Sample Entropy Based Selection of Wavelet Decomposition Level for Finger Movement Recognition Using EMG
This paper reports the recognition of five fingermovements using forearm EMG signals. A relationship between the sample entropy (SampEn) of EMG signals at four wavelet decomposition levels and classification accuracy has been established. Experiments with the EMG at third level of wavelet decomposition can classify the finger movements with a maximum accuracy of 95.5%. These results show that EMG at the decomposition level which possess minimum SampEn produces the maximum classification accuracy. The experimental result shows that this relationship is a very useful criterion for selection of wavelet decomposition level to recognize EMG-based finger movements.
KeywordsWavelet transform Sample entropy SNR EMG
Centre of Excellence in Machine Learning and Big Data Analysis, Tezpur University, funded by Ministry of HRD, Government of India.
- 2.Ravindra, V., Castellini, C.: A comparative analysis of three non-invasive human-machine interfaces for the disabled. Front. Neurobot. (2014)Google Scholar
- 3.Tsenov, G., Zeghbib, A.H., Palis, F., Shoylev, N., Mladenov, V.: Neural networks for online classification of hand and finger movements using surface EMG signals. Neural Netw. Appl. Electr. Eng. 8 (2006)Google Scholar
- 4.Zhao, J., Xie, Z., Jiang, L., Cai, H., Liu, H., Hirzinger, G.: A EMG control for a five-fingered prosthetic hand based on wavelet transform and autoregressive model. In: IEEE International Conference on Mechatronics and Automation, pp. 1097–1102, USA (2006)Google Scholar
- 6.Ouyang, G., Zhu, X., Z.J., Liul, H.: High-density myoelectric pattern recognition toward improved stroke rehabilitation. IEEE Trans. Biomed. Eng. 18(1), 257–265 (2014)Google Scholar
- 8.Kakoty, N.M., Hazarika, S.M.: Recognition of grasp types through PCs of DWT based EMG features. In: IEEE International Conference on Rehabilitation Robotics, pp. 478–482, Zurich (2011)Google Scholar
- 11.Crawford, B., Miller, K., Shenoy, P., Rao, R.: Real-time classification of electromyographic signals for robotic control. Technical Report 2005-03-05, Department of Computer Science, University of Washington (2005)Google Scholar
- 16.Oskoei, M.A., Hu, H.: Myoelectric control systems: a survey. J. Biomed. Signal Process. Control 2, 275–294 (2007)Google Scholar
- 18.Phinyomark, A., Phukpattaranont, P., Limsakul, C.: Optimal wavelet functions in wavelet denoising for multifunction myoelectric control. Trans. Electr. Eng. Electron. Commun. 8(1), 43–52 (2010)Google Scholar
- 19.Phinyomark, A., Limsakul, C., Phukpattaranont, P.: Evaluation of wavelet function based on robust EMG feature extraction. In: The Seventh PSU Engineering Conference, pp. 277–281 (2009)Google Scholar