An Efficient Automatic Face Mask Detection System for Human Safety Based on Deep Learning using Novel YOLOv3 in Comparison of YOLO with Improved Accuracy

The objective is to build an efficient face mask detector using Novel YOLOv3. The algorithm used to detect face masks is Novel YOLOv3 in comparison with YOLO, the dataset used was (Facemask Detection Dataset, no date) with the sample size was 136. Novel YOLOv3 gets an accuracy of 92% and in YOLO it is 88% the increase in accuracy is due to the use of Darknet53 neural network model, the novel YOLOv3 and YOLO are statistically satisfied with the independent sample t-test value
$(\mathrm{P}\unicode{x00A1}{0.001})$
with confidence level of 95%. Face Mask detection in Novel Yolov3 has a significantly better accuracy than YOLO.