, “Automatic Detection of Replay Segments in Broadcast Sports Programs by Detection of Logos in Scene Transitions,” Digital Object Identifier 10.1109/ICASSP Vol. 4, pp. IV3385- IV3388, 2002
[31] O. Trier, A. Jain, and T. Taxt, “Feature Extraction Methods for Character Recognition – A Survey,” Pattern Recognition, Vol. 29, pp. 641-662, 1968
[32] H. Lu and Y.-P. Tan, “Sports Video Analysis and Structuring,” IEEE Fourth Workshop on Multimedia Signal Processing, pp. 45-50, 2001
[33] X.-F. Tong , H.-Q. Lu , Q.-S. Liu, “An Effective and Fast Soccer Ball Detection and Tracking Method, Proceedings of the Pattern Recognition,” 17th International Conference on (ICPR'04), Vol. 4, p.795-798, August 23-26, 2004
[34] 黃泰祥, 具備人臉追蹤與辨識功能的一個智慧型數位監視系統, 私立中原大學電子工程學系碩士學位論文, 2004
[35] ISL eye gaze tracking. http://www.is.cs.cmu.edu/mie/eyegaze.html
[36] Videoland, Sports Channel, http://sport.videoland.com.tw/
[37] ESPN, Sports Channel, http://www.espnstar.com.tw/
[38] Chinese Taipei Baseball Association, http://www.ctba.org.tw/
[39] Major League Baseball, http://mlb.mlb.com/
[40] Chinese Professional Baseball League, http://www.cpbl.com.tw/
[41] Baseball Rules, http://www.baseball-almanac.com/rulemenu.shtml
[42] 棒球規則，http://online.cpbl.com.tw/rule/cpblrules.htm
[43] “Least Square,” Wikipedia, http://en.wikipedia.org/wiki/Least_squares
[44] “YUV, ” Wikipedia, http://en.wikipedia.org/wiki/YUV
[45] 鄭柏左, 色彩理論與數位影像, 臺北縣中和市, 新文京開發, 民93年
[46] 石井藤吉郎, 棒球教室, 台北巿, 聯廣, 民82年
[47] 球類叢書編輯委員會, 棒球入門, 台北巿, 國家, 民69年
[48] 袁定文, 棒球觀點, 台北巿, 聯經, 民84年

簡易檢索 / 詳目顯示

相關論文