www.ijcrt.

org © 2024 IJCRT | Volume 12, Issue 4 April 2024 | ISSN: 2320-2882

AUTOMATED ATTENDANCE SYSTEM USING

FACIAL RECOGNITION
1
Daniel Suresh, 2Anirudhan P M, 3Ashwin C, 4Amal P D, 5Sreeji S
1
Student, 2Student, 3Student, 4Student, 5Assistant Professor (CSE )
1
Computer Science and Engineering Department,
1
Nehru College of Engineering and Research Centre (NCERC), Thrissur, India

Abstract: This project introduces an innovative solution for automating attendance management through
facial recognition technology. By harnessing the power of image processing and machine learning algorithms,
our system offers a seamless and efficient way to track attendance in various educational and organizational
settings. The user-friendly interface allows for easy registration of students or employees, capturing their
facial data for identification during attendance sessions. With real-time recognition capabilities, the system
accurately records attendance without manual intervention, saving time and resources for administrators.
Additionally, comprehensive reporting features enable detailed analysis of attendance trends, contributing to
better decision-making and improved operational efficiency. Overall, our automated attendance system
revolutionizes traditional methods, offering a reliable and convenient solution for attendance management
needs.

Keywords - Automation, Attendance Management, Realtime, Analysis

Introduction
Attendance management is a fundamental aspect of organizational and
educational administration, serving as a cornerstone for monitoring participation, tracking progress, and
ensuring accountability. Traditional methods of attendance tracking, such as manual roll calls or sign-in sheets,
are often labor-intensive, error-prone, and susceptible to manipulation. In response to these challenges, there
has been a growing interest in leveraging technology to automate the attendance management process.
Facial recognition technology has emerged as a promising solution for automating
attendance tracking, offering a non-intrusive and efficient means of identifying individuals based on their
unique facial features. By harnessing the power of image processing and machine learning algorithms, facial
recognition systems can accurately and rapidly match faces against a database of enrolled individuals, thereby
streamlining the attendance recording process.
This paper introduces an innovative automated attendance system based on
facial recognition technology. Our system aims to revolutionize traditional attendance management methods
by providing a reliable, convenient, and secure solution for organizations and educational institutions. Through
a combination of state-of-the-art facial recognition algorithms, intuitive user interfaces, and robust backend
infrastructure, our system offers administrators an efficient tool for monitoring attendance, tracking
participation, and generating comprehensive reports.

In this paper, we present a detailed overview of our automated attendance system, including its architecture,
functionality, and key features. We also discuss the underlying technologies and methodologies employed in
the development of the system, highlighting their strengths and limitations. Furthermore, we provide insights
into the potential benefits of adopting our system, such as improved accuracy, reduced administrative burden,
and enhanced data security.

IJCRT24A4509 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org n104

www.ijcrt.org © 2024 IJCRT | Volume 12, Issue 4 April 2024 | ISSN: 2320-2882
I. LITERATURE REVIEW
[1] The integration of biometric technology, particularly Fingerprint Attendance Systems, has revolutionized
attendance tracking in educational institutions. These systems automate attendance marking, replacing manual
processes with efficient, secure, and accurate methods. Students use hand-held devices equipped with
fingerprint sensors to record their attendance seamlessly, fostering accountability and punctuality. For
educators, this technology reduces administrative burden, frees up teaching time, and provides real-time
attendance data for personalized instruction and improved student success.
[2] The implementation of RFID-based attendance systems addresses the challenge of student irregular
attendance in educational institutions and workplaces. Unlike manual methods, RFID technology offers a
faster, more secure, and efficient alternative by uniquely identifying individuals through RFID tags embedded
in ID cards. This system streamlines attendance tracking processes, ensuring accuracy and security. Users
simply place their ID cards on RFID readers for immediate attendance recording, facilitated by real-time clock
capabilities. The system's versatility extends to various workplaces, supporting seamless integration with
computers via connectivity options like RS232 or USB ports. Recorded attendance data is stored in databases
for efficient management, with accessibility enhanced through integration with HyperTerminal software.
[3] Iris recognition has emerged as a highly reliable method for personal identification in biometrics, leading
to its exploration in various applications, including time attendance systems. This literature review focuses on
the implementation of a Wireless Iris Recognition Attendance Management System, utilizing Daugman's
algorithm. By integrating biometrics and wireless technologies, this system addresses challenges such as
spurious attendance and network infrastructure complexities. The discussion emphasizes the significance of
iris recognition in attendance management and evaluates the effectiveness of the proposed wireless system in
improving user attendance processes.
In [4], The authors suggested using facial recognition as the basis for an attendance system. The system was
implemented using the Support Vector Machine (SVM) classifier and algorithms such as Viola-Jones and
Histogram of Oriented Gradients (HOG) features. The authors took into account a number of real-time
circumstances, including scaling, illumination, occlusions, and position. Peak Signal to Noise Ratio (PSNR)
measurements were the basis for quantitative analysis, which was carried out using the MATLAB GUI.
By comparing the Receiver Operating Characteristics (ROC) curve, the authors of [5] conducted research to
determine which facial recognition algorithm—Eigenface and Fisherface—was optimal for the Open CV 2.4.8.
They subsequently integrated the method into the attendance system. The ROC curve, which was derived from
the trials conducted for this research, demonstrated that Eigenface outperforms Fisherface in terms of results.
A system that used the Eigenface algorithm had an accuracy rate between 70% and 90%.
II. PROBLEM STATEMENT
In traditional academic and organizational settings, manual attendance tracking methods are often time-
consuming, error-prone, and susceptible to various inefficiencies. The existing solutions, such as RFID-based,
fingerprint-based, or manual systems, come with their own set of limitations, including intrusiveness, hygiene
concerns, and scalability issues. These drawbacks highlight the need for a modernized and efficient attendance
tracking system that can overcome these challenges. Furthermore, the ongoing global emphasis on contactless
interactions, heightened privacy concerns related to biometric data, and the demand for scalable solutions in
diverse environments necessitate a technology-driven transformation in attendance management. The objective
is to develop an automated attendance system based on face recognition, leveraging the advancements in
computer vision and deep learning, to address the shortcomings of traditional methods.
III. PROPOSED SYSTEM
Each student in the class must register by providing the necessary information, and their images will then
be taken and added to the dataset. At the conclusion of each class, a list of absentee students will be mailed to
the relevant faculty member overseeing the session. During each session, faces will be detected from live
streaming video of the classroom and compared with images already included in the dataset. If a match is
found, attendance will be marked for that particular student.

