EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Reduction of Vision-Based Models for Fall Detection

Asier Garmendia-Orbegozo iD, Miguel Angel Anton iD, Jose David Nuñez-Gonzalez iD

Sensors · 2024

DOI: 10.3390/s24227256 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2024. Verfasst von Asier Garmendia-Orbegozo und 2 weiteren Autor:innen. Beteiligte Einrichtung: Fundación Tecnalia Research & Innovation, Basque Research and Technology Alliance (BRTA), 20009 San Sebastian, Spain. Hauptthemen laut MEDLINE: Accidental Falls, Neural Networks, Computer. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Diputación Foral de Gipuzkoa, Gipuzkoa Provincial Council.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Due to the limitations that falls have on humans, early detection of these becomes essential to avoid further damage. In many applications, various technologies are used to acquire accurate information from individuals such as wearable sensors, environmental sensors or cameras, but all of these require high computational resources in many cases, delaying the response of the entire system. The complexity of the models used to process the input data and detect these activities makes them almost impossible to complete on devices with limited resources, which are the ones that could offer an immediate response avoiding unnecessary communications between sensors and centralized computing centers. In this work, we chose to reduce the models to detect falls using images as input data. We proceeded to use image sequences as video frames, using data from two open source datasets, and we applied the Sparse Low Rank Method to reduce certain layers of the Convolutional Neural Networks that were the backbone of the models. Additionally, we chose to replace a convolutional block with Long Short Term Memory to consider the latest updates of these data sequences. The results showed that performance was maintained decently while significantly reducing the parameter size of the resulting models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Accidental FallsNeural Networks, ComputerHumansAlgorithmsImage Processing, Computer-AssistedWearable Electronic Devices

Stichwörter der Autor:innen

CNNLSTMfall detectionpruning

Autor:innen und Institutionen

  • Fundación Tecnalia Research & Innovation, Basque Research and Technology Alliance (BRTA), 20009 San Sebastian, Spain.
  • Department of Applied Mathematics, University of the Basque Country UPV/EHU, 20600 Eibar, Spain.

ORCID-Profile: Asier Garmendia-Orbegozo, Miguel Angel Anton, Jose David Nuñez-Gonzalez

Förderung

  • Diputación Foral de Gipuzkoa
  • Gipuzkoa Provincial Council

Zitieren

APA

Garmendia-Orbegozo, A., Anton, M. A., & Nuñez-Gonzalez, J. D. (2024). Reduction of Vision-Based Models for Fall Detection. Sensors. https://doi.org/10.3390/s24227256

Vancouver

Garmendia-Orbegozo A, Anton MA, Nuñez-Gonzalez JD. Reduction of Vision-Based Models for Fall Detection. Sensors. 2024. doi:10.3390/s24227256

BibTeX
@article{GarmendiaOrbegozo2024,
  author = {Garmendia-Orbegozo, Asier and Anton, Miguel Angel and Nuñez-Gonzalez, Jose David},
  title = {Reduction of Vision-Based Models for Fall Detection},
  journal = {Sensors},
  year = {2024},
  doi = {10.3390/s24227256}
}
Kostenlos starten →