Face Detection

Face Mask Detection Alert
System Using Keras and

Tensor-Flow
Under the Guidance Of

Mrs. Rinku Supakar
(Head of the Department, Computer Science and
Engineering)
MEMBERS IN THE
PROJECT
Submitted By:
 AGNIDEEP PODDER (Roll: 25500117065)

 AISHIK SETT (Roll: 25500117065)
 ANIRBAN PAL (Roll: 25500117060)
 NILANJAN MUKHERJEE (Roll: 25500118003)
 RAHUL SAHA (Roll: 25500117031)
SHREYA GHOSH (Roll: 25500117020)
CONTENTS
SERIAL TOPIC SLIDE NO.
NO.
1 ABSTRACT 1
2 INTRODUCTION 2
3 DEEP LEARNING 3
4 COMPUTER VISION 4
5 CONVOLUTION NEURAL NETWORK 5
LAYERS IN CONVOLUTION NEURAL NETWORK 6
6
7 TENSORFLOW KERAS 7
8 NUMPY AND TKINTER 8
9 OPEN-CV 9
10 DATA-SETS(WITHOUT MASK) 10
11 DATA-SETS(WITH MASK) 11
12 RESULTS 12
3
13 CONCLUSION AND FUTURE WORK 13-14
1.
ABSTRACT
 The end of 2019 witnessed the outbreak of
Coronavirus Disease 2019 (COVID-19),which
has continued to be the cause of plight for
millions of lives and businesses even in 2020.
Corona virus disease 2019 has become a
major health problem
 Studies have proved that wearing a face
mask significantly reduces the risk of viral
transmission as well as provides a sense of
protection. So to overcome this problem we
required an efficient mask monitoring system
 By the development of machine learning and
image processing analysis many face
detection algorithms are introduced. By using
image processing analysis and machine
learning method we can implement mask
detection technique.
Fig 1: Face Mask detection
2.
INTRODUCTION
 Rapid advancements in the fields of
Science and Technology have led us to a
stage where we are capable of achieving
feats that seemed improbable a few
decades ago.
 Technologies in fields like Machine
Learning and Artificial Intelligence have
made ow lives easier and provide
solutions to several complex problems in
various areas.
 Face mask detection is a challenging task.
It has been receiving more and more
attention in this era due to the spreading
of corona virus disease.
3.
DEEP LEARNING
 Deep learning is a particular
kind of machine learning that
achieves great power and
flexibility by learning to
represent the world as a nested
hierarchy of concepts.
 Deep learning is an AI function
that mimics the workings of the
human brain in processing data
for use in detecting objects,
recognizing speech, translating
languages, and making
decisions.
Fig 3: DEEP LEARNING

4.
COMPUTER VISION
 Computer Vision, often abbreviated
as CV, is defined as a field of study
that seeks to develop techniques to
help computers “see” and
understand the content of digital
images such as photographs and
videos.
 It largely remains an unsolved
problem based both on the limited
understanding of biological vision
and because of the complexity of
vision perception in a dynamic and
nearly infinitely varying physical
world.
Fig 4: COMPUTER VISION
5.
Convolution Neural Network
 A convolutional neural network is
a feed-forward neural network that
is generally used to analyze visual
images by processing data with
grid-like topology. It’s also known
as a Conv-Net. A convolutional
neural network is used to detect
and classify objects in an image.
 A Convolutional Neural
Network (CNN) is a Deep
Learning algorithm which can take
in an input image, assign
importance (learnable weights and
biases) to various aspects/objects
in the image and be able to
differentiate one from the other.
6.
Layers in Convolution Neural Network

Input Layer: This layer holds the raw input of

image
Convolution Layer: This layer computes the
output volume by computing dot product
between all filters and image patch.
Activation Function Layer: This layer will
apply element wise activation function to the
output of convolution layer.
Pool Layer: . This layer is periodically inserted
in the cov-nets and its main function is to reduce
the size of volume which makes the computation
fast reduces memory and also prevents from
over-fitting
Fully-Connected Layer: This layer is regular
neural network layer which takes input from the
previous layer and computes the class scores and
Fig 6: Fully
outputs the 1-D array of size equal to the number connected layer
of classes.
7.
TENSORFLOW KERAS
 TensorFlow is a Python library for
fast numerical computing created
and released by Google.
 Tensor-Flow can be used to
create Deep Learning models
directly or by using wrapper
libraries that simplify the process
built on top of Tensor-Flow.
 Keras is an open-source software
library that provides a Python
interface for artificial neural
networks.
 Keras acts as an interface for the
TensorFlow library. Fig 7: TENSOR FLOW AND KERAS
8.
NUMPY AND TKINTER

 NumPy is a general-purpose array-
processing package. It provides a high-
performance multidimensional array
object, and tools for working with
these arrays.
 NumPy is the fundamental package
for scientific computing with Python.
 Tkinter is the standard GUI library for Fig 6: NUMPY
Python. Python when combined with
Tkinter provides a fast and easy way to
create GUI applications.
 Tkinter provides a powerful object-
oriented interface to the Tk GUI
toolkit.
Fig 8: TKINTER
9.
OPEN-CV
 OpenCV-Python is a library of
Python bindings designed to solve
computer vision problems.
 OpenCV is a huge open-source
library for computer vision, machine
learning, and image processing.
OpenCV supports a wide variety of
programming Languages like Python,
C++, Java, etc.
 It can process images and videos to
identify objects, faces, or even the
handwriting of a human.
Fig 9: OPEN CV
10.
DATA-SETS
(WITHOUT MASK)
 The data set
consists of 1376
images of faces of
different persons.
 Out of 1376 and
686 images of
persons having no
masks.
FIG 10: PERSONS NOT WEARING

MASKS
11.
DATA-SETS
(WITH MASK)
• Out of
1376
images
there are
690
images of
persons
wearing
FIG 11: PERSONS WEARING MASKS
masks
12.
RESULTS

We have applied Convolution
Neural Network to train the
machine so that it can detect
whether the person is wearing
Mask or not. The machine
shows the correct result when
it gets an unknown data, that
is, the machine has been
trained properly.
FIG 12: RESULT OF FACEMASK

DETECTOR
13.
CONCLUSION
We have studied the performance of convolutional neural
networks in image classification. We found that CNN
architectures are capable of learning powerful features from
weakly-amount and label of dataset that far surpass feature
based methods in performance and that these benefits are
robust to details of the hardware of the architectures in time.
Our training results indicate that while the performance is

highly sensitive to the architectural details and hardware
configuration, especially with GPU processor. However the
confidence level is not too much in difference, with only
around 0.5% between both scenarios, which in return does
not indicate differences in testing result.
14.
FUTURE WORK
 This study is very limited and need
broader context of images or videos
specification. Future work such as face
directions from 180 degrees horizontal or
vertical direction, camera distances, higher
resolution for real life industry use cases.
 The use of Machine Learning in the field
of mobile deployment is rising rapidly.
Hence, we plan to port our models to their
respective Tensor-Flow lite versions
THANK YOU

Face Detection

Uploaded by

Copyright:

Available Formats

Face Detection

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Face Detection

Uploaded by

Copyright:

Available Formats

Face Mask Detection Alert

System Using Keras and

Under the Guidance Of

 AGNIDEEP PODDER (Roll: 25500117065)

Fig 3: DEEP LEARNING

Input Layer: This layer holds the raw input of

NUMPY AND TKINTER

FIG 10: PERSONS NOT WEARING

FIG 12: RESULT OF FACEMASK

Our training results indicate that while the performance is

You might also like