Neural Network Algorithm to detect people wearing face masks and take actions accordingly.
This project detects face masks in real-time using a neural network trained with TensorFlow. When a person without a mask is detected, the system sends a serial command to an Arduino to trigger an action (e.g., LED indicator).
Achieves 98% confidence on the trained model.
Train a CNN on a dataset of masked vs unmasked faces using TensorFlow/Keras.
Works best on Ubuntu 20.04 LTS with GPU acceleration.
Run the face recognition algorithm that uses the trained model to detect masks on faces via webcam.
A pre-trained model (mask_detector.model) is included if you want to skip training.
Serial command triggers an Arduino to switch an LED based on detection state.
pip3 install -r requirements.txt- Arduino board
- Webcam
- LED + resistor
TensorFlow - Keras - imutils - numpy - OpenCV - matplotlib - scipy - argparse - pyserial