Multimodal Fall Detection Wearable Sensor for Elderly People
Umme Tanha Khanum, Md. Tanjeelur Rahman Labib, Shakil Ahmed Shawon, Fariah Mahzabeen · 2025
This paper presents the design and implementation of a fall detection system using the Internet of Things (IoT) techniques to monitor the elderly people. The system integrates multi-modal data analysis, which combines inertia measurement with camera-based visual verification to accurately classify movements as general activities or falls. The sophisticated sensor algorithm is applied for interpreting and handling the data, while video test clips are experimentally applied via image processing techniques. Experimental results confirm the efficiency of the system with sensor detection accuracy as 97.33%, camera analysis 94%, and combining both giving detection accuracy as 96.67%, with a remarkable 100% specificity and precision. Upon a detected fall, the system would alert automatically designated caregivers or emergency service providers. The study describes system architecture including sensor and camera integration, algorithm development and IoT connectivity. Results highlight the strength and reliability of the system, which demonstrate their ability to improve the safety and welfare of elderly individuals through timely detection and intervention.