Abstract
This paper presents methods for developing an intelligent driver fatigue monitoring system using machine learning techniques. Driver fatigue is a major cause of traffic accidents, particularly during long routes and night shifts. The proposed model uses the eye aspect ratio and a support vector machine classifier to detect blinks and other fatigue indicators in real time. Special attention is given to the model's robustness under varying lighting conditions and head orientations, improving reliability in challenging operational environments. The system has been tested on video sequences with different lighting and head poses, demonstrating high accuracy and minimal false detections. The results indicate that the proposed approach is suitable for integration into intelligent transportation systems, providing a practical solution for continuous driver monitoring. The key advantages of the system are its real-time performance and resilience to changing conditions, which significantly enhance the precision and dependability of fatigue detection.