Real-Time Fatigue Analysis of Driver through Iris Recognition

K, Gopalakrishna; S.A., Hariprasad · 2017 · Crossref

DOI: 10.11591/ijece.v7i6.pp3306-3312

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 primary cause of road accidents, which account for approximately 77.5% of all traffic incidents in India. The authors motivate their work by highlighting the limitations of existing fatigue detection methods, such as eye-blink sensors, electrocardiograms (ECG), and electroencephalograms (EEG). These conventional approaches are often invasive, complicated, uncomfortable for drivers, or prone to false alarms. To overcome these drawbacks, the study proposes a non-invasive, real-time fatigue analysis system based on iris recognition, aiming to provide a more accurate and convenient solution for intelligent transport systems. The proposed method utilizes a hardware setup centered on an ARM 7-based LPC2148 microprocessor, interfaced with a web camera equipped with infrared illuminators, an LCD display, and a buzzer for alerts. The system captures facial images and processes them through a multi-stage algorithm. First, face detection is performed, followed by eye region localization using geometric assumptions relative to the face width. Iris detection is achieved using a Circular Hough Transform, which identifies the iris boundary by analyzing edge maps generated via Canny edge detection. The system calculates the visibility ratio of the iris; if the ratio drops below 60%, the system analyzes subsequent frames to distinguish between a normal blink and sustained drowsiness. If the eyes remain closed continuously for four seconds, the microprocessor triggers an alarm to alert the driver. The experimental results demonstrate that the system can effectively differentiate between open and closed eye states under various lighting conditions. The authors report that the proposed iris recognition method achieves an accuracy of approximately 80%, which they claim is superior to existing eye-blink sensor methods. The system successfully identifies drowsiness when eyes are closed for the specified duration, issuing a warning signal without significant computational delay. The study validates the approach through test videos and frame databases, confirming the robustness of the image processing algorithms in detecting fatigue symptoms. The significance of this work lies in its potential to reduce accidents caused by drowsy driving through a low-complexity, non-invasive monitoring system. By leveraging iris visibility rather than complex physiological signals, the method offers a practical solution for real-time implementation in vehicles. The authors conclude that the system provides reliable detection of driver fatigue, helping to keep drivers awake and thereby enhancing road safety. Additionally, the system’s design allows for potential secondary applications in driver security and identification.

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
• 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