留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于球场重建的球员运动数据分析

吉晓琪 宋子恺 于俊清

吉晓琪, 宋子恺, 于俊清等 . 基于球场重建的球员运动数据分析[J]. 北京航空航天大学学报, 2022, 48(8): 1543-1552. doi: 10.13700/j.bh.1001-5965.2022.0131
引用本文: 吉晓琪, 宋子恺, 于俊清等 . 基于球场重建的球员运动数据分析[J]. 北京航空航天大学学报, 2022, 48(8): 1543-1552. doi: 10.13700/j.bh.1001-5965.2022.0131
JI Xiaoqi, SONG Zikai, YU Junqinget al. Player movement data analysis on soccer field reconstruction[J]. Journal of Beijing University of Aeronautics and Astronautics, 2022, 48(8): 1543-1552. doi: 10.13700/j.bh.1001-5965.2022.0131(in Chinese)
Citation: JI Xiaoqi, SONG Zikai, YU Junqinget al. Player movement data analysis on soccer field reconstruction[J]. Journal of Beijing University of Aeronautics and Astronautics, 2022, 48(8): 1543-1552. doi: 10.13700/j.bh.1001-5965.2022.0131(in Chinese)

基于球场重建的球员运动数据分析

doi: 10.13700/j.bh.1001-5965.2022.0131
基金项目: 

国家重点研发计划 2020YFB1805601

详细信息
    通讯作者:

    于俊清, E-mail: yjqing@hust.edu.cn

  • 中图分类号: TP391

Player movement data analysis on soccer field reconstruction

Funds: 

National Key R & D Program of China 2020YFB1805601

More Information
  • 摘要:

    足球比赛中球员运动数据分析对增加观众的观看体验和辅助教练进行球员评估有着重要意义。球员运动数据分析的难点在于如何定位球员在球场上的坐标,即如何确定足球视频中单帧画面出现的缺损球场与标准二维球场之间的映射关系。针对如何在足球比赛中克服相机的高速移动和视角剧烈变化,设计并提出了利用球场重建与球员跟踪来进行球员运动数据分析的方法。球场重建方面,将足球视频中的球场分组为左中右3部分,每组通过球场分割、球场直线检测、球场直线分组、球场中圈点集合识别和球场关键点匹配来实现缺损球场到标准球场的映射;球员跟踪采用核相关滤波(KCF)跟踪算法,得到了球员运动数据统计的可视化结果。结合球场重建和球员跟踪算法定位球员的标准坐标,统计球员的一系列运动数据并进行可视化分析。提出的球员运动数据分析方法能够准确而快速地统计出球员的运动数据,包括球员坐标、运动轨迹、奔跑速度、活动范围和球员间距。球场重建方面采用图像交并进行评估,交并比达到87%,相比于传统的基于字典查询的方法(交并比为83.3%)准确度提升了3.7%。实验结果表明:所提出的球场重建方法能够更准确地表示球场映射关系,为球员运动数据分析统计提供更好的支持。

     

  • 图 1  球员运动数据分析方法整体流程

    Figure 1.  Flow chart of player motion data analysis method

    图 2  半场区域与中场区域的球场分割结果

    Figure 2.  Segmentation results of half field area and midfield area of soccer field

