FACE MASK CHECKER

A Capstone Project

Presented to Information Technology Department

Cebu Technological University

Argao Campus

Argao, Cebu

In Partial Fulfillment of the Requirements for the Degree of Bachelor of

Science in Information Technology

Cabungcag, Maribel P.

Calledo, April Kim S.

Fuentes, Mary Angel R.

Teoganco, Darren D.

2021
 ii

FACE MASK CHECKER

Darren D. Teoganco1, *April Kim S. Calledo2, Maribel P. Cabungcag3, Mary Angel

R. Fuentes4

ABSTRACT

Keywords: face mask detection, SSD, security personnel

Wearing a face mask in public spaces and congested settings is the most
effective prevention method. In 2020, COVID-19 being a pandemic disease had
rapid spread of the virus over the world. The respondents emphasized that they
are in favor of implementing governments action and laws against COVID-19 and
noted to follow. The accuracy rate prediction of face mask checker reached up to
89-90%, using a pre-trained model. The algorithm used by the system was a
deep learning technique which was a convolutional neural network, the Single
Shot Detector (SSD) algorithm. Support Vector Machine, a machine learning
algorithm was not applied in the system because it cannot support tertiary
datasets efficiently for face mask categories namely, with face mask, improper,
and no face mask.

The Face Mask Checker system would be a great help to the Security
Personnel in monitoring students, faculty, and staff who follow health and safety
protocols against COVID-19. The improper face mask detection was not
accurate. It produces a beep alert sound when no face masks were detected,
reminding them to follow minimum health protocol and wear their face masks.

It was recommended that there must be an image capturing feature when

no face mask was detected. Also, reports generated can be stored and
downloaded for security report purposes or able to produce smart reports.
Notification features such as an email or text message will be sent when the area
and risk status is 'danger' which is more than 10 people detected not wearing
face masks.
 iii
APPROVAL SHEET

The Capstone Project entitled Face Mask Checker prepared and submitted by Maribel
P. Cabungcag, April Kim S. Calledo, Mary Angel R. Fuentes, and Darren D. Teoganco has
been examined and is recommended for approval and acceptance.

RECOMMENDED: MELANIE R. ALBARRACIN, Dev.Ed CHRISTINE JOY A. TOMOL

Project Adviser Research Coordinator

======================================================

APPROVED by the Committee on Oral Examination with a grade of

__________________ on April _21_, 2022.

ENGR. RYAN H. TEO MORRIS DAYAGANON

Chairman Member

======================================================

ACCEPTED and APPROVED in partial fulfillment of the requirements in Bachelor of

Science in Information Technology.

CHRISTINE JOY A. TOMOL

Chairperson, IT Department

Date of Oral Defense: April 21 ,2022

iv
 ACKNOWLEDGEMENT

Making and working on this capstone project and research was not easy. Some
people contributed time and effort to the success of this study, which not only improved
this research but made it possible.
We loved to thank our family and parents for giving us the encouragement,
support, and means to reach our aspirations in life. We’re so grateful to the Python
Expert Instructor of the Information Technology department, Dr. Melanie R. Albarracin
as our Project Adviser, who choose our proposed research title and guided us to
improve this research project and system in every chapter.
We would also love to thank our research respondents, who undoubtedly
bestowed their time on answering our prepared questionnaire for this study. We’re so
thankful to our friends, notably to Mr. Mel Jay Llanos suggested this research project
title development.
We would love to thank the faculty and staff of the Cebu Technological
University-Argao Campus. To the cooperation of our team members, who encoded and
decoded the necessary parts and requirements of this capstone project and research.
We would always be grateful for the help of the panelists, Dr. Christine Joy A. Tomol,
Mr. Morris Dayaganon, and Engr. Ryan H. Teo gave suggestions to improve this
research project.
We would love to extend our warm gratitude to our Subject Adviser, Dr.
Christine Joy A. Tomol of Capstone Project and Research, who gave her all-out support
and guided us to enhance and make this research possible throughout the school year.
And especially to our sincere thanks to Almighty God, who gave us strength,
sincerity, and faith to finish this research until the end.
 v

TABLE OF CONTENTS

Page
TITLE PAGE i
ABSTRACT ii
APPROVAL SHEET iii
ACKNOWLEDGEMENT iv
TABLE OF CONTENTS v
LIST OF TABLES viii
LIST OF FIGURES ix
CHAPTER
1 THE PROBLEM AND ITS SCOPE 1
INTRODUCTION 1
Rationale of the Study 1
Theoretical Background 3
Theoretical Framework of the Study 3
Review of Related Literature and Studies 10
Statement of the Problem 21
Scope and Limitations of the Study 22
Significance of the Study 23
RESEARCH METHODOLOGY 24
Design 24
Flow of the Study 25
Locale of the Study 26
Respondents 28
 vi
Data Instruments 28
Treatment of Data 29
SYSTEM METHODOLOGY 30
Requirements Specifications 30
Technical Feasiblity
32
Schedule Feasibility 33
Requirements Modelling (IPO) 34
Object Modelling 36
Flowchart of the System 38
Data Design (ERD) 39
System Architecture 40
List of Modules 41
Testing Plan 42
Definition of Terms and Glossary 44
CHAPTER
2 PRESENTATION, ANALYSIS, AND INTERPRETATION OF DATA 47
CHAPTER
3 SUMMARY OF FINDINGS, CONCLUSION, AND
RECOMMENDATIONS 55
Summary of Findings 55
Conclusion 57
Recommendations 57
System Graphical User Interface (GUI) Design 59
BIBLIOGRAPHY 75
APPENDICES 89
Appendix A Transmittal Letter 90
Appendix B Questionnaire 91
vii
Appendix C Project Working Title Form 93
Appendix D Project Adviser Appointment Form 94
Appendix E Grammarian’s Certification 95
Appendix F Curriculum Vitae 96
 viii

LIST OF TABLES
Table Table Name Page

1 Comparative Matrix for Related Projects 20

1.1.1 Face Mask Checking and Conforming to the Regulation 47
1.1.2 Security Personnel-Recordkeeping 49
1.2.1 Students-Face Mask Checking 50
1.2.2 Students-Conforming to the Regulation 51
2.1.1 Possible Functionalities to Monitor Wearing of Face Mask Policy 53
 ix
LIST OF FIGURES

Figure Figure Name Page

1 The Theoretical Framework of the Study 3

2 Facemask Detection 16
3 No Facemask Detection 16
4 Real-Time Facemask Recognition with Alarm System 17
5 Proposed model of COVID-19 Face Mask Detection 18
6 Using Eigenfaces and LBPH Real-time Mask Detection and Face
Recognition for Attendance System 18
7 Face Mask Detector 19
8 Face Mask Detection 20
9 Flow of the Study 26
10 Location Map of the Research Environment 27
11 Gantt Chart 33
12 Input-Process-Output Diagram 34
13 Use Case Diagram of the Super Admin and Admin 36
14 Face Mask Detection Live Video Stream operated by the Admin 37
15 Flowchart of the System 38
16 Entity - Relationship Diagram 39
17 Network Topology 40
18 Main Landing page of the Web Application, Face Mask Checker 59
19 Super Admin Log in page 60
20 Super Admin Home page 61
21 Add Admin at Admin Account Management page 62
22 View Admin at Admin Account Management page 63
23 Update Admin at Admin Account Management page 64
24 Delete Admin at Admin Account Management page 65
25 Super Admin Account Settings page 66
26 Super Admin System Settings page 67
27 Super Admin Log out pop-up screen 68
28 Admin Log in page 69
29 Admin Home Live stream page 70
30 Admin Account Settings page-View Account 71
31 Admin Account Settings page-Edit Account 72
32 Admin System Settings page 73
33 Admin Log out 74
 1

CHAPTER I

THE PROBLEM AND ITS SCOPE

INTRODUCTION

Rationale of the Study                   

In 2020, COVID-19 being a pandemic disease had rapid spread of the

virus over the world. Wearing a face mask in public spaces and congested

settings is the most effective prevention method. The World Health Organization

(WHO) has developed several guidelines for preventing coronavirus transmission

wherein the first suspected case in the Philippines was investigated in January

2020. Wearing a face mask is essential in all indoor public spaces and covered

areas. Businesses had a difficult time getting started and staying open due to the

COVID-19 virus. Some irresponsible people refuse to wear masks with so many

excuses. COVID-19 is observed to transmit primarily among persons who are

close to one another (almost 6 feet), but it can also be disseminated by persons

who do not have symptoms, are unaware that they are infected, and don't have

protective clothing. 

According to the report of the World Health Organization, the country has

increased health and safety protocols including the wearing of face masks

because of increased cases. All public, private areas, and hospitals are monitored
 2

by security personnel if the people are wearing their face masks when entering

establishments in these areas.

Presently, in Cebu Technological University – Argao Campus, the school’s

university faculty, staff, and students are wearing face masks and face shields to

enter the campus. Monitoring them manually wearing face masks will not be

easy in all areas of the campus. So, an automation process of detecting those

who are wearing and not wearing face masks is one of the solutions. 

In this regard, the researchers are motivated to conduct a study on

developing a face mask detection system that has an alert feature. This aims to

detect or check a person who has a face mask and without a required face mask.

This system, the face mask detector, aims to produce generated reports

accurately based on what is viewed in the camera that is placed in a position that

reaches an area that is passed by many people. The alert feature aims to remind

the security personnel and let people know to wear their face masks preventing

the spread of the virus. This will bridge the gap between the regulation that

keeps people safe from the virus and the people who violate the health and

safety protocol regulation, one of which is wearing face masks.  

Face masks of various types have become synonymous with the

pandemic. People were initially urged not to wear them to conserve the limited

quantity for healthcare professionals. Now that facial coverings are more

commonly available, there is increasing evidence of their utility. Face masks are
 3

an important barrier to preventing the virus from spreading in public settings

(Denton Media, 2020). So, the researchers observed that many people didn't

follow the protocol. This system will help and save more lives of people. It can

make those people responsible as humans, so the spread of viruses will be

minimized.

Theoretical Background

Fig. 1: The Theoretical Framework of the Study

 4

This study will be based on image processing theory. Image processing

started with satellite imagery, wire-photo standards conversion, medical imaging,

videophone, character recognition, and photograph enhancement were all

applications of image processing developed in the 1960s at Bell Laboratories, the

Jet Propulsion Laboratory, Massachusetts Institute of Technology, University of

Maryland, and a few other research facilities. The goal of early image processing

was to boost the image's quality. It was created with the intention of improving

people's visual effects. The input of image processing is a low-quality image, and

the output is a higher-quality image. According to Anbarjafari (2014), it is a sort

of signal processing in which the input is an image and the output is an image or

its associated features.

Image processing involves three steps: Importation, analysis, and

manipulation of the image. Digital image-processing techniques are used by

computers to manipulate digital images. For tangible copies such as prints and

photographs, analog image processing can be utilized. Pre-processing,

augmentation, and display, as well as information extraction, are the three

general processes that all sorts of data must go through when using digital

techniques.

An image processing system consists of a light source to illuminate the

scene, a sensor, commonly a CCD (Charged-coupled device)-camera, and a

 5

sensor-to-computer interface. The interface transforms analog data into digital

data that the computer can understand. This happens in a particular piece of

hardware called a frame grabber, which also serves as a storage device for the

image. Image processing results will be transmitted to the outside world via I/O

interfaces and standard output devices such as monitors, speakers, printers,

projectors, GPS, and so on (Erhardt & Ferron, 2000).

