Face-Mask-Detection

About

The project is a face mask detection system that uses OpenCV for face detection and a Keras/TensorFlow model to classify whether a person is wearing a mask. The project demonstrates real-time video processing and deep learning-based classification. This project was originally developed as part of my college assignment. The current version has been organized and uploaded to GitHub for reference.

⚠️ Heads up: This project is meant mostly for showcasing. If you try to run it, brace yourself for some dependency and setup adventures.

---

Features

• Detect faces in images or video streams.
• Predict if a detected face is wearing a mask or not.
• OpenCV DNN used for face detection.

---

Project Structure

Face-Mask-Detection/
├── dataset/ # Images used for training
├── face_detector/ # Pre-trained face detection models (Caffe)
│ ├── deploy.prototxt
│ └── res10_300x300_ssd_iter_140000.caffemodel
├── detect_mask.py # Real-time mask detection script
├── train_mask_detector.py # Script to train mask detection model
├── mask_detector.h5 # Trained model (generated after training. # Trained model (was originally mask_detector.model, now converted to H5 format)
├── requirements.txt # Python dependencies
└── README.md # Project documentation

---

Setup (if you dare)

1. Clone the repository:

   git clone <repo-url>
   cd Face-Mask-Detection

2. Create a virtual environment and activate it:

   python3 -m venv venv
   source venv/bin/activate  # Linux / macOS
   venv\Scripts\activate     # Windows

3. Install dependencies:

   pip install -r requirements.txt

4. Ensure these files exist:

   face_detector/deploy.prototxt

   face_detector/res10_300x300_ssd_iter_140000.caffemodel

   If missing or corrupted, grab them from OpenCV's GitHub: Face Detector Model Files

5. Run detection (optional):

   python3 detect_mask.py

   If it runs, congratulations! If not, welcome to dependency hell. 😅

---

Notes

• Developed for academic learning and demonstration.

• Training the model is resource-intensive.

• Cloning and running may require some patience and debugging.

• Model training parameters were slightly adjusted from the original college version.