GEETHANJALI INSTITUTE OF SCIENCE AND TECHNOLOGY

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

Real-Time Facemask Recognition

PRESENTED BY
S. Sivasankar - 172U1A0546

B. Deepak Surya -172U1A0503

M. Abhilash -172U1A0530

SD. Hashim Hussain -172U1A0553

UNDER THE GUIDANCE OF

Ms. K. Sree Lakshmi , M.TECH

ABSTRACT
In the background of the COVID-19 pandemic, institutions such as the
academy suffer a great deal from practically closed globally if the current situation is
not going to change. COVID-19 also known as Serious Acute Respiratory Syndrome
Corona virus-2 is an infectious disease that is released from an infected sick person
who speaks, sneezes, or coughs by respiratory droplets. This spreads quickly through
close contact with anyone infected, or by touching objects or surfaces affected with a
virus. There's still currently no vaccine available to protect against COVID-19 and
preventing exposure to the virus seems to be the only method to safeguard ourselves.
Wearing a facemask that covers the nose and mouth in a public setting and often
washing hands or the use of at least 70% alcohol-based sanitizers is one way to avoid
being exposed to the virus. Amid the advancement of technology, DDeep Learning
has proven its effectiveness in recognition and classification through image
processing. The research study uses deep learning techniques in distinguishing facial
recognition and recognize if the person is wearing a facemask or not. The dataset
collected contains 25,000 images using 224x224 pixel resolution and achieved an
accuracy rate of 96% as to the performance of the trained model. The system
develops a Raspberry Pi-based real-time facemask recognition that alarms and
captures the facial image if the person detected is not wearing a facemask. This study
is beneficial in combating the spread of the virus and avoiding contact with the virus.
INTRODUCTION

▪ The prevalence of COVID-19 in the world has manifested itself among the most
dangerous infectious diseases in the world. The situation is currently under attack
and is developing rapidly in WHO-declared countries . Based on epidemics in more
than 114 countries, these flu-like symptoms last 6 days (2-14 days). Millions of
people get sick every day . During this disaster, everyone must be smart and work
alone. This problem requires national governments, social agencies, and workplaces
to meet the required standards by continuously measuring and addressing public
health issues. Millions to millions of handshakes, viral bites and exchanging
accessories with other people . For this problem of violence, we define our work by
identifying masks for those who wear them and those who are neither outside nor in
crowded places.
o Learning is a real-world field for finding images, transforming images,
analyzing output, and buying them electronically. Based on image perception,
it is detected by a numerical number of the power of human perception. Visual
knowledge is associated with YOLO for each type of image detection. Plus,
find YOLO flexible anywhere, from organizations to family members. ,
Colleagues and friends on the move. The WHO announced possible speeches
by keeping distance and inserting . A face mask is required . Put on a face
mask that allows the camera to cover and change only the area of ​the face.
Watching computers is the next step in deep learning, especially in the field of
digital communication networks (CNN).
 LITERATURE SURVEY

AUTHOR METHOD YEAR

Sim, S. W., Moey, K. S. & Tan, N. C The use of facemasks to prevent 2014
respiratory infection: a literature
review in the context of the Health
Belief Model
K. He, X. Zhang, S. Ren, and J. Sun Deep residual learning for image 2016
recognition
A. G. Howard, M. Zhu, B. Chen, D. Efficient Convolutional Neural 2017
Kalenichenko, W. Wang, T. Weyand, Networks for Mobile Vision
M. Andreetto, and H. Adam Applications
A. Kamilaris and F. X. Prenafeta- Deep learning in agriculture: A survey 2018
Boldú
S. Kumar, A. Negi, J. N. Singh, and H. A deep learning for brain tumor MRI 2018
Verma images semantic segmentation using
fcn
 AUTHOR METHOD YEAR
Zhang, K. P., Zhang, Z. P., Li, Z. F. & Joint Face Detection and Alignment 2016
Qiao, Y Using Multi-task Cascaded
Convolutional Networks
Krizhevsky, A., Sutskever, I. & ImageNet Classification with Deep 2017
Hinton, G. E Convolutional Neural Networks.
Communications of the Acm 60
A. Krizhevsky, I. Sutskever, and G. E. Imagenet classification with deep 2012
Hinton convolutional neural networks
Yu, P., Zhu, J., Zhang, Z., & Han, Y A Familial Cluster of Infection 2020
Associated With the 2019 Novel
Coronavirus Indicating Possible
Person-to-Person Transmission During
the Incubation Period
HARDWARE REQUIREMENTS

Processor : Intel Core2Duo

Seed : 2.4GHz
RAM : 2 GB(minimum)
Hard Disk : 180 GB
SOFTWARE REQUIREMENTS

Operating System : Windows

Script : Python
Backend : Tensorflow
Processing : Keras Model
Packages : Open-cv
EXISTING SYSTEM

▪ In existing method we construct the model using only images and predict the result
my giving only as user input.

