This study aimed to explore the application of unmarked motion capture technology in the measurement of standing long jump.RGB-D depth camera was used for motion capture,and image processing techniques such as key frame extraction,image segmentation,and edge detection were used in combination with machine learning algorithms to achieve the accurate measurement of standing long jump distance in an unmarked way.The accuracy and reliability are verified through reliability and validity analysis.Through repeated measurements of 51 standard distances,the Cronbach's alpha coefficient"α"was 0.999 for the test-retest reliability and the intra-class correlation coefficient(ICC)was 0.998,and the accuracy of the retest reached 99.64％.Through 306 actual measurements of standing long jump trials,the Pearson correlation coefficient"r"was 0.999 obtained for criterion-related validity,and the stability and reliability reached 99.69％.It could be seen that this study successfully developed and validated an unmarked motion capture technology-based method for measuring standing long jump distance,which demonstrated its innovative and practical value with high accuracy and reliability.
关键词
无标记动作捕捉/图像处理/机器学习/立定跳远/体育测量
Key words
unmarked motion capture/image processing/machine learning/standing long jump/sports measurement
维普
万方数据