体育视频分析中的运动挖掘方法研究
|
摘要
多媒体数据的日趋激增给数据的快速检索和查询带来了更大的困难。然而,广大用户希望能够快速地从多媒体数据中提取出感兴趣的内容和隐含知识(概念、规则、规律、模式以及关联等),以便利用它们进行快速的检索和查询,同时能提供问题求解层次的决策支持。这种矛盾给多媒体数据挖掘带来了挑战。本文以体育视频中运动目标为研究对象,从理论层次和技术框架等方面进行系统研究,实现从视频的低层运动特征到高层运动语义之间的逐层挖掘,不仅为用户快速查找信息提供支持,而且能为用户解决问题提供辅助决策。
首先研究体育视频运动挖掘的理论和系统框架,其中,理论分为三个层次,分别为:基本特征层,模式、事件层以及知识层。在此基础上对各层次的运动挖掘方法进行了探索研究,重点提出了一种基本运动特征挖掘方法以及一种基于运动的模式、事件和知识(Model, Event and Knowledge, MEK)挖掘的方法。论文主要针对运动目标相对较少,场地相对开阔的隔网对打型比赛,比如网球视频、羽毛球视频等类型的视频进行运动挖掘方法研究。对于基本运动特征挖掘方法,在准确提取运动对象的基础上,运用改进的Camshift算法来解决跟踪丢失问题,为后续轨迹的提取,运动对象位置以及运动方向的获得奠定基础。对于基于运动的MEK挖掘,论文主要以具体的视频实例来进行方法研究。论文的主要研究内容及创新点如下:
提出了体育视频运动挖掘理论层次及相关技术框架。理论层次主要包含三个部分:基本特征层,模式、事件层以及知识层。在此理论层次下,分别对应了相应的技术路线:基本运动特征提取技术,模式、事件挖掘技术等。理论层次中的各层并不是独立的,低层是可以为高层服务的,但它们又都存在各自的特性,能够单独的用于挖掘信息。
研究了一种基本运动特征挖掘方法。在分析体育视频基本运动特征基础上,提出了基本运动特征挖掘的框架,并简要介绍了此框架中各部分功能,以及相应的基本特征提取技术。其中基本特征的提取包括:轨迹的提取,运动对象位置的提取,以及运动方向的提取等。
研究了一种基于运动的模式、事件、知识(MEK)挖掘方法。基于运动的模式挖掘是利用修正轨迹的几何特征来进行分析,获得视频中一些初略的结构化信息以及基于统计的运动习惯性问题等。基于运动的事件探测包含了两个方面的内容:SEIT的构建、基于SEIT的事件匹配。基于运动的知识挖掘是运用传统数据挖掘方法,利用其模式来提取出更高级的知识。
设计并实现了体育视频运动挖掘平台——SVMMP(Sports Video Motion Mining Platform),对体育视频运动挖掘理论层次和技术框架进行了应用和验证。
综上所述,本文提出了体育视频运动挖掘的基本概念、技术框架和理论层次,在经典方法的基础上对部分相关处理技术进行了改进,并通过设计实现体育视频运动挖掘平台,验证了本文的思路。这些研究为视频运动挖掘问题提供了一个新的解决方案,视频运动挖掘技术的不断发展和完善将使其在信息资源的管理和共享以及问题付诸决策等领域发挥越来越大的作用。
Multimedia data is acutely increasing, which makes fast index and query more difficulty. However, most users expect that interested content and hidden knowledge (conception, regulation, rule, mode, relationship, etc.) can be quickly extracted from multimedia data, in order to make index and query faster and to support decision-making for problem solving levels. This conflict makes multimedia data mining meet huge challenge. This paper regards motion objects in sports video as research objects. We study from theory level to technique frame,and achieve step by step mining among video low-level motion feature and high-level motion semantic. This work provides not only fast query, but also decision support for problem.
