【作者】 刘勋； 孟骧龙； 毋立芳；

【Author】 Liu Xun Meng Xianglong Wu Lifang (School of Electronic Information & Control Engineering,Beijing University of Technology,Beijing 100022)

【机构】 北京工业大学电子信息与控制工程学院；

【摘要】 篮球分割和运动跟踪是篮球比赛视频分析的关键技术。本文提出一种基于自适应对象模型的篮球运动跟踪方法。首先利用已分割篮球的基本特征建立对象模板,进一步建立自适应对象模型,然后利用对象模型改进块匹配法跟踪篮球,最后修正跟踪偏差,并建立篮球跟踪有效性检测机制。实验表明本文算法可以有效跟踪篮球对象。更多还原

【Abstract】 Segmentation and tracking of basketball is key issues of basketball video analysis.In this paper,an algorithm based adaptive object model for basketball tracking is put forward.Firstly,object template and adaptive object model are established using the basic characteristic of segmented basketball.Then,block-matching algorithm is improved to track the basketball.Finally,the edge of basketball is revised and tracking validity is validated.Experimental results confirm that our idea is feasible.更多还原