The Project is named Face Mask Detector which is based on the principles of Artificial Intelligence foccused in the areas of Deep Learning and Computer Vision. With the help of this project any camera would autonomously detect in real-time whether a person is wearing a mask or not, and if not will send a message to the console followed by an alarm sound sending an alert if people are not following rules by not wearing a face mask.
It will also be deployable to edge and embedded devises like Google Coral, Rasberry Pi, Nvidia Jetson Nano, etc. since the model would run on a highly efficient architecture.
In the wake of Covid-19, it has almost become madatory to wear a face mask in order to prevent the further spreading of this deadly virus that is highly contagious in nature and since no vaccine has yet been developed so the best precaution one can have is by using a face mask.
Still many people reject this preventive measure and disagree to follow the rules by not wearing a face mask and due to this fact our Government has initiated a fine for those who do not adhere to these guidlines, but to detect all the people not wearing mask in a crowd with the naked eye is nearly impossible so in order to overcome this problem the following project is made which automates the detection process with any device having a camera. Also this detector potentially could be used to help ensure your safety and the safety of others due to the widespread nature of this virus.
Following are the technologies and frameworks that would be used for developing & coding this project:
-
Python – the primary coding language used to build the project from scartch.
-
OpenCV - for accessing the webcam or any camera/vision device, streaming, showing bounding boxes and labels.
-
TensorFlow & Keras – for data augmentation, building a new fully-connected (FC) head, data pre-processing & loading image data and calculating the computation graph.
-
Sklearn - for binarizing class labels, segmenting our dataset, and printing a classification report.
The following project will be made using Convolutional Neural Networks (or CNNs) and other Deep Learning and Computer Vision concepts. It will detect in real-time and will also work on static image as well as video whether a person is wearing a mask or not, and if not will send a message to the console followed by an alarm sound sending an alert if people are not following rules by not wearing a face mask.