In the COVID19 crisis wearing masks is absolutely necessary for public health and controlling the spred of the pandemic. With the power of Deep Learning this system can detect whether a person has worn a mask or not. This model has been trained on the dataset made by prajnasb
Here I have used transfer learning using the MobileNetV2 model which achieved 100% accuracy on the Train, Validation and Test data as can be seen in the notebook
The Haar Cascades Face Detection algorithm here is used to detect faces which is then passed through my model which estimates whether person is wearing a mask or not.
Final Deployment is done using OpenCV that feeds the webcam data as frames to the model.