학술논문

Artificial Vascular Stent-Inspired Bending Sensors Embedded in a Data Glove for Hand Gesture Recognition
Document Type
Periodical
Source
IEEE Sensors Journal IEEE Sensors J. Sensors Journal, IEEE. 23(19):23388-23398 Oct, 2023
Subject
Signal Processing and Analysis
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Robotics and Control Systems
Sensors
Bending
Metals
Data gloves
Support vector machines
Wearable sensors
Intelligent sensors
Bending sensor
data glove
hand gesture recognition (HGR)
multichannel sensors
support vector machine (SVM)
Language
ISSN
1530-437X
1558-1748
2379-9153
Abstract
Owing to the superior conductivity and conductivity stability, the metal filaments are excellent potential materials for wearable sensors. However, the rigidity of metal limits its further application in wearable devices. Inspired by the structure of artificial vascular stents, a new idea for fabricating flexible and stretchable bending sensors based on metal filaments via structural design was presented in this article. Two braiding methods, with the plain-like structure and the loop-like structure, were proposed to fabricate rigid metal filaments into flexible finger sleeves. The designed bending sensors were further integrated into a multichannel and low-cost smart glove for hand gesture recognition (HGR), together with five commercial pressure sensors. The support vector machine (SVM) model was then utilized to recognize the hand gesture from the fusing data composed of five bending signals and five pressure signals, which were collected and transmitted via an embedded electronic system. The experimental results indicate that the smart glove achieves HGR using SVM model with an accuracy of 98.49% in the plain-like structure and 98.15% in the loop-like structure.