Face Mask Detector

Utilizing Transfer Learning, the system works on detecting a face mask in real-time. The model is trained on a balanced dataset of 7,500 samples on a pre-trained MobileNetV2.

Technologies used:

• Python
• OpenCV
• MobileNetV2 (Pre-trained model)
• TensorFlow
• Keras
• NumPy
• sklearn
• matplotlib

Scope:

• Detects for a face mask
• Can detect multiple faces in a single frame
• Has an accuracy of about 98%

Execution:

Training part:

Type python trainMaskDetector.py in command prompt.

Detection part:

Type python maskDetector.py in command prompt.

Output:

Training the model and showing the accuracy:

Mask detection: