Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 33

CHAPTER 4

PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS

PROPOSED SYSTEM

During the creation of the proposed system, the components required included

both hardware and software to build the Face Recognition Attendance System, along

with an explanation of how the system functions. These sections cover the features and

actual design of the prototype for the Face Recognition Attendance System.

Figure 5. Front View of the Actual Design of the System

Shown in Figure 5 is the front view of the system, featuring the display monitor

with a camera mounted on top. The camera, positioned at the top center, is set up for

video input, complementing the overall functionality of the system.

42
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 43

Figure 6. Back View of the Actual Design of the System

Shown in Figure 6 is the back view of the system, highlighting the setup and cable

management. The Raspberry Pi 4B is centrally mounted on the back of the monitor, with

various cables connected for power, network, and display. A camera is mounted at the

top, indicating its use for video input.

HARDWARE COMPONENTS

These are the hardware components used by the researchers for this proposed

Face Recognition Attendance System:

Raspberry Pi

Figure 7. Raspberry Pi 4 Model B


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 44

Powered by a 64-bit quad-core processor, Figure 7 supports dual 4K monitor

output via micro-HDMI; it comes with up to 8GB RAM and 4Kp60 hardware decoding.

It also includes a dual-band wireless, Bluetooth 5.0, and a Gigabit Ethernet connection,

as well as USB 3.0 ports. It also supports PoE on an add-on HAT, delivering performance

like entry-level desktop PCs.1In a face recognition attendance system, Figure 7 functions

as the primary component, responsible for managing the face detection and recognition

processes, maintaining the system's database, hosting the website server, and handling

backup functions of the data.2

Figure 8. SanDisk 32gb Micro SD Card

Figure 8 is used in the Raspberry Pi to provide essential storage for the operating

system, face recognition system, and data associated with the face recognition

attendance system. This includes storing images, attendance logs, database, enabling the

system to operate efficiently and maintain a record of attendance activities. 3

1See Appendix A for the full Raspberry Pi Specification


2See Appendix B for the bullseye operating system installation
3See Appendix C for the full SanDisk 32GB Micro SD Card Specification
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 45

UTP Cable

Figure 9. UTP Cat6 Cable


Figure 9 was used by the researchers to demonstrate why they used UTP Cat6

cable between Figure 7 and the user's device. The cable ensured stable connectivity by

providing reliable data transfer and minimizing interference for uninterrupted

communication.4

Desktop Monitor

Figure 10. NEC Desktop Monitor

4See Appendix D for the full UTP Cat6 Cable Specification


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 46

Figure 10 is used in a face recognition attendance system to display the face

recognition and detection interface.5

Power Adaptor

Figure 11. Okdo Power Adaptor

Figure 11 supplies stable and reliable power to the Raspberry Pi, ensuring it gets

the correct voltage and current for smooth operation and to avoid any power issues that

might damage the board.6

USB Webcam

Figure 12. Rapoo USB Webcam

5See Appendix E for the full NEC Desktop Monitor Specification


6See Appendix F for the full Okdo Power Adaptor Specification
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 47

Figure 12 is used in the face recognition attendance system to capture live images

of individuals. This enables the system to perform real-time face detection and

recognition, essential for verifying identities and logging attendance accurately. 7

Flash Drive

Figure 13. SanDisk 16GB Flash Drive


Figure 13 was used as the hardware backup system used for storing the dataset,

database, and student attendance records for the face recognition attendance system. 8

HDMI to VGA Cable

Figure 14. HDMI to VGA Cable

7See Appendix G for the full Rapoo USB Webcam Specification


8See appendix H SanDisk 16GB Flash Drive
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 48

To connect the display of the Raspberry Pi 4B to the desktop monitor shown in

Figure 10, Figure 14 was used. This setup allowed the monitor to effectively display the

output from the Raspberry Pi 4B.9

Micro HDMI Adapter

Figure 15. Micro HDMI Adapter

To connect the Raspberry Pi 4B to a monitor, figure 14 and figure 15 are used in

converting the Micro HDMI output from the Raspberry Pi to a VGA input suitable for the

desktop monitor in figure 10.10

9See Appendix I for the full HDMI to VGA Cable Specification


10See appendix J Micro HDMI Adapter
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 49

Figure 16. Hardware Components Setup

Figure 16 illustrates the various hardware components connected to the

Raspberry Pi 4B. The Raspberry Pi 4B, located at the center, serves as the main hub to

which all peripherals are connected. A microSD card, inserted into the Raspberry Pi,

provides storage for the operating system and the Face Recognition Attendance System.

