😷 Face Mask Detection

A computer vision project that detects whether a person is wearing a face mask or not using machine learning and image processing techniques.

📌 Features

• Detects face with mask and face without mask
• Uses computer vision & deep learning
• Real-time detection (if webcam enabled)
• Easy to train and test

🛠️ Tech Stack

• Python
• OpenCV
• TensorFlow / Keras
• NumPy
• Matplotlib

📁 Project Structure

Face_Mask_Detection/
│
├── dataset/
│   ├── with_mask/
│   └── without_mask/
├── model/
├── train_model.py
├── detect_mask.py
├── requirements.txt
└── README.md

⚙️ Setup Instructions

1️⃣ Clone the Repository

git clone https://github.com/sofialtamsh/Face_Mask_Detection.git

2️⃣ Create a Virtual Environment (Recommended)

python -m venv venv

3️⃣ Activate the Virtual Environment

Windows:

venv\Scripts\activate

macOS/Linux:

source venv/bin/activate

4️⃣ Install Dependencies

pip install -r requirements.txt

🚀 Usage

🔹 1. Train the Model

python train_model.py

• Loads images from the dataset
• Trains the CNN model
• Saves the trained model in model

🔹 2. Run Face Mask Detection

python detect_mask.py

Uses webcam or video feed, Detects faces in real-time

Displays:

• Green box → Mask detected
• Red box → No mask detected
• Press ESC to quit

🖼️ Example Output

[INFO] Starting video stream...
[INFO] Mask detected with 98.45% confidence

On-screen:

• Bounding box around face
• Label: Mask / No Mask
• Confidence percentage

📊 Dataset Format

Ensure your dataset is structured as follows:

dataset/
├── with_mask/
│   ├── img1.jpg
│   ├── img2.jpg
│   └── ...
└── without_mask/
    ├── img1.jpg
    ├── img2.jpg
    └── ...

Guidelines:

• Images should be clear and front-facing when possible
• Supported formats: .jpg, .png
• Balanced classes improve model accuracy

🧠 Model Architecture

• Convolutional Neural Network (CNN)
• Image preprocessing using OpenCV
• Binary classification:
• With Mask
• Without Mask
• Trained using TensorFlow/Keras
• Model saved in the model/ directory after training

📦 Dependencies

TensorFlow
OpenCV-Python
NumPy
Matplotlib
Scikit-learn

📝 Notes

• Always activate the virtual environment before running scripts
• Ensure proper lighting for better face detection
• You can improve accuracy by adding more training images

🤝 Contributing

Pull requests are welcome!
For major changes, please open an issue first to discuss what you’d like to change.

🪪 License

MIT License © 2025 Sofi Altamsh