Research on Gesture Recognition Technology of Data Glove Based on Joint Algorithm

DOI

10.2991/mecae-18.2018.8How to use a DOI?

Keywords

Gesture recognition, joint algorithm, data glove, bending sensor.

Abstract

A gesture recognition system were fabricated with distributed sensor networks as the core component, and could be wore in the hands. This data glove could detect the posture of fingers using the FLEX2.2 bending sensors and STM32 series single chip. Furthermore, we adopted a gesture recognition joint algorithm which combined template matching with BP neural network. First, the identify data from bending sensors could be obtained by template matching. Then, these data was inputted to the neural network, and got the final gesture recognition rate of joint algorithm. This joint algorithm enhanced minimal template recognition rate, and improved high training volume of BP neural network, showing the higher accuracy and adaptability of recognition. These research exhibit large potential in low cost, accurate, flexible, wearable sign language recognition systems.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Shuran Yin
AU  - Jun Yang
AU  - Yukun Qu
AU  - Wenjun Liu
AU  - Yunfei Guo
AU  - Hongtao Liu
AU  - Dapeng Wei
PY  - 2018/03
DA  - 2018/03
TI  - Research on Gesture Recognition Technology of Data Glove Based on Joint Algorithm
BT  - Proceedings of the 2018 International Conference on Mechanical, Electronic, Control and Automation Engineering (MECAE 2018)
PB  - Atlantis Press
SP  - 41
EP  - 50
SN  - 2352-5401
UR  - https://doi.org/10.2991/mecae-18.2018.8
DO  - 10.2991/mecae-18.2018.8
ID  - Yin2018/03
ER  -