In the field of AI, image processing technology is widely used for image

classification in the machine and deep learning algorithms to train models on

datasets. It’s specifically used for object and face detection or recognition. To

deal with a limited number of samples or datasets, traditional machine learning

algorithms such as multilayer perception machines, support vector machines

(SVM), and so on usually use shallow structures. CNN is good at dealing with

image classification and recognition difficulties and has improved the accuracy of

many machine learning tasks, the convolution neural network (CNN) created in

recent years has been widely used in the field of image processing. It's evolved

into a powerful and widely used deep learning model (Xin & Wang, 2019) for

solving real-life problems.

This study will be anchored by SVM, which was developed by Vladimir

Vapnik with his colleagues (Vapnik et al., 1997), it’s one of the toughest

prediction methods based on statistical learning frameworks or VC theory

 6

proposed by Chervonenkis (1974) and Vapnik (1982, 1995). VC theory seeks to

describe the learning process statistically, as does the theory of building learning

machines.

Before storing the image, some pre-processing steps are performed, such

as masking, illumination, and histogram equalization. Unwanted noise, such as

the background pattern, is removed from the objects of interest during the

masking process. Then histogram equalization is used to manipulate the color

distribution in pictures.

The images of the classes no face mask improper mask, and face mask

are used to train the SVM with the kernel and upper bound margin values. After

obtaining a decision surface through testing, the run-time scheme is applied to

photos that do not include faces, and misclassifications are saved to be used as

negative examples in subsequent training phases.

Support Vector Machines (SVM) are efficient when having a very high

dimensional space. SVM is memory efficient since it uses a subset of training

points to distinguish various objects. Several non-face images can be trained in

SVM for different types of images. Support Vector Machines are adaptable for

different decision functions, it can describe different kernels as long as the

results are right. The selection of the type of kernel that is most productive for

our application is based on the requirements and application.

 7

The study will also use deep learning algorithms of Artificial Intelligence in

general. One of the models that will be used is Convolutional Neural Network

(CNN) for the face mask detection module. The goal is to bring images to finer

pieces into a form that is easier to process without losing features (e.g., eyes,

mask, hair, etc.) (Saha, 2018).

Face detector models in Convolutional Neural Network-based classification

learn directly from the user's data and then apply multiple deep learning

algorithms to it (Ren, et al., 2015). Cascade CNN was developed in 2007 by Li,

Lin, Shen, Brandt, and Hua (2015). Yang, Yan et al. (2015) proposed the concept

of face feature aggregation in the face detection model.

CNN plays a significant part in computer vision-related examples in

recognizing patterns, on account of its less computation cost and also the ability

of spatial extraction. A residual network that can take in personality planning

from the last layer. Mobile Network is proposed for article locators. It's called

"Mobile Network" by He et al., N.D. (He et al.). Mobile Net is a lot lower

compared to networks utilizing standard convolutions.

In addition, as image processing evolved, the health and medical field are

one of its benefactors. As the coronavirus disease of 2019 or known as COVID-

19 worldwide, affects the normal life people had. It affects the health of a person

and its way of transmission is by the droplets of saliva from the infected person
 8

through coughs, sneezes, or by simply talking. The World Health Organization

recommends and mandates everyone to wear a face mask when going public

areas to prevent the spread of the virus, to protect oneself, and to stay safe.

On April 9, 2020, the Department of the Interior and Local Government

(DILG) in coordination with The Inter-Agency Task Force for the Management of

Emerging Infectious Diseases (IATF–EID) established by Executive Order No.168,

s.2014, under the constitutional policy enshrined in Article II, Section 15 of the

1987 Philippine Constitution, which states that the State shall protect and

promote the people's right to health and instill health awareness among them.

The DILG released a Memorandum Circular No. 071-2020, stating a ‘Mandatory

of Wearing a Face Mask or other Protective Equipment in Public Areas’ starting

that day, whenever they are allowed to go out of their homes at all times.

In coordination with IATF also, the Department of Health (DOH) has a

campaign released in 2020 entitled ‘BIDA Solusyon sa COVID-19’, wherein B

stands for ‘Bawal Walang Face Mask, I is for ‘I-sanitize ang mga kamay, iwas-

hawak sa mga bagay’, D for ‘Dumistansya ng isang metro’, and A for ‘Alamin ang

totoong impormasyon’. This is to prevent further increase of the COVID-19 cases

in the country. This is a reminder to all people to follow health protocols and

regulations specifically in wearing face masks in going to public areas. According

to the Philippine President, Pres. Duterte (2021), the authorities must take action
 9

to strictly implement this regulation and those who violate will be punished

according to the penal code.

To the authorities who monitor all people who violate and follow the

regulation, that will be cumbersome since they have to be awake 24/7 (Suresh,

et al., 2021). This research aims to develop a system that requires, checks and

detects people who wear face masks in real-time through cameras and sees to it

if they follow the regulation. This study will also look at the effectiveness of

wearing face mask health regulations of the government. The data sets that are

with masks, no masks, and improper masks to be trained will be from real faces

of people, Kaggle, and other web resources. The system works in real-time

detection of people who wear face masks or not. Then, send a notification alert

to the authorized personnel or admin about the person who violates. Then, the

violator’s facial image will be captured. It will be stored and for recordkeeping

purposes.

During the COVID-19 pandemic, the facemask detection model has

become a must-have system. Since manually tracking whether people are

wearing a facemask in public and crowded areas is cumbersome, this study will

develop a real-time automated model that will be integrated with webcams,

surveillance cameras in public areas and detect whether people are wearing face

masks and report to the appropriate authorities using computer vision.

 10

Review of Related Literature and Studies

This section presents concepts according to various authors and other

studies related to this research which will serve as the foundation of the study.

Review of Related Literature

Technological advancements paved the way for Information Technology,

also abbreviated as IT, in general, refers to the use of computers, software, and

data which is concerned in getting, capturing, analyzing, evaluating, and storing

data.

In the corporate world, Information Technology plays a vital role in the

business industry. According to Stokdyk (2019), its role is to build communication

networks, protect data and information in a company’s databases, guide

employees in troubleshooting mobile devices and computers, and other

necessary tasks to ensure the security and efficiency of business systems.

IT and Pandemic

IT aims to solve real-life problems and make it convenient and easier for

people to do day-to-day tasks. In today’s time of the COVID-19 pandemic, the

implementing regulation of the Philippine government to the public is health and

safety protocols by washing and spraying alcohol to hands, wearing their face

masks when going outside, and maintaining social distance. How about making

use of technology advancement of IT to minimize the number of people who will

 11

be prone to the virus. The wearing of face masks is mandatory and those who

will violate this health regulation will be punished. This research aims to develop

a system that requires, checks and detects people who wear face masks in real-

time through cameras and sees to it if they follow the regulation. This will bridge

the gap between the regulation that keeps people safe from the virus and the

people who violate the said protocol and regulation.

IT and Artificial Intelligence

So, as IT is a solution, it keeps on improving, evolving and innovating. It

needs a collaboration or a partner technology to achieve it, which is Artificial

Intelligence. There were so many studies and propositions about Artificial

Intelligence way back before until Alan Turing published "Computing Machinery

and Intelligence" in 1950, in which he suggests an imitation game, later known

as the "Turing Test." The proposition is that if a computer could have a

conversation (via teleprinter) that was indistinguishable from a conversation with

a person, it was fair to conclude that the machine was "thinking" (Press, 2016).

So, the beginning of Artificial Intelligence (AI) was like a science fiction story,

away from reality and an impossible before however today, it’s part of reality and

is possible such as face recognition, speech recognition, automation, and many

more. As stated by My Computer Career (2021), AI has become an important

part of the innovation of the future as well as IT because many corporate

industries depend on it.

 12

Face Masks detection system is one of the talks amidst this pandemic

because many researchers wanted to help prevent the rise of COVID-19 cases.

According to Deb (2020), one of the pioneering researchers from India of the

Face Mask detection system, her motivation was that the public must ensure

safety and there were no efficient face mask detection applications that

transportation, residential districts, enterprise, and manufacturers need.

Based on the work of Rosebrook (2020), when his system project, COVID

19: Face Mask detection is deployed correctly, this has the potential to help

ensure everyone’s safety but he will let the medical professionals decide on if

they want to implement it in their establishments.

Face Mask detection systems use AI, which is an emerging technology,

whose role is to build intelligent systems that are capable of learning, reasoning,

adapting, and performing tasks similar to humans. In this research, let’s say a

case scenario that a security guard or authorized personnel is a human who

monitors and sees if the people wear face masks in the surveillance camera

while what if there is a computer system that acts as the guard or personnel

artificially, that is the Face Mask Checker. Since not all the time the guard or

personnel have an eye on the people 24/7.

AI can modernize old methods of IT systems to do optimized tasks. It is

one of the keys of the IT industry to innovate their systems to intelligent ones

applying it in their IT operations. The core operations in the collaboration of IT

 13

and AI are automation and optimization. The following are data security, building

better information systems, and processing automation (SoulPage, 2020).

AI for IT operations (AIOps) uses machine learning and big data. Machine

learning is an application of AI that automates data processing from past

examples and new examples of data to build a model that can predict results.

According to SoloLearn (n.d.), one example is when a credit card company gives

a call to a user when they notice suspicious activity in the user’s account, they

use a model based on past activity to know this anomalous behavior.

Face Mask Datasets

In the case of face mask detection, there are existing data sets that will

be trained by the machine learning model. The data sets are images of people

with face masks, improper face masks, and no face masks. The data sets will be

searched on the web and from Kaggle, the Face Mask Level Dataset (FMLD),

Real-world Masked Face Dataset (RMFD), or pre-trained datasets. Kaggle is a

Google LLC affiliate, is an online group of data scientists and machine learning

practitioners. In a web-based data science environment, Kaggle allows users to

find and publish data sets, as well as explore and create models (Wikipedia,

2021).
 14

Machine Learning Techniques

Machine learning gives a way to teach a computer to recognize what it

sees and let the computer learn in the same way that a human can do. That’s

the core of machine learning in Artificial Intelligence (Moroney, 2019).

Support Vector Machine (SVM) is a machine learning model that is a

supervised learning model associated with other learning algorithms to analyze

data for regression analysis and classification. SVM was developed by Vladimir

Vapnik with his colleagues (Vapnik et al., 1997), it’s one of the toughest

prediction methods based on statistical learning frameworks or VC theory

proposed by Chervonenkis (1974) and Vapnik (1982, 1995).

Support Vector Machine model classifier will be used as a binary classifier

to recognize between the face mask, improper face masks, and no face mask

faces. For a lesser amount of data sets, SVM is suitable however if there are

many data sets, it needs other deep learning algorithms or random forests to

function. There was a research that uses a hybrid algorithm in their machine

learning application and that one of which, they use SVM with 3 datasets to

train. This method will efficiently work with face mask recognition or detection

because its accuracy rate reached 99.5 to 100% (Pan, et al., 2020).
 15

Deep Learning Techniques

Deep learning is an evolution of machine learning of Artificial Intelligence

and it’s the same as neural networks. The idea is it has multiple neuron layers

like the human brain. Each layer is learning a different type of representation

(Gordon, 2021). CNN implementations are found in image processing, computer

vision, self-driving systems, and many more (Militante and Dionisio, 2020).

