Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

A Decade of Progress in Drowsiness Detection Using Facial Features: A Comprehensive Bibliometric Analysis and Thematic Mapping

Rektorat UDINUS 2026 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2026
Goals
Record
905d6d59-a40e-44cc-8f20-e61ec805dbcc
Published
2026-10-30

Universitas Dian Nuswantoro conducted a comprehensive bibliometric analysis of research trends in drowsiness detection using facial features. The activity involved Subowo, Moh Hadi; Andono, Pulung Nurtantio; Shidik, Guruh Fajar; and Santoso, Heru Agus as participants. The study examined publications from 2014 to 2024 and applied a PRISMA-guided methodology to extract and analyze 347 documents from the Scopus database.

The analysis identified patterns in research productivity, citation impact, geographical distribution, institutional contributions, and thematic evolution. Publication volume showed an annual growth rate of 18.59%, with a notable increase after 2019 coinciding with the widespread adoption of deep learning approaches. Asian countries dominated research output. India recorded 182 publications and China recorded 100 publications, while China had the highest citation impact with 1370 citations.

Co-occurrence network analysis identified four research clusters: physiological insights via neural networks, computer vision-based drowsiness detection, multi-modal fatigue detection for accident prevention, and deep learning for biomedical signal analysis. Temporal keyword analysis indicated movement from traditional machine learning toward deep neural networks, Internet of Things integration, and real-time monitoring systems. Thematic mapping categorized CNNs and eye aspect ratio as basic themes; driver fatigue detection and cloud computing as motor themes; 3D head pose estimation and behavioral measurement as niche themes; and pupil detection and blink detection systems as emerging or declining themes.

The activity also identified deployment-related gaps. Only 3.3% of top-cited papers reported frames per second, and none reported latency or time-to-alarm, despite frequent claims of real-time capability. In multimodal research, 74% of studies focused exclusively on facial features, while 26% incorporated physiological or vehicular signals. The analysis further identified gaps concerning deployment readiness, ethnicity-specific considerations, and low-light environments. Through this health-related research activity, UDINUS documented the evolution and research needs of drowsiness detection systems in relation to Goal 3.

Reference https://www.scopus.com/pages/publications/105036247853?origin=resultslist