Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM69090
Volume:
10 |
11 |
Nov 2026
Pages:
1-5
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

Department of Electronics and Computer Engineering
Aditya College of Engineering and Technology

Department of Electronics and Computer Engineering
Aditya College of Engineering and Technology

Department of Electronics and Computer Engineering
Aditya College of Engineering and Technology
Professor & Head Department of Electronics and Communication Engineering
Aditya College of Engineering and Technology Bangalore, India
This study outlines an AI-supported approach to observing the activity of older adults who may spend periods of time without a caregiver nearby. Video from a camera is analysed with computer-vision methods to recognise selected patterns that could need attention, including a possible fall, an atypical body position, extended inactivity, or a noticeable departure from usual activity. A YOLO-based model is proposed for locating a person and, if the chosen model provides the function, estimating human pose. Rather than treating one image as a confirmed event, a decision stage considers detections across time. If a potentially concerning event is identified, the application stores its time and may notify an authorised caregiver with a short clip or image so that the situation can be checked by a person. The application is intended to assist monitoring and communication; it is not a medical diagnostic system. The paper sets out the need, design, workflow, data requirements, notification process, anticipated outcomes, constraints, and a plan for evaluation. No accuracy or performance figures are stated because these must be established through prototype tests.
KEYWORDS
This paper has presented a proposed system for supporting elderly care through camera-based vision analysis, review of activity patterns over time, caregiver messaging, and a dashboard containing event records. A YOLO-based model may be used to locate a person and, when a suitable pose model is selected, estimate body keypoints for posture analysis. A separate decision stage examines the sequence of observations and can raise a candidate alert with an associated video extract. The purpose is to improve caregiver awareness, not to identify or diagnose a medical condition. Further development should include building and checking a dataset collected or used with appropriate consent, evaluating alternative detection and pose-estimation models, refining the analysis of events over time, and measuring the frequency of incorrect alerts. Other work includes implementing secure caregiver access and testing the complete application in realistic indoor settings. The completed project should publish actual measurements and clearly identify the model, dataset, computing hardware, and notification service used.
[1] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779–788. doi: 10.1109/CVPR.2016.91. [2] Ultralytics, “Pose Estimation,” Ultralytics YOLO Documentation. Available: https://docs.ultralytics.com/tasks/pose/ (accessed 14-Nov-2026). [3] Ultralytics, “Ultralytics YOLO Documentation,” Available: https://docs.ultralytics.com/ (accessed 14-Nov-2026).
Publication Date
11/17/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
1
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...