A Review for Feature Extraction of EMG Signal Processing


In this paper, we introduce a new time-evolved spectral analysis-SLEX for analyzing the EMG signal. First we had review on four other common ways for feature extraction of EMG signal and last of all we focus on SLEX. Smooth Localized Complex Exponential (SLEX), is a kind of time dependent spectral analysis. Diverse from conventional Fourier method, be appropriated by two particular smooth windows on Fourier basis function and has the capability to be simultaneously orthogonal and localized. In this study we tried to show the application of the FFT, Wavelet Transform, Autoregressive, and PSE in EMG feature extraction. Each of which of these techniques for feature extraction has its own pros and cons which we brought it to the note. This method can overcome the shortcoming of conventional Fourier-based spectral analysis and accurately describe the time-dependent statistical property of EMG signal. Our conclusion table shows the best way for EMG feature extraction which is SLEX-based EMG recognition systems and it represent good performance in classifying 8 wrist motions. The classification accuracy of the 4channels is about 98% superior to the other methods.

  • Abstract
  • Key Words
  • 1. Introduction
  • 2. Fourier Transform (Short Time Fourier Transform)
  • 3. Power Spectral Density (PSD) of EMG Signal
  • 4. Wavelet Transform
  • 5. Auto Regression (AR)
  • 6. Smooth Localized Complex Exponential (SLEX)
  • 7. Comparison Between Whole Techniques for EMG Feature Extraction
  • 8. Conclusion
  • References

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In