Overview / Usage

In the wake of the COVID-19 pandemic, wearing masks has become an essential practice to help prevent the spread of the virus. However, ensuring compliance with mask-wearing guidelines in public places can be challenging, especially in high-traffic areas.

The machine learning project aims to implement machine learning algorithms to detect whether a person in a given image is wearing a mask or not.The algorithm used in this project is Convolutional Neural Network (CNN) which is a type of neural network that is specifically designed to process and analyze visual data such as images or videos.The dataset used for this project consists of images of people wearing masks and people not wearing masks: https://www.kaggle.com/datasets/omkargurav/face-mask-dataset.

The project will be useful in various situations such as:

📍 Public places such as airports, hospitals, and shopping malls to enforce mask-wearing policies

📍 Educational institutions to enforce mask-wearing policies among students

📍 Industries such as construction and manufacturing to ensure the safety of workers in hazardous areas.

Methodology / Approach

The objective of the project is to accurately identify whether the person in an input image is wearing a mask or not.To achieve this, we will use a Convolutional Neural Network (CNN) architecture that is very effective for image classification tasks.

The preprocessing stage of the project includes steps such as labellng the images with mask and without mask, resizing and converting the images into numpy arrays which will be stored in a list which will be used to train the CNN model.

The dataset will be split into training, validation, and testing sets. The training and validation sets will be used to train and tune the CNN model, respectively. The testing set will be used to evaluate the performance of the trained model on new data.

The CNN model is designed and trained to classify images as either containing a person wearing a mask or not.The model includes 2 convolutional layers, 2 max-pooling layers, and 2 fully dense layers

The output layer has 2 neurons (one for each class). The output will be a probability distribution over the 2 classes, with which we will arrive at a binary classification decision that indicates whether a mask is present in the image or not.

Technologies Used

This project uses oneDNN to optimize the performance and efficiency of the CNN model. The tensorflow framework which is integrated with OneDNN is used to achieve lower memory consumption,higher accuracy,faster training times and better utilization of hardware resource.

Intel(R) Extension for Scikit-learn is also used which provides a seamless way to speed up the Scikit-learn application.

The other libraries used are

• tensorflow
• sklearn
• matplotlib
• cv2
• keras

Documents and Presentations

Repository

https://github.com/mukunth26/Facemask-Detection-OneAPI