For network connectivity, an Ethernet cable is connected to the Ethernet port, which is

also linked to the admin device. The admin device is used to manage and operate the face

recognition attendance system. A webcam, connected to a USB port on the Raspberry Pi,

captures video input. Additionally, a USB flash drive is plugged into one of the USB ports

for hardware backup of the database, dataset, and attendance records. The monitor is

connected to the Raspberry Pi's HDMI port via a Micro HDMI adapter and a VGA to

HDMI cable, allowing the display of the Raspberry Pi’s output. The Micro HDMI adapter

links the Raspberry Pi's Micro HDMI port to the VGA to HDMI cable, which then

connects to the monitor. Finally, a power adapter provides power to the Raspberry Pi

through its power input port, completing the setup.11

11See Appendix K for the total cost of expenditures of the entire system
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 50

SOFTWARE COMPONENTS

These are the software components used by the researchers for this proposed

Face Recognition Attendance System:

RealVNC Viewer

Figure 17. RealVNC Viewer

Figure 17 showed why the researchers chose RealVNC Viewer among all software

options: it allowed users to remotely access and control the graphical interface of a

Raspberry Pi from another device. This facilitated remote management and operation of

the Raspberry Pi, enabling tasks such as accessing files, running programs, and

troubleshooting without the need for physical access to the Raspberry Pi itself.12

Raspberry Pi Imager

Figure 18. Raspberry Pi Imager Client

12See Appendix L for the RealVNC Viewer installation


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 51

Figure 18 shows a tool created by the Raspberry Pi Foundation. The researchers

chose the Raspberry Pi Imager client because it simplifies the process of installing

operating systems onto SD cards for Raspberry Pi computers. This functionality was

essential for the face recognition attendance system, as it ensured the Raspberry Pi was

equipped with the necessary software for implementing facial recognition and managing

attendance records.13

PuTTY

Figure 19. PuTTY

Figure 19 shows the PuTTY application. The researchers chose PuTTY because it

allowed users to remotely access their Raspberry Pi from another computer using the

SSH (Secure Shell) protocol. With PuTTY, users could execute commands, configure

settings, and manage their Raspberry Pi without needing a dedicated monitor and

keyboard.14

13See Appendix M for the Raspberry Pi Imager Client installation


14See Appendix N for the PuTTy installation
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 52

SQLAlchemy

Figure 20. Application of SQLAlchemy to the Flask Framework

The researchers used SQLAlchemy as the database for the face recognition

attendance system because it effectively managed student information and their

corresponding attendance records. SQLAlchemy was chosen for its ability to facilitate

management and tracking of attendance, offering support for complex queries and

seamless integration with the system's architecture. Additionally, SQLAlchemy is paired

well with Flask, a lightweight web framework, because it simplifies database interactions.15

Face Detection and Recognition

Figure 21. Application of Face Detection and Recognition

15See Appendix O for the Full Code


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 53

Face recognition attendance system utilizes Histogram of Oriented Gradients

(HOG) for face detection, a technique that extracts features from facial images, while

employing dlib and face_recognition libraries for recognizing and identifying individuals

from the detected faces based on pre-trained models, enabling accurate attendance

tracking through facial recognition technology. Sample output is shown in figure 21.

Flask Framework

Figure 22. Face Recognition Attendance System Website Dashboard

Flask is a lightweight and flexible Python web framework that simplifies the

development of web applications by providing tools and libraries for building web

services and APIs, as shown in Figure 22 which is a website dashboard using flask

framework.
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 54

Figure 23. Sample output of the system recognizing an individual student.