The task of CNN is to bring images to finer pieces into a form that is

easier to process without losing its features (e.g., eyes, mask, hair, etc.) which is

crucial to get good prediction results (Saha, 2018). This research will be applied

in this research with data sets to train in the CNN model and get accurate

predictions when a person has a mask or no face masks.

Google's MobileNetV2 is a general-purpose computer vision with

convolutional neural network algorithms. It will do the feature extraction of faces

with masks and no face masks for detection and segmentation, Sandler and

Howard (2018) said. The research was part of the TensorFlow-slim image

classification library. In addition, mobileNetV2 will do the fine-tuning process to

the model that has already been trained for a task by adding extra classification

layers (Fine-tune in MobileNetV2 from Kaggle, 2020). Then, training the entire

model with image net weights to make it perform a second similar task (Joshi,

2020), and so on.

 16

Review of Related Studies

The Facemask and Physical Distancing Detection with Alarm Systems,

research from the University of Antique by Militante and Dionisio (2020), detects

a person who is wearing a facemask or not and detects if the person in an area

does observe physical distancing. They use a convolutional neural network (CNN)

algorithms model. Focusing on their Face Mask Detection with an alarm system,

another module feature it comprises is that it signals an alarm then captures

facial images of the person detected who does not wear a mask. It reached an

accuracy rate of 97%.

Fig. 2: Facemask Detection Fig. 3: No Facemask Detection

In the Real-Time Facemask Recognition with Alarm System using Deep

Learning by still Militante and Dionisio (2020) from the University of Antique, this

system uses a Raspberry Pi-based real-time facemask recognition module, with a

face mask and no face mask detection. It also signals an alarm then captures

facial images of the person detected who does not wear a mask. The proposed
 17

model they use is convolutional neural network (CNN) algorithms. It reached an

accuracy rate of 96% with a face mask and no face mask detection which would

be beneficial in preventing the spread of the COVID-19 virus.

Fig. 4: Real-Time Facemask Recognition with Alarm System

 18

The COVID-19 Face Mask Detection of Information Technology Medi-Caps

University at India, research by Maurya, et al. (2021), is a system that has

modules on Face Mask

Detection and beep alert

generation. Their system

aims to screen a person not

wearing a face mask. When

a person is detected not wearing a face mask, a beep alert sound can be heard

and stops when they wear their face mask on and properly. The proposed model

they use combines hybrid deep learning and classical machine learning, a simple

convolutional neural network (CNN) algorithm. The trained model reached an

accuracy rate of 95%.

Fig. 5: Proposed model of COVID-19 Face Mask Detection

The Real-time Mask Detection and Face Recognition using eigenfaces and

LBPH for Attendance System of Kuching Community College at Malaysia,

research by Suhaimin, et al. (2021), system consists of modules on mask

 19

detection, face detection, face recognition, and attendance report generation.

The face and face mask detection uses the haar cascade classifier model in

machine learning. When the face mask is detected, the 'Using Mask' message will

be shown that indicates the person has to remove her mask to allow face

recognition for attendance.

Fig. 6: Using Eigenfaces and LBPH Real-time Mask Detection and Face
Recognition for Attendance System
The Face Mask Detector by Purohit (2020), has face mask detection and

alert system modules. The proposed model he used is deep Learning models

such as YOLOv2, YOLOv3, and YOLOv4. YOLO means You Only Look Once, is

another real-time object detection algorithm of deep learning. The alert system

has been composed of mask count, no mask count, and status of the area which

were ‘safe’, ‘warning’, and ‘danger’. His test model reached an accuracy rate of

78.95% - 93.95% while the trained model

reached 83.83% – 99.65%.

 20

Fig. 7: Face

Mask

Detector

The Face

Mask

Detection using

machine

learning models by Pan, et al. (2020) from Stanford University uses Multinomial

Logistic Regression model and Support Vector Machine (SVM) algorithms such as

linear, RBF, and polynomial kernels. Their system’s data sets are a mask, no

mask, and incorrect. The trained model reached an accuracy rate of 93.61%.
 21

Fig. 8: Face Mask Detection

Table 1.
Comparative Matrix for Related Projects
Real-time
Facemask Real-Time
mask
and Face-
COVID- detection
Physical mask Face Face Face
System / 19 Face and face
Distancing Recogniti Mask Mask Mask
Features Mask recognitio
Detection on with Detector Detection Checker
Detection n for
with Alarm Alarm
attendanc
Systems System
e system
Displays Date and Time ✔ ✔ ❌ ✔ ❌ ❌ ✔
Face Mask Detection-3 ❌ ❌ ❌ ❌ ❌
categories or classes (With (With and No (With and (With and (With and (With and
✔ ✔
Mask, Improperly Worn Mask Face Mask No Face No Face No Face No Face
and No Mask) only) Mask only) Mask only) Mask only) Mask only)
Live Count of With Mask and
❌ ❌ ❌ ❌ ✔ ❌ ❌
No Mask
No Mask Image
Capture
✔ ✔ ❌ ❌ ❌ ❌ ❌

Alarm/Beep alert ✔ ✔ ✔ ❌ ❌ ❌ ✔

Alert Area & Risk Status

features (Safe, Warning,
Danger-SMS Alert
❌ ❌ ❌ ❌ ✔ ❌ ❌
to Admin)

Beep Alert of No &

Improper face ❌ ❌ ❌ ❌ ❌ ❌ ✔
masks
System Generated
❌ ❌ ❌ ✔ ❌ ❌ ❌
reports
Total Percentage 37.50% 37.50% 12.50% 25% 25% 12.50% 50%
Legend: ✔ - Applicable ❌ - Not Applicable

The system, Face Mask Checker will include the following major modules;

Face Mask Detection (data sets: With Mask, Improper Mask and No Mask) with a

display of date and time, live count of the three datasets, an alert system that

displays the area status and risk status then also those who do not wear their
 22

mask properly and no mask will be captured, and the system can have generated

reports of the data managed by the administrator.

The above-mentioned studies will be considered essential to the present

study because they confirmed the use of Face Mask Detection systems in this

time of pandemic to help protect people and prevent the spread of the COVID-19

virus by wearing a face mask when going outside.

Statement of the Problem

This study aims to design and develop a system that will check and detect

the person on a real-time video stream at CTU-Argao Campus, wearing a face

mask, the improper wearing of a mask, and no mask in webcam or surveillance

cameras using computer vision, machine, and deep learning algorithm.

This study will answer the following questions:

1. What is the current status of health and safety regulation specifically

wearing of face mask at school in terms of:

1.1. Face Mask Checking;

1.2. Conforming to the regulation;

1.3. Recordkeeping?

2. What functionalities that can be implemented to help the CTU-Argao

campus monitor the face mask policy?

 23

3. Based on the findings of the study, what system can be designed and
implemented?

Scope and Limitations of the study

The proposed system aims to detect the person on a real-time video

stream at CTU-Argao Campus wearing a face mask, the improper wearing of a

mask, and no mask in webcam or surveillance cameras combined with Computer

Vision to detect those face mask categories implementing machine learning and

deep learning algorithms of AI Technology. The proposed system will be user-

friendly, allowing the user admin to see who is wearing a face mask by the video

stream captured by the camera frame.

In multi-face mask detection, it can predict the face mask categories

within 3-5 meters. Then, when people are detected with no face mask, the

system will beep as an alert notification reminding them to wear their face masks

properly. The Admin can take appropriate actions to make them follow the health

protocols specifically the wearing of face masks.

The system’s other scope is that it will be connected with computer/s with

webcams at an institution’s premises. This system will use an internet connection

for IP cameras or other cameras and the system’s functionalities since it’s a web

application. For the system’s limitations, the proposed system will not use face

recognition to the faces of the person captured by the system then sends a
 24

notification message reminding them to wear face masks properly. It cannot

capture the image of the people who don’t follow the proper wearing of face

masks.

Significance of the Study

The proposed system of this study aims to detect the person wearing a

face mask, the improper wearing of a mask, and no mask in webcam or

surveillance cameras and sees to it if they follow the regulation through a real-

time video stream at CTU-Argao Campus. Here are the significant end-users who

will benefit from the proposed system is expected to provide:

CTU-Argao Campus. This system will help students, faculty, and staff is aware

of wearing their face masks to be safe, monitored by authorized personnel in

surveillance.

Health Institution. This system will help the health workers, patients, and staff

to be aware to always wear their face masks regulated by authorized personnel

in surveillance.

General Public. The demand for protection from the virus is exceedingly great.

Even in public settings, especially at unguarded entrances, crowds must be

reminded and able to wear masks. Face mask checker and detection will be

particularly useful for authorized staff in surveillance.

 25

Future Researchers. They can use this research as a guide in conducting

related studies of this system as a reference and as to improve this research.

RESEARCH METHODOLOGY

The study's research methodology will be covered in this chapter. The

researchers outline the research design, environment of the study, research

method, method of data collection, research process, and ethical considerations.

Design

This research will utilize the descriptive research design. A descriptive

design is used to describe the current state of events, people, or topics. It

usually makes some comparisons and contrast linkages that can occasionally be

established to some level in carefully planned and choreographed researchers

(Alceso, 2011). This is to look for an overview of the study's variables (Bhandari,

2021) by an interview and observation in the project environment. This study will

also follow the software development life cycle, the agile software development

which is flexible in responding to change. This framework is an evolutionary and

iterative approach to software creation. It is broken down into several sub-

modules, or mini-projects, or also called sprints. Each sub-module is delegated to

its assigned team or member and goes through the entire production process,

from concept to development, testing, and documentation (Bhupinder, 2018).

 26

Flow of the Study

Input. The research strategic flow of the study commenced with the current

face mask checking at the CTU-Argao campus. It will also cover the data, record,

and report that shows who abides by the regulation.

Process. The process stage will begin with observation of the current status of

face mask regulation and gathering through personal interviews of the Security

personnel and selected students at CTU-Argao Campus and interpreting the

data. The planning phase will be completed to determine why a face mask

detection and checker should be developed. After planning, analysis was carried

out to determine who uses the system, what it is, what it does, and where and

when it is used. To determine exactly how the system works, a design is created.

(Teo, 2016), then followed by testing.

Output. The output of the study is a proposed Face Mask Checker system

design.
 27

INPUT PROCESS OUTPUT

•Data Gathering • Proposed Face
•Cuurent Face
Mask Checking through Mask Checker
•Abides Face observation and
mask regulation personal interview
•Recordkeeping of the Security
personnel and
some students of
CTU-Argao
Campus
•Interpretation of
data
•Planning
•Analysis
•Design
•Testing

Fig. 9: Flow of the Study

Locale of the study

This study will be conducted at Cebu Technological University in Argao,

Cebu, which is about 1km from the national highway. It is located directly to the

right of Secretary Cerge Remonde Sports and Cultural Complex. The campus is

bordered on the north by Dr. TS. Kintanar Street, on the south by Bonifacio

Street, on the east by Ed. Kintanar Street, and on the west by Columbus Street.

