article · Sensors
Driver fatigue remains a leading cause of vehicle collisions and fatalities globally. Existing detection tools are frequently intrusive or slow to respond to onset sleepiness. To resolve these issues, a non-intrusive system uses facial analysis paired with machine learning to identify drowsy drivers in real time. The approach was systematically evaluated across three widely recognised public datasets: NTHUDDD, YawDD, and UTA-RLDD. Testing encompassed traditional classifiers, deep neural networks, and computer vision architectures, including K-Nearest Neighbours, support vector machines, convolutional networks, and several object detection models. The K-Nearest Neighbours model achieved 98.89 per cent accuracy on the UTA-RLDD dataset, while the YOLOv5 and YOLOv8 vision models delivered perfect precision and recall scores of 100 per cent. By accurately detecting physical signs of tiredness, the system offers an effective basis for issuing proactive, real-time safety warnings.
Drowsy driving contributes to thousands of fatal road accidents each year, making timely intervention essential. Developing non-intrusive monitoring tools that accurately detect fatigue without hindering the driver allows safety systems to issue prompt warnings. This research demonstrates that vision-based machine learning can achieve near-perfect detection accuracy, helping prevent fatigue-related crashes and save lives on the road.
The system shows clear utility for automotive manufacturers, transport fleet operators, and in-cabin safety technology developers seeking non-intrusive monitoring solutions. By validating models such as YOLOv5, YOLOv8, and K-Nearest Neighbours on public datasets, the research demonstrates tested algorithmic feasibility for real-time proactive alert systems. However, because testing relied on curated public datasets rather than in-vehicle trials, the technology remains at an applied research stage requiring embedded hardware validation before market deployment.
AI-generated from the published abstract. Always read the original work before citing.
Drowsy driving poses a significant challenge to road safety worldwide, contributing to thousands of accidents and fatalities annually. Despite advancements in driver drowsiness detection (DDD) systems, many existing methods face limitations such as intrusiveness and delayed reaction times. This research addresses these gaps by leveraging facial analysis and state-of-the-art machine learning techniques to develop a real-time, non-intrusive DDD system. A distinctive aspect of this research is its systematic assessment of various machine and deep learning algorithms across three pivotal public datasets, the NTHUDDD, YawDD, and UTA-RLDD, known for their widespread use in drowsiness detection studies. Our evaluation covered techniques including the K-Nearest Neighbors (KNNs), support vector machines (SVMs), convolutional neural networks (CNNs), and advanced computer vision (CV) models such as YOLOv5, YOLOv8, and Faster R-CNN. Notably, the KNNs classifier reported the highest accuracy of 98.89%, a precision of 99.27%, and an F1 score of 98.86% on the UTA-RLDD. Among the CV methods, YOLOv5 and YOLOv8 demonstrated exceptional performance, achieving 100% precision and recall with mAP@0.5 values of 99.5% on the UTA-RLDD. In contrast, Faster R-CNN showed an accuracy of 81.0% and a precision of 63.4% on the same dataset. These results demonstrate the potential of our system to significantly enhance road safety by providing proactive alerts in real time.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3390/s25030812
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