As shown in the figure 23 the yellow box around the individual's face indicates that

the system has detected a face within the frame. This box is typically drawn in real-time

by the face recognition software to highlight the region of the image where a face has

been found. When a student is positioned within the camera's frame with facial features

clearly visible, as demonstrated by Jake M. Najera, the system will not only identify the

individual but also determine their grade and section, and then log their attendance.

Figure 24. Sample output of the system recognizing two students.


As shown in Figure 24, the system can detect and recognize two students simultaneously.
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 55

Figure 25. Sample output of the system recognizing three students.

As shown in Figure 25, the system is capable of detecting and recognizing three

students at the same time. Additionally, it correctly identified the student Lyzel Van G.

Manluctao, even though she is not directly facing the camera and is wearing glasses.

However, it is recommended that students avoid wearing facial coverings to ensure the

system's highest accuracy in recognition.

Figure 26. Sample output of the system recognizing four students.

As shown in Figure 26, the system can detect and recognize four students

simultaneously.
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 56

Figure 27. Sample result when the system recognizes multiple individuals.

Figure 27 shows the interface of the system when multiple individuals are present

in front of the camera. It demonstrates the system's capability to accurately recognize

and distinguish between individuals, even in scenarios with multiple people.

ACCURACY OF THE SYSTEM

The accuracy of the system is classified in terms of distance and accuracy metrics.

DISTANCE (m)
0.5 1 1.5 2 TOTAL
SUCCESSFUL
1 1 1 1 4
RECOGNITION
UNSUCCESSFUL
0 0 0 0 0
RECOGNITION

Total 1 1 1 1 4

Table 5. Accuracy in distance


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 57

Figure 28. Testing of accuracy in different distances

Figure 28 and Table 5 display the data on the system's accuracy at different

distances. The study examined four distances: 0.5 meters, 1 meter, 1.5 meters, and 2

meters. At each distance, the researchers tested four individuals, and the system

successfully recognized the students.

Data Gathered in Accuracy in Terms of Face Recognition

Table 6. St. Andrew Face Recognition Accuracy Result

Metric Count

True Positive 28

True Negative 1

False Positive 1

False Negative 0

Accuracy (%) 96.67%

Table 6 indicates an accuracy rate of 96.67% for the St. Andrew Face Recognition

system. With 28 true positives and only 2 errors (1 false positive and 0 false negatives),

the system demonstrates accurate performance in face recognition. This affirms the
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 58

findings of Shetty et al. (2021), which suggest that high true positive rates, coupled with

low error rates, indicate a system's high accuracy.

Table 7. St. Bernadette Face Recognition Accuracy Result

Metric Count

True Positive 33

True Negative 0

False Positive 3

False Negative 0

Accuracy (%) 91.67%

Table 7 illustrates the Face Recognition Accuracy Result, indicating a

commendable accuracy rate of 91.67%. With 33 true positives and only 3 false positives,

the system demonstrates strong performance in identifying correct matches, though

there were no true negatives or false negatives recorded. This supports the study by

Shetty et al. (2021), emphasizing that high rates of true positives, when combined with

minimal errors, strongly indicate a system's high accuracy.

Figure 29. Sample output of the system false positive


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 59

The system has mistakenly identified the individual in the image as James Lincoln

M. Flores, despite him not being registered on the database. This false identification

likely happened because the individual is not directly facing the camera and wearing

glasses. It's advised that for better accuracy, individuals should remove glasses or any

items covering their face when using the system. The sample output is shown in figure

29.

Figure 30. Sample result when a spoof detected.

When a student attempts to take attendance using a picture, the system first

identifies the individual in front of the camera. It then verifies whether the attempt is

genuine or a spoof. If a spoof is detected, the system promptly deletes the associated

attendance record and captures an image of the detected spoof, storing it in a designated

spoof folder for the teacher to review instances of attempted attendance fraud.

