2021 2nd Asia Conference on Computers and Communications (ACCC)

QR Code-Based Student Attendance System

Khang Jie Liew Tee Hean Tan
Centre for American Education Centre for American Education
Sunway University Sunway University
5, Jalan Universiti, Bandar Sunway, 47500 Petaling Jaya, 5, Jalan Universiti, Bandar Sunway, 47500 Petaling Jaya,
Selangor, Malaysia. Selangor, Malaysia.
khangjiel@sunway.edu.my teeheant@sunway.edu.my

Abstract—Student attendance in higher education institutions Malaysian higher education institutions as most institutions
is a factor that impacts academic performance. Some higher utilise the conventional method of tracking attendance.
education institutions have penalised students for poor attendance
2021 2nd Asia Conference on Computers and Communications (ACCC) | 978-1-6654-0743-4/21/$31.00 ©2021 IEEE | DOI: 10.1109/ACCC54619.2021.00009

records. Most universities, however, have not implemented an The barcode-based attendance system developed by [3], is
automated attendance-taking system to deter poor attendance, delivered through an Android mobile application, employing the
leaving the task instead to instructors who manually record use of a barcode from the student’s identity card. Nonetheless,
students’ attendances into the system; a time-consuming and the system exhibits limitations; it is slow in capturing an image,
tedious undertaking, particularly, with a large number of students. image processing, and analysing the captured image. [4]
This study primarily aimed to propose a Quick Response (QR) recommend a QR code-based system, whereby the QR code
code-based attendance system, complete with several features to is displayed for students at the beginning or during each class.
prevent attendance cheating. Students at the Centre for American The system tracks students through facial verification before
Education, Sunway University tested a mobile application confirming their attendance. The QR code contains such
designed to scan QR codes, specifically generated for each information as the subject and class times, and smartphone
classroom. The system checked three categories of data: subject location data. A weakness identified in this system is that the
class hour, registered mobile device, and geolocation. Once the instructor is only able to detect attendance fraud in the updated
information was verified, the student’s attendance was recorded attendance sheet by checking both facial image matching and
into the system. The proposed system succeeded in overcoming
distance calculations. A study from [5], highlights that radio-
limitations encountered in the university’s existing student
frequency identification (RFID) and biometric-based
attendance-taking system.
attendance systems require additional costs for hardware and
Keywords—QR code, student attendance, system, mobile maintenance. Contrarily, an Android-based system that
application operates with a barcode is cost-efficient. A comparison study
conducted on various attendance systems supports this.
I. INTRODUCTION Based on the barcode, RFID, QR code, biometrics, and
The possible correlation between student attendance and its Bluetooth technology, [6] conclude that the biometric-based
impact on academic performance has concerned researchers. system has better accuracy than others; with relatively high
According to [1], studies show that attendance and academic authentication but is vulnerable to security and privacy risks.
performance positively correlate but the relationship is generally The QR code-based system, once again, stands out for its
weak; further supporting attendance as a factor able to influence easy implementation, especially when mobile phones are
student performance. The findings from [2], reveal that equipped with the ability to read QR codes. Meanwhile, a
absenteeism from class had a profound impact on students’ fingerprint biometric attendance system recently implemented
Calculus final exam scores. The absenteeism percentage in a Malaysian college university to monitor student attendance
generally reduces when the mobile attendance application encounters a glitch when the system was unavailable as a
replaces the manual attendance system. Furthermore, the mobile application [7].
traditional attendance-taking procedure is time-consuming and An automated multi-factor analytic is suggested by [8],
may be recorded inaccurately by instructors. There is a to prevent attendance cheating, applying the function of four
possibility for attendance cheating when instructors circulate the components: the QR code, international mobile equipment
attendance sign-in sheet around the lecture room. In Malaysian identity (IMEI), attendance time, and students’ geolocations. A
higher education institutions, student attendance in lectures and need to incorporate biometrics into the system is, nonetheless,
tutorials is mandatory. Penalties are imposed when students’ established to prevent identity forging. An iBeacon, which
attendances are poor; students are barred from sitting an exam centers on Bluetooth Low Energy technology, is proposed by
or are required to withdraw a subject by the higher education [9], to monitor and manage student attendance. The iBeacon
institution authority. International students in Malaysia must operates as an indoor positioning system and it is incorporated
achieve at least 80% and above for their overall attendance into the student card. Attendance is automatically marked in
record, in order to be able to renew their student passes, besides the classroom when the iBeacon is connected to the server,
having their academic performance monitored. Therefore, an any attendance forgery is identified and an alert is sent to the
efficient automated attendance system is urgently needed in staff but the system does not discard the forgery notification
automatically.

