Applied Mathematics & Information Sciences
Abstract
While most golfers concern about the path that golf club travels on swing plane or the angle of the club face when addressing, rather the detrimental part may lie in golfer’s body rotation. This research delineate a measure in capturing golfer’s swing motions in which the Kinect motion tracking by Microsoft is adopted to validate the proposed measure. Subjects were hired to perform golf swings in front of the Kinect skeleton tracking system where subject’s 20 joint positions were recorded. The derived data were analyzed with ANOVA and multivariate clustering. Without clustering, the three investigated factors (gender, age and experience) depicted no impacts on subject’s swing performance. With the clustering, the differences were revealed in the light of the resulted two groups for distinguished characteristics. One group of less golfing experience which was mainly composed of female, younger and less golfing practice subjects achieved lower scores on swing correctness. Another group of experienced one which was with majority of male, senior and longer golfing tenure achieved higher swing scores. This research contributes in proposing an analysis framework that reveals the embedded information of golf swing motion. Implications for athletic practices were drawn accordingly.
Digital Object Identifier (DOI)
http://dx.doi.org/10.18576/amis/100235
Recommended Citation
Wang, Wen-Cheng; Ku, Hao-Hsiang; and Tsang, Seng-Su
(2016)
"Using Clustering Algorithm to Validate the Golf Backswing Action Simulated by Microsoft Kinect Motion-Sensing Photography,"
Applied Mathematics & Information Sciences: Vol. 10:
Iss.
2, Article 35.
DOI: http://dx.doi.org/10.18576/amis/100235
Available at:
https://digitalcommons.aaru.edu.jo/amis/vol10/iss2/35