Abstract

To assist individuals in sports activities is one of the emerging areas of wearable applications. Among various kinds of sports, detecting tennis strokes faces unique challenges. we propose an approach to detect three tennis strokes (backhand, forehand, serve) by utilizing a smartwatch. In our method, the smartwatch is part of a wireless network in which inertial data file is transferred to a laptop where data prepossessing and classification is performed. The data file contains acceleration and angular velocity data of the 3D accelerometer and gyroscope.

Instructions:

We used Fitbit Ionic(TM) smartwatch as a wearable sensor to
collect inertial motion data. Our smartwatch dataset contains
three main tennis strokes of serve, backhand, and forehand.
The sampling rate for inertial sensors accelerometer and
gyroscope is set to 50Hz. Each subject has repeated every
stroke 4 times. Eight right-handed subjects’ recordings were
selected from the dataset. Therefore, the selected dataset
contains 96 data sequences. Data recordings were done under a
local area wireless network in the Pervasive Lab at the Shahid
Beheshti University in Tehran.

Data Descriptor Article DOI:

https://ieeexplore.ieee.org/document/8964775