Drowsiness Detection and Real-Time System for Monitoring Driver Vigilance Using Raspberry Pi

Naidu, Marisarla Ganesh Bhagavan · 2025 · Crossref

DOI: 10.22214/ijraset.2025.67319

archive: archived pipeline: cataloged

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

Summary

This paper addresses the critical road safety issue of driver drowsiness, a leading cause of traffic accidents resulting from impaired reaction time and attention. The study aims to develop a non-intrusive, real-time computer vision system to monitor driver vigilance, moving away from existing solutions that rely on intrusive wearable sensors or complex physiological monitoring. The proposed system utilizes a Raspberry Pi 4 board paired with a Pi Camera to capture and analyze driver facial features in real-time, offering a more convenient and scalable alternative to traditional Arduino-based or sensor-heavy approaches. The experimental design centers on a hardware prototype comprising a Raspberry Pi 4, a Pi Camera, a 64GB SanDisk SD card, a USB-C power adapter, a buzzer for alerts, and jumper wires for connectivity. The software implementation employs image processing techniques, specifically using Haar Cascade or Dlib facial landmark detection to identify the driver’s face and eyes. The system calculates six key visual parameters to characterize vigilance: Percent Eye Closure (PERCLOS), eye closure duration, blink frequency, nodding frequency, face position, and fixed gaze. These parameters are derived using the Eye Aspect Ratio (EAR) technique and are fused through a fuzzy classifier to infer the driver’s level of inattentiveness. When drowsiness is detected, the system triggers an audible alert via a buzzer and optionally displays status on an LCD screen. In advanced configurations, the system can interface with vehicle controls to slow the vehicle if the driver remains unresponsive. The results indicate that the system successfully detects facial features such as eye closure and head movement with minimal delay. Experimental testing was conducted under various conditions, including initial states, wearing spectacles, dim lighting, and microsleep scenarios. The system achieved a reported accuracy of 90% in recognizing drowsy states. The authors note that while performance may vary slightly based on environmental factors, the integration of machine learning algorithms enhances accuracy and reduces false detections compared to single-parameter methods. The system operates continuously, stopping alerts when the driver regains alertness and resuming them if drowsiness persists. The significance of this work lies in its demonstration of a cost-effective, non-intrusive solution for real-time driver monitoring. By replacing wearable sensors with a camera-based approach, the system improves user comfort and practicality for long-term use. The use of the Raspberry Pi provides sufficient processing power for real-time image analysis, making the solution scalable for broader automotive applications. The study concludes that such systems can significantly enhance road safety by proactively preventing fatigue-related accidents, with future potential for deeper integration into vehicle control mechanisms.

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 qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.

Topics

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

• drowsiness detection algorithms

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).

• Methodological Resource: tool software