Firstly, this paper studies sports video motion mining (SVMM) theory level and system frame, in which theory level contains three levels which are basic feature level, model and event level, and knowledge level. We study motion mining methods for each level, and emphatically propose a method of basic motion feature mining (BMFM) and a method of Model, event, and knowledge mining (MEKM). This paper mostly studies racket games which contain few motion objects and open field, such as tennis, badminton, and so on. For BMFM, we use improved Camshift algorithm to solve track-lose problem after accurately extracting motion object, which confirms base for track extraction, motion object position, and motion orientation. The main content and innovations are as follows:
Propose SVMM theory level and corresponding technique frame. The theory level contains three parts: basic feature layer, model and event layer, and knowledge layer. Under the theory level, we propose technique route: basic motion feature extraction technique, model and event detection technique, and so on. Each layer in this theory level is not absolute. Low layer serve for high layer, and each layer has its characteristic, which could absolutely mine information.
Study a BMFM method. After analyzing sports video basic motion feature, we propose BMFM basic frame, and briefly introduce each function in the frame and basic feature extraction technique. Basic feature extraction contains: tract extraction, position extraction, and orientation extraction.
Study a MEKM method.Model mining based on motion uses the geometrical feature of modificatory track (MT) to analyze, and achieve rude structure information and motion habit problem based on statistics. Event detection based on motion contains two aspects: the creation of the SEIT and event matching method based on SEIT. Knowledge mining based on motion extracts more advanced knowledge by traditional data mining method and the result of model mining.
Design and achieve the SVMMP, which apply and validate the SVMM theory level and corresponding technique frame.
To summarize, this paper proposes SVMM basic concept, technique frame and theory level. We improve the classical disposal technique, design the SVMMP, which validate the technique route. These researches provide a new solution for sports motion mining problem. The development on video technique will bring into play important role in the field of information resource management and share and problem decision-making.
首先研究体育视频运动挖掘的理论和系统框架,其中,理论分为三个层次,分别为:基本特征层,模式、事件层以及知识层。在此基础上对各层次的运动挖掘方法进行了探索研究,重点提出了一种基本运动特征挖掘方法以及一种基于运动的模式、事件和知识(Model, Event and Knowledge, MEK)挖掘的方法。论文主要针对运动目标相对较少,场地相对开阔的隔网对打型比赛,比如网球视频、羽毛球视频等类型的视频进行运动挖掘方法研究。对于基本运动特征挖掘方法,在准确提取运动对象的基础上,运用改进的Camshift算法来解决跟踪丢失问题,为后续轨迹的提取,运动对象位置以及运动方向的获得奠定基础。对于基于运动的MEK挖掘,论文主要以具体的视频实例来进行方法研究。论文的主要研究内容及创新点如下:
提出了体育视频运动挖掘理论层次及相关技术框架。理论层次主要包含三个部分:基本特征层,模式、事件层以及知识层。在此理论层次下,分别对应了相应的技术路线:基本运动特征提取技术,模式、事件挖掘技术等。理论层次中的各层并不是独立的,低层是可以为高层服务的,但它们又都存在各自的特性,能够单独的用于挖掘信息。
研究了一种基本运动特征挖掘方法。在分析体育视频基本运动特征基础上,提出了基本运动特征挖掘的框架,并简要介绍了此框架中各部分功能,以及相应的基本特征提取技术。其中基本特征的提取包括:轨迹的提取,运动对象位置的提取,以及运动方向的提取等。
研究了一种基于运动的模式、事件、知识(MEK)挖掘方法。基于运动的模式挖掘是利用修正轨迹的几何特征来进行分析,获得视频中一些初略的结构化信息以及基于统计的运动习惯性问题等。基于运动的事件探测包含了两个方面的内容:SEIT的构建、基于SEIT的事件匹配。基于运动的知识挖掘是运用传统数据挖掘方法,利用其模式来提取出更高级的知识。
设计并实现了体育视频运动挖掘平台——SVMMP(Sports Video Motion Mining Platform),对体育视频运动挖掘理论层次和技术框架进行了应用和验证。
综上所述,本文提出了体育视频运动挖掘的基本概念、技术框架和理论层次,在经典方法的基础上对部分相关处理技术进行了改进,并通过设计实现体育视频运动挖掘平台,验证了本文的思路。这些研究为视频运动挖掘问题提供了一个新的解决方案,视频运动挖掘技术的不断发展和完善将使其在信息资源的管理和共享以及问题付诸决策等领域发挥越来越大的作用。
Multimedia data is acutely increasing, which makes fast index and query more difficulty. However, most users expect that interested content and hidden knowledge (conception, regulation, rule, mode, relationship, etc.) can be quickly extracted from multimedia data, in order to make index and query faster and to support decision-making for problem solving levels. This conflict makes multimedia data mining meet huge challenge. This paper regards motion objects in sports video as research objects. We study from theory level to technique frame,and achieve step by step mining among video low-level motion feature and high-level motion semantic. This work provides not only fast query, but also decision support for problem.
