A Survey on State-of-the-Art Drowsiness Detection Techniques

Ramzan, Muhammad; Khan, Hikmat Ullah; Awan, Shahid Mahmood; Ismail, Amina; Ilyas, Mahwish; Mahmood, Ahsan · 2019 · OpenAlex-citations

DOI: 10.1109/access.2019.2914373

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 presents a comprehensive survey of state-of-the-art techniques for detecting driver drowsiness, motivated by the significant safety risks and economic losses associated with fatigue-related road accidents. Citing data from the U.S. National Highway Traffic Safety Administration, the authors highlight that drowsy driving causes approximately 100,000 accidents and over 1,500 deaths annually. The study aims to categorize existing detection methods, evaluate their effectiveness, and identify research gaps to guide future development in automated fatigue detection systems. The authors conducted a systematic literature review, initially identifying 1,020 research papers through searches on platforms like IEEE Explore, ACM, and Google Scholar using keywords related to drowsiness, fatigue, and specific detection modalities. Through a multi-stage selection process involving title, abstract, and full-text analysis, 41 relevant papers were selected for detailed evaluation. The survey classifies drowsiness detection techniques into three primary categories: behavioral, vehicular, and physiological parameters. Additionally, the paper reviews supervised learning classification methods, such as Support Vector Machines (SVM) and neural networks, used to process data from these categories. The findings detail specific implementations within each category. Behavioral techniques, which are non-invasive, utilize computer vision to monitor eye closure ratios (PERCLOS), yawning, and head posture. For instance, systems using eye tracking and dynamic template matching achieved accuracy rates up to 99.01%, while methods combining eye closure and head posture estimation reached 80% accuracy. Vehicular techniques analyze driving patterns, such as steering wheel angle variability and lane deviation, though the authors note these can be influenced by external factors like road geometry and weather. Physiological techniques monitor biological signals like heart rate and body temperature, offering high reliability but requiring invasive sensors. The survey also highlights hybrid approaches that combine multiple data sources to improve detection robustness. The significance of this work lies in its structured comparison of detection methodologies, providing a clear overview of the pros and cons of each approach. The authors conclude that while behavioral methods are widely adopted due to their non-invasive nature, they are susceptible to environmental variables like lighting conditions. Vehicular methods offer indirect detection but lack specificity, whereas physiological methods provide accurate data but face practical implementation challenges. The paper serves as a foundational resource for researchers, outlining current limitations and suggesting potential future work in developing more reliable, real-time drowsiness detection systems to enhance road safety.

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-prismaquant on 2026-06-26; verification: verified_with_issues.

Topics

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

• drowsiness detection algorithms
• drowsiness
• distraction detection algorithms
• microsleep

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