Recognition of Voluntary Blink and Bite Base on Single Forehead EMG
With the development of intelligent wearable technology, the need for a more effective and practical means of human-computer interaction is becoming increasingly urgent. In this paper, we used only one forehead Electromyogram (EMG) channel to accurately recognize at least 6 different voluntary blink and bite patterns as output interactive commands. Differential square moving average (DSMA) and square moving average (SMA) were used to distinguish blink and bite, voluntary blink and natural blink, respectively. Then, random forests classifier was employed to classify the 6 blink and bite patterns with extracted time-domain features. The accuracy of 92.60 ± 2.55 was obtained for the dataset of 10 subjects. It provides an effective human-computer interaction method with the advantages of rich commands, good real-time performance, low cost and small individual differences. The method proposed can be conveniently embedded in wearable device as an alternative of interaction.
KeywordsHuman-computer interaction Wearable device EMG Blink recognition Bite recognition
This work was supported in part by the National Natural Science Foundation of China under Grant 61100102 and Grant 61473110 and Grant 61633010, in part by the International Science and Technology Cooperation Program of China under Grant 2014DFG12570.
- 3.Wu, W., Hsu, C., Yang, S.: Smart wearable device design for bicycle lights direction control with speech-recognition function. In: International Symposium on Computer, Consumer and Control, Xian, pp. 53–56. IEEE (2016)Google Scholar
- 9.Ning, B., Li, M., Liu, T., Shen, H., Hu, L., Fu, X.: Human brain control of electric wheelchair with eye-blink electrooculogram signal. In: Su, C.-Y., Rakheja, S., Liu, H. (eds.) ICIRA 2012, Part I. LNCS (LNAI), vol. 7506, pp. 579–588. Springer, Heidelberg (2012). doi: 10.1007/978-3-642-33509-9_58 CrossRefGoogle Scholar