Firstly, this paper studies sports video motion mining (SVMM) theory level and system frame, in which theory level contains three levels which are basic feature level, model and event level, and knowledge level. We study motion mining methods for each level, and emphatically propose a method of basic motion feature mining (BMFM) and a method of Model, event, and knowledge mining (MEKM). This paper mostly studies racket games which contain few motion objects and open field, such as tennis, badminton, and so on. For BMFM, we use improved Camshift algorithm to solve track-lose problem after accurately extracting motion object, which confirms base for track extraction, motion object position, and motion orientation. The main content and innovations are as follows:
Propose SVMM theory level and corresponding technique frame. The theory level contains three parts: basic feature layer, model and event layer, and knowledge layer. Under the theory level, we propose technique route: basic motion feature extraction technique, model and event detection technique, and so on. Each layer in this theory level is not absolute. Low layer serve for high layer, and each layer has its characteristic, which could absolutely mine information.
Study a BMFM method. After analyzing sports video basic motion feature, we propose BMFM basic frame, and briefly introduce each function in the frame and basic feature extraction technique. Basic feature extraction contains: tract extraction, position extraction, and orientation extraction.
Study a MEKM method.Model mining based on motion uses the geometrical feature of modificatory track (MT) to analyze, and achieve rude structure information and motion habit problem based on statistics. Event detection based on motion contains two aspects: the creation of the SEIT and event matching method based on SEIT. Knowledge mining based on motion extracts more advanced knowledge by traditional data mining method and the result of model mining.
Design and achieve the SVMMP, which apply and validate the SVMM theory level and corresponding technique frame.
To summarize, this paper proposes SVMM basic concept, technique frame and theory level. We improve the classical disposal technique, design the SVMMP, which validate the technique route. These researches provide a new solution for sports motion mining problem. The development on video technique will bring into play important role in the field of information resource management and share and problem decision-making.
引文
[1]陈剑赟.基于BSU的视频挖掘:博士学位论文.长沙:国防科学技术大学,2005.
[2]胡军涛.视频高层结构分析和挖掘:硕士学位论文.长沙:国防科学技术大学,2002.
[3]代科学,武德峰,付畅俭,李国辉,李惠佳.视频挖掘技术综述.中国图像图形学报,2006,11(4):451~457.
[4]谢毓湘,栾悉道,吴玲达,老松杨.视频摘要技术研究.计算机应用研究.2004,21(11):1~4.
[5]辛国娟.视频数据挖掘的方法研究及应用:硕士学位论文.上海:上海交通大学,2007.
[6] Zaiane O R.Resource and knowledge disvovery from the Internet and Multimedia repositories:博士学位论文.Canada:Simon Fraser University,1999.
[7] Kitamoto Asanobu.Data mining for Typhoon image collection.MDM/KDD 2001,San Francisco,CA,USA,2001.
[8] Wijesekera D,Barbara D.Mining cinematic knowledge:An Extended Abstract.MDM/KDD 2000,Boston,MA,USA,2000.
[9] Lin T,Zhang H J.Automatie Video Seene Extraetion by Shot Grouping.International Conferenee on Pattern Reeognition,Bareelona,SPain,Sept.2~8,2000.
[10] Chong-wah Ngo,Ting Chuen,Hong jiang,Zhang Pong.On Clustering and Retrieval of video Shots.ACM Multimedia 2001,ottawa,Canada,September 30~October 5,2001.
[11]胡军涛,武德峰,李国辉,甘亚莉.多媒体数据挖掘的体系结构和方法.计算机工程.2003, 29(9):149~151.
[12] KimShearer,ChitraDorai,SvethaVenkatesh.Ineorporating Domain Knowledge with Video and Voice Data Analysis in News Broadeasts.MDM/KDD 2000.
[13]马力,张茂军,胡小佳,刘少华.视频对象分类特征评价与选择方法.小型微型计算机系统.2009,30(10):2062~2068.
[14]付畅俭,李国辉,代科学.视频结构挖掘的概念及应用.计算机应用研究.2007,2:20~23.
[15]代科学,付畅俭,武德峰,李国辉.视频挖掘:概念、技术与应用.计算机应用研究.2006,1:1~4.
[16]罗李.基于数据挖掘的镜头分类技术研究:硕士学位论文.重庆:重庆邮电大学,2007.
[17]贺玲,吴玲达,蔡益朝.数据挖掘中的聚类算法综述.计算机应用研究.2007,1:10~13.
[18]焦波,李国辉,涂丹,汪彦明.一种视频序列中二值图像的快速聚类算法.湖南大学学报(自然科学版).2008,35(8):73~77.
[19] Zhu X,Wu X,A.Elmagarmid,etc.Video data mining:Semantic indexing and event detection from the association perspective.IEEE Trans.Knowledge and dataengineering.2005,17(5):665~677.
[20] Zhu X O.Mining video associations for efficient database management.Multimedia Systems,2003,9(6):31~53.
[21] Zhu X Q.Sequential association mining for video summarization.In:IEEE International Conference on Multimedia & Expo,Baltimore,MD,USA,2003,3:333~336.
[22]老松杨,白亮,刘海涛等.基于感知概念和有限状态机的体育视频语义内容分析模型.小型微型计算机系统.2009,30(6):1137~1141.
[23] Chen S C,Shyu M L,Zhang C C,etc.Multimedia data mining for traffic video sequences.San Francisco:ACM SIGKDD Conference,2001.
[24] Luo M,Ma Y F,Zhang H J.Pyramid wise structuring for soccer highlight extraction.IEEE Pacific Rim Conf.Multimedia,Singapore,2003.
[25] Ekin A,Tekalp A M.Robust dominant color detection and color based application for sports video.IEEE Int’l Conf.Image Processing,Barcelona,Spain,2003.
[26] Yu X G,Tian Q,Wan K W.A Novel Ball Detection Framework for Real Soccer Video.IEEE International Conference on Multimedia and Expo,2003:2~265.
[27] Jiang S Q,Ye Q X, Gao W.A new method to segment playfield and its applications in match analysis in sports video.ACM Mult imedia,New York,NY,2004.
[28] Wang L,Zeng B Y,Lin S,et al.Automatic extraction of semantic colors in sports video.IEEE Int’l Conf.Acoustics,Speech,and Signal Processing,Quebec,Canada,2004.
[29]刘扬,黄庆明,高文,叶齐祥.自适应高斯混合模型球场检测算法及其在体育视频分析中的应用.计算机研究与发展.2006,43(7) : 1207~ 1215.
[30]张龙飞,曹元大,张明杰等.体育视频中的对象检测与分割.计算机工程.2006,32(3):229~231.
[31]阎刚,崔国栋,于明.基于高斯混合模型的背景建模球员检测算法.计算机仿真.2010,27(9):258~262.
[32]张龙飞,曹元大,周艺华等.基于支持向量机元分类器的体育视频分类.北京理工大学学报.2006,26(1):41-44.
[33] Liu J,Tong X F,Li W L,ect.Automatic Player Detection.Labeling and Tracking in Broadcast Soccer Video.
[34]吴玥,田兴彦.基于单目视频运动跟踪的三维人体动画.微计算机应用.2008,29( 5):42~47.
[35]高书秀,黄剑华,唐降龙,等.基于多区域的人体运动跟踪研究与应用.微计算机信息.2008,24 (19):267~269.
[36]马波,张田文.一种新颖的轮廓线跟踪算法.信号处理.2004,20(2) .
