Abstract: This research presents the design and implementation of a cyber-physical system based on IoT technologies and computer vision aimed at the early detection of drowsiness in land transport drivers with applications in the field of occupational e-Health. The system integrates a camera, a facial analysis model in Python using Mediapipe and OpenCV, and embedded devices such as the ESP32 for continuous monitoring of visual signals to detect eye closure or yawning. The proposed architecture includes real-time communication via the MQTT protocol, triggering local alerts (sound and light), storing visual evidence, and sending notifications to Telegram. In addition, a remote monitoring interface using Node-RED and a mobile application in Kotlin for local event visualization are implemented. Functional tests were conducted, demonstrating the viability of the system as a preventive tool for reducing fatigue-related risks in the context of monitoring ground vehicle drivers. The solution's modular and portable approach allows for adaptation to other biomedical monitoring environments and suggests a low-cost alternative for digital health applications.

Keywords: Telecare, Work Fatigue, OpenCV, Mediapipe, Android, Facial analysis, Driver monitoring

Pages: 85-94