Argao is a 1st municipal income class municipality in the province of Cebu,

the Philippines located 68 kilometers from Cebu City. The rectangle that is Argao

– roughly 20 km x 40 km (12 mi x 25 mi) – is flanked in the west by Badian,

 28

Alcantara, Ronda, and Dumanjug; in the north by Sibonga; and in the east by

the Bohol Strait, in that order (TheTravelLad, 2016). Argao is the southern part

of Cebu with a population of more or less than 62,000 people. The municipality

occupies a total land of approximately 20, 753 hectares. Barangay Lamacan is

where Cebu Technological University is located.

Fig. 10: Location Map of the Research Environment

 29

Respondents

The respondents will be the security personnel and students of CTU -

Argao campus. There will be 5 students who will be interviewed randomly across

the BSIT department or other departments. The interview will be done via phone

call, messenger chat, or call. According to Albiso, et. al (2018), researchers

mostly collected notes to aid in the analysis of the obtained data. Respondents

were permitted to voice any thoughts or ideas they wanted throughout the

interview, even if they weren't on the list of questions.

Data Instruments

This study uses interviews and descriptive observation as instruments.

The researchers will interview the security personnel and students of the school

consisting of close and open-ended questions based on the current situation of

the research environment. An interview is a dialogue in which information is

gathered. An interviewee and an interviewer are involved in a research interview.

The interviewer manages the interaction and asks questions, while the

interviewee responds to the questions. Interviews might take place in person or

over the phone. The internet is also becoming more popular as a tool for

conducting interviews (Easwaramoorthy & Fataneh Zarinpoush, 2001).

 30

Treatment of Data

This descriptive research will be conducted in these phases: firstly, is to

gather the evidence to create a framework capable of creating the system;

secondly, check possible changes can be done based on the interview of the

school admin in surveillance in CTU-Argao Campus; thirdly, is to search relevant

datasets for face mask checking and detection - with face mask, the improper

wearing of face mask, and no face mask classes or categories; fourthly, to

choose the appropriate number of datasets or pre-trained datasets for each

category that matches the face masks used at school with the use of face

shields.

In addition, the findings will be validated through documentation. The

method takes an SDLC agile approach, in which questions to be addressed in the

interview can make possible changes required by the system development. To

interpret the close-ended interview questions, this study will also utilize Thematic

Content Analysis (TCA) to determine the qualitative responses of the research

respondents that will be generated from the interview. Interview transcripts from

research participants or other identifiable writings that reflect experientially on

the topic of investigation are examples of qualitative data (Anderson, 2007). The

results will be represented by tables, figures, or graphs to illustrate their

 31

responses based on the statement of the problem, the system’s report

generation feature, and interpretation of the study.

SYSTEM METHODOLOGY

Requirements Specifications

The following will be the requirements specification aiming to be applied

to the system, Face Mask Checker and Detection System.

Non-Functional Requirements

User-Friendliness

The system's user interface will be simple, easy to understand, and will provide a

pleasant experience for the user.

Interoperability

With the bare minimum of technical requirements, the proposed system will work

smoothly.

Performance

Even with multiple simultaneous log-ins, the suggested system will run smoothly

as long as it is connected to the same network.

 32

Robustness

This system’s face mask checker and detection system algorithm will

automatically detect the face mask categories, with mask, improper face mask,

and no face mask. In the admin section, the proposed system will validate all

inputs and data. Request assistance can't move on to the next step unless

they've satisfied or fulfilled all of the requirements.

Maintainability

The system will be programmed in a way that the following developers can

comprehend for future changes, development, and error tracking.

Reusability

The system will be built to anticipate future changes in time and will be

accessible to researchers and users in the future.

Security

The system will contain a login/logout functionality for security reasons, allowing

only account administrators to access it.

 33

Technical Feasibility

The following software and hardware specifications will be used during the

development process.

Software Specification

Operating System: Windows 10 and higher

Hardware Specification

Central Processing Unit (CPU) Processor: Intel® i3 processor or higher

Graphics Processing Unit (GPU): Intel UHD Graphics, AMD

Programming Development Environment

Front End. Django and OpenCV – Django is a python framework for web

applications composed of HTML, CSS, and Javascript scripting languages with

Model-View-Template (MVT) architectural design pattern. Furthermore, the

system will use OpenCV (Open Computer Vision) python library for identification

of face masks in human faces using MobileNet V2.

Back End. Python, TensorFlow, and Keras – Python is a high level programming

language ideal for building machine and deep learning applications. Tensorflow

and Keras are python libraries used for machine learning and deep learning

models implementation, specifically image classification and processing of No

face mask, Improper face mask, and With face mask datasets. The proposed

system will be developed in PyCharm Integrated Development Environment

(IDE). In addition, MySQL is used to communicate with databases. MySQL Server

 34

and Workbench serves as the medium for interacting with the database and will

be used to locally host the server.

Schedule Feasibility

Fig. 11: Gantt Chart

The Gantt chart figure above shows the researchers’ timeline and

management of deliverables that will be submitted. From planning to the

proposal. Analyzing the stage of the research and development coding of the

project takes the most time because this will be the catalyst to finding the

problems and possible solutions the system can provide.

 35

Requirements Modelling (IPO)

INPUT PROCESS OUTPUT

•Current problem •Gather training or • Proposed Face
on Face Mask pre-trained Mask Checker
Detection and samples/datasets
checking Systems •Face Detection
in Pandemic and Mask
•Pre-trained Classification-test
model for datasets dataset models
to be used which with SVM and CNN
are with Face algorithms
Mask, Improper, •Analysis on
and No Face Mask prediction or
detection of face
mask categories
•Check the results

Fig. 12: Input-Process-Output Diagram

Input. The research strategic flow of the study commenced with the current

problem of face mask detection and checking systems in pandemic. It also

covered datasets to be used with Mask, Improper Wearing of Mask, and No

Mask.

Process. The process stage begins with the gathering of training datasets and

samples. The data sets will be pre-trained model or Face-Mask Level datasets
 36

available in GitHub and Kaggle (Batagelj, B., Peer, P. et al., 2021). The training

phase which is also the design phase of the dataset models for Face Detection

and Mask Classification will be implemented using Support Vector Machine (SVM)

and Convolutional Neural Network (CNN) algorithms. Afterward, an analysis will

be done on how to fill in the gap in this research which is to generate reports in

the system. In addition to that is to add the important features on the face mask

detection main module. When the above features will be implemented, checking

the functionality of the system will be done to test the system’s accuracy rate if it

will reach good results.

Output. The output of the study is the report generation and its proposed Face-

Mask Checker system design.

 37

Object Modelling

Fig. 13: Use Case Diagram of the Super Admin and Admin

The figure above illustrates the allowable actions the Super Admin and

Administrator can do. The Super Admin has to create the account of the Admin

before he/she can access the system in face mask checking and detection live

video stream. Then, give the credentials to the authorized Security Personnel

Admin. The Super Admin can view the designated area in surveillance and can

manage the accounts of the Admins. The Admin has access to all privileges and
 38

can perform all the functionalities of admin user level. Additionally, Admin can

view the face mask detection video stream and edit the system settings of the

system. On the other hand, the Super Admin can have full power access to the

entire network and system and control to the Admins of the system.

Fig. 14: Face Mask Detection Live Video Stream operated by the Admin

The figure above illustrates after logging in of the Security Personnel

Admin. He/She can monitor those who wear face masks when in the Home page

of the system where the Face Mask detection live video stream is placed. The

Admin has access to all privileges and can perform admin user level

functionalities.
 39

Flowchart of the System

Data and Process Modelling

Fig. 15: Flowchart of the System

The system flowchart depicts the system's operation. It's used to lay down

the logic of the system before starting the coding to show it's automated

procedure. It can assist in demonstrating how code is organized, visualizing code

 40

execution within a program, displaying the structure of an application, and

understanding how users navigate the software.

Data Design (ERD)

 41

Fig. 16: Entity - Relationship Diagram

The figure above is the table entity-relationship diagrams that the

proposed system will utilize. Database implementation will be made possible by

Python, and MySQL local server and workbench.

 42

System Architecture

Fig. 17: Network Topology

In a hybrid approach, analog cameras will be used with webcam in

computers or laptops, or analog CCTV transmission to a control point or control

room, where video compression will be applied. The control points are connected

via an IP network (EFSIC Global, 2010). In essence, a hybrid solution gives one

of the best network architecture.

 43

List of Modules

The following are the list of modules in Face Mask Checker and Detection:

SUPER ADMIN

● LOG IN
● LOG OUT
● HOME

Ø View the total number of Admins

● ADMIN ACCOUNT MANAGEMENT

Ø Add admin

Ø View admin

Ø Update admin

Ø Delete admin

● SUPER ADMIN ACCOUNT SETTINGS

Ø Edit account

● SYSTEM SETTINGS

Ø Edit system settings

ADMIN

● LOG IN
● LOG OUT
 44

● HOME

Ø Face Mask Detection live video stream page

Ø Beep Alert Notification

● ADMIN ACCOUNT SETTINGS

Ø View account

Ø Edit account

● ADMIN SYSTEM SETTINGS

Ø Edit system settings

Testing Plan

Test Plan. The scope, technique, resources, and timing of proposed testing

operations can all be described in a test plan. The test planning process

produces a high-level test plan document that describes the software objects to

be tested, the degree of tester independence, the test environment, the test

case design, and test measurement techniques to be employed, and the

reasoning behind their selection (Ahamad, 2012).

Unit Testing. Unit testing's basic purpose is to separate the smallest bit of

tested software in an application, separate the rest of the code, and see if it

functions exactly as the developer intended. The researcher utilized this method

to test each component of the system to confirm that each unit or module
 45

performs as planned before being connected with the others. This testing

procedure has demonstrated its worth by detecting a high percentage of

problems during its application (Figueiredo, 2016).

Integration Testing. This technique of testing is used to evaluate the system's

compliance with specified requirements on a complete, integrated system. The

researchers used this strategy to ensure that the user could complete the task

using the program. The system went through this procedure to see if there were

any flaws with the client's understanding of how the system worked. The

researchers assessed each tool used in the system's development to determine

software and hardware compatibility, as well as to discover faults in the system

as a whole. 

Acceptance Testing. The proponents utilized this testing method to evaluate

whether or not the system matched its acceptance criteria, allowing the

respondent to decide whether or not to accept it.

Usability Testing. Usability testing focuses on a user's ease of use, the

application's flexibility in handling controls, and the system application's capacity

to accomplish its functions. For the purpose of determining how simple and user-

friendly the system is and to expose usability flaws, a small group of intended

end-users uses the system or software program (Hamilton, 2021).

Definition of Terms and Glossary

 46

Admin. A user of the system. The admin of this study is the Security Personnel

assigned to monitor a specific area in an institution.

Artificial Intelligence (AI). Emerging technology in Science, whose role is to

build intelligent systems that are capable of learning, reasoning, adapting, and

performing tasks similar to humans.

Computer Vision. It is a branch of artificial intelligence (AI) that allows

computers and systems to extract useful information from digital photos, videos,

and other visual inputs, as well as to conduct actions or make recommendations

based on that data (IBM, n.d.). Furthermore, it is a python library for the

identification of face masks in human faces using MobileNet V2.