[37] Numm Iaro K,Koller-M Eier E,Van Gool L.A adaptive color based particle filter.Image and Vision Computing.2003,21(1): 99~110.
[38]朱胜利.MeanShift及相关算法在视频跟踪中的研究:博士学位论文.杭州:浙江大学,2006.
[39]李金,胡文广.基于颜色的快速人体跟踪及遮挡处理.智能系统学报.2010,5(4):353~359.
[40]陈家树,程小平.基于SSD的核粒子滤波器下的球类运动员跟踪.西南师范大学学报(自然科学版).2008,33(2):110~113.
[41] G Sudhir,J C M Lee,and A K Jain."Automatic classification of tennis video for high-level content-based retrieval" in Proc.Int.Workshop on Content-Based Access of Image and Video Databases,Bombay,India,1998:81~90.
[42] G S Pingali,Y Jean,and I Carlbom."Real time tracking for enhanced tennis broadcasts," in Proc.Conf.Computer Vision and Pattern Recognition,Santa Barbara,CA,1998:260~265.
[43]张良.一种视频关键语义对象的检测方法.北京信息科技大学学报.2010,25(2):75~78.
[44] P Perez,C Hue,J Vermaak,M Gangnet."Color-based probabilistic tracking" ECCV.2002, 01(i):661~675.
[45] Y Tan,D Saur,S Kulkami,P Ramadge."Rapid estimation of camera motion from compressed video with application to video annotation" In IEEE Tram.On Circuits and SyStems for Video Technology,Vohane ,2000,10(i) :133~146.
[46] Y Gong,T S LiIIl,H C Chua."Automatic parsing of TV soccer programs" in Proc.IEEE Int.Conf.on Multimedia Computing and Systems,May,1995.
[47] L Xie,S Chang,A Divakaran,H Sun."Structure analysis of soccer video with hidden Markov models" in Proc.tat.Conf.Acoustic,Speech and Signal Processing.Orlando,FL,2002.
[48] N Babaguchi,Y Kawai,T Kitahashi."Event based indexing of broadcasted sports video by intermodal collaboration",In IEEE Trans.OH Multimedia,2002, 4(1):68~75.
[49] D Zhang,R Raj,S Chang."General and domain specific techniques for detecting and recognizing superimposed text in video",In Proc of IEEE International Conference on Image Processing, 2002, l:593~596.
[50] Y Rui,A Gupta,and A Acero."Automatically extracting hiflaiflats for TV baseball programs" ,In Proe.ACM hat.Conf.Multimedia,ACM Press,2000:105-115.
[51] M Xu,L Duan,C Xu,Q Tiall."A fusion sheme of visual and auditory modalities for event detection in sports video",in Proc.Int.Conf.Acoustic,Speech and Sigllal Processing,2003:189~192.
[52] Z Xiong,R Radhakrishnan,A Divakaran,T Huang."Audio events detection based highlights extraction from baseball , golf and soccer games in a unified framework",in Proc.IEEE Int.Conf.Acoustics,Speech,And Signal Processing,2003,5:632~635.
[53] S Dagtas,M Mottaleb."Extraction of TV highlights using multimedia features",In Proe.of IEEE Fourth Workshop on Multimedia signal Processing,2001:91~96.
[54] Y Gong,M Han,W Hua,W Xu."Maximum entropy model based baseball highlight detection and classification",Computer Vision and Image Understanding.2004, 96 (2):181~199.
[55] R Leonardi,P Migliorati,M Prandini."Semantic indexing of soccer audio-visual sequences:a multimodal approach based on controlled Markov chains",IEEE Trans.Circuits Syst.Video Tech,2004,14(5):634~643.
[56] M Pctkovic,V Mihajlovic,W Jonker,S Kajan."Multi Modal extraction ofhighlights from TV formula programs",In Proe of IEEE Int.Conf.on Multimedia and Expo, 2002,I:817~820.
[57] Ebadollahi Shahram,Xie Lexing,Chang Shih—Fu,Smith John R.Visual event detection using multi-dimensional concept dynamics.Proceedings of the International Conference on Multimedia & Expo.Toronto,Ontario,Canada, 2006:881-884.
