A deep learning based model that detects whether a person is wearing a face mask or not with nearly 100% accuracy.

Framework : Keras

Pretrained model used : MobileNetV2

Trainable parameters : 3M

Number of classes : 2

Train Accuracy : 99.29%

Validation Accuracy : 100%

Dependencies

• Tensorflow
• Keras
• OpenCV
• NumPy

Overview

Techniques used to reduce overfitting :-

• Data Augmentation
• BatchNormalization
• Dropout
• Early Stopping

Click here to checkout the code.

To run the code :-

• For training refer mask_detection.ipynb
• For testing on your own image run detect_facemask.py