Paper
18 July 2023 Research on gesture recognition based on vision and sEMG signals
Guohua Cao, Zhisheng Wang, Shuo Wang, Hongsheng Liu, Xianyu Meng
Author Affiliations +
Proceedings Volume 12722, Third International Conference on Mechanical, Electronics, and Electrical and Automation Control (METMS 2023); 127222A (2023) https://doi.org/10.1117/12.2679574
Event: International Conference on Mechanical, Electronics, and Electrical and Automation Control (METMS 2023), 2023, Hangzhou, China
Abstract
Aiming at the problem of inaccurate hand positioning caused by the Kinect sensor placed too far away, we propose an advanced motion classification method. Combining Kinect with sEMG to classify human upper limb movements.Collect human body depth images and sEMG signals, extract human skeleton coordinate points to complete joint angle calculation. Perform data processing on sEMG signals to construct feature vector vector space.Then perform BP neural network training to classify 6 kinds of gestures. The experimental results show that the classification accuracy of arm gesture recognition is 98.5%, and the classification accuracy of hand gesture recognition is 97.6%. Therefore, our method has good recognizability for detailed motion and can meet the requirements of gesture motion recognition classification.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Guohua Cao, Zhisheng Wang, Shuo Wang, Hongsheng Liu, and Xianyu Meng "Research on gesture recognition based on vision and sEMG signals", Proc. SPIE 12722, Third International Conference on Mechanical, Electronics, and Electrical and Automation Control (METMS 2023), 127222A (18 July 2023); https://doi.org/10.1117/12.2679574
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Gesture recognition

Sensors

Action recognition

Education and training

Neural networks

Neurons

Muscles

Back to Top