图卷积网络的行为识别技术在空手道技战术分析中的应用  

The Application of Behavior Recognition Technology Based on Graph Convolutional Network in Karate Tactics Analysis

作　　者：郭建平[1] 李希 刘俊明 徐大宏 谢健 GUO Jian-ping;LI Xi;LIU Jun-ming;XU Da-hong;XIE Jian(College of Physical Education,Hunan Normal University,Changsha 410081;College of Information Science and Engineering,Hunan Normal University,Changsha 410081,China)

机构地区：[1]湖南师范大学体育学院,中国长沙410081 [2]湖南师范大学信息科学与工程学院,中国长沙410081

出　　处：《湖南师范大学自然科学学报》2021年第1期81-87,共7页Journal of Natural Science of Hunan Normal University

基　　金：国家国防科技工业局国防基础科研计划资助项目(WDZC20205500114)。

摘　　要：研究运动员训练和比赛视频,是空手道等竞技类体育运动中进行技战术分析的重要手段和方法。随着人工智能技术的日益成熟,人工智能技术与体育技战术分析相融合无疑是创新与提高技战术水平的重要途径。本论文建立了一种新的图卷积模型,用于对空手道运动员的技术动作识别、动作频度统计及其轨迹跟踪的自动智能分析。该技术有效地解决了传统方法中人力成本高、数据丢失严重、延时长、精度低等问题,结果表明新的拓扑图构建方法对提高行为识别精确度效果显著,同时也为技战术分析奠定了基础。It is an important method of technical and tactical analysis in Karate and other competitive sports to study athletes'training and competition video.With the rise of artificial intelligence technology,the integration of artificial intelligence technology and sports technical and tactical analysis is undoubtedly an important way to innovate and improve the level of technology and tactics.In this paper,a new graph convolution model is established for the automatic intelligent analysis of Karate athletes'technical action recognition,action frequency statistics and trajectory tracking.This technology effectively solves the problems of high labor cost,serious data loss,long delay and low precision in traditional methods.The results show that the new topology construction method has a significant effect on improving the accuracy of behavior recognition,and also lays a foundation for technical and tactical analysis.

关 键 词：行为识别 轨迹跟踪 空手道 技战术分析 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]