Face Mask Detection System using machine learning
Capstone project report
[Link] Name Enrollment Number
1. K Ashish 1713114018
2. Tarun Aggarwal 1713114049
3. Loveneesh Singh 18021140047
Under the Supervision of
Ashok Kumar Yadav
Professor
School of Computing Science and Engineering
Greater Noida, Uttar Pradesh
Fall 2020 - 2021
SCHOOL OF COMPUTING AND SCIENCE AND ENGINEERING
BONAFIDE CERTIFICATE
Certified that this project report “Face Detction using machine learning” is
the bonafide work of “Tarun Aggarwal(1713114049), k Ashish
(1713114018), Lovneesh singh(18021140047)” who carried out the project
work under my supervision.
Ashok Kumar Yadav
SIGNATURE OF DEAN SIGNATURE OF SUPERVISOR
Approval Sheet
This thesis/dissertation/report entitled Face detection using maching learning by Tarun Aggarwal, k
Ashish , lovneesh singh is approved for the degree of Computer Science and
engineering (Degree details).
Examiners
_____________
_____________
Supervisor (s)
Ashok Kumar Yadav
____________
____________
____________
Chairman
____________
Date:__11-05-2020_______
Place:___Greater Nodia______
Statement of Project Report Preparation
1. Thesis title: Face Detection using machine learning
1. Degree for which the report is submitted: Computer Science and Engineering
3 Project Supervisor was referred to for preparing the report.
4. Specifications regarding thesis format have been closely followed.
5. The contents of the thesis have been organized based on the guidelines.
6. The report has been prepared without resorting to plagiarism.
7. All sources used have been cited appropriately.
8 The report has not been submitted elsewhere for a degree.
Tarun Aggarwal,K Ashish, Lovneesh
(Signature of the student)
Tarun Aggarwal K Ashish Loveneesh singh
1713114049 1713114018 18021140047
Statement of Preparation:
Every student have to submit the statement of thesis preparation
TABLE OF CONTENT
[Link] Particulars
1. Abstract
2. Literature Reviews
3. Problem Formulation
4. Required tools
5. Feasibility Analysis
6. Complete work plan layout
7. Future Scope
ABSTRACT
We all are witnessing a pandemic, where in order to keep ourselves safe from the COVID
19 virus which has taken many lives till now and cases are increasing rapidly. Although
the government has issued guidelines to protect us from this virus .We can only fight this
by taking precautions. We constantly have to maintain social distance, wear masks and
wash hands as many times as possible, but sometimes we fail to do so and. Our
application helps identify these defaulters and helps in tracking them in real time.
Defaulters inside of a public space get identified by cameras and to keep a check on it our
application takes use of the existing surveillance system and with ease helps maintain the
[Link] our system, public spaces can be utilised with ease.
Keywords : Machine Learning, Artificial Intelligence, Python, PyQT5/PyQT4 , DBMS, Natural
Language Processing(NLP),Computer networking, Data .
Literature survey :
Face recognition has become one of the most active applications of visual pattern recognition
due to its potential value for law enforcement, surveillance, and human-computer interaction.
Progress has advanced to the point that face recognition systems are being demonstrated in real-
world [Link] rapid development of face recognition is due to a combination of factors:
active development of algorithms, the availability of a large database of facial images, and a
method for evaluating the performance of face recognition algorithms.
We have used the face recognition algorithm in our project. As the whole world is suffering from
Covid-19 pandemic and till now the virus has taken many lives. The only solution we have got is
to take precautions. Precaution includes wearing a mask and sanitizing hands regularly. But
everyone doesn't follow these precautions. We have come up with a solution, that people who are
not wearing a mask will get identified by cameras and to keep a check on it our application takes
use of the existing surveillance system. We will use face recognition techniques to accomplish
[Link] this simple method we can help to reduce the spread of coronavirus to the community.
Problem Formulation:
To flatten the curve, we have to follow the guidelines. But somehow people fail to do so.
Sometimes the defaulters forgot to wear a mask or some think it will not be an issue. They take
this lightly but they should understand the seriousness of this pandemic. Our application can
somehow help in this pandemic as our application will identify the people not wearing a mask
and will give alerts to the security out there and restrict the people by roaming out in public
without a mask.
Required Tools
● Front-End - PyQT5/PyQt4
● Back-End – Cv2
● Database
FEASIBILITY STUDY:
➢ Face Detection and recognition
○ Almost every premises uses CCTV camera
○ This app takes use of the already existing CCTV cameras
○ This app uses the CCTV cameras for detecting the defaulters
➢ Guards
○ Also every premises has security guards
○ Our app will communicate with the guards using their phones
○ Guards will receive alerts on their phone
○ They will receive the live image of the defaulters
➢ Technical Feasibility
○ Our app uses python which is a free, open-source programming language that is
available for everyone to use.
○ There are many libraries available in python which is also free to use
○ Our application uses PyQT5/PyQT4 for front-end which is a part of python which
is also free to use
○ The app can uses a machine learning algorithm for detecting the defaulters
➢ Alert
○ Almost everyone have mobile phones
○ This app uses the guard’s mobile phone to alert them
○ Even if the guard doesn’t have a smartphone our alert system will work through
by just sending a SMS.
○ It can help in directly contacting the guards via alarms or message.
➢ Future Scope
○ We can add thermal imaging which can help to find a covid infected person.
○ Alert system for any health emergency work by using AI.
○ To also check social distance is amongst them.
COMPLETE WORK PLAN LAYOUT:
● Phase -1 Basic Implementation (Image processing)
This will be the basic implementation of the software where system takes a photo using
cctv cameras and then after system process the image in identifying the face in photo
using cv2 module in python and then after it detect a separate eyes ,nose ,mouth and ear
and after this system if detect eye and ear but no nose and mouth then it wear a mask or if
it detect eye , ear and also detect nose and mouth then the person is not wearing a mask
and it will be unauthorized to access . One more algorithm is used to detect a mask by
making a dot image of the face and check ,If there is any rectangle or not ,if there is no
such a thing then it moves to phase 2.
● Phase -2 Alert
After the above implementation, the next phase will include the implementation of
Machine learning ,Natural language processing(NLP) and connectivity with databases.
We store the image of a person if he/she has not worn a mask and alert the guard by
sharing it live image feed( the image will access only on the same network ) so the guard
can catch the person and give a mask or make him/her out from the place.
This system has a simple UI which helps the admin to use. In the UI the admin can check the
time and database .
After compilation of these two phases, the app will be all ready to perform the same as
mentioned.
ER Diagram:
Future Scope:
. We can add thermal imaging which can help to find a covid infected person.
. Control entry gates by making them automated with the system
. Alert system for any health emergency work by using AI.
. To also check social distance is amongst them.