Conversely, if the individual is deemed authentic, the system proceeds to mark their

attendance, as shown in Figure 30.


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 60

Figure 31. Sample result when an unregistered person attempts to take attendance.

When an unregistered individual tries to take attendance within the system, it

successfully labels the person as "unknown" since they are not registered. The system

then captures a picture of the unidentified individual and stores it in the "unknown"

folder. This feature enables the teacher to review the image and identify the person who

attempted to take attendance without being registered, as shown in Figure 31.

Figure 32. Delay observed by the face recognition system on Raspberry Pi.
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 61

Figure 33. Delay observed in the face recognition system on a Desktop Computer.

Figure 34. Specification of the said Desktop Computer

Based on Figures 32 and 33, it has been noted that there is a delay of 3 to 5

seconds in the face detection and recognition processes. Specifically, a delay of 3 seconds

occurs when detecting motion, with an additional 2 seconds before the results are

displayed on the screen to allow the system time to identify whether the student is

attempting a spoofing attack. Additionally, the system is operated on a desktop computer

equipped with a Ryzen 5 5600G processor and 16GB of RAM, as shown in Figure

34. This setup shows no


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 62

significant difference in delay, suggesting that the hardware configuration does not

influence the delay of these processes.

IMPLEMENTED HISTORY TRACKING FEATURE

The system's history tracking feature provides teachers with comprehensive tools

to access, filter, search, update, delete, and download attendance records, ensuring

accurate record-keeping and communication.

Figure 35. History tracking feature.

Figure 36. History tracking feature individual searching.


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 63

Figures 30 and 31 showcase the system's history tracking feature, which enables

teachers to access past attendance records. By selecting the desired start and end dates

along with the relevant section, the system filters the data accordingly. This feature has a

search function that allows teachers to view attendance records for a specific student.

Additionally, the system offers update and delete functions, enabling teachers to modify

the date and time or attendance status of a student and to manually delete attendance

records. The system has also a feature to allow teachers to download the Master list in a

PDF file and attendance records in both PDF and CSV formats. This new functionality

provides a convenient way for teachers to maintain and share comprehensive attendance

documentation, facilitating better record-keeping and communication regarding student

attendance.

The data presented below was gathered using a detailed questionnaire, ensuring

accurate and relevant information for the study.16 17 The respondents include Sir Ric

Garcia the TLE teacher, and the Grade 10 students from St. Andrew and St. Bernadette.

The survey uses a 5-point Likert scale, with 5 being the highest and 1 being the lowest.

This chapter discusses, presents, analyzes, and interprets the results of the survey

questionnaire.

RELIABILITY OF THE SYSTEM

The reliability of the system refers to its consistency and dependability in

identifying and verifying individuals' faces and attendance records. This characteristic

ensures that the system maintains its performance, delivering consistent and dependable

recognition capabilities.

16See Appendix P for the Questionnaire Basis


17See Appendix Q for the Questionnaire
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 64

Presentation of Data Collected in Grade 10 - St. Andrew


Table 8. Reliability in Grade 10 – St. Andrew

Indicators Weighted Mean Descriptive Equivalent

1. The system performs its


function without failure 3.97
Agree
for a reasonable amount
of time.

2. The system can


effectively manage
various changes in
student positioning or
posture during 4.18
Agree
attendance capture for a
considerable duration.

3. Even in scenarios where


multiple individuals are
present in the frame,
the system reliably
maintains accurate
attendance records for a 4. 34
significant duration. Strongly Agree

4.17
Average Weighted Mean Agree
LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 8 shows that the system has an average weighted mean of 4.17. According

to the scale described in the study by Salac (2020), this mean falls within the range of

3.41 to 4.20, corresponding to the descriptive equivalent of 'Agree.' This indicates that

participants generally agree on the system’s consistent performance and accurate

capture of attendance, highlighting its strong reliability. As stated by Salac (2020),

reliability means the system is dependable for consistent use, delivers accurate and

reliable attendance data, and operates according to its stated specifications without

failure.
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 65

Presentation of Collected in Grade 10 - St. Bernadette


Table 9. Reliability in Grade 10 – St. Bernadette

Indicators Weighted Mean Descriptive Equivalent

1. The system performs its


function without failure
4.56 Strongly Agree
for a reasonable amount
of time.

