The Journal of China Universities of Posts and Telecommunications ›› 2022, Vol. 29 ›› Issue (4): 106-116.doi: 10.19682/j.cnki.1005-8885.2022.2023

• Others • Previous Articles    

Lighting control with Myo armband based on customized classifier

Jiang Yujian, Yang Xue, Zhang Junming, Song Yang   

  1. 1. Key Laboratory of Acoustic Visual Technology and Intelligent Control System, Communication University of China, Beijing 100024, China
    2. Beijing Key Laboratory of Modern Entertainment Technology, Communication University of China, Beijing 100024, China
    3. School of Information and Communication Engineering, Communication University of China, Beijing 100024, China
  • Received:2021-07-12 Revised:2021-10-25 Accepted:2021-12-07 Online:2022-08-31 Published:2022-08-31
  • Contact: Jiang Yujian, E-mail:
  • Supported by:
    This work was supported by the National Key R&D Program of China (2021YFF0307603).

Abstract: This paper focuses on gesture recognition and interactive lighting control. The collection of gesture data adopts the Myo armband to obtain surface electromyography (sEMG). Considering that many factors affect sEMG, a customized classifier based on user calibration data is used for gesture recognition. In this paper, machine learning classifiers k-nearest neighbor (KNN), support vector machines (SVM), and naive Bayesian (NB) classifier, which can be used in small sample sets, are selected to classify four gesture actions. The performance of the three classifiers under different training parameters, different input features, including root mean square (RMS), mean absolute value (MAV), waveform length (WL), slope sign change (SSC) number, zero crossing (ZC) number, and variance (VAR) are tested, and different input channels are also tested. Experimental results show that: The NB classifier, which assumes that the prior probability of features is polynomial distribution, has the best performance, reaching more than 95% accuracy. Finally, an interactive stage lighting control system based on Myo armband gesture recognition is implemented.

Key words: Myo armband, gesture recognition, surface electromyography, customized classifier, lighting control

CLC Number: