Treffer: Skeletal point analysis to determine the accuracy of forehand smash shots played by badminton players.

This study aims to address the scarcity of scientific research on badminton performance analysis, specifically the accuracy of forehand smash shots. The authors propose the use of a skeletal coordinates-based technology to analyze a badminton player's biomechanics. To achieve this, specific techniques, such as formulating a quantitative description of badminton smash biomechanics based on the available literature, collecting video footage of badminton rallies and processing them using a MediaPipe-powered Python program, were followed. Three main approaches were considered for the analysis, defining a dynamic mathematical model, creating a player-to-player comparison model, and developing a machinelearning model. Preliminary results suggest that the use of three-dimensional points in comparison to two-dimensional points provides more accuracy in detecting the angle between three skeletal points from any camera perspective. This research also proposes a novel approach to compare two players and evaluate their skills based on a set of key parameters. The study explores the integration of machine learning algorithms to classify and predict player performance accurately. All three proposed methods enable coaches and players to identify and improve upon their weaknesses, enhancing their overall performance, as these findings have the potential to reduce subjectivity in measuring shot accuracy during training and to provide players with a more objective means of evaluating their performance. The proposed methodology and results contribute to a better understanding of badminton biomechanics and have implications for future research in this field. [ABSTRACT FROM AUTHOR]

Copyright of Journal of the National Science Foundation of Sri Lanka is the property of National Science Foundation of Sri Lanka and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)