COVID-19. In December 2019, in Wuhan, China, a new critical respiratory

illness has infected millions of people and lost millions of helpless lives in more

than 200 countries based to the World Health Organization (WHO) and declared

a global pandemic.

Dataset. The datasets are images of people with face masks, improper face

masks, and no face masks. Possibly, the data sets will be searched on the web

and from Kaggle.

Deep learning. Evolution of machine learning in Artificial Intelligence and it’s

the same as neural networks. The idea is it has multiple neuron layers like the

human brain. Each layer is learning a different type of representation (Gordon,

2021). In this study, it takes input through an image, then entrusts essential
 47

various aspects or objects which refer to the features (e.g., eyes, face mask,

hair, etc.) in the image and can differentiate from the other (Saha, 2018).

Django. Django is an open-source python framework for web applications

integrated with HTML, CSS, and Javascript scripting languages with a Model-

View-Template (MVT) architectural design pattern. While working at the

Lawrence Journal-World newspaper, Python programmers Adrian Holovaty and

Simon Willison created it. The Django Software Foundation maintains it

(Wikipedia, 2022). It enables the creation of secure and maintained websites in a

short amount of time.

Face Mask. The Philippine government mandated the use of a face mask. A

piece of personal protective equipment is worn by people nowadays during the

pandemic and by health professionals. The mask blocks the movement of

pathogens shed in respiratory droplets and aerosols into and from the wearer's

mouth and nose.

Machine learning. It gives a way to teach a computer to recognize what it sees

and let the computer learn in the same way that a human can do. That’s the

core of machine learning in Artificial Intelligence (Moroney, 2019). In this study,

this model will be used as a binary classifier to recognize between a face mask,

the improper wearing of a face mask, and no face mask.

MobileNet V2. MobileNetV2 is a substantial advancement over MobileNetV1 in

terms of classification, object identification, and semantic segmentation for

 48

mobile visual recognition. The TensorFlow-Slim Image Classification Library

includes MobileNetV2 (Sandler and Howard, 2018).

Python. A scripting language that is an interpreted, general-purpose, high-level

programming language would be used in this study.

Single-Shot Detector (SSD). A Single-Shot Detector is made up of two parts:

a backbone model and an SSD head. As a feature extractor, the backbone model

is commonly a pre-trained image classification network. This is a MobileNet that

has had the final fully linked classification layer removed (ArcGIS, n.d.).

Super Admin. The superuser of this study is the Security Personnel who

manages the accounts of the Admin.

CHAPTER II
 49

PRESENTATION, ANALYSIS, AND INTERPRETATION OF DATA

This chapter covers the presentation, analysis, and interpretation of the

data gathered on the current status of health and safety regulation specifically

wearing of face masks and functionalities of the proposed system to help

monitor the face mask policy by the Security Personnel and students of CTU-

Argao Campus. The responses of the respondents are translated into English

language for readability purposes. Moreover, since this study is also

developmental in nature, this chapter also presents the basis for the final list of

modules based on the assessment of the current status.

I. CURRENT STATUS OF HEALTH AND SAFETY REGULATION

SPECIFICALLY WEARING OF FACE MASK AT SCHOOL

1.1. Security Personnel

Table 1.1.1.
Face Mask Checking and Conforming to the Regulation

Theme Responses

Prone to the virus infection if not “For me, I agree because that’s for
following health protocols our health, and safety in this COVID-
19 situation. It’s difficult when we’re
at work and gets the virus, we have
homes to go back from work. And
that’s for safety as well for us, so I
agree to that.”
 50

Face mask policy is monitored “Yes Ma’am, that’s our protocol.

mandated by the government That is what we are monitoring now
and we’re just following the
protocol mandated from the IATF
memorandum for that.”

Some of them remove their face mask “For us Ma’am, once they enter the
when inside the school premises school, they are required to wear face
masks. There are tendencies that
while they’re inside, they will remove
it but if we see them, we will ask them
why they remove their mask. They
shall follow the policy.”

Table 1.1.1 shows that the Security Personnel gave his comment

regarding the current status of health and safety regulations specifically wearing

face masks in terms of face mask checking and conforming to the regulation or

protocol in CTU-AC. It was found out that when they do not wear face masks

when going outside, they are prone to virus infection if not able to follow health

protocols especially in wearing a face mask and always consider the social

distancing policy. Another is that the face mask policy is monitored and

mandated by the government because wearing a face mask reduces the amount

of virus we breathe in and breathe out.

According to Harvard Health Publishing (2021), they also always urged

unvaccinated people to wear a mask indoors, as well as anyone at greater risk,

regardless of the extent of community transmission. The last conforming

regulation is that some of them remove their face mask when inside the school
 51

premises, they can't deny the fact that they remove their facemask when inside

the school because it causes heat and disruption of their breathing.

Since the current practice turned out to be prone to the virus infection if

not following health protocols and some of the people who go inside the school

premises remove their face masks, a system should be in place to address the

problem. The Image Processing theory is a process of performing operations on

an image to create a better image or extract important information (Anbarjafari,

2014). In artificial intelligence, image processing technology is widely used for

image classification in the machine and deep learning algorithms to train models

on datasets to create useful smart systems. It’s specifically used for object and

face detection or recognition.

Table 1.1.2.
Security Personnel-Recordkeeping

Theme Responses

“For me Ma’am, we are more in favor

Not all the time on the watch because of that. Yes, we are in favor.”
they have other duties
“It will greatly help us Ma’am to
It will help them lessen their work lessen our work in monitoring and we
can conduct other work that is still
related to our duties not just focusing
on looking if they wear face masks.
We have other tasks such as roving,
checking, and securing other school
properties. It can take time for us just
because of monitoring them.”
 52

Table 1.1.2 shows that the Security Personnel commented on the current

status of health and safety regulations on wearing face masks in terms of

recordkeeping. It was known that the security personnel did not have time to

watch every person because they had other duties, and it would help them

lessen their work. The security personnel did not keep records of those who

wore face masks and did not wear face masks.

Therefore, a system should be in place to address the problem of applying

the image processing technology used for object and face detection or

recognition on people wearing face masks.

1.2. Students

Table 1.2.1.
Students-Face Mask Checking

Theme Responses

Wearing of face mask in 100%

school

Type of face Surgical face mask 100%

mask they wear
N95 20%

Washable face mask 20%

Table 1.2.1 shows that the five respondents commented on the current

status of health and safety regulations on wearing face masks in terms of face

mask checking when they went to the university.

 53

It was found out that students agreed to the system if it will be

implemented at school. Wherever they are at school, 100% of the students are

always wearing face masks to be fair to others and prevent the spread of

viruses. Since then, wearing a face mask not only protects others from expelled

respiratory droplets but can also protect them from other people’s illnesses.

Wearing a face mask is a must, so people need to use face masks for their

health and safety the protection from COVID-19. So, the researchers found out

that a Surgical mask is the most preferable to use by the students with 100%,

but rarely N95 and washable face masks, both are 20% worn by them. A face

mask significantly reduces virus transmission in the community by preventing

anyone, including those who are unwittingly carrying the virus, from transmitting

it to others.

Table 1.2.2.
Students-Conforming to the Regulation

Theme Responses

Following the health and safety 100%

protocol inside the school

Removing face mask when inside the

school because causes heat and 99.9%
disruption of breathing

Trust and support to the government 100%

actions to COVID-19
 54

Table 1.2.2 shows that the five respondents commented on the current

status of health and safety regulations, specifically wearing face masks in terms

of whether they conform to the health regulation or protocol.

It was found out that all of the students are always wearing their face

masks with 100% preference when they go inside the school and are following

the health and safety protocol.

Since wearing a face mask is a must, the students can’t deny the fact that

there are times they will not wear or remove their face mask when inside the

school with 99.9% preference because it causes heat and disruption of their

breathing. With the intent of preventing further increase of COVID-19, DILG

strictly implemented a mandatory wearing of a face mask to all residents in

public areas. All respondents agreed to the memorandum of DILG about the

wearing of face masks with 100% preference because wearing face masks help

prevent the spread of the virus. It is also a way of protecting their body health,

in a way that they can inhale air that has a droplet of a COVID-19 virus, so a

face mask can filter throughout the mouth. This implementation gives the

student’s a big help to protect not just themselves, but also the safety of others.

Wearing a face mask is vital, anyone is obliged and must obey the protocols

because it is a way to mitigate the spread of the virus.

 55

II. FUNCTIONALITIES THAT CAN BE IMPLEMENTED TO HELP

MONITOR THE FACE MASK POLICY IN CTU-AC

2.1. Security Personnel

Table 2.1.1
Possible Functionalities to Monitor Wearing of Face Mask Policy

Theme Responses

“It will greatly help us Ma’am to lessen

our work in monitoring and we can
conduct other work that is still related to
our duties not just focusing on looking if
they wear face masks. We have other
Face mask checking tasks such as roving, checking, and
and detecting securing other school properties. It can
take time for us just because of
monitoring them.”

“Yes Ma’am, that will be useful and so

important if there is something like
that.”

Table 2.1.1 shows that the Security Personnel gave his comment

regarding the possible functionality of the proposed system to help monitor the

face mask policy in CTU-AC. It was found out that to monitor the wearing of face

masks policy is through face mask checking with automation on detection.

Since the current practice turned out that they do not keep on watch all

students, staff, and other visitors at school, a system should be in place to

address the problem in the application of image processing technology used for
 56

checking objects and face detection or recognition on people wearing face

masks.
 57

CHAPTER III

SUMMARY OF FINDINGS, CONCLUSION, AND RECOMMENDATIONS

Chapter three consists of three parts: (1) Summary of Findings, (2)

Conclusion, and (3) Recommendations. In addition, it presents the Graphical

User Interface (GUI) of the system showing the screenshots of the design.

Summary of Findings

The respondent perceived that if not following the health protocols,

specifically wearing face masks, the students, faculty, and staff are prone to

virus infection of COVID-19 when they go inside the school premises and remove

their face masks.

It was also mandated by the government to follow health and safety

protocols during this pandemic. Also, the respondent clarified that he/she cannot

monitor the whole time those who wear face masks or not because he/she and

other Security Personnel like him/her have other duties and they didn’t keep a

record of them. So, a system can be used to address their concern in the

monitoring of wearing face masks. The respondents emphasized that they are in

favor of implementing governments action and law against COVID-19 and noted

to follow.
 58

The study utilized the descriptive research design. This is used to describe

the current state of events, people, or topics. This is to look for an overview of

the study’s variables by an interview and observation in the project environment.

This study also follows the software development life cycle, the agile software

development which is flexible to change.

Moreover, it covered the design, gathering of pre-trained datasets and

models, development, testing, implementation, and checking the accuracy results

of face mask checking and detection. The accuracy rate prediction of face mask

checker reached up to 89-90%, using a pre-trained model. In addition, the

algorithm used by the system is a deep learning technique which is a

convolutional neural network, the Single Shot Detector (SSD) algorithm. Support

Vector Machine, a machine learning algorithm is not applied in the system

because it can not support tertiary datasets efficiently for face mask categories

namely, with face mask, improper, and no face mask.

The integrated functionalities of the proposed system are face mask

detection, admin account management, and system settings management. The

mentioned functionalities made up the Face Mask Checker system.

