Two-Channel Surface Electromyography for Individual and Combined Finger Movements

Publisher:
IEEE
Publication Type:
Conference Proceeding
Citation:
35th Annual International Conference of the IEEE EMBS 2013, 2013, pp. 4961 - 4964
Issue Date:
2013-01
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2012007705OK.pdf515.8 kB
Adobe PDF
This paper proposes the pattern recognition system for individual and combined finger movements by using two channel electromyography (EMG) signals. The proposed system employs Spectral Regression Discriminant Analysis (SRDA) for dimensionality reduction, Extreme Learning Machine (ELM) for classification and the majority vote for the classification smoothness. The advantage of the SRDA is its speed which is faster than original LDA so that it could deal with multiple features. In addition, the use of ELM which is fast and has similar classification performance to well-known SVM empowers the classification system. The experimental results show that the proposed system was able to recognize the individual and combined fingers movements with up to 98 % classification accuracy by using only just two EMG channels.
Please use this identifier to cite or link to this item: