Fatigue Detection on Face Image Using FaceNet Algorithm and K-Nearest Neighbor Classifier

Background: The COVID-19 pandemic has made people spend more time on online meetings than ever. prolonged looking at the monitor may cause fatigue, which can subsequently impact mental and physical health. A fatigue detection system is needed to Internet users well-being. Previous research related used a fuzzy system, but accuracy was below 85%. In this research, machine learning improve accuracy.Objective: This examines combination of FaceNet algorithm with either k-nearest neighbor (K-NN) or multiclass support vector (SVM) accuracy.Methods: study, we UTA-RLDD dataset. features for come from face, so dataset segmented using Haar Cascades method, then resized. feature extraction process uses FaceNet's pre-trained algorithm. extracted are classified into three classes—focused, unfocused, fatigue—using K-NN SVM method.Results: between K-NN, value resulted in better polynomial kernel (at 94.68% 89.87% respectively). processing speed both combinations methods allowed real-time data processing.Conclusion: provides an overview early while working computer that limit staring screen too long switch places maintain health our eyes.