Research on Gesture Definition and Electrode Placement in Pattern Recognition of Hand Gesture Action SEMG
The goal of this study is to explore the effects of electrode place-ment on the hand gesture pattern recognition performance. We have conducted experiments with surface EMG sensors using two detecting electrode channels. In total 25 different hand gesture
- PDF / 954,301 Bytes
- 8 Pages / 430 x 660 pts Page_size
- 66 Downloads / 201 Views
Department of Electronics Science & Technology University of Science & Technology of China, Hefei, China, 230027 [email protected] 2 Nokia Research Center, Interaction CTC, Interacting in Smart Environments P.O. Box 407, FI-00045 Nokia Group, Finland [email protected] 3 Nokia Research Center, NOKIA (China) Investment CO., LTD., Beijing, 100013
Abstract. The goal of this study is to explore the effects of electrode placement on the hand gesture pattern recognition performance. We have conducted experiments with surface EMG sensors using two detecting electrode channels. In total 25 different hand gestures and 10 different electrode positions for measuring muscle activities have been evaluated. Based on the experimental results, dependencies between surface EMG signal detection positions and hand gesture recognition performance have been analyzed and summarized as suggestions how to define hand gestures and select suitable electrode positions for a myoelectric control system. This work provides useful insight for the development of a medical rehabilitation system based on EMG technique.
1 Introduction In myoelectric control systems, surface electromyographic (sEMG) sensors are used for measuring the activities of the muscular system in a non-intrusive fashion. Detected muscle activities are identified as motion commands which can be used for controlling an externally powered device [1], especially including the artificial prosthesis for rehabilitation [2], [3]. So far, the control motions considered in EMG recognition research have been simple hand gestures with a large scale. Typically, the number of classifiable gestures has been limited between four and eight [3], [4]. In order to realize a highly accurate multi-gesture pattern recognition system, it is important to explore both the sEMG signal processing algorithms and the underlying myoelectric controlling mechanism [5]. Though much excellent work [3], [5] has been done on multi-gesture sEMG pattern recognition, the research effort has been focused mainly on the classification algorithm development. The relationship between the signal detection positions, muscles involved with the execution of the hand gestures, and the classification performance has been ignored. So far, there have not been systematic investigations on how to define usability-wise a good set of hand gestures D. Zhang (Ed.): ICMB 2008, LNCS 4901, pp. 33–40, 2007. © Springer-Verlag Berlin Heidelberg 2007
34
X. Zhang et al.
and select the appropriate EMG signal detecting positions for achieving a myoelectric control application with high recognition performance. The main objective of this paper is to conduct a systematic study on the definition of gestures and the selection of signal measuring positions for a sEMG-based control system. On the basis of sEMG pattern recognition experiments with various kinds of hand gestures and various electrode positions, we have analyzed the dependencies between the electrode detecting positions and hand gesture recognition performance. These resu
Data Loading...
