JIITA

Swing gesture recognition with wearable sensors for playing a table tennis game

JIITA, vol. 3, no. 1, pp.253-256, 2019

DOI: SOON

Swing gesture recognition with wearable sensors for playing a table tennis game

Sangwoo Noh 1), Ahyoung Choi 2,*)
1) Samsung Electronics, Suwon, Republic of Korea
2) Department of Software, Gachon University, Seongnam, Republic of Korea

Abstract: Recently, a fitness tracker has been developed to monitor various kinds of exercise. However, most of the trekkers still provide the function to monitor exercise time, distance, and type of exercise. However, the technique of analyzing the activities of the ballgame is not developed yet. In this paper, we propose an algorithm to distinguish the behavior of table tennis games in order to monitor the movement of ball games. In a controlled environment, the analytical accuracy was found to be about 100%, 95%, and 90% accurate for the forehead, backhand, and none swing activity respectively.

Keyword: Activity recognition; accelerometer; gyroscope; wearable sensors