Human Locomotion Activity and Speed Recognition Using Electromyography Based Features

Seyed Mostafa Rezayat Sorkhabadi, Prudhvi Tej Chinimilli, Daniel Gaytan-Jenkins, Wenlong Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Human locomotion recognition methods based on electromyography (EMG) signals have not shown robust and accurate classification performance. This is due to the limitations of EMG signals such as its stochastic nature and sensitivity to placement of the sensors, as well as the number of sensors, feature extraction and classification algorithms. In this paper, a robust classification approach with only two features derived from EMG signals is developed to recognize locomotion activities and detect changing speeds. The root means square (RMS) and energy of the EMG signals are the features adopted in this method. The energy of the EMG signal is extracted using energy kernel method. The proposed approach uses a low number of sensors and features, online unsupervised classification, and is generalizable to different lower-limb muscle groups. To evaluate the benefits of the proposed approach, it is initially tested on a public dataset of five participants with two EMG sensors on biceps femoris and gastrocnemius, doing separate trials on the treadmill at various speeds and slopes. We performed additional experiments on two participants with EMG sensors on vastus laterialis and vastus medialis, as treadmill speeds changed online within each trial. The proposed approach achieved significant classification accuracy (above 90%) using the standard unsupervised K-means clustering, for both locomotion activity and speed recognition with the public dataset and our collected data.

Original languageEnglish (US)
Title of host publication2019 Wearable Robotics Association Conference, WearRAcon 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages80-85
Number of pages6
ISBN (Electronic)9781538680568
DOIs
StatePublished - May 21 2019
Externally publishedYes
Event2019 Wearable Robotics Association Conference, WearRAcon 2019 - Scottsdale, United States
Duration: Mar 25 2019Mar 27 2019

Publication series

Name2019 Wearable Robotics Association Conference, WearRAcon 2019

Conference

Conference2019 Wearable Robotics Association Conference, WearRAcon 2019
CountryUnited States
CityScottsdale
Period3/25/193/27/19

Fingerprint

Electromyography
Locomotion
Sensor
energy
Sensors
Exercise equipment
Unsupervised Classification
Unsupervised Clustering
K-means Clustering
Kernel Methods
Energy Method
Classification Algorithm
Energy
Muscle
Mean Square
Placement
Feature Extraction
Slope
Human
Roots

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Control and Optimization
  • Human Factors and Ergonomics

Cite this

Rezayat Sorkhabadi, S. M., Chinimilli, P. T., Gaytan-Jenkins, D., & Zhang, W. (2019). Human Locomotion Activity and Speed Recognition Using Electromyography Based Features. In 2019 Wearable Robotics Association Conference, WearRAcon 2019 (pp. 80-85). [8719626] (2019 Wearable Robotics Association Conference, WearRAcon 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WEARRACON.2019.8719626

Human Locomotion Activity and Speed Recognition Using Electromyography Based Features. / Rezayat Sorkhabadi, Seyed Mostafa; Chinimilli, Prudhvi Tej; Gaytan-Jenkins, Daniel; Zhang, Wenlong.