 59

Conclusion

In conclusion, the Face Mask Checker system will be a great help to the

Security Personnel in monitoring students, faculty, and staff who follow health

and safety protocols against COVID-19 mandated by the IATF, DOH, DILG, and

the government. As well, it helps to monitor health workers, patients, and staff

at health institutions and other public areas.

Additionally, this allows Security Personnel to monitor and see through the

live video stream, to those people within the frame of a certain area. It can be

placed in any other areas where people usually pass by. The improper wearing

of face mask detection is not consistent in prediction. Thus, it produces a beep

alert sound when no face masks were detected, reminding them to follow

minimum health protocol and wear their face masks.

Recommendations

1. For hardware recommendation, build this deep learning web application

app or system in an NVIDIA GPU- supported computer, a CPU of Intel i3

and higher or Ryzen 3 and higher;

2. For the software recommendations, fully implement the Face Mask

Checker system;

3. Implement YOLO algorithm to display the live count of face mask

 60

categories - with face mask, improper, and no face mask for fast face

mask detections and so that the accuracy rate of prediction and detection

increases;

4. Notification feature such as an email or text message to the Admin

Security Personnel will be sent when area and risk status is ‘danger’ which

is more than 10 people detected of not wearing face masks;

5. Super Admin Security Personnel can view all the live video cameras of

their Admin Security Personnel;

6. Add an image capturing feature of no face mask people in the frame;

7. Produce smart reports;

8. Reports generated can be stored and downloaded for security report

purposes;

9. Conduct use case scenarios evaluation to maximize systems functionalities

to improve it;

10. Maintain the system periodically to know possible changes to implement

which are asked by the users.

 61

System Graphical User Interface (GUI) Design

Landing Page

This page is the main landing page wherein the users can navigate and

choose user-level accounts to log in as either a Super Admin or an Admin

Security Personnel.

Fig. 18: Main Landing page of the Web Application,

Face Mask Checker
 62

SUPER ADMIN

Log in Page

The log-in page allows the Super Admin Security Personnel, who was the

superuser of the system allowing them to input their credentials which were their

registered email and password, enabling them to access the system features.

Fig. 19: Super Admin Log in page

 63

Home Page

This page is displayed after successfully logging in to the system. It shows

the total number of Admin Security Personnel and the importance of wearing a

face mask, as well as the details of the government mandating the use of a face

mask, as well as the Department of Interior and Local Government (DILG)

Memorandum No. 071-2020 in coordination with the Inter-Agency Task Force for

the Management of Emerging Infectious Diseases (IATF-EID) with Executive

Order No. 168, s.2014 and the Department of Health (DOH) with the campaign

released ‘BIDA Solusyon sa COVID-19’.

Fig. 20: Super Admin Home page

 64

Admin Account Management

This module consists of pages where the Super Admin can add a new

account of an Admin Security Personnel, view, edit, and delete Admins’ accounts.

Add Admin

The Super Admin can add a new account of an Admin Security Personnel

then savers the Admin’s information.

Fig. 21: Add Admin at Admin

Account Management page

View Admin
 65

This module allows the Super Admin the view the saved information of the

Admin Security Personnel.

Fig. 22: View Admin at Admin

Account Management page

Update Admin
 66

This module allows the Super Admin to edit and update the accounts of

Admin Security Personnels.

Fig. 23: Update Admin at Admin

Account Management page

 67

Delete Admin

This module allows the Super Admin delete the Admin Security

Personnels’ accounts.

Fig. 24: Delete Admin at Admin

Account Management page

 68

Super Admin Account Settings

This module allows the Super Admin to change their log-in credentials,

username, and name. The email cannot be altered in the front-end design unless

there will be tinkering in the source codes at the back-end.

Fig. 25: Super Admin Account Settings page

 69

System Settings

This module allows the system's color layout design to be customizable by

the Super Admin.

Fig. 26: Super Admin System Settings page

 70

Log out

The Super Admin can log out of the system using this module by

navigating at the upper right corner, then clicking the username with an icon of

the super admin profile.

Fig. 27: Super Admin Log out

 71

ADMIN

Log in Page

This page allows the Admin Security Personnel user to input their

credentials which are their registered email and password given by the Super

Admin to log in and access the system. It also gives users the option of changing

their passwords. If forgotten, the system will send an email to change their

password.

Fig. 28: Admin Log in page

 72

Home Page

This page is displayed after the Admin successfully logs in to the system.

It shows the current date, time, camera ID number, and place assigned in

monitoring face masks of the Security Personnel. When no face masks and

improper face masks are detected, the box colors are red and blue. Then, it will

beep in real-time, reminding the people in the frame to wear their face masks

properly. When the box color is green, indicating proper face masks are

detected.

Fig. 29: Admin Home Live stream page

 73

Admin Account Settings

View Account

The Admin can view their information in this module.

Fig. 30: Admin Account Settings page-View Account

 74

Edit Account

The Admin can modify their login credentials, username, and name using

this module.

Fig. 31: Admin Account Settings page-Edit Account

 75

Admin System Settings

This module allows the Admin to customize the color layout design of the

system.

Fig. 32: Admin System Settings page

 76

Log out

This module allows the Admin to log out into the system found in the

upper right corner of the page.

Fig. 33: Admin Log out

 77

BIBLIOGRAPHY

Journal

Batagelj, B., Peer, P., et al. (2021). How to Correctly Detect Face-Masks for

COVID-19 from Visual Information? Applied Sciences, 11(5), 1–24.

https://doi.org/10.3390/app11052070

Hussain, S., Yu, Y., et al. (2021). IoT and Deep Learning Based Approach for

Rapid Screening and Face Mask Detection for Infection Spread Control of
COVID-19. Applied Sciences, 11(8), 1–16.
https://doi.org/10.3390/app11083495

Jignesh Chowdary, G., Punn, N. S., et al. (2021). Face Mask Detection Using

Transfer Learning of InceptionV3. International Conference on Big Data

Analytics, 12581(2020), 81–90. https://doi.org/10.1007/978-3-030-66665-
1_6

Li, T., & Liu, Y. (2020). Mask or no mask for COVID-19: A public health and

market study. PLoS ONE, 15(8), 1–12.

https://doi.org/10.1371/journal.pone.0237691

Loey, M., Manogaran, G., et al. (2021). A hybrid deep transfer learning model

with machine learning methods for face mask detection in the era of the
COVID-19 pandemic. Measurement, 167(108288), 1–2.
https://doi.org/10.1016/j.measurement.2020.108288

Matusiak, Ł., et. al, (2020). The use of face masks during the COVID ‐19
pandemic in Poland: A survey study of 2315 young adults. PMC.
Published. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7361243/
 78

Maurya, P., Nayak, S., & Vijayvargiya, S. (2021). COVID-19 Face Mask Detection.

Icarset 2nd International Conference on Advanced Research in Science,

Engineering, and Technology , 1–5. https://www.dpublication.com/wp-
content/uploads/2021/03/28-418.pdf

Meivel, S., Indira Devi, K., et al. (2021). Real time analysis of unmask face

detection in human skin using tensor flow package and IoT algorithm.
Materials Today: Proceedings , 1.
https://doi.org/10.1016/j.matpr.2020.12.864.

Militante, S. V., & Dionisio, N. V. (2020). Deep Learning

Implementation of Facemask and Physical Distancing Detection with Alarm

Systems. 2020 Third International Conference on Vocational Education
and Electrical Engineering (ICVEE), 1–5.
https://doi.org/10.1109/ICVEE50212.2020.9243183

Militante, S. V., & Dionisio, N. V. (2020). Real-Time Facemask

Recognition with Alarm System using Deep Learning. 2020 11th IEEE
Control and System Graduate Research Colloquium (ICSGRC) , 106–110.
https://doi.org/10.1109/ICSGRC49013.2020.9232610

Nagrath, P., Jain, R., et al. (2021). SSDMNV2: A real time DNN-based

face mask detection system using single shot multibox detector and
MobileNetV2. Sustainable Cities and Society, 66(102692), 1–9.
https://www.sciencedirect.com/science/article/pii/S2210670720309070

Pooja, S., & Preeti, S. (2021). Face Mask Detection Using AI.
SpringerLink,293–305.https://link.springer.com/chapter/10.1007/978-981-
33-4236-1_16

Suhaimin, M. S. M., Hijazi, M. H. A., et al. (2021). Real-time mask detection and
 79

face recognition using eigenfaces and local binary pattern histogram for
attendance system. Bulletin of Electrical Engineering and Informatics ,
10(2), 1105–1113. https://doi.org/10.11591/eei.v10i2.2859

Vinitha, V., & Velantina, V. (2020). COVID-19 FACEMASK DETECTION WITH

DEEP LEARNING AND COMPUTER VISION. International Research Journal

of Engineering and Technology (IRJET) , 7(8), 3127–3132.
https://www.irjet.net/archives/V7/i8/IRJET-V7I8530.pdf

Xin, M., & Wang, Y. (2019). Research on image classification model based on
deep convolution neural network. EURASIP Journal on Image and Video
Processing, 40.
https://jivp-eurasipjournals.springeropen.com/articles/10.1186/s13640-
019-0417-8

Websites

A. (2005, November 30). What is a Test Plan? Software Testing - Share

Knowledge About Software Testing.

https://testingsoftware.blogspot.com/2005/11/what-is-test-plan_30.html

Affandi, A. (2021, June 5). How do I set a condition in HTML button tags to stop
my Webcam Video Stream on webpage | Python Django . Stackoverflow.
Retrieved April 16, 2022, from
https://stackoverflow.com/questions/67850593/how-do-i-set-a-condition-
in-html-button-tags-to-stop-my-webcam-video-stream-on-w

AI, T. (2020, October 14). Build Face Mask Detection Model In 5 Minutes .

Towards DataScience. https://towardsdatascience.com/you-can-do-it-too-

build-face-mask-detection-model-in-5-minutes-727d83497a8c

Alceso, M. A. (2011, July 10). Descriptive research. Slideshare.

https://www.slideshare.net/Angilo/descriptive-research
 80

Año, E. (2020, April 9). Mandatory Wearing of Face Masks or other Protective

Equipment in Public Areas. DILG.

https://dilg.gov.ph/issuances/mc/Mandatory-Wearing-of-Face-Masks-or-
other-Protective-Equipment-in-Public-Areas/3195

Argao trip planner. (n.d.). Inspirock. Retrieved September 19, 2021, from

https://www.inspirock.com/philippines/argao-trip-planner

Artificial Intelligence is Changing the Information Technology Sector. (2018,

December 28). Idexcel Technologies.

https://www.idexcel.com/blog/artificial-intelligence-is-changing-the-
information-technology-sector/

B. (2020, March 11). Opinion about wearing of face mask. E-Edukasyonph.

https://e-edukasyonph.com/english/opinion-about-wearing-of-face-mask-
6193911

Banerjee, S. (2020, June 13). Face Mask Detection using Google Colab. Analytics

Vidhya. https://medium.com/analytics-vidhya/face-mask-detection-using-
google-colab-160f7b20cdea

Basic evaluation measures from the confusion matrix. (2016). Wordpress.

https://classeval.wordpress.com/introduction/basic-evaluation-measures/

#:~:text=Accuracy%20is%20calculated%20as%20the,dataset%20(P
%20%2B%20N).

