If You Must Mask

A facemask detection using Deep Learning and Computer Vision

The goal of this project is to build a face mask detection system using deep learning algorithsm and computer vision .
The project uses the face mask detection images dataset from Kaggle that contains 11,800 images of face with mask and without mask, evenly distributed. The binary classification model was built with Keras/Tensorflow using CNN algorithm and had an accuracy score of 0.99. Using Haar Cascade Classifier as face detector, the face mask detection applies the binary CNN model as mask detector to the area of face and shows the prediction result. The application is built with Streamlit and deployed on Heroku.

Tools

Python (Numpy, Pandas)
Google Colab (Cloud computing)
Keras/Tensorflow
OpenCV
Matplotlib,seaborn
Streamlit, Heroku

Techniques/Algorithms

Convolutional Neural Network (Mask Detector)
Haar Cascades Classifier (Face Detector)
Computer Vision (Show prediction)

Application Usage

Face mask detection on Images

Face mask detection on Webcam

To Learn More, Check Out My:

Note: The app can take awhile to load... please be patient :)