[58] Dong Xu,Change Shih—Fu.Video event recognition using kernel methods with multilevel temporal alignment.IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(11):1985~1997.
[59]焦李成,周伟达,张莉等著.智能目标识别与分类.北京:科学出版社,2010.
[60]边肇祺,阎平凡,杨存荣.模式识别.北京:清华大学出版社,1998.
[61]王鹏,蔡锐,李彬,杨士强.网球视频分析的运动向量场变换算法.电子学报.2005,33(5):935~938.
[62] A Yilmaz,O Javed and M Shah.Object tracking:A survey.ACM association for computing machinery,2006,38(4):1~45.
[63]施伯乐,丁宝康,汪卫.数据库系统教程.北京:高等教育出版社,2003.
[64]王晓峰,张大鹏,王绯,史忠植.基于语义轨迹的视频事件探测.计算机学报.2010,33(10):1845~1858.
[65]孟祥茹.视频镜头边界检测的研究:硕士学位论文.沈阳:沈阳工业大学,2007.
[66] Duan L,Xu M,Chua T,T ian Q,Xu C.A mid-level representation framework for semantic sports video analysis.Proceedings of the ACM Mul timedia. Berkeley,CA,USA,2003:33~44.
[2]胡军涛.视频高层结构分析和挖掘:硕士学位论文.长沙:国防科学技术大学,2002.
[3]代科学,武德峰,付畅俭,李国辉,李惠佳.视频挖掘技术综述.中国图像图形学报,2006,11(4):451~457.
[4]谢毓湘,栾悉道,吴玲达,老松杨.视频摘要技术研究.计算机应用研究.2004,21(11):1~4.
[5]辛国娟.视频数据挖掘的方法研究及应用:硕士学位论文.上海:上海交通大学,2007.
[6] Zaiane O R.Resource and knowledge disvovery from the Internet and Multimedia repositories:博士学位论文.Canada:Simon Fraser University,1999.
[7] Kitamoto Asanobu.Data mining for Typhoon image collection.MDM/KDD 2001,San Francisco,CA,USA,2001.
[8] Wijesekera D,Barbara D.Mining cinematic knowledge:An Extended Abstract.MDM/KDD 2000,Boston,MA,USA,2000.
[9] Lin T,Zhang H J.Automatie Video Seene Extraetion by Shot Grouping.International Conferenee on Pattern Reeognition,Bareelona,SPain,Sept.2~8,2000.
[10] Chong-wah Ngo,Ting Chuen,Hong jiang,Zhang Pong.On Clustering and Retrieval of video Shots.ACM Multimedia 2001,ottawa,Canada,September 30~October 5,2001.
[11]胡军涛,武德峰,李国辉,甘亚莉.多媒体数据挖掘的体系结构和方法.计算机工程.2003, 29(9):149~151.
[12] KimShearer,ChitraDorai,SvethaVenkatesh.Ineorporating Domain Knowledge with Video and Voice Data Analysis in News Broadeasts.MDM/KDD 2000.
[13]马力,张茂军,胡小佳,刘少华.视频对象分类特征评价与选择方法.小型微型计算机系统.2009,30(10):2062~2068.
[14]付畅俭,李国辉,代科学.视频结构挖掘的概念及应用.计算机应用研究.2007,2:20~23.
[15]代科学,付畅俭,武德峰,李国辉.视频挖掘:概念、技术与应用.计算机应用研究.2006,1:1~4.
[16]罗李.基于数据挖掘的镜头分类技术研究:硕士学位论文.重庆:重庆邮电大学,2007.
[17]贺玲,吴玲达,蔡益朝.数据挖掘中的聚类算法综述.计算机应用研究.2007,1:10~13.
[18]焦波,李国辉,涂丹,汪彦明.一种视频序列中二值图像的快速聚类算法.湖南大学学报(自然科学版).2008,35(8):73~77.
[19] Zhu X,Wu X,A.Elmagarmid,etc.Video data mining:Semantic indexing and event detection from the association perspective.IEEE Trans.Knowledge and dataengineering.2005,17(5):665~677.
