Real-Time Driver Fatigue Detection for Enhanced Road Safety

Sujathakumari B A; Rakshith kumar M; Raghunandana V Mulgund · 2025 · Crossref

DOI: 10.47392/irjaem.2025.0441

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 driver fatigue, a major contributor to road accidents and fatalities globally, particularly in India. The authors highlight that unlike alcohol impairment, drowsiness develops gradually and often goes undetected until a critical incident occurs. While physiological methods like EEG are accurate, they are intrusive and impractical for daily use, and vehicle-based methods like lane departure monitoring often react too late. To bridge this gap, the study proposes a real-time, non-invasive system that combines ocular monitoring with environmental sensing to detect drowsiness and sleep states proactively, aiming to enhance road safety through early warning. The proposed system is built around an Arduino Uno microcontroller, which serves as the central processing unit. It utilizes an infrared (IR) eye-blink sensor to monitor eyelid movements, tracking metrics such as blink frequency and duration to calculate the Percentage of Eyelid Closure (PERCLOS). An alert driver typically exhibits 10–20 blinks per minute, whereas a drowsy driver falls below 10 blinks per minute. Additionally, a photoresistor measures ambient light levels to improve detection accuracy in varying lighting conditions, with low-light readings (below 200 analog units) reinforcing the likelihood of a sleep state. The system’s logic continuously compares sensor inputs against defined thresholds; if the eye remains closed for more than three seconds, the system classifies the driver as "Sleeping." A 200 ms delay between sensor readings is implemented to minimize false positives caused by noise. The system was validated through hardware testing in simulated scenarios. Results demonstrated that the system accurately distinguishes between awake, drowsy, and sleeping states. When the driver’s eyes were open, the system correctly displayed "Awake" on the LCD and kept the alert LED off, confirming the sensor’s ability to reset the internal timer and avoid false alarms. Conversely, when eyes were closed beyond the three-second threshold, the system immediately triggered a red LED and a buzzer, while the serial monitor displayed "Sleeping." The integration of visual and audible alerts ensured immediate driver notification. The hardware-software integration showed consistent response times within milliseconds, with minimal error rates in distinguishing states. The study concludes that the proposed Arduino-based system is a practical, scalable, and reliable solution for real-world driver safety applications. By leveraging non-invasive behavioral indicators and environmental data, the system offers an improvement over existing methods in terms of accuracy and real-time applicability. The findings validate the effectiveness of combining IR-based monitoring with light-level detection for robust drowsiness recognition, suggesting potential for integration into driving assistance systems or wearable health devices to prevent fatigue-related accidents.

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
• microsleep
• dms validation
• drowsiness
• gaze based attention detection
• truck driver fatigue

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, validation psychometrics