Driver Drowsiness Detection System

Kushwaha, Abhishek · 2026 · Crossref

DOI: 10.22214/ijraset.2026.81391

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

Driver drowsiness is a critical contributor to road accidents, causing significant fatalities and economic losses globally. This paper addresses the need for effective, non-intrusive systems to detect fatigue in real-time, motivated by statistics indicating that human error accounts for 94% of accidents and drowsy driving causes nearly 100,000 crashes annually in the U.S. alone. The authors propose a web-based Driver Drowsiness Detection System designed to monitor alertness and prevent fatigue-related incidents through immediate alerts. The system utilizes computer vision and machine learning techniques, specifically OpenCV and Google’s MediaPipe Face Mesh, to process real-time webcam video feeds. It extracts precise facial landmarks to calculate three key metrics: Eye Aspect Ratio (EAR) for blink rate analysis, Mouth Aspect Ratio (MAR) for yawning frequency detection, and head pose estimation (pitch and yaw) to identify distracted driving behaviors. Unlike traditional standalone scripts, this solution is deployed as a full-stack web application built with React and Flask. It integrates a Firebase database to provide a secure user dashboard, allowing drivers and fleet managers to track historical alerts and driving statistics. The paper also reviews existing drowsiness evaluation scales, such as the Karolinska Sleepiness Scale and the Wierwille and Ellsworth five-level scale, to contextualize the detection of physiological signs like micro-sleeps and frequent blinking. The study compares geometric-based detection methods with deep learning approaches. While deep learning models like Convolutional Neural Networks (CNNs) offer high accuracy, they require substantial computational resources and training data. In contrast, the proposed system relies on rule-based geometric thresholding using EAR and MAR, which provides efficient real-time performance without heavy processing demands. The system triggers alarms or visual warnings when specific thresholds for eye closure, yawning, or head deviation are exceeded. The significance of this work lies in its provision of a cost-effective, scalable, and accessible solution for improving road safety. By leveraging widely available hardware like webcams and standard libraries, the system offers a practical alternative to expensive embedded solutions. The integration of a web-based dashboard enhances its utility for both individual drivers and transportation industries by enabling long-term monitoring of fatigue patterns. The authors conclude that this approach effectively balances accuracy and efficiency, offering a viable tool for reducing drowsy driving incidents with potential for integration into modern vehicle safety systems.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• drowsiness detection algorithms
• drowsiness
• distraction detection algorithms
• microsleep
• dms validation
• gaze based attention detection

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: physiological data
• Methodological Resource: tool software, measurement protocol