2. The system can


effectively manage
various changes in
student positioning or
posture during
4.47 Strongly Agree
attendance capture for a
considerable duration.

3. Even in scenarios where


multiple individuals are
present in the frame, the
system reliably maintains
accurate attendance 4. 58 Strongly Agree
records for a significant
duration.

Average Weighted Mean 4.54 Strongly Agree


LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 9 underscores the system's reliability, with respondents strongly agreeing

that it operates consistently over time, effectively manages variations in student

positioning during attendance capture, and maintains accurate attendance records even

with multiple individuals in the frame. The average weighted mean of 4.54, falling under

"Strongly Agree.” According to Salac (2020), the scale described in the study by, this

mean falls within the range of 4.21 to 5.00, corresponding to the descriptive equivalent of

"Strongly Agree." This indicates that participants generally agree on the system’s

consistent performance and accurate capture of attendance, highlighting its strong

reliability. The study of Salac (2020) defines reliability as the system consistently

performing its functions without failure over a reasonable period, highlighting its
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 66

dependable operation. The system's reliability is well-supported by its consistent

performance and adaptability to varying conditions.

Presentation of Results Collected to the Grade 10 TLE -ICT Teacher

Table 10. Reliability in TLE – ICT Teacher


Indicators Weighted Mean Descriptive Equivalent

1. The system has a data


recovery plan in place to
5 Strongly Agree
address instances of data
loss.

2. The dashboard maintains


accurate information
regarding the time,
5 Strongly Agree
name, and status of each
student for a
considerable duration.

3. The history tracking


feature reliably presents
accurate information, 5 Strongly Agree
even when accessing
older data.

Average Weighted Mean 5 Strongly Agree


LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 10 focuses on the system's reliability from the perspective of TLE-ICT

teacher Sir Ric Garcia. He strongly agrees that the system has a data recovery plan to

address data loss, the dashboard maintains accurate student information over time, and

the history tracking feature reliably presents accurate older data. The average weighted

mean of 5, categorized as "Strongly Agree," confirms the scale in the study of Salac

(2020), this mean falls within the range of 4.21 to 5.00 system's exceptional reliability.

As stated by Rollon (2023), reliability shows preparedness for data loss scenarios

through a data
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 67

recovery plan, ensuring that data can be recovered and reliability is maintained even in

adverse conditions.

LEVEL OF ACCEPTABILITY OF THE SYSTEM

Usability and functionality of the system are combined result to determine the

level of acceptability of the system.

Level of Acceptability of the Implemented Face Recognition Attendance System in St.

Andrew:

Table 11. Usability in Grade 10 – St. Andrew


Indicators Weighted Mean Descriptive Equivalent
1. Registering my face
with the face recognition
attendance system was 4.74
Strongly Agree
straightforward and
relatively simple.

2. Taking attendance
with the face recognition
4.61
attendance system was Strongly Agree
straightforward and
relatively simple.

3. The live display is 4.47


visually appealing. Strongly Agree

Average Weighted Mean 4.61 Strongly Agree


LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 11 indicates that respondents found the face recognition attendance system

highly user-friendly and visually appealing. The average weighted mean of 4.61,

categorized as "Strongly Agree," affirms the scale in the study of Salac (2020), this mean

falls within the range of 4.21 to 5.00 system's exceptional confirms the system's ease of

use and aesthetic appeal. According to Salac (2020), usability involves processes that are

easy to learn and understand, easy to use and operate, and usable by every user.
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 68