Bhandari, P. (2020, June 19). An introduction to qualitative research. Scribbr.

https://www.scribbr.com/methodology/qualitative-research/

Bhupinder. (2018, October 1). 9 reasons to choose Agile Methodology for Mobile
 81

App Development. Packt. https://hub.packtpub.com/9-reasons-to-choose-

agile-methodology-for-mobile-app-development/

BIDA Solusyon sa COVID-19. (2020). Department of Health. Retrieved

September 20, 2021, from https://doh.gov.ph/bidasolusyon

Braulio. (2020, July 9). Face mask detection in street camera video streams

using AI: behind the curtain. Tryo Labs.

https://tryolabs.com/blog/2020/07/09/face-mask-detection-in-street-
camera-video-streams-using-ai-behind-the-curtain/

CCTV Network Topology, Synthesis, and Network-Realization . (n.d.).

WebTechnologies. Retrieved September 19, 2021, from

https://www.webtechnologies.co.tz/networking-solutions/cctv-network-
topology-synthesis-and-network-realization.html

Andrew, C. (2021, September 27). Face-Mask-Detector. GitHub. Retrieved April

15, 2022, from https://github.com/achen353/Face-Mask-Detector
Cisco. (n.d.). File Server Cisco Network Topology Icons 3015. Vecta.Io. Retrieved
March 14, 2022, from https://vecta.io/symbols/240/cisco-network-
topology-icons-3015/78/file-server

Dickson, B. (2020, January 6). What are convolutional neural networks (CNN)?

TechTalks. https://bdtechtalks.com/2020/01/06/convolutional-neural-
networks-cnn-convnets/#:~:text=Convolutional%20neural%20networks
%2C%20also%20called,a%20postdoctoral%20computer%20science
%20researcher.&text=The%20early%20version%20of%20CNNs,)%2C
%20could%20recognize%20handwritten%20digits.

Digital Image Processing. (2014, January). University of Tartu.

https://sisu.ut.ee/imageprocessing/book/1

Django (web framework). (2022, March 24). Wikipedia. Retrieved March 29,
2022, from https://en.wikipedia.org/wiki/Django_(web_framework)
 82

Escarlate, A. (2020, June 9). My quarantine project: a real-time face mask

detector using Tensorflow. Towards DataScience.

https://towardsdatascience.com/my-quarantine-project-a-real-time-face-
mask-detector-using-tensorflow-3e7c61a42c40

Face Mask Detection System using Artificial Intelligence . (n.d.-b). LeewayHertz.

Retrieved September 18, 2021, from https://www.leewayhertz.com/face-

mask-detection-system/

Face Mask Detection System using Artificial Intelligence . (n.d.). Trident.

Retrieved September 18, 2021, from https://www.tridentinfo.com/face-

mask-detection-systems/

Face masks mandatory in several places. (n.d.). Government of the Netherlands.

Retrieved September 19, 2021, from

https://www.government.nl/topics/coronavirus-covid-19/face-masks-
mandatory-in-all-indoor-public-spaces

Feng, J. (n.d.). Importance of Face Mask Compliance and Safety for Businesses .

Turing. Retrieved September 19, 2021, from https://turingvideo.com/face-

mask-compliance-and-safety-for-businesses/

Figueiredo, R. (2016, April 27). What exactly is a “unit” in unit testing? The

Blinking Caret. https://www.blinkingcaret.com/2016/04/27/what-exactly-

is-a-unit-in-unit-testing/

Gordon, J. (2021, March 5). Intro to Deep Learning (ML Tech Talks). YouTube.

https://youtu.be/AhE8RhPGH1A

Grossfeld, B. (2020, January 23). Deep learning vs. machine learning: a simple
 83

way to learn the difference. Zendesk Blog.

https://www.zendesk.com/blog/machine-learning-and-deep-learning/

Guest view: A critical roadblock to spreading the virus. (2020, June 16). Denton

Record-Chronicle. https://dentonrc.com/opinion/guest-view-a-critical-
roadblock-to-spreading-the-virus/article_1c937c2e-ff37-5a7d-843c-
5a5ac4186510.html

Gupta, K. (2021, July 16). Display date and time in videos using OpenCV –

Python. GeeksforGeeks. Retrieved January 17, 2022, from

https://www.geeksforgeeks.org/display-date-and-time-in-videos-using-
python-opencv/

Gurucharan, M. K. (2020, May 30). COVID-19: Face Mask Detection using

TensorFlow and OpenCV. Towards Datascience.

https://towardsdatascience.com/covid-19-face-mask-detection-using-
tensorflow-and-opencv-702dd833515b

Haddad, J. (2020, May 22). How I built a Face Mask Detector for COVID-19
using

PyTorch Lightning (updated PL V.1.3.5) . TowardsDataScience.

https://towardsdatascience.com/how-i-built-a-face-mask-detector-for-
covid-19-using-pytorch-lightning-67eb3752fd61

Hamilton, T. (2021, August 27). What is Usability Testing? UX(User Experience)

Testing Example. Guru99. https://www.guru99.com/usability-testing-

tutorial.html

History of artificial intelligence. (2020, August 30). Wikipedia.

https://en.wikipedia.org/wiki/History_of_artificial_intelligence
 84

How single-shot detector (SSD) works? (n.d.). ArcGIS Developer. Retrieved

March
29, 2022, from https://developers.arcgis.com/python/guide/how-ssd-
works/#:~:text=SSD%20uses%20a%20matching%20phase,object’s
%20class%20and%20its%20location.

Importance Of Artificial Intelligence (AI) In Information Technology . (2020,

February 18). Soulpage. https://soulpageit.com/importance-of-artificial-

intelligence-ai-in-information-technology/

javascript month day year format. (n.d.). Grepper. Retrieved April 15, 2022, from
https://www.codegrepper.com/code-examples/javascript/
javascript+month+day+year+format

Jayadas, A. (2021, July 14). Face Mask Detection. GitHub. Retrieved March 20,

2022, from https://github.com/AISWARYAJAYADAS/facemask

Joshi, N. (2020, January 14). How to fine-tune your artificial intelligence

algorithms. Allerin. https://www.allerin.com/blog/how-to-fine-tune-your-

artificial-intelligence-algorithms

Joshi, S. (2020, August 21). CovidFaceMaskDetection. GitHub. Retrieved March

15, 2022, from https://github.com/sauravjoshi23/CovidFaceMaskDetection

Mittal, L. (2021, September 15). Mask Detector. GitHub. Retrieved January 17,

2022,from
https://github.com/laksh-1/mask-detector/tree/837f0a5b58e8e653068157
2260a25aa7cd46a563

Moroney, L. (2019, August 31). Intro to Machine Learning (ML Zero to Hero -

Part 1). YouTube. https://youtu.be/KNAWp2S3w94

 85

Mujtaba, H. (2021, January 17). Real-time Face detection | Face Mask Detection

using OpenCV. Great Learning.

https://www.mygreatlearning.com/blog/real-time-face-detection/

OpenCV putText() – Writing Text on Images. (n.d.). AskPython. Retrieved

January 17, 2022, from

https://www.askpython.com/python-modules/opencv-puttext

Palmer, J. (2020, August 31). Face Mask Detection at the Edge. IBM Cloud.

https://www.ibm.com/cloud/blog/face-mask-detection-at-the-edge

Pan, C., et al. (2020, November 18). Face Mask Detection Using Machine

Learning Models. GitHub.

https://github.com/dastratakos/Face-Mask-Detection

Press, G. (2013, December 30). A Very Short History Of Artificial Intelligence

(AI). Forbes. https://www.forbes.com/sites/gilpress/2016/12/30/a-very-

short-history-of-artificial-intelligence-ai/?sh=5ce8690c6fba

Preventing the spread of the coronavirus. (2022, January 11). Harvard Health

Publishing. Retrieved January 17, 2022, from

https://www.health.harvard.edu/diseases-and-conditions/preventing-the-
spread-of-the-coronavirus

Promoting mask-wearing during the COVID-19 pandemic. (2020, August).

Prevent Epidemics.
https://preventepidemics.org/wp-content/uploads/2020/08/Annex-
I_Sample-questionnaire-on-mask-usage.pdf
 86

Purohit, A. (2020, July). Face Mask Detector. GitHub.

https://github.com/adityap27/face-mask-detector

Python Get Today’s Current Date and Time. (n.d.). NixCraft. Retrieved January

17, 2022, from https://www.cyberciti.biz/faq/howto-get-current-date-

time-in-python/

Ramachandran, K. (2020, May 28). Face Mask Detector using Deep Learning

(PyTorch) and Computer Vision (OpenCV). Blog Home Categories.

https://www.ideas2it.com/blogs/face-mask-detector-using-deep-learning-
pytorch-and-computer-vision-opencv/

Rosebrock, A. (2020, May 4). COVID-19: Face Mask Detector with OpenCV,

Keras/TensorFlow, and Deep Learning. Pyimagesearch.

https://www.pyimagesearch.com/2020/05/04/covid-19-face-mask-
detector-with-opencv-keras-tensorflow-and-deep-learning/#download-the-
code

Saha, S. (2018, December 16). A Comprehensive Guide to Convolutional Neural

Networks — the ELI5 way. TowardsDataScience.

https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-
neural-networks-the-eli5-way-3bd2b1164a53

Sandler, M., & Howard, A. (2018, April 3). MobileNetV2: The Next Generation of
On-Device Computer Vision Networks. Google AI Blog. Retrieved March
29, 2022, from https://ai.googleblog.com/2018/04/mobilenetv2-next-
generation-of-on.html

Sawardekar, S. (2020, May 3). Django_VideoStream. GitHub. Retrieved March

15,

2022, from https://github.com/sawardekar/Django_VideoStream

S.C. (2020, May 17). Student Management System in Django. GitHub. Retrieved
 87

March 20, 2022, from

https://github.com/hackstarsj/student_management_system_in_django

Shetty, A., Lara,et. al. (n.d.). Face Mask Detection. Yashlara. Retrieved

September 18, 2021, from https://yashlara.github.io/Face-Mask-

Detection/

Singh, J. (2015, February 6). Dependency, Generalization, Association,

Aggregation, Composition in Object Oriented Programming. C-

Sharpcorner.
https://www.c-sharpcorner.com/UploadFile/b1df45/dependency-
generalization-association-aggregation-compos/

Sriram, G. (2020). Fine Tune MobileNet V2. Kaggle.

https://www.kaggle.com/sgcuber24/fine-tune-mobilenet-v2

Srivastava, B. (2017, January 16). What is agile methodology? Disadvantage of

waterfall model in software development. Bikeshsrivastava Blogspot.

https://bikeshsrivastava.blogspot.com/2017/01/part-43what-is-agile-
methodology.html

Stokdyk, D. (2019, October 29). What is Information Technology (IT)? Southern

New Hampshire University.

https://www.snhu.edu/about-us/newsroom/stem/what-is-information-
technology

Support-vector machine. (2021, September 6). Wikipedia.

https://en.wikipedia.org/wiki/Support-vector_machine
 88

The CCTV transmission networks explained. (2010, November 10). IFSEC Global.

https://www.ifsecglobal.com/cctv-2/the-cctv-transmission-network-the-
glue-that-ties-it-all-together/#:~:text=The%20topologies%20usually
%20employed%20are,see%20Figure%201%20at%20right.

