篮球手臂投篮视频图像跟踪提取仿真

发布时间：2018-01-24 15:56

本文关键词： 投篮动作视频图像轮廓跟踪　出处：《计算机仿真》2017年11期 　论文类型：期刊论文

【摘要】：对篮球手臂投篮视频图像的跟踪提取,能够更好的掌握好投篮技术。对投篮视频图像的跟踪提取,需要获取动作的差异性误差向量,得到手臂动作像素信息特征仿射变换,完成视频图像轮廓跟踪提取。传统方法采用亮点模型衍射的方法实现视觉穿透,给出投篮动作手臂技术特征,但忽略了动作像素信息特征仿射的变换,完成投篮动作视频图像的跟踪提取。提出基于camshift的篮球投篮动作手臂视频图像轮廓跟踪提取方法。对原始视频图像进行滤波处理,利用最大方差比阈值法对投篮的手臂运动区域进行划分,得到初始变形下投篮动作坐标,给出篮球球员投篮举球、托球动作的差异性误差向量,获取手臂动作像素信息特征仿射变换,在图像序列中对篮球球员投篮动作手臂动态轨迹进行跟踪,完成对篮球投篮动作手臂视频图像轮廓跟踪提取。实验结果证明,所提方法标记精度较高,大幅度提升了球员的投篮技术。
[Abstract]:To the basketball arm shooting video image tracking extraction, can better grasp the shooting technology, the shooting video image tracking extraction, need to obtain the movement difference error vector. Get the affine transformation of arm action pixel information, complete the contour tracking extraction of video image. The traditional method uses the bright spot model diffraction method to achieve visual penetration, and gives the technical features of shooting action arm. But the affine transformation of action pixel information is ignored. To complete the shooting action video image tracking extraction. Based on camshift basketball shooting action arm video image contour tracking extraction method. The original video image filtering processing. The maximum variance ratio threshold method is used to divide the arm movement area of the shooting, and the coordinate of the shooting action under the initial deformation is obtained, and the difference error vector of the basketball players' shooting lifting and holding action is given. Acquisition of arm action pixel information feature affine transformation, in the image sequence of basketball players to track the dynamic trajectory of the shooting action arm. The experimental results show that the proposed method has high marking accuracy and greatly improves the shooting technique of players.
【作者单位】：中北大学朔州校区;
【分类号】：G841;TP391.41
【正文快照】： 1引言在国内,随着计算机科技水平的不断提升,目标图像跟踪和人体动力学等学科实现了对篮球投篮动作特征数据分析和采集,成为当前体育运动学科和计算机应用学科研究的核心课题[1~3]。在投篮比赛中,投篮得分的多少决定着一场比赛的胜负,所有掌握好投篮技术并不断地提高投篮命中

【相似文献】