Hand Gesture Data Collection Procedure Using a Myo Armband for Machine Learning

Report No. ARL-TN-0699
Authors: Michael Lee and Nikhil Rao
Date/Pages: September 2015; 20 pages
Abstract: In 2015, the Battlefield Information Processing Branch investigated using machine learning (ML) to identify military hand gestures. A Naïve Bayes model was trained and created as an initial test. Naturally, the capability of the model relies on the quality and quantity of the training data. Therefore, the data collection is an important and necessary step in building a ML classifier. This report describes the procedure used to collect arm gesture movement data using a Myo armband. The source code for this work is included as an Appendix.
Distribution: Approved for public release
  Download Report ( 0.540 MBytes )
If you are visually impaired or need a physical copy of this report, please visit and contact DTIC.

Last Update / Reviewed: September 1, 2015