The Future of IT and Artificial Intelligence . (n.d.). MyComputer Career. Retrieved

September 18, 2021, from https://www.mycomputercareer.edu/the-

future-of-i-t-and-artificial-intelligence/#:~:text=AI%20for%20IT
%20operations%20refers,both%20historical%20and%20online%20data.

TheTraveLad. (2016, July 24). Travel Guide | Argao, Cebu. TheTraveLad.

https://thetravelad.wordpress.com/2016/07/24/travel-guide-argao-cebu/

Transfer learning and fine-tuning. (n.d.). TensorFlow. Retrieved September 18,

2021, from https://www.tensorflow.org/tutorials/images/transfer_learning

Uri, E. (2020, June 8). Real time face mask detection in Android with TensorFlow

Lite. Medium. https://medium.com/@estebanuri/real-time-face-mask-

recognition-in-android-with-tensorflow-lite-215df6327265

Urooj, W. (2019, November 26). How to fetch and modify Date and Time in

Python? Edureka. Retrieved January 17, 2022, from

https://www.edureka.co/blog/date-time-in-python/

Vapnik–Chervonenkis theory. (n.d.). Wikipedia. Retrieved August 5, 2021, from

https://en.wikipedia.org/wiki/Vapnik
%E2%80%93Chervonenkis_theory#:~:text=Vapnik
%E2%80%93Chervonenkis%20theory%20(also%20known,a
%20statistical%20point%20of%20view.
 89

Watson, R. (2018, October 21). How to enable face detection in images using

the Python data science platform Anaconda . Maker Pro.

https://maker.pro/custom/projects/enable-face-detection-using-python-
and-the-anaconda-platform

What is computer vision? (n.d.). IBM. Retrieved March 29, 2022, from
https://www.ibm.com/ph-en/topics/computer-vision#:~:text=Computer%
20vision%20is%20a%20field,recommendations%20based%20on%20that
%20information.

E-Books

Albiso, A., Malenab, M., & Saldo, J. (2019). Municipal Public Vehicle Information

System Cebu Technology University Argao Campus (Revised ed.). Cebu

Technological University.

Anderson, R. (2007). Thematic Content Analysis (TCA) 1 Descriptive

Presentation

of Qualitative Data.
https://rosemarieanderson.com/wp-content/uploads/2014/08/ThematicCo
ntentAnalysis.pdf

Erhardt-Ferron, A. (2000). Theory and Applications of Digital Image Processing.

University of Applied Sciences Offenburg.
https://doc.lagout.org/science/0_Computer%20Science/9_Others/1_Digita
l%20Signal%20Processing/Theory%20and%20Applications%20of
%20Digital%20Image%20Processing.pdf

Easwaramoorthy, M., & Zarinpoush, F. (2006). INTERVIEWING FOR RESEARCH.

Canada Volunteerism Initiative.
http://sectorsource.ca/sites/default/files/resources/files/tipsheet6_intervie
wing_for_research_en_0.pdf

Hasan, H. (2019). A Comparison Between Support Vector Machine (SVM) and

 90

Convolutional Neural Network (CNN) Models For Hyperspectral Image

Classification (357th ed., Vol. 012035). IOP Publishing.
https://iopscience.iop.org/article/10.1088/1755-1315/357/1/012035/pdf

APPENDICES
 91

Appendix A Transmittal Letter

April 28, 2021

DR. EINGILBERT C. BENOLIRAO

Campus Director
Cebu Technological University – Argao Campus
Ed Kintanar Street, Lamacan Argao Cebu, Philippines

Dear Dr. Benolirao:

In partial fulfillment of the requirements on the subject, Capstone Project

and Research I and II, the undersigned are conducting a study entitled FACE
MASK CHECKER. The main objective of this study is to design and develop a
system that will check and detect a person who has a face mask, improper face
mask, and without a required face mask. The alert feature aims to remind the
security personnel and let people know to wear their face masks preventing the
spread of the virus and ensuring safety. 
 92

In this connection, we would like to humbly request your permission to allow

us to conduct this study inside the school. We assure you that the information
obtained will be treated with the utmost confidentiality. We are hoping for your
positive response on this matter.

Truly yours,
MARY ANGEL R. FUENTES
DARREN D. TEOGANCO
MARIBEL P. CABUNGCAG
APRIL KIM S. CALLEDO
         Researchers

Noted by:
DR. CHRISTINE JOY A. TOMOL, Ph.D.
Adviser, Capstone Project, and Research I & II

Approved by:
DR. EINGILBERT C. BENOLIRAO, DEV. Ed. D.
              Campus Director
Appendix B Questionnaire

Interview Questions

Security Personnel/Guard

1. In this time of pandemic, do you agree on wearing a face mask and face
shield?
2. the students and staff of the school always wear face masks when they
enter the gate?
3. To those who enter the school, do they wear face masks even though
they are no longer at the gate and are already inside the school
premises?
4. Would it be nice to have a system or application that you can use to
monitor the wearing and face mask? Yes, or No?
5. What would you like this system to do to help you watch or monitor
those who are not wearing face masks or wearing a face mask?
6. What do you think if there’s a system like this, would it be useful to all
Security Personnel? Yes, or no?

Students
 93

1. When going and entering the school, do you wear a face mask?
1.1. When you're inside of the school?
1.2. Are there times that you remove it while inside the school premises?
1.3. What type of face mask do you wear?
• surgical mask
• respirator/N95/N99
• Dust mask
• Washable cloth masks
• others

2. Do you agree about the mandatory memorandum of DILG (Executive

Order No. 168, s. 2014) in the coordination of IATF and DOH's 'BIDA
Solusyon' sa Covid-19 enforcing to wear face masks in public areas and
other protective equipment? Yes or No. Why yes? Why not?
2.1. Do you agree that wearing face masks in public areas should be
mandatory, specifically at school? (Yes or No)

3. If there is a technological system that enables to detect and check people

wearing face masks at your school monitored by the Security personnel,
will you agree to have this kind of system or not?
 94

Appendix C Project Working Title Form

 95

Appendix D Project Adviser Appointment Form

96
 97

Appendix E Grammarian’s Certification

 98

Appendix F Curriculum Vitae

DARREN D. TEOGANCO
Lamacan, Sibonga, Cebu
09919622150
[email protected]
March 11, 2000

Education
● Lamacan Elementary School, 2006-2012, Elementary Graduate
● Don Julian Enad Memorial National High School, 2012-2018,
General Academic Strand (GAS), Senior High School Graduate
● Cebu Technological University – Argao Campus, 2018-Present,
Bachelor of Science in Information Technology

Relevant Skills
● Self-Learning Skills
● Knowledge of Microsoft Office Suites
● Interpersonal communication
● Adaptable and eager to try new things, as well as a desire to improve
efficiency on assigned jobs of the company
● Ability to understand and carry out oral and written instructions and
request clarification when needed
● Familiar with computer languages such as C, Python, PHP, MySQL, HTML,
CSS, JavaScript, C#, Visual Basic, Java
● Basic knowledge on Git version control

Seminar, Webinar, and Training Experiences

● 1st PSITE-7 Regional ICT Student Congress 2021, March 6, 2021 via
Zoom at Cebu City, Cebu
● 2021 Startup Thursdays – Part 3, “Fruits of Stubbornness”, June 17, 2021
via Google Meet conducted by Department of Information and
Communications Technology (DICT) - Visayas Cluster 2
 99

APRIL KIM S. CALLEDO

Guiwanon, Argao, Cebu
09463612188
[email protected]
April 02, 2000

Education
● Guiwanonon Elementary School, 2006-2012, Elementary Graduate
● Simala National High School, 2012-2018, Accounting Businesses and
Management (ABM), Senior High School Graduate
● Cebu Technological University – Argao Campus, 2018-Present,
Bachelor of Science in Information Technology

Relevant Skills
● Knowledge of Microsoft Office Suites
● Computer Literate
● Good Interpersonal
● Analytical
● Willingness to learn and be trained
● Adaptable and eager to try new things, as well as a desire to improve
efficiency on assigned jobs of the company
● Able to set up a computer system according to instructions
● Setting up hardware and software
● Familiar with computer languages such as C, Python, MySQL, HTML, CSS,
Javascript, C#, Java

Seminar, Webinar, and Training Experiences

● Barkada Kontra Droga (BKD) Seminar Workshop,
at DepEd Ecotech Center
● 1st PSITE-7 Regional ICT Student Congress 2021, March 6, 2021 via
Zoom at Cebu City, Cebu
● 2021 Startup Thursdays – Part 3, “Fruits of Stubbornness”, June 17, 2021
via Google Meet conducted by Department of Information and
Communications Technology (DICT) - Visayas Cluster 2

MARIBEL P. CABUNGCAG
 100

Tabon, Dalaguete, Cebu

09754190993
[email protected]
October 9, 1999

Education
● Tabon Elementary School, 2006-2012, Elementary Graduate
● Mantalongon National High School, 2012-2018,
Technical Vocational (ICT strand), Senior High School Graduate
● Cebu Technological University – Argao Campus, 2018-Present,
Bachelor of Science in Information Technology

Relevant Skills
● Knowledge of Microsoft Office Suites
● Analytical
● Able to set up a computer system according to instructions
● Setting up hardware and software
● Interpersonal communication
● Adaptable and eager to try new things, as well as a desire to improve
efficiency on assigned jobs of the company

Seminar, Webinar, and Training Experiences

● 2021 Startup Thursdays – Part 3, “Fruits of Stubbornness”, June 17, 2021
via Google Meet conducted by Department of Information and
Communications Technology (DICT) - Visayas Cluster 2
● 1st PSITE-7 Regional ICT Student Congress 2021, March 6, 2021 via
Zoom at Cebu City, Cebu
● Barkada Kontra Droga (BKD) Seminar Workshop,
at DepEd Ecotech Center

MARY ANGEL R. FUENTES

Talo-ot, Argao, Cebu
 101

09151738232
[email protected]
April 5, 1999

Education
● Talo-ot Central Elementary School, 2006-2012, Elementary Graduate
● Simala National High School, 2012-2018,
General Academic Strand (GAS), Senior High School Graduate
● Cebu Technological University – Argao Campus, 2018-Present,
Bachelor of Science in Information Technology

Relevant Skills
● Arts and Creativity
● Knowledge of Microsoft Office Suites
● Analytical
● Interpersonal communication
● Adaptability
● Computer hardware and software systems installation according to
instructions
● Familiar with computer languages such as C, Python, MySQL, HTML, CSS,
JavaScript, Visual Basic, C#, Java
● Basic knowledge on Git version control

Seminar, Webinar, and Training Experiences

● SSG Division Federation Election, Orientation, and Org. Workshop focused
on Barkada Kontra Droga, July 20-21,2015 at DepEd Ecotech Center
● 6th ASLC Adelante Student Leaders’ Congress 2016, February 13-14, 2016
at University of San Jose-Recoletos
● Reserve Officers’ Training Corps (ROTC), August 2018-May 2019
at Cebu Technological University-Argao Campus
● 1st PSITE-7 Regional ICT Student Congress 2021, March 6, 2021 via
Zoom at Cebu City, Cebu
● 2021 Startup Thursdays – Part 3, “Fruits of Stubbornness”, June 17, 2021
via Google Meet conducted by Department of Information and
Communications Technology (DICT) - Visayas Cluster 2