Mid Sweden University

miun.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Evaluation of 2D-/3D-Feet-Detection Methods for Semi-Autonomous Powered Wheelchair Navigation
Mid Sweden University, Faculty of Science, Technology and Media, Department of Electronics Design.
Mid Sweden University, Faculty of Science, Technology and Media, Department of Electronics Design. IMMS Institut für Mikroelektronik- und Mechatronik-Systeme gemeinnützige GmbH, Ilmenau, Germany.
Mid Sweden University, Faculty of Science, Technology and Media, Department of Electronics Design. University of Ontario Institute of Technology, Canada.
Mid Sweden University, Faculty of Science, Technology and Media, Department of Electronics Design.
2021 (English)In: Journal of Imaging, ISSN 2313-433X, Vol. 7, no 12, article id 255Article in journal (Refereed) Published
Abstract [en]

Powered wheelchairs have enhanced the mobility and quality of life of people with special needs. The next step in the development of powered wheelchairs is to incorporate sensors and electronic systems for new control applications and capabilities to improve their usability and the safety of their operation, such as obstacle avoidance or autonomous driving. However, autonomous powered wheelchairs require safe navigation in different environments and scenarios, making their development complex. In our research, we propose, instead, to develop contactless control for powered wheelchairs where the position of the caregiver is used as a control reference. Hence, we used a depth camera to recognize the caregiver and measure at the same time their relative distance from the powered wheelchair. In this paper, we compared two different approaches for real-time object recognition using a 3DHOG hand-crafted object descriptor based on a 3D extension of the histogram of oriented gradients (HOG) and a convolutional neural network based on YOLOv4-Tiny. To evaluate both approaches, we constructed Miun-Feet—a custom dataset of images of labeled caregiver’s feet in different scenarios, with backgrounds, objects, and lighting conditions. The experimental results showed that the YOLOv4-Tiny approach outperformed 3DHOG in all the analyzed cases. In addition, the results showed that the recognition accuracy was not improved using the depth channel, enabling the use of a monocular RGB camera only instead of a depth camera and reducing the computational cost and heat dissipation limitations. Hence, the paper proposes an additional method to compute the caregiver’s distance and angle from the Powered Wheelchair (PW) using only the RGB data. This work shows that it is feasible to use the location of the caregiver’s feet as a control signal for the control of a powered wheelchair and that it is possible to use a monocular RGB camera to compute their relative positions.

Place, publisher, year, edition, pages
2021. Vol. 7, no 12, article id 255
Keywords [en]
3D object recognition, YOLO, YOLO-Tiny, 3DHOG, histogram of oriented gradients, ModelNet40, feature descriptor, Intel RealSense, depth camera, wheelchair
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
URN: urn:nbn:se:miun:diva-43838DOI: 10.3390/jimaging7120255ISI: 000737580000001Scopus ID: 2-s2.0-85121390982OAI: oai:DiVA.org:miun-43838DiVA, id: diva2:1614008
Available from: 2021-11-24 Created: 2021-11-24 Last updated: 2025-09-25Bibliographically approved
In thesis
1. Semi-Autonomous Navigation of Powered Wheelchairs: 2D/3D Sensing and Positioning Methods
Open this publication in new window or tab >>Semi-Autonomous Navigation of Powered Wheelchairs: 2D/3D Sensing and Positioning Methods
2021 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Autonomous driving and assistance systems have become a reality for the automotive industry to improve driving safety in the car. Hence, the cars use a variety of sensors, cameras and image processing techniques to measure their surroundings and control their direction, braking and speed for obstacle avoidance or autonomously driving applications.Like the automotive industry, powered wheelchairs also require safety systems to ensure their operation, especially when the user has controlling limitations, but also to develop new applications to improve its usability. One of the applications is focused on developing a new contactless control of a powered wheelchair using the position of a caregiver beside it as a control reference. Contactless control can prevent control errors, but it can also provide better and more equal communication between the wheelchair user and the caregiver

This thesis evaluates the camera requirements for a contactless powered wheelchair control and the 2D/3D image processing techniques for caregiver recognition and position measurement beside the powered wheelchair. The research evaluates the strength and limitations of different depth camera technologies for caregiver feet detection above the ground plane to select the proper camera for the application. Then, a hand-crafted 3D object descriptor is evaluated for caregiver feet recognition and compared with respect to a state-of-the-art deep learning object detector. Results for both methods are good, however, the hand-crafted descriptor suffers from segmentation errors and consequently, their accuracy is lower. After the depth camera and image processing techniques evaluation, results show that it is possible to use only an RGB camera to recognize and measure his or her relative position.

Place, publisher, year, edition, pages
Sundsvall: Mid Sweden University, 2021. p. 64
Series
Mid Sweden University doctoral thesis, ISSN 1652-893X ; 354
Keywords
3D object recognition, YOLO, YOLO-Tiny, 3DHOG, Histogram-of-Oriented-Gradients, ModelNet40, Feature descriptor, Intel RealSense, Depth camera, Wheelchair
National Category
Computer graphics and computer vision
Identifiers
urn:nbn:se:miun:diva-43829 (URN)978-91-89341-32-6 (ISBN)
Public defence
2021-12-09, O102, Mittuniversitetet, Sundsvall, 16:06 (English)
Opponent
Supervisors
Note

Vid tidpunkten för disputationen var följande delarbeten opublicerade: delarbete 5 inskickat.

At the time of the doctoral defence the following papers were unpublished: paper 5 submitted.

Available from: 2021-11-24 Created: 2021-11-23 Last updated: 2025-09-25Bibliographically approved

Open Access in DiVA

fulltext(45238 kB)788 downloads
File information
File name FULLTEXT01.pdfFile size 45238 kBChecksum SHA-512
c36142ba12624e939b6c9ced33309e874c3b00c31b30bd1c651cf3d3254b1e5f1ce4f18fa0d041d606f32c47a33a9d52cc3f03c57544efade8e421acf68cf6cc
Type fulltextMimetype application/pdf

Other links

Publisher's full textScopus

Authority records

Vilar, CristianKrug, SilviaO'Nils, Mattias

Search in DiVA

By author/editor
Vilar, CristianKrug, SilviaO'Nils, Mattias
By organisation
Department of Electronics Design
Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 789 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 458 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf