Simulation study of deep-learning-based gait classification of young/elderly adults using Doppler radar
Deep-learning-based gait classification of young and elderly adults using micro-Doppler radar (MDR) is presented in this paper. The MDR signal data were accurately simulated using an open motion-capture gait dataset, and deep-learning classification of the time-velocity distribution (i.e., spectrogr...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2021-12-01
|
Series: | Measurement: Sensors |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2665917421000660 |
id |
doaj-2831d506e97249aab2d4b60a45d34812 |
---|---|
record_format |
Article |
spelling |
doaj-2831d506e97249aab2d4b60a45d348122021-09-17T04:37:51ZengElsevierMeasurement: Sensors2665-91742021-12-0118100103Simulation study of deep-learning-based gait classification of young/elderly adults using Doppler radarToshiyuki Hoshiga0Kenshi Saho1Keitaro Shioiri2Masahiro Fujimoto3Yoshiyuki Kobayashi4Department of Intelligent Robotics, Toyama Prefectural University, Imizu, JapanCorresponding author.; Department of Intelligent Robotics, Toyama Prefectural University, Imizu, JapanDepartment of Intelligent Robotics, Toyama Prefectural University, Imizu, JapanHuman Augmentation Research Center, National Institute of Advanced Industrial Science and Technology (AIST), Kashiwa, JapanHuman Augmentation Research Center, National Institute of Advanced Industrial Science and Technology (AIST), Kashiwa, JapanDeep-learning-based gait classification of young and elderly adults using micro-Doppler radar (MDR) is presented in this paper. The MDR signal data were accurately simulated using an open motion-capture gait dataset, and deep-learning classification of the time-velocity distribution (i.e., spectrogram) images calculated with the generated data are presented. Utilizing a simulation, we also investigated the body parts deemed most efficient for classification based on their generation of good MDR data. As a result, the classification rate using whole-body data was 74%. However, this classification rate of using only leg data showed an accuracy of 91%, which indicates that the thighs and shanks are efficient target body parts for the gait classification of both young and elderly adults.http://www.sciencedirect.com/science/article/pii/S2665917421000660Gait classificationMicro-DopplerDeep learningConvolutional neural networkSpectrogram |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Toshiyuki Hoshiga Kenshi Saho Keitaro Shioiri Masahiro Fujimoto Yoshiyuki Kobayashi |
spellingShingle |
Toshiyuki Hoshiga Kenshi Saho Keitaro Shioiri Masahiro Fujimoto Yoshiyuki Kobayashi Simulation study of deep-learning-based gait classification of young/elderly adults using Doppler radar Measurement: Sensors Gait classification Micro-Doppler Deep learning Convolutional neural network Spectrogram |
author_facet |
Toshiyuki Hoshiga Kenshi Saho Keitaro Shioiri Masahiro Fujimoto Yoshiyuki Kobayashi |
author_sort |
Toshiyuki Hoshiga |
title |
Simulation study of deep-learning-based gait classification of young/elderly adults using Doppler radar |
title_short |
Simulation study of deep-learning-based gait classification of young/elderly adults using Doppler radar |
title_full |
Simulation study of deep-learning-based gait classification of young/elderly adults using Doppler radar |
title_fullStr |
Simulation study of deep-learning-based gait classification of young/elderly adults using Doppler radar |
title_full_unstemmed |
Simulation study of deep-learning-based gait classification of young/elderly adults using Doppler radar |
title_sort |
simulation study of deep-learning-based gait classification of young/elderly adults using doppler radar |
publisher |
Elsevier |
series |
Measurement: Sensors |
issn |
2665-9174 |
publishDate |
2021-12-01 |
description |
Deep-learning-based gait classification of young and elderly adults using micro-Doppler radar (MDR) is presented in this paper. The MDR signal data were accurately simulated using an open motion-capture gait dataset, and deep-learning classification of the time-velocity distribution (i.e., spectrogram) images calculated with the generated data are presented. Utilizing a simulation, we also investigated the body parts deemed most efficient for classification based on their generation of good MDR data. As a result, the classification rate using whole-body data was 74%. However, this classification rate of using only leg data showed an accuracy of 91%, which indicates that the thighs and shanks are efficient target body parts for the gait classification of both young and elderly adults. |
topic |
Gait classification Micro-Doppler Deep learning Convolutional neural network Spectrogram |
url |
http://www.sciencedirect.com/science/article/pii/S2665917421000660 |
work_keys_str_mv |
AT toshiyukihoshiga simulationstudyofdeeplearningbasedgaitclassificationofyoungelderlyadultsusingdopplerradar AT kenshisaho simulationstudyofdeeplearningbasedgaitclassificationofyoungelderlyadultsusingdopplerradar AT keitaroshioiri simulationstudyofdeeplearningbasedgaitclassificationofyoungelderlyadultsusingdopplerradar AT masahirofujimoto simulationstudyofdeeplearningbasedgaitclassificationofyoungelderlyadultsusingdopplerradar AT yoshiyukikobayashi simulationstudyofdeeplearningbasedgaitclassificationofyoungelderlyadultsusingdopplerradar |
_version_ |
1717377629681614848 |