Author
Listed:
- Sri Surya Dharmaraj V
- Dr
- V. S. Anita Sofia
- Dr
- L. Thara
Abstract
Driver drowsiness is an important traffic safety problem, as reduced alertness is known to affect visual attention, decision making and reaction time thereby increasing the risk of unsafe driving. Continuous monitoring of driver behavior can help identify visual signals that are associated with drowsiness and provide feedback to the driver in a timely manner. In this paper, we present a web CV driver monitoring system that focuses on drowsiness detection based on facial landmarks and temporal eye behavior. We detect the driver face and obtain 68 facial landmark points to inspect eye part by this method. The EAR (Eye Aspect Ratio) is a common ratio we can calculate from the eye landmakrs to find out how open are the eyes. Since any isolated frame displaying decreased eye openness may only reflect natural blinking or transient saccadic eye movement, the system applies exponential moving average (EMA) smoothing and temporal decision logic to assess shifts between contiguous observations. Finally the last setting maintains a 0.283 EAR threshold and a 0.16 EMA smoothing coefficient. The broader monitoring framework further provides auxiliary capabilities for head-pose and gaze detection, road-scene processing, and safety-oriented monitoring. For the second case, i.e., quantitative evaluation of drowsiness component, 168 samples were used in total for labeling: 29 are labeled as alert and 139 are labeled as drowsy. The assessment produced an accuracy of 90.48%, precision, and recall of 99.00%, F1-score of 93.89% and balanced accuracy — of nearly ~94.24%. The confusion matrix contained 29 true negatives, 123 true positives, 16 false negatives and no false positives. In the evaluation configuration, it took the detector approximately 41.08 ms to process a single frame and achieved a throughput of 24.34 FPS (frames per second). These findings illustrate how the implemented facial-landmark-based approach performs on the available labeled dataset, while also showing its computational feasibility for a prototype monitoring application. The study acknowledges limitations concerning the relatively small and imbalanced evaluation dataset, variations in environment and subjects, and the difference between frame-level evaluation and continuous temporal video analysis. Additional validation with larger, subject-independent, continuous driving-video datasets is needed to determine the robustness and generalizability of the proposed approach.
Suggested Citation
Sri Surya Dharmaraj V & Dr & V. S. Anita Sofia & Dr & L. Thara, 2026.
"A Computer Vision-Based Driver Drowsiness Detection System Using Facial Landmarks and Temporal Eye Analysis,"
International Journal of Scientific Research in Computer Science, Engineering and Information Technology, International Journal of Scientific Research in Computer Science, Engineering and Information Technology, vol. 12(5), pages 224-232, September.
Handle:
RePEc:jbh:ijsrcs:v12:y2026:i5:id:2171
DOI: 10.32628/CSEIT2612535
Note: Article URL: https://ijsrcseit.com/home/article/view/CSEIT2612535
Download full text from publisher
Corrections
All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:jbh:ijsrcs:v12:y2026:i5:id:2171. See general information about how to correct material in RePEc.
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
We have no bibliographic references for this item. You can help adding them by using this form .
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Pankaj Sharma (email available below). General contact details of provider: https://ijsrcseit.com/home .
Please note that corrections may take a couple of weeks to filter through
the various RePEc services.