DOZR is a real-time drowsiness detection system designed to improve road safety by identifying drowsy drivers using YOLOv5 and OpenCV. It processes webcam input to detect signs of drowsiness and triggers alerts to prevent accidents.

Built with PyTorch, DOZR supports custom dataset training and is compatible with both CPU and GPU for efficient inference. With high precision and recall, it offers a reliable solution for drivers, fleet managers, and researchers aiming to enhance safety and study drowsiness patterns.

DOZR leverages YOLOv5 for real-time object detection, integrated with PyTorch for deep learning and OpenCV for computer vision tasks. The system processes webcam or video input to detect drowsy states, using annotated datasets created with LabelImg.

The architecture supports custom training with user-provided datasets and runs efficiently on both CPU and GPU. With a modular design, DOZR is extensible for future features like dashboards and mobile alerts, making it a versatile tool for road safety applications.

DOZR is a powerful tool for enhancing road safety through AI-driven drowsiness detection. Explore the project on GitHub to try it out, contribute, or integrate it into your safety solutions.