Integrating Attention Mechanisms and ResNet-50 For Enhanced Driver Sleepiness Detection

One of the main causes of traffic accidents is driver exhaustion, making the development of advanced systems crucial for real-time monitoring and preventive measures. This work presents a novel framework for intelligent information aggregation, enhancing decision-making in sleepiness detection by leveraging attention mechanisms and ResNet-50, a state-of-the-art deep convolutional neural network. The proposed system reliably classifies driver states by integrating logical inference with visual data elements such as head movements, eye closure patterns, and face recognition Experimental evaluations on the Driver Drowsiness Dataset (DDD) demonstrate the model’s effectiveness, achieving an accuracy of 93.5%, a precision of 94.2%, a recall of 92.7%, and an F1-score of 93.4% These results highlight the synergy between artificial intelligence and attention mechanisms in improving classification performance. This research provides a robust and scalable foundation for AI-driven decision-making, contributing to safer and more intelligent transportation systems.