Related Experiment Video
Updated: Dec 27, 2025

A Method for Quantifying Upper Limb Performance in Daily Life Using Accelerometers
Published on: April 21, 2017
A Real-Time Wearable Assist System for Upper Extremity Throwing Action Based on Accelerometers
Kuang-Yow Lian1, Wei-Hsiu Hsu1, Deepak Balram1
1Department of Electrical Engineering, National Taipei University of Technology, Taipei 10608, Taiwan.
This study introduces a real-time wearable system using inertial measurement unit (IMU) sensors to analyze and improve baseball throwing actions. The system accurately recognizes throwing phases, aiding players in learning, correction, and rehabilitation with over 90% accuracy.
Area of Science:
- Sports Science
- Biomechanics
- Wearable Technology
Background:
- Baseball throwing involves complex, continuous upper extremity movements across six distinct phases.
- Accurate analysis of throwing mechanics is crucial for player performance, injury prevention, and rehabilitation.
- Existing action recognition systems often lack the precision required for nuanced, continuous movement analysis.
Purpose of the Study:
- To develop a real-time wearable assist system for analyzing and improving upper extremity throwing actions.
- To enable precise recognition of six serial throwing phases (wind-up to follow-through) using inertial measurement unit (IMU) sensors.
- To provide players with immediate feedback for learning, rectification, and rehabilitation of throwing mechanics.
Main Methods:
- Utilized accelerometers from IMU sensors to capture three-axial acceleration signals during throwing actions.
- Applied Kalman filtering for noise reduction, followed by leveling and labeling for data processing.
- Employed the Longest Common Subsequence (LCS) method to recognize the six serial phases of the throwing action by comparing sequence data.
- Integrated intelligent functions for status recognition, movement initiation, and posture transition analysis.
Main Results:
- Achieved an average recognition accuracy of 95.14% for all three users in experimental trials.
- Demonstrated successful real-time recognition of complex, continuous throwing actions with high precision.
- The developed system effectively analyzed posture and provided precise comments for improving throwing actions.
Conclusions:
- The developed real-time wearable assist system accurately recognizes upper extremity throwing actions.
- The system's high accuracy (average 95.14%) validates its effectiveness for baseball players.
- This technology offers a valuable tool for enhancing throwing performance, learning, and rehabilitation in sports.
More Related Videos
13:44Haptic/Graphic Rehabilitation: Integrating a Robot into a Virtual Environment Library and Applying it to Stroke Therapy
Published on: August 8, 2011
06:25Capturing Representative Hand Use at Home Using Egocentric Video in Individuals with Upper Limb Impairment
Published on: December 23, 2020