    图 3  球场直线分组

    Figure 3.  Straight-line grouping in soccer field

    图 4  概率决策树模型

    Figure 4.  Probabilistic decision tree model

    图 5  球场中圈点集合识别

    Figure 5.  Recognition of center circle point set in soccer field

    图 6  球场关键点

    Figure 6.  Key points in soccer field

    图 7  非球场直线剔除实验结果

    Figure 7.  Off-soccer field straight line culling

    图 8  直线匹配距离计算

    Figure 8.  Distance calculation of matching straight line

    图 9  中圈关键点映射

    Figure 9.  Key point mapping in center circle

    图 10  球场映射结果

    Figure 10.  Soccer field mapping results

    图 11  球场映射实验对比

    Figure 11.  Experimental comparison of soccer field mapping

    图 12  球员活动范围参数计算

    Figure 12.  Calculation of player motion range parameters

    图 13  球员跑动距离-时间统计

    Figure 13.  Player running distance-time statistics

    图 14  球员进球时运动类型统计

    Figure 14.  Statistics of player's movement types when scoring a goal

    图 15  球员位置热力图

    Figure 15.  Player position heat map

    表  1  速度离散化说明

    Table  1.   Description of velocity discretization

    运动类型分类 速度间隔
    站立 小于0.2 m/s
    0.2~2.1 m/s
    慢跑 2.1~3.8 m/s
    快跑 3.8~6.1 m/s
    冲刺 大于6.1 m/s
    下载: 导出CSV
  • [1] BALLARD D H. Generalizing the Hough transform to detect arbitrary shapes[J]. Pattern Recognition, 1981, 13(2): 111-122. doi: 10.1016/0031-3203(81)90009-1
    [2] HENRIQUES J F, CASEIRO R, MARTINS P, et al. High-speed tracking with kernelized correlation filters[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(3): 583-596. doi: 10.1109/TPAMI.2014.2345390
    [3] SHARMA R A, BHAT B, GANDHI V. Automated top view registration of broadcast football videos[C]//Proceedings of IEEE Winter Conference on Applications of Computer Vision. Piscataway: IEEE Press, 2018: 305-313.
    [4] MIRZA M, OSINDERO S. Conditional generative adversarial nets[EB/OL]. (2014-11-06)[2022-03-01]. https://arxiv.org/abs/1411.1784.
    [5] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2005, 1: 886-893.
    [6] CHEN J, LITTLE J J. Sports camera calibration via synthetic data[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Piscataway: IEEE Press, 2019: 2497-2504.
    [7] LUCAS B. An iterative image registration technique with an application to stereo vision (DARPA)[C]//Proceedings of DARPA Image Understanding Workshop, 1981: 121-130.
    [8] HOMAYOUNFAR N, FIDLER S, URTASUN R. Sports field localization via deep structured models[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2017: 4012-4020.
    [9] DETONE D, MALISIEWICZ T, RABINOVICH A. Deep image homography estimation[EB/OL]. (2016-01-13)[2022-03-01]. https://arxiv.org/abs/1606.03798.
    [10] NGUYEN T, CHEN S W, SHIVAKUMAR S S, et al. Unsupervised deep homography: A fast and robust homography estimation model[J]. IEEE Robotics and Automation Letters, 2018, 3(3): 2346-2353.
    [11] JIANG W, HIGUERA J, ANGLES B, et al. Optimizing through learned errors for accurate sports field registration[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. Piscataway: IEEE Press, 2020: 201-210.
    [12] CANNY J. Collision detection for moving polyhedra[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(2): 200-209.
    [13] FENG N, SONG Z, YU J, et al. SSET: A dataset for shot segmentation, event detection, player tracking in soccer videos[J]. Multimedia Tools and Applications, 2020, 79(39): 28971-28992.
    [14] PUWEIN J, ZIEGLER R, VOGEL J, et al. Robust multi-view camera calibration for wide-baseline camera networks[C]//2011 IEEE Workshop on Applications of Computer Vision (WACV). Piscataway: IEEE Press, 2011: 321-328.
    [15] BOLME D S, BEVERIDGE J R, DRAPER B A, et al. Visual object tracking using adaptive correlation filters[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2010: 2544-2550.
  • 加载中
图(15) / 表(1)
计量
  • 文章访问数:  56
  • HTML全文浏览量:  11
  • PDF下载量:  13
  • 被引次数: 0
出版历程
  • 收稿日期:  2022-03-09
  • 录用日期:  2022-03-25
  • 刊出日期:  2022-03-31

目录

    /

    返回文章
    返回
    常见问答