Real-time drowsiness detection using computer vision

Drowsiness poses a significant risk across multiple domains, including transportation, learning, healthcare, and industrial sectors. Addressing this critical issue requires the development of an effective drowsiness detection system paired with robust alertness assessment mechanisms. In this project, we utilize cutting-edge technologies such as machine learning, computer vision, and signal processing to offer real-time detection and mitigation of drowsiness-related risks. Webcam offers a costeffective means for real-time monitoring and detecting drowsiness. Our approach introduces an innovative method focusing on the classification of eye states using transfer learning and computer vision techniques. Leveraging the MRL Eye Dataset, our method trains a system to identify the eye state of a driver through webcam-based monitoring. The proposed system demonstrates robustness in detecting drowsiness, accommodating both eyeglass wearers and varying lighting conditions, with an impressive accuracy rate of 97% and low computational complexity. This project represents a significant advancement in drowsiness detection technology, providing a real-time, cost-effective, and accurate solution that can be seamlessly integrated into vehicles to promote safer transportation for all. Moreover, our solution is adaptable and customizable to meet the unique requirements of various applications and industries. Through collaboration with partners and stakeholders, we aim to deploy our drowsiness detection technology in commercial vehicles, healthcare facilities, and industrial workplaces, thereby advancing safety and efficiency on a global scale.