Drowsy Guard research proposal¶
An IoT research project from the Master of Information Technology Engineering (MITE), Cohort 19 at the Royal University of Phnom Penh (RUPP).
Complete A4 proposal¶
The full proposal follows the provided research template and includes the problem statement, objectives, research questions, methodology, system design, evaluation plan, timeline, budget, risks and references.
Project information¶
| Item | Details |
|---|---|
| Project title | Drowsy Guard |
| Research area | Driver drowsiness detection |
| Subject | IoT: Technology and Application |
| Lecturer | Dr. Chey Chan Oeurn |
| Programme | Master of Information Technology Engineering (MITE) |
| Cohort | 19 |
| Shift | Evening – Weekdays |
| Institution | Royal University of Phnom Penh (RUPP) |
| Project code | |
| Pilot location |
Research team¶
| Role | Name |
|---|---|
| Project leader | Seng Phirum |
| Project member | Seang Chanviwath |
| Project member | Suon Mesa |
| Project member | Theng Rathrongroeung |
| Project member | Lov Kimheng |
Proposal at a glance¶
Drowsy Guard investigates whether a low-cost ESP32-S3 vision device can detect early signs of driver drowsiness and issue a timely local alert without relying on cloud processing. The system combines eye closure, long blinks, yawning and head movement, then applies a time-based risk filter before speaking a warning.
The study will measure detection performance, alert delay, false-alert rate and on-device speed under controlled conditions. Video is processed locally; the diagnostic dashboard supports testing but is not required for the alert.
Research use only
Drowsy Guard is a research prototype, not a certified automotive safety device. It must not be treated as a substitute for rest or safe driving.