Development of a Driver’s Activity Monitoring System Using Facial and Heartbeat Parameters

Abstract

Driver fatigue and distraction significantly contributes to road accidents worldwide, as such monitoring drivers' physical and mental states could mitigate these risks of fatigue and natural distraction. The existing driver monitoring systems are often complex, and inaccessible to drivers. The real-time driver monitoring device presented in this study is intended to improve road safety by identifying indicators of driver weariness and drowsiness. An ESP32-CAM module is integrated into the system to record live video, which is then processed using computer vision algorithms to identify signs of exhaustion, such as yawning and eye closing. More so, a MAX30102 heart rate sensor was incorporated with an Arduino Nano to monitor the driver's pulse rate, providing a complementary method for identifying signs of drowsiness. The system was able to process the data from both video and sensor inputs to issue alerts whenever any fatigue scenarios were detected. The system was tested successfully in a simulated scenario, demonstrating its effectiveness in detecting signs of drowsiness and alerting the driver accordingly. The system accurately detected driver fatigue to be 87% accurate and drowsiness to be 79% accurate through facial expression analysis and the designed web app effectively alerted drivers of any of these options. The heart rate of the driver was also measured, and the monitoring results shows a strong correlation with driver stress levels to about 66%. This driver monitoring system demonstrated feasibility and effectiveness in detecting driver fatigue and distraction. Its simplicity, and high accuracy makes it suitable for widespread adoption while further research prospect should consider artificial intelligent approach for optimization.