Drowsiness Detection in Automobile Drivers

The Drowsiness Detection in Automobile Drivers project addresses the critical issue of driver fatigue, a major cause of road accidents globally. Using deep learning techniques, including YOLOv3, VGG19, CNN, and Viola Jones algorithms, the project developed a system capable of predicting driver drowsiness with high accuracy. The system utilizes real-time video feeds from a car's webcam to analyze facial expressions and eye movements, detecting signs of fatigue such as prolonged eye closure and yawning. The YOLOv3 algorithm excels in real-time performance and object detection, while VGG19 and CNN are leveraged for their precision in identifying patterns and features related to drowsiness. The Viola Jones algorithm enhances the system's ability to detect eye states accurately. An application built using these models provides a seamless interface for real-time drowsiness detection, ensuring timely alerts to prevent accidents. This innovative approach significantly enhances road safety by integrating advanced machine learning techniques into practical applications for driver monitoring ​