IJCRT24A4509 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org n105

www.ijcrt.org © 2024 IJCRT | Volume 12, Issue 4 April 2024 | ISSN: 2320-2882

Fig.1 : system architecture

This paper presents a detailed overview of an Automatic Attendance System using Facial Recognition,
outlining its four key stages: Dataset Creation, Face Detection, Face Recognition, and Attendance Updation.

3.1 Dataset Creation

In the Dataset Creation stage, students register themselves by inputting required details, following
which their images are captured and stored in a dataset. This process involves multiple images of each student
acquired with varied gestures and angles using a webcam. The captured images undergo preprocessing,
including cropping to obtain the Region of Interest (ROI), resizing to a specific pixel position, and conversion
to grayscale. Each image is then saved with the respective student's name in a designated folder.
3.2 Face Detection
Face Detection is performed using the Haar-Cascade Classifier with OpenCV, trained to detect human
faces. The classifier utilizes features extracted from training data, enabling the creation of rectangles around
detected faces in images. Parameters such as scaleFactor, minNeighbors, and minSize are adjusted to optimize
face detection accuracy. This stage ensures swift and accurate identification of faces from live streaming
video.

3.3 Face Recognition

The Face Recognition stage involves three key steps: preparing training data, training the face
recognizer, and prediction. Training data comprises images from the dataset, each assigned a unique integer
label corresponding to the student. The Local Binary Pattern Histogram face recognizer is employed,
generating histograms from Local Binary Patterns (LBPs) of entire faces. During recognition, the histogram
of the face to be recognized is compared with precomputed histograms, yielding the best-matched label.
3.4 Attendance Updation
After face recognition, recognized faces are marked as present in an Excel sheet, while unrecognized
faces are marked as absent. A list of absentees is generated and emailed to the respective faculties for further
action. Faculties receive monthly attendance sheets for comprehensive updates, ensuring accurate attendance
tracking and management.

IV. RESULTS AND DISCUSSION

The users interact with the system through a Graphical User Interface (GUI) developed using Python Tkinter.
The interface offers three primary options: Register, Take Attendance, and View Attendance. The Register
option allows users to input student details such as enrolment number and name. Notably, the system captures
approximately 200 images of each individual continuously to ensure enhanced accuracy. These images
undergo Region of Interest (ROI) cropping and are stored in the database for subsequent model training. This
meticulous data collection process forms the foundation for robust attendance tracking.

IJCRT24A4509 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org n106

www.ijcrt.org © 2024 IJCRT | Volume 12, Issue 4 April 2024 | ISSN: 2320-2882

Fig 2 : GUI for registration

In the Capture Attendance section, users have the capability to input the name of the section or subject, against
which attendance is marked. Leveraging the facial recognition technology integrated into the system,
attendance is efficiently recorded based on facial matches with the stored dataset. This streamlined process
eliminates the need for manual attendance taking, thereby saving time and reducing errors.

Fig 2.1 : Training image

Furthermore, the View Attendance option provides users with an intuitive interface to access and review
attendance records. Through this feature, users can effortlessly check attendance status and access attendance
sheets for detailed analysis. The system's comprehensive functionality empowers administrators and educators
to monitor attendance effectively, identify patterns, and make informed decisions to enhance overall
organizational performance.

Fig 2.2 : Attendance marking

IJCRT24A4509 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org n107

www.ijcrt.org © 2024 IJCRT | Volume 12, Issue 4 April 2024 | ISSN: 2320-2882
IV. CONCLUSION
This system uses facial recognition algorithms to provide an efficient class attendance system. The
suggested solution has the capability to record attendance using facial recognition. Through the use of a
webcam, it will first identify faces. Following acknowledgment, the attendance record will be updated and the
recognized student's attendance will be noted.

REFERENCES
[1] B. K. Mohamed and C. Raghu, “Fingerprint attendance system for classroom needs,” in India Conference
(INDICON), 2012 Annual IEEE. IEEE, 2012, pp. 433–438.
[2] T. Lim, S. Sim, and M. Mansor, “Rfid based attendance system,” in Industrial Electronics & Applications,
2009. ISIEA 2009. IEEE Symposium on, vol. 2. IEEE, 2009, pp. 778–782.
[3] S. Kadry and K. Smaili, “A design and implementation of a wireless iris recognition attendance
management system,” Information Technology and control, vol. 36, no. 3, pp. 323–329, 2007.
[4] Rathod, Hemantkumar, et al. "Automated attendance system using machine learning approach." 2017
International Conference on Nascent Technologies in Engineering (ICNTE). IEEE, 2017.
[5] Siswanto, Adrian Rhesa Septian, Anto Satriyo Nugroho, and Maulahikmah Galinium.
"Implementation of face recognition
algorithm for biometrics based time attendance system." 2014 International Conference on ICT For
Smart Society(ICISS).IEEE, 2014.

IJCRT24A4509 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org n108