Road Safety by Detecting Drowsiness while Driving using Machine Learning

Pardhi, Piyush · 2025 · Crossref

DOI: 10.22214/ijraset.2025.72725

archive: archived pipeline: cataloged verified

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

Summary

This paper addresses the critical issue of road safety by proposing a real-time driver drowsiness detection system. Driver fatigue is identified as a major contributor to accidents, impairing reaction times and decision-making, particularly during long-distance travel. While traditional awareness campaigns have limited effectiveness, the authors leverage advancements in computer vision and embedded systems to create a non-intrusive, intelligent driver assistance solution. The primary objective is to design an efficient system that monitors driver alertness and issues immediate warnings to prevent accidents before they occur. The system is implemented using Python, integrating libraries such as OpenCV for image processing, dlib for facial landmark detection, and pygame for audio alerts. It operates via a standard webcam or camera module, tracking facial landmarks to analyze specific indicators of drowsiness, including eye aspect ratio (EAR), blinking behavior, and head pose estimation. The hardware requirements are modest, suggesting compatibility with dual-core processors, 4–8 GB of RAM, and embedded computing boards like the Raspberry Pi or NVIDIA Jetson Nano. When the system detects signs of fatigue, such as prolonged eye closure or slow blinking, it triggers an audible alarm to alert the driver. Experimental results indicate that the system performs reliably under normal lighting conditions, offering a practical and cost-effective method for improving road safety. The authors highlight several advantages, including real-time monitoring with minimal delay, non-intrusive operation without wearables, and scalability for various vehicle types. However, the study acknowledges limitations, noting that detection accuracy may decrease in poor lighting, with the use of sunglasses or headgear, or during extreme head movements. Privacy concerns regarding continuous monitoring are also noted as a potential disadvantage. The significance of this work lies in its potential application across personal vehicles, public transport, and fleet logistics, where it can reduce accident risks and liability. The paper outlines future scope for enhancing the system through IoT integration for remote monitoring, multimodal inputs combining eye tracking with heart rate or steering behavior data, and the adoption of deep learning techniques like Convolutional Neural Networks (CNNs) to improve robustness across different lighting conditions and facial features. Ultimately, the study demonstrates that computer vision-based drowsiness detection is a viable, affordable technology for enhancing intelligent transportation 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
• distraction detection algorithms
• drowsiness

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