Raymond Yeo Deng Shun Universiti Teknologi PETRONAS
This project designs and deploys a machine learning-based system to detect mental fatigue levels using facial cues and HRV. Facial features are captured via camera to compute Eye Aspect Ratio (EAR) and Mouth Opening Ratio (MOR), while HRV is measured using electrocardiogram (ECG) to extract frequency features like LF/HF and Total Power (TP). The data is classified into three fatigue levels—normal, alert, and fatigue—using Random Forest (for facial data) and SVM (for HRV). Based on data collected, the model achieved 93% accuracy with facial cues and 76% accuracy with HRV, showing strong potential for fatigue detection and classification.