978-1-6654-0743-4/21/$31.00 ©2021 IEEE 10

DOI 10.1109/ACCC54619.2021.00009
Authorized licensed use limited to: Ayushy Limbani. Downloaded on October 16,2024 at 08:18:05 UTC from IEEE Xplore. Restrictions apply.
 [10] proposes another QR code-based system that Fidelity (Wi-Fi) network, for online payment, and contact
automatically refreshes the QR code every 10 seconds to prevent tracing during the current Covid-19 pandemic.
students who are not in the classroom from scanning the code.
The study from [11], affirms that the QR-based attendance B. The Proposed System
system works to track and record student attendance but is not A prototype of the proposed system was implemented as a
designed to identify attendance forgery. customised 2-tier application comprising of server and mobile
modules. The server module, a web-based module using the
For this study, a QR code-based attendance system is Hypertext Preprocessor (PHP) programming language, was
proposed. The system is developed as an Android mobile hosted on the Tomcat web server and accessed data from the
application with a QR code function containing information on back-end MySQL database. It processed students’ attendances,
classroom venue, and is selected for its ease of use as part of a complete with student details, subject, timetable, and interacted
mobile application or mobile phone camera. The proposed with the mobile module on students’ smartphones. Whereas the
system addresses attendance cheating by integrating several mobile module was implemented on Android platform using
verification factors such as subject class hours, registered mobile Java programming language. It consists of two main modules,
devices, and students’ geolocations through the Global which are student registration module and attendance-taking
Positioning System (GPS). In addition, each student can only module. Upon installation, the mobile module will request the
register one mobile device for attendance purposes. user permissions on using the GPS tracking as well as camera to
The brief introduction to QR code and the system design will scan the QR code, and these two permissions will automatically
be described in Section 2, along with the implementation be enabled every time the student uses the mobile module.
algorithms. Section 3 presents the interface of the mobile Fig. 1 shows the overall architecture of the proposed system,
application and description on the implementation of the which involves a server module that connects to the database;
proposed attendance system in order to illustrate the algorithms. and a mobile module that can track GPS locations, scan the
Lastly, the study is concluded in Section 4. classroom QR code, and interact with the server module.
II. MATERIAL AND METHODS
A. QR Code
A QR code is a two-dimensional barcode developed in 1994
for use in the automobile industry by Japanese company Denso
Wave. The QR code is displayed in a distinctive design,
consisting of black-coloured squares arranged in a square grid
on a white background; and information is encoded through four
standardised modes, numeric, alphanumeric, 8-bit bytes (binary),
and Kanji. Nowadays, the QR code is readable at high speed via
mobile applications and mobile cameras from any direction at Fig. 1. System architecture of the proposed system.
360◦, achieved through position detection patterns located at the
three corners of the QR code. Fig. 2 displays a sample screen of the server module’s main
The QR code carries a version range from one to 40, with the page.
version number indicating the volume of data that can be stored
within it. Each version of the QR code has a specific number of
rows and columns. For example, version 1 has 21 rows and 21
columns. The QR Code employs the Reed- Solomon error
correction code system to restore data even if data is found to be
partially dirty or damaged. There are four error correction levels,
denoted as L (Low), M (Medium), Q (Quartile), and H (High);
the amount of correction is based on the error correction levels
measured at 7%, 15%, 25%, and 30%, respectively. According
to [12], the size of the QR code symbol and environmental
conditions are taken into account when selecting the levels of
error correction. If the QR code, for instance, contains a large
amount of data and is used in a clean environment, then the level Fig. 2. Main page of server module.
L is sufficient but in most cases, the level M is generally used.
There are two types of QR codes: static, and dynamic. The Each student was required to download the mobile module
encoded information cannot be altered in a static QR code, from the university’s play store and install it on their
whereas the dynamic QR code allows for editing. The type of smartphone devices. First-time login instructions involved
QR code to apply to a system depends on its purpose, either for registering student identity (ID), mobile numbers, and mobile
personal or business usage. The QR code application is widely device IDs; the registered information was sent to the server
used in various instances such as on restaurant menus, survey module for verification. It must be observed that students
forms, for product information, connecting to a Wireless were only allowed to register one mobile phone number and
one mobile device ID to prevent attendance cheating. The
Sunway University individual research grant: GRTIN-IRG-103-2021.