[20] Zhu X O.Mining video associations for efficient database management.Multimedia Systems,2003,9(6):31~53.
[21] Zhu X Q.Sequential association mining for video summarization.In:IEEE International Conference on Multimedia & Expo,Baltimore,MD,USA,2003,3:333~336.
[22]老松杨,白亮,刘海涛等.基于感知概念和有限状态机的体育视频语义内容分析模型.小型微型计算机系统.2009,30(6):1137~1141.
[23] Chen S C,Shyu M L,Zhang C C,etc.Multimedia data mining for traffic video sequences.San Francisco:ACM SIGKDD Conference,2001.
[24] Luo M,Ma Y F,Zhang H J.Pyramid wise structuring for soccer highlight extraction.IEEE Pacific Rim Conf.Multimedia,Singapore,2003.
[25] Ekin A,Tekalp A M.Robust dominant color detection and color based application for sports video.IEEE Int’l Conf.Image Processing,Barcelona,Spain,2003.
[26] Yu X G,Tian Q,Wan K W.A Novel Ball Detection Framework for Real Soccer Video.IEEE International Conference on Multimedia and Expo,2003:2~265.
[27] Jiang S Q,Ye Q X, Gao W.A new method to segment playfield and its applications in match analysis in sports video.ACM Mult imedia,New York,NY,2004.
[28] Wang L,Zeng B Y,Lin S,et al.Automatic extraction of semantic colors in sports video.IEEE Int’l Conf.Acoustics,Speech,and Signal Processing,Quebec,Canada,2004.
[29]刘扬,黄庆明,高文,叶齐祥.自适应高斯混合模型球场检测算法及其在体育视频分析中的应用.计算机研究与发展.2006,43(7) : 1207~ 1215.
[30]张龙飞,曹元大,张明杰等.体育视频中的对象检测与分割.计算机工程.2006,32(3):229~231.
[31]阎刚,崔国栋,于明.基于高斯混合模型的背景建模球员检测算法.计算机仿真.2010,27(9):258~262.
[32]张龙飞,曹元大,周艺华等.基于支持向量机元分类器的体育视频分类.北京理工大学学报.2006,26(1):41-44.
[33] Liu J,Tong X F,Li W L,ect.Automatic Player Detection.Labeling and Tracking in Broadcast Soccer Video.
[34]吴玥,田兴彦.基于单目视频运动跟踪的三维人体动画.微计算机应用.2008,29( 5):42~47.
[35]高书秀,黄剑华,唐降龙,等.基于多区域的人体运动跟踪研究与应用.微计算机信息.2008,24 (19):267~269.
[36]马波,张田文.一种新颖的轮廓线跟踪算法.信号处理.2004,20(2) .
[37] Numm Iaro K,Koller-M Eier E,Van Gool L.A adaptive color based particle filter.Image and Vision Computing.2003,21(1): 99~110.
[38]朱胜利.MeanShift及相关算法在视频跟踪中的研究:博士学位论文.杭州:浙江大学,2006.
[39]李金,胡文广.基于颜色的快速人体跟踪及遮挡处理.智能系统学报.2010,5(4):353~359.
[40]陈家树,程小平.基于SSD的核粒子滤波器下的球类运动员跟踪.西南师范大学学报(自然科学版).2008,33(2):110~113.
[41] G Sudhir,J C M Lee,and A K Jain."Automatic classification of tennis video for high-level content-based retrieval" in Proc.Int.Workshop on Content-Based Access of Image and Video Databases,Bombay,India,1998:81~90.
[42] G S Pingali,Y Jean,and I Carlbom."Real time tracking for enhanced tennis broadcasts," in Proc.Conf.Computer Vision and Pattern Recognition,Santa Barbara,CA,1998:260~265.
[43]张良.一种视频关键语义对象的检测方法.北京信息科技大学学报.2010,25(2):75~78.
[44] P Perez,C Hue,J Vermaak,M Gangnet."Color-based probabilistic tracking" ECCV.2002, 01(i):661~675.
