Design and Implementation of an EMG-Based Wrist Band Device with Imperfect Hand Gesture Recognition for Patients 


Vol. 46,  No. 12, pp. 2280-2290, Dec.  2021
10.7840/kics.2021.46.12.2280


PDF
  Abstract

EMG-based hand gesture recognition, which can recognize natural hand gestures in detail, is actively studied as an interface technology for patients. Existing research on gesture recognition does not take into account the characteristics of patients and hospital environments. Because patients have weak muscles and constrained range of motion, it is difficult to fully perform the specified gestures and easy to perform imperfect hand gestures. In addition, a device considering the convenience and hygiene for patients is needed. In this paper, we design and implement a wrist band device with imperfect hand gesture recognition for patients. The proposed method allows hand gesture recognition for patients by collecting an EMG dataset including imperfect hand gestures and training a lightweight deep neural network model. The implemented devices using low-power MCU and dry-type EMG sensors can recognize hand gestures in real-time. In addition, we study the performance of imperfect hand gesture recognition according to model architecture, quantization, sensor configuration, users, and sessions. As a result, the proposed method shows an accuracy of 87.68% and an accuracy of 84.98% after quantization.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

J. Lim, J. Seo, Y. Baek, "Design and Implementation of an EMG-Based Wrist Band Device with Imperfect Hand Gesture Recognition for Patients," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 12, pp. 2280-2290, 2021. DOI: 10.7840/kics.2021.46.12.2280.

[ACM Style]

Jaebong Lim, Juhee Seo, and Yunju Baek. 2021. Design and Implementation of an EMG-Based Wrist Band Device with Imperfect Hand Gesture Recognition for Patients. The Journal of Korean Institute of Communications and Information Sciences, 46, 12, (2021), 2280-2290. DOI: 10.7840/kics.2021.46.12.2280.

[KICS Style]

Jaebong Lim, Juhee Seo, Yunju Baek, "Design and Implementation of an EMG-Based Wrist Band Device with Imperfect Hand Gesture Recognition for Patients," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 12, pp. 2280-2290, 12. 2021. (https://doi.org/10.7840/kics.2021.46.12.2280)