11

Authorized licensed use limited to: Ayushy Limbani. Downloaded on October 16,2024 at 08:18:05 UTC from IEEE Xplore. Restrictions apply.
server module registered the pertinent information into the
respective databases. Students discovered having existing
registered mobile numbers or mobile device IDs, had their
registrations rejected; and these students had to approach the
course administrator to update or change their information.
Fig. 3 shows the flowchart of the registration process with
the mobile module. An image of the screen for the mobile
module without registration is captured in Fig. 4.

Begin

Student registers his

or her mobile number

Mobile module sends the

registered information
to the server module

Server module receives

the registered information
Fig. 4. Mobile module screen without registration.

Server module [Invalid] Server module returns an error

validates the message to mobile module
information

[Valid]

Server module returns Server module displays

a successful message the error message
to mobile module

Mobile module
displays the message

End

Fig. 3. Flowchart on the student registration by the mobile module.

Fig. 5. Mobile module shows that the student is on the campus and is ready to
III. RESULTS AND DISCUSSION scan the QR code for attendance.
In this section, the process for a student to take to
attendance is being described in detail. Each classroom had at least two QR code posters available
within the classroom parameter. The error correction level L, of
When a student launched the mobile module, the GPS 7%, was selected because the QR code posters were placed
tracking on the module ensured that the student was on campus indoors and less exposed to harsh, unpredictable elements. The
by retrieving the student’s live coordinates, latitude and QR code consisted of the classroom code, floor level, and
longitude, comparing it against campus latitude and longitude building name as shown in Fig. 6.
positions. The Sunway campus latitude and longitude are set to
3.0681 and 101.6041, respectively. If the difference between
the longitude and the latitude values, d, was determined as
d > 0.005, it indicated that the student was not on campus.
(Note that, 0.005 is a user-defined value.) As a result, the
mobile module displayed an error message to inform the
student. Otherwise, the ‘SCAN THE QR CODE’ button on
the mobile module was enabled if the student was found to be
within the university campus, as shown in Fig. 5.
Fig. 6. QR code for a classroom.

12

Authorized licensed use limited to: Ayushy Limbani. Downloaded on October 16,2024 at 08:18:05 UTC from IEEE Xplore. Restrictions apply.
 For attendance of a specific class to be taken, the following check-in from anywhere as long as they gain access to the
stages were needed: students had to scan the QR code poster in random code.
the classrooms through their mobile modules; the mobile
module, first, interpreted the QR code to extract each classrooms
information; then connected the information to the server
module, either via the local Wi-Fi coverage offered by the
university or through the internet; once the connection was
successful, the mobile module sent the categories of information
to the server module, which were date, time (in hours and
minutes), classroom name, and student mobile device ID.

Fig. 7. Mobile module shows that the student has successfully check-in.

Upon receiving the information, the server module began

processing the information by verifying it against data stored in
the database. The server module, first, retrieved the student
information through the student mobile device ID before
validating it against the subjects the student had registered for
and the timetables for the registered subjects. Students were
allowed to sign in for their attendances to be taken 15 minutes
before or after the scheduled class. If all the information is
correct, the server module recorded it in the database and if the
scanned time exceeded the stipulated 15 minutes, the server
added a late flag to the attendance. Post processing, the server
responded to the mobile module with a message. The mobile
module, in return, displayed the received response to the student,
as shown in Fig. 7. An error message sent by the server if a
student had accidentally scanned their attendance multiple times
or when a student scanned the wrong classroom QR code for
attendance. Next, in Fig. 8 a flowchart summarises the flow of
the attendance-taking process.
Currently, the attendance tracking system at Sunway
University requires students to enter a five-digit random code
generated for the class by the instructor through the attendance
server. Students subsequently log in to their student portal
server to enter the given code, which is validated by the server,
matching it to the list of subjects registered to each student. If
there is a match, the server updates the student’s attendance,
given the student signs in within 15 minutes after the random
code is generated. Otherwise, students would require their End
instructor’s assistance to update their attendance manually. The
two shortcomings of the existing attendance system are as Fig. 8. Flowchart visualises the process for a student to take attendance.
follows: the system is not an automated system and students can

