SUMMARY
The use of face masks in public spaces is an effort by the government to suppress the spread of COVID-19. Currently, supervision of the use of face masks is still carried out manually, namely officers who carry out direct monitoring in the field, so monitoring cannot be done all the time. In monitoring the use of face masks, there have been many developments, especially in the field of computer vision using various detection methods, namely YOLO, Convolutional Neural Network (CNN), Viola-Jones, or Haar Cascade, and Hybrid Deep Transfer Learning.From several applications of detection methods for the use of face masks, previous research has not detected the accuracy of the use of face masks, namely the face mask is used under the nose, the chin is visible, it is used not to cover the mouth and nose, placed under the chin, worn around the neck, removed when speaking, so there is still a chance for spread. virus.The purpose of this research is to create a face mask detection system using CNN and Haar Cascade algorithms. This study resulted in a detection system with an accuracy value of 97% with a total of 15 epochs. The results of the evaluation of the model using the confusion matrix resulted in an average F1-score of 0.97.