Iot Based System for Covid-19

Arjun Dange, Pranjal Barhate, Puja Bhole, Pooja Dabholkar, Priyanka Kad

Department of Computer Engineering,

Smt. Kashibai Navale College of Engineering,
Savitribai Phule Pune University,Maharashtra
[email protected], [email protected], [email protected],
[email protected], [email protected].

ABSTRACT - Preprocessing involves From the very basic hygiene standards to

dividing an image into various regions the treatments in the hospitals, people are
or into faces based on head-poses, such doing all they can for their own and the
as a 3D geometry or front/back face society’s safety; face masks are one of the
geometry, since algorithms, such as face personal protective equipment. People
recognition, often enable the input to be wear face masks once they step out of
present in different views. people use their homes and authorities strictly ensure
masks to shield themselves from the that people are wearing face masks while
effects of the COVID pandemic, which they are in groups and public places. To
means they wear facial masks to shelter monitor that people are following this
much of their faces Although some basic safety principle, a strategy should be
commonly used algorithms cannot be developed. A face mask detector system
extended to head pose estimation in the can be implemented to check this. Face
current case, there are a few that work mask detection means to identify whether
almost as well. and subsequent testing a person is wearing a mask or not. In our
various image-processing algorithms project our system is also check the
were able to eliminate the head issues, a temperature of person and pulse rate also.
method of classifying head poses by
using color images and a line-based face
image was created. The proposed II. MOTIVATION
approach involves the use of the HLS
color space as well as well as the face- we are seeking to identify a single
and grey images, and after that, uses a individual or a single thing, which is why
CNN to identify features for we have called it the in addition, our
classification. The results on the MAFA system will send a message to all users
dataset reveal that the suggested about sanitation Finally, one of our goals
approach has better efficiency than the of our project is to mirror the expansion
landmark-based algorithms. decreasing the number of involved
[entangled] patients Our primary
KEYWORDS: MAFA, CNN, COVID- motivation is relatively straightforward to
19 discover by means of image analysis. For
this mission, sensors will be examined:
I. INTRODUCTION Temperature and Oxygen Level
measurements will be taken. Instead of
The COVID19 pandemic is the most life- drawing attention to themselves, it avoids
changing event which has startled the attention to other users.
world since the year began. Affecting the
health and lives of masses, COVID-19 has III. LITERATURE SURVEY
called for strict measures to be followed in
order to prevent the spread of disease. The creator in [1]. In situations where

1
people pose ten different images of people and basic tasks are everywhere
in front of the camera at the same time, highlighted with the first function
certain packages of software include extraction in this task the reference
preprocessing such as the Head category implementation of this paper offers is an
for faces; software, for this reason, expansion of "Supervised and
incorporates faces into ten images of Unsupervised.
varying views. The creator in [2] / Fast-tracked Identification with Boosted
Explainable AI and Mass Surveillance Simple Attributes" is available in Paul
System-based Healthcare Framework to Viola & Jones' paper on Supervised and
Combat COVID-19 like Pandemics”. we Quick attribute cascade expansion Work
suggest the B3G5 network monitoring via this article in the basic understanding
scheme, using the low-latency and high of machine learning (EML) with more-
bandwidth to watch for diaphoresis’s, backed learning would benefit you, but
along with X-ray or CT images. you can simply follow along with the
The creator in [3], risk factors simple instructions without a prior
coupled with real time, indoor, personal knowledge of the concept. However, you
localization data for potential As Present can still go without learning things, when
we present a system that encapsulates all you just look at the overall situation to
the mentioned functionality via readily get a complete understanding of the big
available components (both hardware and picture.
software) and is further enhanced with
preliminary RTLS data acquisition, V. SYSTEM ARCHITECTURE
enabling post symptom detected.
The creator in [4], A Novel AI-
enabled Framework to Diagnose Corona
virusCOVID19 using Smartphone
Embedded Sensors. The creator in [5]
we propose a portable non-contact
method to screen the health conditions
of people wearing masks through
analysis of the respiratory characteristics
from RGB-infrared sensors.
The creators in [6] This report Fig V. System Architecture
addresses a topic of high cur- rent
interest - the ultraviolet susceptibility of VI. MODULE
SARS-CoV-2, the virus that causes
COVID-19 disease. Pre-processing is a common name for
operations with images at the lowest level
IV. PROPOSED SYSTEM of abstraction both input and output are
intensity images. The aim of p r e -
Haar cascade algorithm: - processing is an improvement of the
The system created in the name of the image data that suppresses unwanted
Hungarian physicist László Harai distortions or enhances some image
(spelled: László Harai) for face detection features important for further processing.
has for the first time reached real-time Feature Extraction Feature extraction is
effectiveness. The object recognition a part of the dimensionality reduction
technique in the system uses is called process, in which, an initial set of the raw
Haar, which can recognize objects in an data is divided and reduced to more
image or a collection of frames. Features manageable groups. So, when you want
can be used rapidly and easily in simple to process it will be easier.

2
Classification is a supervised machine covid-19 is higher compared to sars
learning approach, in which the algorithm coronavirus,” Journal of travel medicine,
learns from the data input provided to it 2020.
and then uses this learning to classify S. Feng, C. Shen, N. Xia, W. Song, M.
new.CONCLUSION Fan, and B. J. Cowling, “Rational use of
face masks in the covid-19 pandemic,”
Who will tell the person whether they can The Lancet Respiratory Medicine, 2020.
remove their ask to speak or keep it on?
We build a model that decides whether the
user has a virus or not, and whether the
mask is needed, but does not know
whether they are speaking or in the first
place.

VII. REFERENCES
Shuang Li, Xin Ning, Lina Yu, Liping
Zhang, Xiaoli Dong (2020),” Multi-angle
Head Pose Classification when Wearing
the Mask for Face Recognition under the
COVID19 Corona virus Epidemic.”
M. Shamim Hossain, Ghulam Muhammad,
and Nadra Guizani (2017),” Explainable
AI and Mass Surveillance System-based
Healthcare Framework
J. Barabas, R. Zalman and M. Kochlan
(2019),” Automated evaluation of COVID-
19 risk factors.
HalgurdS. Maghdid, Kayhan Zrar
Ghafoor, Ali Safaa Sadiq, Kevin Curran,”
A Novel AI-enabled Framework
Zheng Jiang, Menghan Hu, Zhongpai Gao
(2017) Detection of Respiratory Infections
using RGB- infrared sensors on
Portable Device.
Wladyslaw.J.. Kowal- ski (2020). COVID-
19 Coronavirus Ultraviolet Susceptibility.
P. A. Rota, M. S. Oberste, S. S. Monroe,
W. A. Nix, R. Campagnoli, J. P. Icenogle,
S. Penaranda, B. Bankamp, K. Maher, M.-
h. Chen et al., vol. 300, no. 5624, pp.
1394–1399, 2003.
Z. A. Memish, A. I. Zumla, R. F. Al-
Hakeem, A. A. Al- Rabeeah, and G. M.
Stephens, “Family cluster of middle east
respiratory syndrome coronavirus
infections,” New England Journal of
Medicine, vol. 368, no. 26, pp. 2487–2494,
2013.
Y. Liu, A. A. Gayle, A. Wilder-Smith, and
J. Rocklöv, “The reproductive number of