Description of Invention
Driver drowsiness is one of the leading causes of road accidents, and early detection is critical to improving road safety. This study explores and compares the performance of three deep learning models YOLO (You Only Look Once), Vision Transformer (ViT), and Swin Transformer in driver drowsiness detection. Each model is evaluated based on its ability to classify “awake” vs “drowsy” states using a standardized image dataset. Performance is analysed through key metrics such as accuracy, precision, recall, and F1 score. The study provides insight into the strengths and weaknesses of CNN-based and transformer-based architectures for safety critical applications.