2019 Wearable Robotics Association Conference, WearRAcon 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 80-85 8719626 (2019 Wearable Robotics Association Conference, WearRAcon 2019).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rezayat Sorkhabadi, SM, Chinimilli, PT, Gaytan-Jenkins, D & Zhang, W 2019, Human Locomotion Activity and Speed Recognition Using Electromyography Based Features. in 2019 Wearable Robotics Association Conference, WearRAcon 2019., 8719626, 2019 Wearable Robotics Association Conference, WearRAcon 2019, Institute of Electrical and Electronics Engineers Inc., pp. 80-85, 2019 Wearable Robotics Association Conference, WearRAcon 2019, Scottsdale, United States, 3/25/19. https://doi.org/10.1109/WEARRACON.2019.8719626
Rezayat Sorkhabadi SM, Chinimilli PT, Gaytan-Jenkins D, Zhang W. Human Locomotion Activity and Speed Recognition Using Electromyography Based Features. In 2019 Wearable Robotics Association Conference, WearRAcon 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 80-85. 8719626. (2019 Wearable Robotics Association Conference, WearRAcon 2019). https://doi.org/10.1109/WEARRACON.2019.8719626
Rezayat Sorkhabadi, Seyed Mostafa ; Chinimilli, Prudhvi Tej ; Gaytan-Jenkins, Daniel ; Zhang, Wenlong. / Human Locomotion Activity and Speed Recognition Using Electromyography Based Features. 2019 Wearable Robotics Association Conference, WearRAcon 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 80-85 (2019 Wearable Robotics Association Conference, WearRAcon 2019).
@inproceedings{dce8f6e3963b4eada397c5506eb047c3,
title = "Human Locomotion Activity and Speed Recognition Using Electromyography Based Features",
abstract = "Human locomotion recognition methods based on electromyography (EMG) signals have not shown robust and accurate classification performance. This is due to the limitations of EMG signals such as its stochastic nature and sensitivity to placement of the sensors, as well as the number of sensors, feature extraction and classification algorithms. In this paper, a robust classification approach with only two features derived from EMG signals is developed to recognize locomotion activities and detect changing speeds. The root means square (RMS) and energy of the EMG signals are the features adopted in this method. The energy of the EMG signal is extracted using energy kernel method. The proposed approach uses a low number of sensors and features, online unsupervised classification, and is generalizable to different lower-limb muscle groups. To evaluate the benefits of the proposed approach, it is initially tested on a public dataset of five participants with two EMG sensors on biceps femoris and gastrocnemius, doing separate trials on the treadmill at various speeds and slopes. We performed additional experiments on two participants with EMG sensors on vastus laterialis and vastus medialis, as treadmill speeds changed online within each trial. The proposed approach achieved significant classification accuracy (above 90{\%}) using the standard unsupervised K-means clustering, for both locomotion activity and speed recognition with the public dataset and our collected data.",
author = "{Rezayat Sorkhabadi}, {Seyed Mostafa} and Chinimilli, {Prudhvi Tej} and Daniel Gaytan-Jenkins and Wenlong Zhang",
year = "2019",
month = "5",
day = "21",
doi = "10.1109/WEARRACON.2019.8719626",
language = "English (US)",
series = "2019 Wearable Robotics Association Conference, WearRAcon 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "80--85",
booktitle = "2019 Wearable Robotics Association Conference, WearRAcon 2019",

}

TY - GEN

T1 - Human Locomotion Activity and Speed Recognition Using Electromyography Based Features

AU - Rezayat Sorkhabadi, Seyed Mostafa

AU - Chinimilli, Prudhvi Tej

AU - Gaytan-Jenkins, Daniel

AU - Zhang, Wenlong

PY - 2019/5/21

Y1 - 2019/5/21

N2 - Human locomotion recognition methods based on electromyography (EMG) signals have not shown robust and accurate classification performance. This is due to the limitations of EMG signals such as its stochastic nature and sensitivity to placement of the sensors, as well as the number of sensors, feature extraction and classification algorithms. In this paper, a robust classification approach with only two features derived from EMG signals is developed to recognize locomotion activities and detect changing speeds. The root means square (RMS) and energy of the EMG signals are the features adopted in this method. The energy of the EMG signal is extracted using energy kernel method. The proposed approach uses a low number of sensors and features, online unsupervised classification, and is generalizable to different lower-limb muscle groups. To evaluate the benefits of the proposed approach, it is initially tested on a public dataset of five participants with two EMG sensors on biceps femoris and gastrocnemius, doing separate trials on the treadmill at various speeds and slopes. We performed additional experiments on two participants with EMG sensors on vastus laterialis and vastus medialis, as treadmill speeds changed online within each trial. The proposed approach achieved significant classification accuracy (above 90%) using the standard unsupervised K-means clustering, for both locomotion activity and speed recognition with the public dataset and our collected data.

AB - Human locomotion recognition methods based on electromyography (EMG) signals have not shown robust and accurate classification performance. This is due to the limitations of EMG signals such as its stochastic nature and sensitivity to placement of the sensors, as well as the number of sensors, feature extraction and classification algorithms. In this paper, a robust classification approach with only two features derived from EMG signals is developed to recognize locomotion activities and detect changing speeds. The root means square (RMS) and energy of the EMG signals are the features adopted in this method. The energy of the EMG signal is extracted using energy kernel method. The proposed approach uses a low number of sensors and features, online unsupervised classification, and is generalizable to different lower-limb muscle groups. To evaluate the benefits of the proposed approach, it is initially tested on a public dataset of five participants with two EMG sensors on biceps femoris and gastrocnemius, doing separate trials on the treadmill at various speeds and slopes. We performed additional experiments on two participants with EMG sensors on vastus laterialis and vastus medialis, as treadmill speeds changed online within each trial. The proposed approach achieved significant classification accuracy (above 90%) using the standard unsupervised K-means clustering, for both locomotion activity and speed recognition with the public dataset and our collected data.

UR - http://www.scopus.com/inward/record.url?scp=85067103804&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067103804&partnerID=8YFLogxK

U2 - 10.1109/WEARRACON.2019.8719626

DO - 10.1109/WEARRACON.2019.8719626

M3 - Conference contribution

T3 - 2019 Wearable Robotics Association Conference, WearRAcon 2019

SP - 80

EP - 85

BT - 2019 Wearable Robotics Association Conference, WearRAcon 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -