Face Mask Detection using Deep Learning

This project can detect whether a person is wearing a face mask or not using a webcam or image.

It uses Deep Learning (MobileNetV2) and OpenCV to work in real time.

---

What this project does

The program:

1. Looks at a person’s face through a webcam or an image.
2. Detects if the person’s face has a mask or no mask.
3. Shows the result on the screen with a label and box around the face.

---

How it works

1. Dataset — Around 4000 images of people with and without masks were collected.

2. Model Training —

   • The images are split into 80% training and 20% testing.
   • A MobileNetV2 deep learning model (pre-trained on ImageNet) is used.
   • Only the last few layers are trained to detect masks.

3. Saved Model — After training, a file called mask_detector_model.h5 is created.

4. Detection —

   • The webcam or an image is given to the model.
   • It checks the face area and predicts “Mask” or “No Mask”.

---

🖥️ How to run the project

Step 1: Install the requirements

Make sure you have Python installed, then open Command Prompt in your project folder and type:

pip install -r requirements.txt

Step 2: Train your own model (optional)

If you want to train it yourself:

python train_model.py

This will create a new model file — mask_detector_model.h5.

Step 3: Run the detector

After the model is ready:

python mask_detection_main.py

It will open your webcam and start detecting faces with or without masks.

---

---

🧪 Technologies Used

• Python 3
• TensorFlow / Keras
• OpenCV
• MobileNetV2 (Pretrained CNN)
• NumPy & Matplotlib

---

⚡ Features

✅ Works in real time using a webcam ✅ Detects faces accurately ✅ Trained on a custom dataset ✅ Can easily be improved with more mask types and colors

---

💡 Future Improvements

• Add third category → “Improperly worn mask”
• Improve accuracy with more training images
• Deploy the model as a web or mobile app

---

👨‍💻 Made By

Sameet Patro Undergraduate Student — IIIT Sonepat