Table 12. Functionality in Grade 10 – St. Andrew

Indicators Weighted Mean Descriptive Equivalent


1. The system successfully
performs face recognition 53
accurately. 4. Strongly Agree

2. The system recognizes


4.63
registered students. Strongly Agree

3. The system recognizes 3.39


not registered students. Agree
4.18
Average Weighted Mean Agree

LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 12 illustrates the system's functionality, with respondents' strong agreement on its

accurate face recognition and recognition of registered students, scoring 4.53 and 4.63

respectively. While the ability to identify non-registered students scored lower at 3.39

categorized as 'Agree', the average weighted mean of 4.18 still indicates that the system

operates generally as intended, confirming the scale in the study of Salac (2020), this

mean falls within the range of 4.21 to 5.00 system's usability. According to Rollon

(2023), functionality entails performing its intended specifications, carrying out its

functions with less time and effort, and supporting the needs of every user.

Table 13. Systems Acceptability in Grade 10 – St. Andrew

System Quality Weighted Mean Descriptive Equivalent


Functionality 4.18 Agree
Usability 4.61 Strongly Agree

Average Weighted Mean 4.40 Strongly Agree


LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

The combined results of system functionality and usability determined the

system's acceptability. Table 13 shows a weighted mean of 4.18 for functionality, falling

under "Agree," and a weighted mean of 4.61 for usability, falling under "Strongly Agree."
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 69

The average weighted mean of 4.40, also falling under "Strongly Agree," confirms the

system's high overall acceptability. The scale presented in Salac (2020) study, this

average falls within the 4.21 to 5.00 range, indicating the system's high acceptability.

The findings suggest that while there may be room for slight improvements in

functionality, the excellent usability of the system significantly enhances its overall

acceptability. According to the principles outlined in Rollon (2023), the system

effectively supports the needs of its users, operates efficiently, and is straightforward to

use, all of which are crucial for a positive user experience and the success of a system.

Level of Acceptability of the Implemented Face Recognition Attendance System in St.


Bernadette:

Table 14. Usability in Grade 10 – St. Bernadette

Indicators Weighted Mean Descriptive Equivalent


1. Registering my face with
the face recognition
attendance
system was straightforward 4. 69 Strongly Agree
and relatively simple.

2. Taking attendance with the


face recognition attendance
4.67
system was straightforward Strongly Agree
and relatively simple.

3. The live display is visually 4.75


Strongly Agree
appealing.
4.70
Average Weighted Mean Strongly Agree
LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 14 highlights the system's usability, with respondents strongly agreeing that

registering their face and taking attendance with the face recognition system are

straightforward and relatively simple tasks. Additionally, they found the live display to

be visually appealing. The average weighted mean of 4.70, falling under "Strongly

Agree," confirms the system's high usability. The scale presented of Salac (2020), this

average falls
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 70

within the 4.21 to 5.00 range, indicating the system's high usability. According to Rollon

(2023), usability encompasses processes that are straightforward to learn and

comprehend, simple to use and operate, and accessible to all users.

Table 15. Functionality in Grade 10 - St. Bernadette


Indicators Weighted Mean Descriptive Equivalent
1. The system successfully
performs face recognition 53
4. Strongly Agree
accurately.

2. The system recognizes


4.67
registered students. Strongly Agree

3. The system recognizes 4.31


not registered students. Strongly Agree
4.50
Average Weighted Mean Strongly Agree
LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 15 focuses on the system's functionality, with respondents strongly agreeing

on its accurate face recognition for both registered and non-registered students. The

average weighted mean of 4.49, falling under "Strongly Agree," affirms the system's

overall effectiveness and reliability in performing its intended functions. The scale

presented of Salac (2020), this average falls within the 4.21 to 5.00 range, indicating the

system's high functionality. As stated by Salac (2020), functionality is evaluated based

on the system's ability to operate according to its designed specifications, and meet the

requirements of all users.

