Fall Detection for Elderly People Using Computer Vision and Artificial Intelligence

Fuente: Zenodo
Guardado en:
Detalles Bibliográficos
Autores principales: Tanvi Rajesh Misal, Lekha Gadpade
Formato: Recurso digital
Publicado: Zenodo 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866901236000751616
author Tanvi Rajesh Misal
Lekha Gadpade
author_facet Tanvi Rajesh Misal
Lekha Gadpade
contents <p>Among the elderly in the world, falls are considered to be the major causes of injuries, hospitalization, and death. Quick response can greatly decrease the number of medical complications as the fall incidents can be detected in time. Conventional fall detection systems use wearable sensors that are usually awkward or obtrusive and not always convenient to use in their entirety. The proposed research paper suggests a non-invasive fall detection system of the elderly using data provided by cameras and implements MediaPipe-based human pose estimation, OpenCV-velocity of movement, angles and temporal shifts of the posture to enhance the rate of detection and false alarms. The system will incorporate an automated alert system to alert the caregivers or medical staff members in real time when a fall takes place. Experimental evidence shows that the method proposed is highly accurate, sensitive and robust in different environmental conditions which make it applicable in real world application of smart home and assisted living facility.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_19667695
institution Zenodo
language
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle Fall Detection for Elderly People Using Computer Vision and Artificial Intelligence
Tanvi Rajesh Misal
Lekha Gadpade
Fall Detection, Computer Vision, Artificial Intelligence, MediaPipe, OpenCV, Elderly Care, Pose Estimation, Smart Healthcare
<p>Among the elderly in the world, falls are considered to be the major causes of injuries, hospitalization, and death. Quick response can greatly decrease the number of medical complications as the fall incidents can be detected in time. Conventional fall detection systems use wearable sensors that are usually awkward or obtrusive and not always convenient to use in their entirety. The proposed research paper suggests a non-invasive fall detection system of the elderly using data provided by cameras and implements MediaPipe-based human pose estimation, OpenCV-velocity of movement, angles and temporal shifts of the posture to enhance the rate of detection and false alarms. The system will incorporate an automated alert system to alert the caregivers or medical staff members in real time when a fall takes place. Experimental evidence shows that the method proposed is highly accurate, sensitive and robust in different environmental conditions which make it applicable in real world application of smart home and assisted living facility.</p>
title Fall Detection for Elderly People Using Computer Vision and Artificial Intelligence
topic Fall Detection, Computer Vision, Artificial Intelligence, MediaPipe, OpenCV, Elderly Care, Pose Estimation, Smart Healthcare
url https://doi.org/10.5281/zenodo.19667695