AI-Powered Deep Learning Model for Driver Behavior Analysis in Intelligent Cyber-Physical Transport Systems

This paper aims to enhance road safety and traffic efficiency by leveraging deep learning to monitor driver behavior within Public Transit Optimization. It focuses on detecting signs of fatigue and distraction by analyzing behaviors such as eye closure, yawning, and overall alertness. This study presents a novel deep learning-based approach for detecting driver drowsiness using YOLOv8, a state-of-the-art object detection model. The system is designed to process both live video streams and recorded footage, accurately identifying key indicators of fatigue, including prolonged eye closure and yawning. Through advanced data preprocessing, model optimization, and the integration of real-time alert mechanisms, the system delivers rapid responses to potential drowsiness, thereby enhancing road safety. By utilizing deep learning techniques, the model aims to detect, interpret, and predict driving behaviors and anomalies that may indicate hazardous or inefficient driving. This system represents a crucial advancement in intelligent and adaptive transportation technologies in vehicles.