Advanced

Pulsed Millimeter Wave Radar for Hand Gesture Sensing and Classification

Fhager, Lars Ohlsson LU ; Heunisch, Sebastian LU ; Dahlberg, Hannes ; Evertsson, Anton and Wernersson, Lars Erik LU (2019) In IEEE Sensors Letters 3(12).
Abstract

A pulsed millimeter wave radar operating at a frame rate of 144 Hz is utilized to record 2160 scattering signatures of 12 generic hand gestures. Gesture recognition is achieved by machine learning, utilizing transfer learning on a pretrained convolutional neural network. This yields excellent classification results with a validation accuracy of 99.5%, based on a 60% training versus 40% validation split. The corresponding confusion matrix is also presented, showing a high level of classification orthogonality between the tested gestures. This is the first demonstration where data from a pulsed millimeter wave radar is used for gesture recognition by machine learning. It proves that the range-time envelope representation of high... (More)

A pulsed millimeter wave radar operating at a frame rate of 144 Hz is utilized to record 2160 scattering signatures of 12 generic hand gestures. Gesture recognition is achieved by machine learning, utilizing transfer learning on a pretrained convolutional neural network. This yields excellent classification results with a validation accuracy of 99.5%, based on a 60% training versus 40% validation split. The corresponding confusion matrix is also presented, showing a high level of classification orthogonality between the tested gestures. This is the first demonstration where data from a pulsed millimeter wave radar is used for gesture recognition by machine learning. It proves that the range-time envelope representation of high frame-rate data from a pulsed radar is suitable for hand gesture recognition. Further improvements are expected for more complex detection schemes and tailored neural networks.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
classification, convolutional neural network, gesture sensing, hand gesture recognition, machine learning, Microwave/millimeter wave sensors, millimeter wave radar, pulsed radar, transfer learning
in
IEEE Sensors Letters
volume
3
issue
12
article number
3502404
external identifiers
  • scopus:85082634141
ISSN
2475-1472
DOI
10.1109/LSENS.2019.2953022
language
English
LU publication?
yes
id
1c34f121-df71-43a7-b39c-5caf5090917c
date added to LUP
2020-04-28 16:20:32
date last changed
2020-04-29 04:23:47
@article{1c34f121-df71-43a7-b39c-5caf5090917c,
  abstract     = {<p>A pulsed millimeter wave radar operating at a frame rate of 144 Hz is utilized to record 2160 scattering signatures of 12 generic hand gestures. Gesture recognition is achieved by machine learning, utilizing transfer learning on a pretrained convolutional neural network. This yields excellent classification results with a validation accuracy of 99.5%, based on a 60% training versus 40% validation split. The corresponding confusion matrix is also presented, showing a high level of classification orthogonality between the tested gestures. This is the first demonstration where data from a pulsed millimeter wave radar is used for gesture recognition by machine learning. It proves that the range-time envelope representation of high frame-rate data from a pulsed radar is suitable for hand gesture recognition. Further improvements are expected for more complex detection schemes and tailored neural networks.</p>},
  author       = {Fhager, Lars Ohlsson and Heunisch, Sebastian and Dahlberg, Hannes and Evertsson, Anton and Wernersson, Lars Erik},
  issn         = {2475-1472},
  language     = {eng},
  number       = {12},
  series       = {IEEE Sensors Letters},
  title        = {Pulsed Millimeter Wave Radar for Hand Gesture Sensing and Classification},
  url          = {http://dx.doi.org/10.1109/LSENS.2019.2953022},
  doi          = {10.1109/LSENS.2019.2953022},
  volume       = {3},
  year         = {2019},
}