Face Mask Detector

Due to the recent pandemic, we all were locked down in our homes but now we have to accept the situation and carry on our lives with this "new normal". One of the important steps to minimize the spreading of the virus is using mask. Whether you are in office, school, public place, or anywhere, you should wear masks to keep yourself and others safe. The goal of this project is to ensure that people wear masks whenver they enter any building. This system can be installed in offices, restaurants, metro stations, or in any buildings entrance to make sure that the person who enters the building is wearing a mask.

Details about the structure of the codebase

• The main python scripts that is used to detect the mask is in the mask_detector folder. This folder contains 2 python scripts namely, mask_detect_img.py and mask_detect_vid.py. The mask_detect_img.py is used to detect masks in an image file and the mask_detect_vid.py is used to detect masks in a video stream.
• The notebooks folder contains a jupyter notebook that is used to train the model for mask detection.
• Face Detection i.e enclosing the face of a person inside a bounding box is done by using a pre-trained face detection model which is stored in face_detection folder.
• Images folder contains 2 test images that can be used to test the mask_detect_img.py script.

Prerequisites

• Python 3 or above
• All other requirements are inside the requirements.txt file

Installation

1. Clone the repository

$ git clone https://github.com/Ayush-Parhi/FaceMaskDetection.git

2. Go to the project directory

$ cd FaceMaskDetection

3. Create a virtual environment. (Here, I am using pipenv but you can use virtualenv also)

$ pipenv shell

4. Install the dependencies.

$ pip install -r requirements.txt

Usage

1. Detect masks in images.

$ python mask_detector/mask_detect_image.py --image images/pic1.jpeg

2. Detect masks in real-time video

$ python mask_detector/mask_detect_vid.py