Drowsiness detection and alerting system for drivers using single board computer

Mohd Ghani, Norisa Shafika (2022) Drowsiness detection and alerting system for drivers using single board computer. Project Report. Melaka, Malaysia, Universiti Teknikal Malaysia Melaka. (Submitted)

	Text (24 Pages) Drowsiness detection and alerting system for drivers using single board computer.pdf - Submitted Version Download (634kB)
	Text (Full Text) Drowsiness detection and alerting system for drivers using single board computer.pdf - Submitted Version Restricted to Repository staff only Download (8MB)

Abstract

Due to the massive growth in traffic, road accidents have become a major concern. Drowsiness of drivers during the night is the leading cause of accidents. Fatigue and drowsiness are two of the most common causes of serious accidents. The only way to solve this problem is to detect tiredness and notify the driver. The objectives of this project is to study the implementation of IOT for this project which is to Drowsiness Detection And Alerting System For Drivers Using Single Board Computer, make the system to alert the drivers by measure the drivers’ drowsiness level based on the state of eyes and to analyze how much this system’s accuracy and reliabilty of this proposed project. This project will be built using the Raspberry Pi 3 Model B, buzzer, web-camera and an LED. This is because drowsiness will be detected by the image acquired by the web-camera, which will send the data to the microcontroller. The microcontroller then will process the data to indicate the driver's status. This will form part of a driver drowsiness detection and alerting system for drivers. The fundamental of this system is to use web-camera to capture picture and the Haar Cascade algorithm to detect the driver's face and both eyes movements, and if the driver is feeling tired, the system will send out a warning message via a loud buzzer alert and the LED will be lighted up. The drowsiness detection warning message also will be sent to the Telegram along with the captured picture of being drowsy by the web-camera. As a result, this system have 90% accuracy to detect drowsiness of drivers since the system can detect the drowsiness in real time when the driver is wearing the glasses or not wearing them accurately as long the brightness of the image in the video is not too bright or too dark. As a conclusion, the system is successfully detect drowsiness for drivers and alerting them to wake them up so that they can avoid the accidents when driving on the road.

Actions (login required)

View Item