Table 16. System Acceptability in Grade 10 – St. Bernadette


System Quality Weighted Mean Descriptive Equivalent
Functionality 4.49 Strongly Agree

Usability 4.70 Strongly Agree

Average Weighted Mean 4.60 Strongly Agree


LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 71

Based on the combined results of system functionality and usability as shown in

Table 16, the system's acceptability is high. Both functionality and usability received strong

agreement, resulting in an average weighted mean of 4.60, indicating strong overall

acceptability. The scale presented in Salac (2020), study, this average falls within the

4.21 to 5.00 range, indicating the system's high acceptability. In Salac's (2020) study,that

the combination of functionality and usability determines the level of acceptability

toward users. The system's ability to operate according to its designed specifications and

complete tasks efficiently, along with its ease of use and user-friendly interface, are

critical factors in achieving high user acceptance.

Level of Acceptability of the Implemented Face Recognition Attendance System in TLE-


ICT Teacher:
Table 17. Usability in TLE – ICT Teacher
Indicators Weighted Mean Descriptive Equivalent
1. The systems navigation can 5 Strongly Agree
be used easily.

2. The system is user-friendly. 5 Strongly Agree

3. The website usage is easy to


understand.
5 Strongly Agree
Average Weighted Mean 5 Strongly Agree
LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 17 shows that TLE-ICT Instructor Sir Ric Garcia finds the system highly

usable. He strongly agrees that the system's navigation, user-friendliness, and website

usage are all excellent. The average weighted mean of 5 confirms this high usability

rating based on the scale of Salac (2020) study. Based on Rollon (2023), usability is

characterized by processes that are intuitively easy to learn and understand. The systems

are designed to be user-friendly and efficient in operation, ensuring that they can be

effectively used by everyone, regardless of their skill level. This approach to usability

ensures that the


CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 72

technology is inclusive, accommodating a wide range of users with varying degrees of

technological proficiency.

Table 18. Functionality in TLE – ICT Teacher


Indicators Weighted Mean Descriptive Equivalent
1. The system menu items are
working. 5 Strongly Agree

2. The system is successfully


performing the required
tasks, including dashboard and
export functionalities, student 5 Strongly Agree
management, and attendance
history tracking.

3. The system ensures accuracy


across its content, including
student attendance records,
5 Strongly Agree
cloud-based backup, and
external storage backup.

Average Weighted Mean 5 Strongly Agree


LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 18 highlights the system's functionality, with Sir Ric Garcia strongly

agreeing that the system's menu items work, it successfully performs required tasks like

dashboard functions and attendance tracking and ensures accuracy across content and

backups. The average weighted mean of 5 confirms the system's exceptional

functionality. The scale presented in Salac (2023) study, this average falls within the 4.21

to 5.00 range, indicating the system's high functionality. According to Uy et al. (2022),

The functionality of a website pertains to its capacity to meet user needs effectively. It

should execute necessary tasks by offering precise and current information. Additionally,

the website should be quick to load, simple to navigate, and compatible across various

browsers. High functionality significantly enhances the overall usability of the website.
CHAPTER 4: PRESENTATION, ANALYSIS, AND INTERPRETATION OF FINDINGS 73

Table 19. System Acceptability


System Quality Weighted Mean Descriptive Equivalent
Functionality 5 Strongly Agree
Usability 5 Strongly Agree

Average Weighted Mean 5 Strongly Agree


LEGEND: Strongly Agree - 4.21 – 5.00 Agree - 3.41 – 4.20 Neutral - 2.61 – 3.40 Disagree - 1.80 – 2.60 Strongly Disagree - 1.00 – 1.79

Table 19 combines the results of functionality and usability to determine the

system's acceptability. With a weighted mean of 5 for both functionality and usability,

falling under "Strongly Agree," the average weighted mean of 5 confirms the system's

outstanding acceptability. The scale presented in Salac (2023), this average falls within

the 4.21 to 5.00 range, indicating the system's high acceptability. In Salac's (2020) study,

the combination of functionality and usability determines the level of acceptability

toward users. The system's ability to operate according to its designed specifications and

complete tasks efficiently, along with its ease of use and user-friendly interface, are

critical factors in achieving high user acceptance. The systems are designed to be user-

friendly and efficient in operation, ensuring that they can be effectively used by everyone,

regardless of their skill level.

You might also like