Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 13

REAL TIME FACE

MASK DETECTION
OBJECTIVE:-

 In this project we created a COVID-19 face mask detector using openCV,


Keras,Tensorflow

To create our face mask detector, we trained a two-class model of people
wearing masks and people not wearing masks.

The face mask recognition system, which is based on CNN model uses dataset
consists of different facial images with and without mask. The same model can be
used for different purposes related to image processing  using dataset containing
images related to that task.

The main objective is to create a mask detection system, able to recognize face
masks both in images and in real time video, drawing bounding box around faces.

We tuned CNN model on our mask/no mask dataset and obtained a classifier
that is accurate
MOTIVATION:-

• To ensure the safety of the citizens during the global pandemic.

• To effectively implement image processing approaches for the purpose of facial


mask detection.
INTRODUCTION:-

The whole world today, is facing the COVID-19 pandemic research has found that
wearing a mask is one way to reduce the risk of viral transmission. In this project, we
propose face mask detection via webcam cameras using image recognition which is one
of the high-accuracy and efficient face mask detector. This proposed system consists of
three methods to detect face mask:

1) Image preprocessing
2) Face detection and crop
3) Classification of face mask.

This system adopts the frameworks of deep learning, TensorFlow, Keras, and OpenCV
libraries to detect face masks in real time. This user-friendly model can be incorporated
with several existing technologies such as face detection, biometric authentication and
facial expression detection for further advancements in the future.
CONVOLUTIONAL NEURAL NETWORK(CNN)

Convolution Neural Network is a key aspect in modern computer vision task like pattern
object detection, image classification, pattern recognition tasks, etc

.
The layers of CNN consists of an input layer, ouput layer and hidden layer
EXISTING SYSTEM Vs PROPOSED SYSTEM:-

EXISTING SYSTEM:
The existing system is so heavily dependent onadvanced technical
concepts like machine learning that grasping the idea of it has become
atleast slightly more difficult. The existing system also requires more time
and monetary cost which are both highly valued.The existing systems all
use RGB color algorithms and the architecture diagram is morecomplicated.
It also makes use of proximity sensors and thermostats which are quite
expensiveto use in large scale. It suggests to use networks like RESTNET
50 and VGG16

PROPOSED SYSTEM:-
In order to predict whether a person has put on a mask, the model requires
learning from a well-curated dataset. The proposed system uses
Convolution Neural Network layers (CNN) as its backbone architecture to
create different layers. Along with this, libraries such as OpenCV, Keras,
and Tensorflow are also used.
REQUIREMENT SPECIFICATION:-

• Hardware Specification:-

processor:pentium Dual Core 2.3GHz


speed : 900MHz
Ram : 2GB(Min)
Hard Disk : processor 250GB or Higher
Mointor : 15” color CRT monitor
Keyboard : 104 keys

• Software Specification:-

Back end : Python


Operating System : Windows 7 or higher
Language used : python(pandas,Numpy,keras,Tensorflow)
Tools : Anaconda,JupyterNotebook,Spyder.
Work Flow:-

Load face mask dataset

Build CNN

Train the CNN model

Test the model on nee


image/live video
SYSTEM ARCHITECTURE:-

The System Architecture aims at classifying whether a person is wearing mask


or not by taking input from live video streaming.

The proposed system consists dataset of size 10,013 images out of which

5012 are with mask


5001 are without mask
Data Collection

Data preprocessing

Classification of CNN

Convocation
Convocation Layer with Min pool
Layer with Max pool
Flatten the image

Detecting Output

NO MASK MASK
OUTCOME:-
CONCLUSION:-

This manuscript proposes a face mask recognition system for real-time video
that automatically identifies if a person is wearing a mask which is an
excellent solution for the spread of the COVID-19 pandemic. By using Keras,
OpenCV, and CNN, the proposed system is able to detect the presence or
absence of a face mask and the model gives precise and quick results. The
trained model yields an accuracy of around 98%.

You might also like