부산대학교 도서관

COVID-19 pandemic has quickly inflated health crises all over the world and has effects on regular mode. The best way to decrease the spread of corona is by using a facemask. Many reports proved that carrying face masks at work mostly reduces the danger of spread. Monitoring manually if the people are carrying facemasks properly andto apprize the person not wearing a mask publically and jammed areas could also be a troublesome task. The present situation demands AN economical mask Detection Model. It's AN object detection and classification drawback with 2 totally different categories (Mask and while not Mask). Our proposed approach will detect the faces in an image using haar features and check whether it has a mask or not. The dataset used in the above model contains images of faces, in some images the faces are with masks and some are without masks. We created our own dataset for the model. We collected many images which contain faces, some with masks and some without masks. This model uses machine learning and image processing to detect faces in an image. This model can be used anywhere, whereas getting into the place everybody ought to scan their face therefore entry guaranteeing they have a mask with them. If anyone is found to be while not wearing a mask, a label of not carrying a mask is generated. Using image processing, faces are detected in an image. When a face is detected, the image will be reshaped and cropped only keeping a face. Then that image is converted into an array and checks whether the face contains a mask or not. According to the array values, a square box is displayed around the face and a label is displayed saying 'mask' whenthere is a mask or 'no mask' when there is no mask. [ABSTRACT FROM AUTHOR]