[45] Y Tan,D Saur,S Kulkami,P Ramadge."Rapid estimation of camera motion from compressed video with application to video annotation" In IEEE Tram.On Circuits and SyStems for Video Technology,Vohane ,2000,10(i) :133~146.
[46] Y Gong,T S LiIIl,H C Chua."Automatic parsing of TV soccer programs" in Proc.IEEE Int.Conf.on Multimedia Computing and Systems,May,1995.
[47] L Xie,S Chang,A Divakaran,H Sun."Structure analysis of soccer video with hidden Markov models" in Proc.tat.Conf.Acoustic,Speech and Signal Processing.Orlando,FL,2002.
[48] N Babaguchi,Y Kawai,T Kitahashi."Event based indexing of broadcasted sports video by intermodal collaboration",In IEEE Trans.OH Multimedia,2002, 4(1):68~75.
[49] D Zhang,R Raj,S Chang."General and domain specific techniques for detecting and recognizing superimposed text in video",In Proc of IEEE International Conference on Image Processing, 2002, l:593~596.
[50] Y Rui,A Gupta,and A Acero."Automatically extracting hiflaiflats for TV baseball programs" ,In Proe.ACM hat.Conf.Multimedia,ACM Press,2000:105-115.
[51] M Xu,L Duan,C Xu,Q Tiall."A fusion sheme of visual and auditory modalities for event detection in sports video",in Proc.Int.Conf.Acoustic,Speech and Sigllal Processing,2003:189~192.
[52] Z Xiong,R Radhakrishnan,A Divakaran,T Huang."Audio events detection based highlights extraction from baseball , golf and soccer games in a unified framework",in Proc.IEEE Int.Conf.Acoustics,Speech,And Signal Processing,2003,5:632~635.
[53] S Dagtas,M Mottaleb."Extraction of TV highlights using multimedia features",In Proe.of IEEE Fourth Workshop on Multimedia signal Processing,2001:91~96.
[54] Y Gong,M Han,W Hua,W Xu."Maximum entropy model based baseball highlight detection and classification",Computer Vision and Image Understanding.2004, 96 (2):181~199.
[55] R Leonardi,P Migliorati,M Prandini."Semantic indexing of soccer audio-visual sequences:a multimodal approach based on controlled Markov chains",IEEE Trans.Circuits Syst.Video Tech,2004,14(5):634~643.
[56] M Pctkovic,V Mihajlovic,W Jonker,S Kajan."Multi Modal extraction ofhighlights from TV formula programs",In Proe of IEEE Int.Conf.on Multimedia and Expo, 2002,I:817~820.
[57] Ebadollahi Shahram,Xie Lexing,Chang Shih—Fu,Smith John R.Visual event detection using multi-dimensional concept dynamics.Proceedings of the International Conference on Multimedia & Expo.Toronto,Ontario,Canada, 2006:881-884.
[58] Dong Xu,Change Shih—Fu.Video event recognition using kernel methods with multilevel temporal alignment.IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(11):1985~1997.
[59]焦李成,周伟达,张莉等著.智能目标识别与分类.北京:科学出版社,2010.
[60]边肇祺,阎平凡,杨存荣.模式识别.北京:清华大学出版社,1998.
[61]王鹏,蔡锐,李彬,杨士强.网球视频分析的运动向量场变换算法.电子学报.2005,33(5):935~938.
[62] A Yilmaz,O Javed and M Shah.Object tracking:A survey.ACM association for computing machinery,2006,38(4):1~45.
[63]施伯乐,丁宝康,汪卫.数据库系统教程.北京:高等教育出版社,2003.
[64]王晓峰,张大鹏,王绯,史忠植.基于语义轨迹的视频事件探测.计算机学报.2010,33(10):1845~1858.
[65]孟祥茹.视频镜头边界检测的研究:硕士学位论文.沈阳:沈阳工业大学,2007.
[66] Duan L,Xu M,Chua T,T ian Q,Xu C.A mid-level representation framework for semantic sports video analysis.Proceedings of the ACM Mul timedia. Berkeley,CA,USA,2003:33~44.