Bangun-lah: Behavioural Based Real-Time Drowsiness Detection for Drivers

Ng, Xu Xiang (2025) Bangun-lah: Behavioural Based Real-Time Drowsiness Detection for Drivers. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RDS_Ng Xu Xiang_Fulltext.pdf Restricted to Registered users only Download (2MB)

Abstract

"Bangun-lah" is a mobile application tailored specifically for drivers, offering real-time drowsiness detection through yawn scanning and more, using smartphone cameras and advanced computer vision algorithms such as Histogram of Oriented Gradient (HOG), Support Vector Machine (SVM) and Regression Tree. By analysing facial movements, particularly yawning patterns, it dynamically assesses drivers' alertness levels and issues timely warnings such as an alarm sound to get the driver’s attention when signs of drowsiness are detected. With customisable a user-friendly interface for drivers, Bangun-lah aims to prevent accidents and promote road safety by helping drivers stay awake and alert during their journeys.