EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

An IoT-Based Real-Time Human Fall Detection and Notification System with Instance Segmentation Deep Learning

Wiwat Su-hren1, Tanawat Srirugsa, Saowanee Singsarothai, Supachai Kaewpoung, Tawat Chuchit

ASEAN Journal of Scientific and Technological Reports · 27.02.2026 · 29 (3) · e260412

DOI: 10.55164/ajstr.v29i3.260412 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ASEAN Journal of Scientific and Technological Reports, online erschienen am 27.02.2026. Verfasst von Wiwat Su-hren1 und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This research presents a person fall detection system based on deep learning using the Instance Image Segmentation technique with the YOLOv11-seg model, which achieves high speed and accuracy in object detection. The developed system aims to individually detect a person's posture in an image, enabling accurate analysis of their falling posture characteristics. And there is a notification in the application to the administrator or relevant person within 10 seconds, allowing quick help. Using a dataset of 10,169 images, 8622 for training, 994 for inspection, and 553 for testing (with a ratio of 85:10:5 for training, inspection, and testing). The system performs impressively by using all 3 YOLOv11-seg models: YOLOv11s-seg, YOLOv11n-seg, and YOLOv11m-seg. The training dataset showed excellent performance with a YOLOv11s-seg model, achieving a precision of 0.966, a recall of 0.910, and an F1 Score of 0.88. The results show that the developed system can detect falls and issue real-time alerts via an IoT-based framework. It improves safety and reduces the risk for the elderly or patients at risk of falling.

Zitieren

APA

Su-hren1, W., Srirugsa, T., Singsarothai, S., Kaewpoung, S., & Chuchit, T. (2026). An IoT-Based Real-Time Human Fall Detection and Notification System with Instance Segmentation Deep Learning. ASEAN Journal of Scientific and Technological Reports, 29(3), e260412. https://doi.org/10.55164/ajstr.v29i3.260412

Vancouver

Su-hren1 W, Srirugsa T, Singsarothai S, Kaewpoung S, Chuchit T. An IoT-Based Real-Time Human Fall Detection and Notification System with Instance Segmentation Deep Learning. ASEAN Journal of Scientific and Technological Reports. 2026;29(3):e260412. doi:10.55164/ajstr.v29i3.260412

BibTeX
@article{Suhren2026,
  author = {Su-hren1, Wiwat and Srirugsa, Tanawat and Singsarothai, Saowanee and Kaewpoung, Supachai and Chuchit, Tawat},
  title = {An IoT-Based Real-Time Human Fall Detection and Notification System with Instance Segmentation Deep Learning},
  journal = {ASEAN Journal of Scientific and Technological Reports},
  year = {2026},
  volume = {29},
  number = {3},
  pages = {e260412},
  doi = {10.55164/ajstr.v29i3.260412}
}
Kostenlos starten →