Face-Mask-Detection

Face Mask Detection

In recent trend in world wide Lockdowns due to COVID19 outbreak, as Face Mask is became mandatory for everyone while roaming outside, approach of Deep Learning for Detecting Faces With and Without mask were a good trendy practice. Here I have created a model that detects face mask trained on 7553 images with 3 color channels (RGB). On Custom CNN architecture Model training accuracy reached 94% and Validation accuracy 96%.

Data set used for this project is uploaded on Kaggle and can be downloaded from below link:-

https://www.kaggle.com/omkargurav/face-mask-dataset

Kaggle Code Submission:-> https://www.kaggle.com/omkargurav/face-mask-detection

With transfer learning using MobieNetV2 architecture training accuracy achieved 98% and validation accuracy 99%.

Algorithm : Convolutional Neural Network

Framework : Tensorflow

Accelerator : GPU

Dataset : 7553 RGB Images (With Mask/Without Mask)

Dataset Credit:

I am going to use my own data set of Total 7553 images.

I have taken 1776 images including both With and Without Face Mask images from Prajna Bhandary's Github account

https://github.com/prajnasb/observations

Remaining 5777 images are collected and filtered from Google search engine.

3725 Images of Face with Mask

3828 Images of Face without Mask.

Outputs:->

YouTube Video :->