首页|运动员潜力评测系统在学生体能训练中的应用

运动员潜力评测系统在学生体能训练中的应用

扫码查看
针对目前运动员测评系统无法挖掘学生潜在体能的问题,本文结合视频分析技术、LSTM技术、高斯混合、视觉时空评价等方法开发了一套适合运动员能力预测评估的系统.该系统在第一人称视频中提取元事件进行预测分析,并构建高度非线性视觉空间,通过卷积神经网络学习,对运动员行为特征进行评估.实验结果表明,本文评估方法相比其他第一人称视角评估模型的预测精度更高,能有效识别学生体能训练的贡献行为和负面行为,并以此为依据对学生体能潜能进行预测评价.
The Application of Athlete Potential Evaluation System in Student Physical Training
In response to the problem that the current athlete evaluation system cannot explore students'potential physical fitness,this article combines video analysis technology,LSTM technology,Gaussian mixture,visual spatiotemporal evaluation and other methods to develop a system suitable for athlete ability prediction and evaluation.The system extracts meta events from first person videos for predictive analysis and constructs a highly nonlinear visual space.Through convolutional neural network learning,it evaluates the behavioral characteristics of athletes.The experimental results show that the evaluation method proposed in this paper has higher prediction accuracy compared to other first person perspective evaluation models.It can effectively identify the contribution behavior and negative behavior of students'physical training,and use this as a basis to predict and evaluate students'physical potential.

evaluate performanceprediction systemLSTM technologyconvolutional networksstudent physical training

张冰

展开 >

山西铁道职业技术学院,山西太原 030000

评估性能 预测系统 LSTM技术 卷积网络 学生体能训练

2024

软件
中国电子学会 天津电子学会

软件

影响因子:1.51
ISSN:1003-6970
年,卷(期):2024.45(7)