Detection and Alerting of Driver Drowsiness using a Machine Learning Approach

Driver fatigue is a significant causative factor to accidents worldwide. To signify this issue, we have created a prototype system for detecting tiredness in drivers. This system monitors the driver’s level of drowsiness by tracking their eye movements and issuing an alert if prolonged eye closure is detected. Our focus is on improving driver safety while maintaining a non-intrusive approach. In this project, we utilize the Haarcascade library in OpenCV to detect facial features and accurately detect eye blinks. Through research in the field of computer engineering, we aim to prevent accidents caused by fatigued or sleepy drivers. The project discusses various strategies, their outcomes, and the implemented fixes. By providing insights into the system’s functionality and potential adjustments, the article aims to enhance the system’s overall effectiveness. Furthermore, the project offers observations and recommendations for future optimization in this field, with the goal of maximizing utility at a lower cost and promoting safer roads. The article provides a practical understanding of the system’s operation and serves as a valuable resource for advancing research and development efforts in driver drowsiness detection.