Human motion analysis and action scoring technology for sports training based on computer vision features

IF 1.5 Q2 COMPUTER SCIENCE, THEORY & METHODS

International Journal of Fuzzy Logic and Intelligent Systems Pub Date : 2021-01-01 DOI:10.3233/JIFS-219092

Yongheng Bai, Yinggang Chen

{"title":"Human motion analysis and action scoring technology for sports training based on computer vision features","authors":"Yongheng Bai, Yinggang Chen","doi":"10.3233/JIFS-219092","DOIUrl":null,"url":null,"abstract":"With the advent of the information age, computer-related application research has become more and more extensive, human motion analysis and action scoring based on computer vision have gradually become the focus of attention. In order to adapt to the development of the times and solve the problems related to the analysis of human motion, the experiment analyzed the similarity of eight common human movement behaviors, analyze the movement speed of men and women under sports training, and analyzed the accuracy of the human body motion recognition model in the two cases of the original gray data and the frame difference channel, finally, the denoising performance of four different algorithms of SMF, EMF, RAMF and median filter algorithm in digital image processing is analyzed. The final result shows that there is a big similarity between the same kind of human movement behavior, the accuracy rate of the frame difference channel human body recognition model is higher than that of the original gray data recognition model, and digital image processing median filter algorithm has good image denoising performance.","PeriodicalId":44705,"journal":{"name":"International Journal of Fuzzy Logic and Intelligent Systems","volume":"8 1","pages":"1-9"},"PeriodicalIF":1.5000,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Fuzzy Logic and Intelligent Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3233/JIFS-219092","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, THEORY & METHODS","Score":null,"Total":0}

引用次数: 5

Abstract

With the advent of the information age, computer-related application research has become more and more extensive, human motion analysis and action scoring based on computer vision have gradually become the focus of attention. In order to adapt to the development of the times and solve the problems related to the analysis of human motion, the experiment analyzed the similarity of eight common human movement behaviors, analyze the movement speed of men and women under sports training, and analyzed the accuracy of the human body motion recognition model in the two cases of the original gray data and the frame difference channel, finally, the denoising performance of four different algorithms of SMF, EMF, RAMF and median filter algorithm in digital image processing is analyzed. The final result shows that there is a big similarity between the same kind of human movement behavior, the accuracy rate of the frame difference channel human body recognition model is higher than that of the original gray data recognition model, and digital image processing median filter algorithm has good image denoising performance.

查看原文本刊更多论文

基于计算机视觉特征的运动训练人体动作分析与动作评分技术

随着信息时代的到来，与计算机相关的应用研究越来越广泛，基于计算机视觉的人体动作分析和动作评分逐渐成为人们关注的焦点。为了适应时代的发展，解决人体运动分析的相关问题，实验分析了八种常见的人体运动行为的相似性，分析了运动训练下男性和女性的运动速度，分析了原始灰度数据和帧差通道两种情况下人体运动识别模型的准确性，最后分析了SMF、EMF、分析了数字图像处理中的RAMF算法和中值滤波算法。最终结果表明，同类人体运动行为之间存在较大的相似性，帧差通道人体识别模型的准确率高于原始灰度数据识别模型，数字图像处理中值滤波算法具有良好的图像去噪性能。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

International Journal of Fuzzy Logic and Intelligent Systems COMPUTER SCIENCE, THEORY & METHODS-

CiteScore

2.80

自引率

23.10%

发文量

期刊介绍： The International Journal of Fuzzy Logic and Intelligent Systems (pISSN 1598-2645, eISSN 2093-744X) is published quarterly by the Korean Institute of Intelligent Systems. The official title of the journal is International Journal of Fuzzy Logic and Intelligent Systems and the abbreviated title is Int. J. Fuzzy Log. Intell. Syst. Some, or all, of the articles in the journal are indexed in SCOPUS, Korea Citation Index (KCI), DOI/CrossrRef, DBLP, and Google Scholar. The journal was launched in 2001 and dedicated to the dissemination of well-defined theoretical and empirical studies results that have a potential impact on the realization of intelligent systems based on fuzzy logic and intelligent systems theory. Specific topics include, but are not limited to: a) computational intelligence techniques including fuzzy logic systems, neural networks and evolutionary computation; b) intelligent control, instrumentation and robotics; c) adaptive signal and multimedia processing; d) intelligent information processing including pattern recognition and information processing; e) machine learning and smart systems including data mining and intelligent service practices; f) fuzzy theory and its applications.