Real-time remote monitoring of the elderly and hand gesture recognition with computer vision

This research presents a real-time remote monitoring system tailored for the elderly, integrating advanced hand gesture recognition using computer vision technology. The system aims to facilitate remote monitoring of elderly individuals’ activities and well-being, enabling intuitive interaction between the elderly and their caregivers or family members. Leveraging computer vision algorithms, the system accurately detects and interprets hand gestures in real-time. Experimental evaluations demonstrate the feasibility and effectiveness of the proposed approach in practical scenarios. Results suggest the potential of the system to enhance elderly care and promote independent living while improving communication and interaction between elderly individuals and their caregivers. This research aims to study Internet of Things (IoT) technology and other related theories to develop a remote elderly monitoring system using Python language and the Mediapipe library. The system utilizes pose landmark detection to track the positions and orientations of 32 body landmarks and hand landmarks detection to track the positions and orientations of hand landmarks through real-time camera feed. It can accurately detect and classify body postures and hand gestures and display appropriate messages.Testing was conducted in four different environments: low-light simulated environment, medium-light simulated environment, high-light simulated environment, and a varied simulated environment. The results showed that the system achieved an accuracy of at 90% least in classifying body postures and hand gestures in all tested environments. This capability enables accurate monitoring and timely response to elderly individuals, enhancing convenience in their care and ensuring rapid response to their needs.