Real-time face mask detector

Completed in November 2021 in a post-pandemic context.
This project implements a computer vision model capable of detecting in real-time via webcam whether a person is wearing a face mask or not.

❓ Issue

How to design a lightweight and responsive solution that visually detects mask-wearing in public environments (transport, businesses, etc.) to reinforce health measures?

🛠️ Implemented solution

• 📸 Use of OpenCV to access the camera and detect faces

• 🧠 Training of a convolutional neural network (CNN) with Keras to classify faces as "masked" or "unmasked"

• 🔄 Real-time integration: video stream analysed live with visual feedback (green/red outline)

• 🖥️ Creation of a simple interface allowing local detection to be initiated

Link to the project's GitHub repository: Real-time Face Mask Detector

⚙️ Technical stack

• Language : Python

• Libraries : OpenCV · Keras · TensorFlow · scikit-learn

• Techniques : Computer Vision · Binary Classification · Real-time Video Processing

• Environment : Jupyter Notebook

• Interface : Tkinter

• Prerequisites : Deep Learning · CNN · Computer Vision