13

Authorized licensed use limited to: Ayushy Limbani. Downloaded on October 16,2024 at 08:18:05 UTC from IEEE Xplore. Restrictions apply.
 When the experiment was carried to evaluate the proposed REFERENCES
system, few errors were found. Firstly, it is the minimum [1] S. Bu¨chele, “Evaluating the link between attendance and performance
requirement of the Android version. The mobile application is Assessment & Evaluation in Higher Education, vol. 46, no. 1, pp. 132–
required to run on Android 7 and above. Thus, students and 150, 2021.
instructors with older version of Android mobile phone are [2] N. Ahmad, S. A. Mohamed, A. Z. Ul-Saufie, H. Ahmat, and F. A. Alias,
unable to install the system. Next, the insensitivity of the GPS “Enhancing attendance and student exam score based on mobile
attendance application,” ESTEEM Journal, vol. 16, pp. 38–46, 2020.
detection of mobile phone. GPS detection on some mobile
phones is not very sensitive, students have to try for a few times [3] W. H. Woo, “Student attendance recording system.” [Online]. Available:
http://eprints.utar.edu.my/2199/1/SE-2015-1207679-1.pdf.
and move to different locations in the classroom.
[4] F. Masalha and N. Hirzallah, “A students attendance system using
qr code,” International Journal of Advanced Computer Science and
IV. CONCLUSION Applications, vol. 5, no. 3, pp. 75–79, 2014.
In this study, the proposed automated QR code-based [5] S. A. M. Noor, N. Zaini, M. F. A. Latip, and N. Hamzah, “Android- based
attendance system overcame the mentioned shortcomings by attendance management system,” in 2015 IEEE Conference on Systems,
preventing the attendance cheating through the verification Process and Control (ICSPC). IEEE, 2015, pp. 118–122.
process of the categories of information. Several [6] V. M. Vinod, S. Thokaiandal, C. Sindhuja, V. Mekala, M. Manimegalai,
and N. Prabhuram, “A comprehensive study on academic and industry
enhancements can be further integrated into the system to authentication and attendance systems,” International Journal of
improve and refine the limitations encountered. For example, Scientific & Technology Research, vol. 9, no. 3, 2020.
the prototype mobile module only supported Android mobile [7] N. Z. Lamin, W. A. W. Jusoh, J. Zainudin, and H. Samad, “Imple-
operating systems. Therefore, students using other mobile menting student attendance system using fingerprint biometrics for kolej
operating systems were unable to use the system. Besides, the universiti poly-tech mara,” in IOP Conference Series: Materials Science
prototype server module only printed out the daily attendance and Engineering, vol. 1062, no. 1. IOP Publishing, 2021, p. 012037.
reports and instructors were unable to obtain attendance [8] S. Tachmammedov, Y. K. Hooi, and K. S. Kalid, “Automated multifactor
percentages for every student. Additionally, the geolocation analytics for cheat-proofing attendance-taking,” in Proceedings of the
2018 7th international conference on software and computer applications,
tracking was restricted to students within the Sunway 2018, pp. 183–188.
University campus; but this was deemed sufficient as the [9] S. Limkar, S. Jain, S. Kannurkar, S. Kale, S. Garsund, and S. Desh- pande,
university grounds were small. Therefore, students outside the “ibeacon-based smart attendance monitoring and management system,”
range with the threshold would be considered absent but the in First International Conference on Artificial Intelligence and Cognitive
accuracy of geolocation tracking will be the future research Computing. Springer, 2019, pp. 637–646.
direction. Due to the Covid-19 pandemic, instructors had to [10] W. L. Yong, “Smart attendance system using qr code.” [Online].
show the QR code to students during online learning, whereby Available: http://eprints.utar.edu.my/3386/1/FYP report corrected.pdf
the geolocation feature in the proposed system was disabled or [11] A. Nuhi, A. Memeti, F. Imeri, and B. Cico, “Smart attendance system
using qr code,” in 2020 9th Mediterranean Conference on Embedded
the students’ locations were recorded for attendance to be taken Computing (MECO). IEEE, 2020, pp. 1–4.
remotely. The system was designed for physical classes at the
[12] S. Tiwari, “An introduction to qr code technology,” in 2016 international
university, and it is hoped that it can be implemented once conference on information technology (ICIT). IEEE, 2016, pp. 39–44.
students return to campus in the near future.
A CKNOWLEDGMENT
Khang Jie Liew would like to extend his gratitude to
Sunway University for the financial support involving the
individual research grant code GRTIN-IRG-103-2021

14

Authorized licensed use limited to: Ayushy Limbani. Downloaded on October 16,2024 at 08:18:05 UTC from IEEE Xplore. Restrictions apply.