Portfolio Detail

"YOLOv7_FaceMask" aims to detect and classify whether a person is wearing a face mask in real-time video streams. The project is based on the YOLOv7 object detection algorithm, which is a state-of-the-art deep learning model that can detect and classify objects with high accuracy and speed.

The project's main goal is to help prevent the spread of COVID-19 by enforcing face mask policies in public places. With the YOLOv7_FaceMask model, it becomes easier to monitor compliance with face mask regulations in public spaces such as airports, train stations, shopping malls, and other areas where large crowds gather.

1. Download YOLOv7

2. Process the dataset: 1)Create a dictionary - label encoding: file name (file), image width (width), image height (height), label (with_mask, without_mask, mask_wearded_incorrect), coordinates of the bounding box (xmin, ymin, xmax, ymax). 2)Unpacked dataset - image adjustment: 90% training set, 7% test set and 3% validation set.

3. Model training - YOLOv7-X - epochs: 50

4. Evaluation (Confusion matrix, F1 curve, Precision curve, Precision-Recall curve, Recall curve).