EUVIMEDEUVIMED entdecken
Crossref

Unobtrusive High-Generalization Fall Detection: A Domain-Generalization Framework

Yicheng Yao, Hao Zhang, Zhongrui Bai, Pan Xia, Changyu Liu, Fanglin Geng, Lidong Du, Xianxiang Chen, Zhenfeng Li, Peng Wang, Huadong Zhu, Yecheng Liu und 1 weitere

IEEE Sensors Journal · 01.10.2024 · 24 (19) · 31512-31523

DOI: 10.1109/jsen.2024.3443532 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 01.10.2024. Verfasst von Yicheng Yao und 12 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 55 Literaturangaben. Gefördert von: National Key Research and Development Project, CAMS Innovation Fund for Medical Sciences, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

It is essential to find and treat the elderly in time after they fall. A millimeter-wave radar sensor not only solves the problem of invasion of privacy caused by the camera but also avoids letting patients wear wearable devices, which has received more and more attention in the field of fall detection and motion recognition. Due to the different modulations of radar signals by different users and environments, the performance of the fall detection model is inevitably reduced. In this article, we propose a highly generalized radar-based fall detection framework to improve the generalization of the fall detection system in the new domain. We use a feature extractor with a ResNet structure constructed from 3-D convolution to extract features of the signal and a multilayer perceptron as a classifier for fall detection. We measure the cross-domain cross entropy of the features by setting subclassifiers for each domain to measure the cross-domain invariance of the features, enabling the feature extractor to extract domain-invariant features. We propose a specific data augmentation technology for radar time-series signals, which is combined with the proposed model to achieve more generalized fall detection. We have conducted sufficient experiments to explore the relationship between the number of users and domains and the model’s generalization. The proposed method achieves an ${F}1$ score of 0.9747 on the proposed dataset containing ten scenarios, 21 subjects, 52 nonfalling actions and 12 falling actions, and an ${F}1$ score of 0.9949 on the public dataset, which shows superior effectiveness.

Förderung

  • National Key Research and Development Project
  • CAMS Innovation Fund for Medical Sciences
  • National Natural Science Foundation of China
  • National High Level Hospital Clinical Research Funding

Zitieren

APA

Yao, Y., Zhang, H., Bai, Z., Xia, P., Liu, C., Geng, F., Du, L., Chen, X., Li, Z., Wang, P., Zhu, H., Liu, Y., & Fang, Z. (2024). Unobtrusive High-Generalization Fall Detection: A Domain-Generalization Framework. IEEE Sensors Journal, 24(19), 31512-31523. https://doi.org/10.1109/jsen.2024.3443532

Vancouver

Yao Y, Zhang H, Bai Z, Xia P, Liu C, Geng F, et al. Unobtrusive High-Generalization Fall Detection: A Domain-Generalization Framework. IEEE Sensors Journal. 2024;24(19):31512-31523. doi:10.1109/jsen.2024.3443532

BibTeX
@article{Yao2024,
  author = {Yao, Yicheng and Zhang, Hao and Bai, Zhongrui and Xia, Pan and Liu, Changyu and Geng, Fanglin and Du, Lidong and Chen, Xianxiang and Li, Zhenfeng and Wang, Peng and Zhu, Huadong and Liu, Yecheng and Fang, Zhen},
  title = {Unobtrusive High-Generalization Fall Detection: A Domain-Generalization Framework},
  journal = {IEEE Sensors Journal},
  year = {2024},
  volume = {24},
  number = {19},
  pages = {31512--31523},
  doi = {10.1109/jsen.2024.3443532}
}
Kostenlos starten →