▪ There is no affective detection it compares with images and get the output by using
image processing
PROPOSED SYSTEM

▪ The proposed methodology involves video monitoring by using CNN with deep
learning frame works

▪ In this research study, deep learning techniques are applied to construct a classifier
that will collect images of a person wearing a face mask and not from the database
and differentiate between these classes of facemask-wearing and not facemask-
wearing .

▪ The artificial neural network has been demonstrated to be a vigorous procedure for
feature extraction from unprocessed data . This study proposes the use of a
convolutional neural network to design the facemask classifier and to include the
effect of the number of the convolutional neural layer on the prediction accuracy.
MODULE

 The first step of the real-time facemask recognition system is image acquisition. High-
quality images of the person posing with facemask wearing and not wearing facemask are
obtained through digital cameras, cellphone cameras, or scanners
 A Knowledge-based dataset is created by proper labeling of the captured images with
unique classes
 The obtained images that will be engaged in a preprocessing step is further enhanced
specifically for image features during processing. The segmentation process divides the
images into several segments and utilized in the extraction of facemask covered areas in
the person's face from the background.
 This section involves the convolutionary layers that obtain image features from the resize
images and is also joined after each convolution with the ReLU. Max and average pooling
of the feature extraction decreases the size. Ultimately, both the convolutional and the
pooling layers act as purifiers to generate those image characteristics
 The final step is to classify images, to train deep learning models along with the labeled
images to be trained on how to recognize and classify images according to learned visual
patterns. In this we are using CNN model for classification

CNN(Convolution Neural Network)

It 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. The pre-processing required in a ConvNet is much lower as compared to
other classification algorithms. While in primitive methods filters are hand-engineered, with
enough training, ConvNets have the ability to learn these filters/characteristics

A ConvNet is able to successfully capture the Spatial and Temporal dependencies in an

image through the application of relevant filters

The Convolutional Layer and the Pooling Layer, together form the i-th layer of a Convolutional
Neural Network
DATA DESIGN

Image Aquation

Annoted Dataset

Image Processing

Feature Extraction

Classification
UML DESIGN
RESULT
SCREENSHOTS
CONCLUSION

real-time facemask recognition through deep learning techniques by way of

Convolutional Neural Networks. This process gives a precise and speedily results for
facemask detection. The test results show a distinguished accuracy rate in detecting
persons wearing a facemask and not wearing a facemask. The trained model was able
to perform its undertaking using the VGG-16 CNN model achieving a 96% result for
performance accuracy. Moreover, the study presents a useful tool in fighting the spread
of the COVID-19 virus by detecting a person who wears a facemask or not and sets an
alarm if the person is not wearing a facemask.
The integration of several models of CNNs and compare each model with the highest
performance accuracy during training to increase the performance in detecting and
recognizing people wearing facemasks is suggested. Also, the researchers recommendd
a different optimizer, enhanced parameter settings, fine-tuning, and using adaptive
transfer learning models.
REFERENCES
▪ [1] Yu, P., Zhu, J., Zhang, Z., & Han, Y. (2020). A Familial Cluster of Infection Associated With the
2019 Novel Coronavirus Indicating Possible Person-to-Person Transmission During the Incubation
Period. The Journal of infectious diseases, 221(11), 1757–1761. https://doi.org/10.1093/infdis/jiaa077
▪ [2] Chavez, S., Long, B., Koyfman, A. & Liang, S. Y. Coronavirus Disease (COVID-19): A primer for
emergency physicians. Am J Emerg Med, https://doi:10.1016/j.ajem.2020.03.036 (2020).
▪ [3] World Health Organization. Coronavirus disease 2019 (COVID-19) Situation Report– 142, 2020,
[cited 10 June 2020], https://www.who.int/docs/default-source/coronaviruse/situationreports/
20200610-covid-19-sitrep-142.pdf?sfvrsn=180898cd_6
▪ [4] Bai, Y., Yao, L., Wei, T., Tian, F., Jin, D. Y., Chen, L., & Wang, M. (2020). Presumed
Asymptomatic Carrier Transmission of COVID-19. JAMA, 323(14), 1406–1407. Advance online
publication. https://doi.org/10.1001/jama.2020.2565
▪ [5] Centers for Disease Control and Prevention. Interim Infection Prevention and Control
Recommendations for Patients with Suspected or Confirmed Coronavirus Disease 2019 (COVID-19)
in Healthcare Settings. 2020 [cited 5 June 2020].
https://www.cdc.gov/coronavirus/2019-ncov/hcp/infection-controlrecommendations.html
THANK YOU