NUR FIRZANAH BINTI MOHD ZAIDI
Rafidah Md Noor
M Haikal Bin Saripan
Road safety is a pressing concern in Malaysia, with fatigue-related accidents contributing significantly to traffic fatalities, particularly among heavy vehicle drivers. Fatigued driving impairs concentration, reduces reaction time, and increases the likelihood of accidents. Existing drowsiness detection systems, while valuable, often suffer from high implementation costs, complexity, and reliance on generic thresholds that fail to account for individual differences, such as facial feature variations across races. This project addresses these limitations by developing an affordable, accurate, and non-intrusive drowsiness monitoring and alert system tailored for bus drivers. The proposed system utilizes Internet of Things (IoT) technology combined with open-source libraries such as OpenCV and Dlib for real-time facial feature detection. Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) calculations are employed to monitor physical reactions, including eye closure and yawning, as indicators of drowsiness. The system consists of two core components: the Driver System, which performs in-vehicle monitoring, and the Central Monitoring System, which processes data remotely to issue timely alerts. This dual-component approach ensures continuous monitoring and effective intervention to enhance road safety. By aligning with Sustainable Development Goal (SDG) 3.6, which aims to halve global road traffic fatalities by 2030, the project contributes to reducing accidents caused by driver fatigue. This report highlights the system's design, methodologies, and performance evaluation, demonstrating its potential to significantly improve safety standards for long-distance drivers and the transportation industry at large.