File Exchange

image thumbnail

EMG Feature Extraction Toolbox

version 1.1.1 (7.9 KB) by Jingwei Too
This toolbox offers 17 feature extraction methods (EMAV, EWL, MAV, WL, SSC, ZC ...) for Electromyography (EMG) signals applications.


Updated 01 Aug 2019

View License

This toolbox offers 17 types of EMG features
(1) Enhanced Mean absolute value (EMAV)
(2) Enhanced Wavelength (EWL)
(3) Mean Absolute Value (MAV)
(4) Slope Sign Change (SSC)
(5) Zero Crossing (ZC)
(6) Waveform Length (WL)
(7) Root Mean Square (RMS)
(8) Average Amplitude Change (AAC)
(9) Difference Absolute Standard Deviation Value (DASDV)
(10) Log Detector (LD)
(11) Modified Mean Absolute Value (MMAV)
(12) Modified Mean Absolute Value 2 (MMAV2)
(13) Myopulse Percentage Rate (MYOP)
(14) Simple Square Integral (SSI)
(15) Variance of EMG (VAR)
(16) Willison Amplitude (WA)
(17) Maximum Fractal Length (MFL)

The "Main" demos how the feature extraction methods can be applied by using the generated sample signal.

[1] J. Too, A. R. Abdullah, N. Mohd Saad, and W. Tee, “EMG Feature Selection and Classification Using a Pbest-Guide Binary Particle Swarm Optimization,” Computation, vol. 7, no. 1, 2019.
[2] J. Too, A. R. Abdullah, and N. Mohd Saad, “Classification of Hand Movements based on Discrete Wavelet Transform and Enhanced Feature Extraction,” Int. J. Adv. Comput. Sci. Appl., vol. 10, no. 6, 2019.

Cite As

Jingwei Too (2020). EMG Feature Extraction Toolbox (, MATLAB Central File Exchange. Retrieved .

Comments and Ratings (7)


very good!

Jingwei Too

Dear Sai Krishna,

You may set the value of the threshold based on research articles. Normally, we test with several values like 0.001, 0.01, 0.1 .... and apply the optimal one.

Sai Krishna

What should be the values of threshold for EMG Data??

Jingwei Too

Dear Polo Joachín,

You need to read some related paper before you started. To use it, you need an EMG signal.
In the "main" file, you just replace the "X" with your EMG signal.

How do I use it. I am new.

Jingwei Too

Dear Marcus Schneider, Thank you for the information. I have updated the program.

Unfortunately, the calculation of zero crossings and sign slope changes are wrong.



Add MFL feature




Add paper




Add another 10 feature extraction methods




(1) The SSC and ZC have been corrected.
(2) Add RMS feature



MATLAB Release Compatibility
Created with R2018a
Compatible with any release
Platform Compatibility
Windows macOS Linux