Drowsy driver detection system - via facial recognition and driving data / Nur Iman Kamila Azharudin, Hafizah Mansor and Shaila Sharmin
According to the National Highway Traffic Safety Administration, an estimated 17.6% of all fatal crashes in the years 2017–2021 involved a drowsy driver. This study proposes a drowsy driver detection system that uses both facial recognition and vehicular data to detect if a driver is feeling sleepy...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universiti Teknologi MARA Press (Penerbit UiTM)
2024
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/105182/1/105182.pdf https://ir.uitm.edu.my/id/eprint/105182/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | According to the National Highway Traffic Safety Administration, an estimated 17.6% of all fatal crashes in the years 2017–2021 involved a drowsy driver. This study proposes a drowsy driver detection system that uses both facial recognition and vehicular data to detect if a driver is feeling sleepy behind the wheel. We aim to address a lack of works in the literature that combine data measured from the driver (image or biological data) and vehicular data for drowsy driver detection. Our primary data was collected from simulated driving sessions in which a camera was used to record test drivers’ faces while driving a virtual car in the CARLA simulator in both drowsy and non-drowsy states. The collected data consists of video of test drivers' faces from the camera and vehicular data from the simulated car. The video data was used to obtain facial features such as Mouth Over Eyes (MOE), Eyes Aspect Ratio (EAR), and Mouth Aspect Ratio (MAR), while the vehicle data yielded features such as speed, steering wheel movement and pedal readings. These features were used to train Support Vector Machine (SVM) and Random Forest (RF) models to detect drowsy drivers. The results indicate that RF is a better model to be used as compared to SVM in predictions of drowsiness in drivers with an accuracy of 96.24% and 86.85% respectively. |
|---|