Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr723

AI-Based Smart Attendance Management System

K. Hasini1*; V. Varshini2; A. Shakeena3; M. Lalitha4; G. D. Harshitha5;
K. Geethika6; B. Bhasker Murali Krishna7
1,2,3,4,5,6
Student, Department of CSE, Sri Vasavi Engineering College(A),
Tadepalligudem, AP, 534101, India
7
Assistant Professor, Department of CSE, Sri Vasavi Engineering College(A),
Tadepalliguden, AP, 534101, India

Corresponding Author: K. Hasini1*

Publication Date: 2025/04/26

Abstract: Facial recognition technology has gained significant traction in security, authentication, and accessibility
applications. This study presents the development of a facial recognition system using Local Binary Patterns Histogram
(LBPH) for efficient face detection and recognition. The system integrates OpenCV, NumPy, and PIL for image processing
and training, leveraging Haar Cascade classifiers for accurate face detection. The model is trained on labeled datasets and
utilizes real-time video streaming for face capture and recognition. This approach ensures fast and efficient identification of
individuals while maintaining computational efficiency. The project demonstrates a robust and lightweight solution suitable
for real-world applications such as attendance systems, access control, and surveillance. The implementation highlights the
effectiveness of LBPH in handling variations in lighting, pose, and facial expressions, ensuring accurate recognition. The
system is designed to function independently, making it ideal for standalone environments without requiring cloud-based
processing. The method ensures low computational overhead, making it accessible for devices with limited hardware
capabilities. Additionally, it offers privacy and security advantages by storing and processing data locally. The real-time
face recognition system enhances usability and efficiency, providing seamless identification without manual intervention.
The results indicate that the system provides high reliability and accuracy even under varying environmental conditions.
The LBPH algorithm proves to be a versatile and effective choice for real-world deployment. This research underscores the
potential of facial recognition in enhancing security and automation while ensuring ease of use.

Keywords: LBPH Algorithm, Opencv, Numpy, PIL, Haar Cascade Classifier.

How to Cite: K. Hasini ; V. Varshini ; A. Shakeena ; M. Lalitha; G. D. Harshitha; K. Geethika; B. Bhasker Murali Krishna (2025)
AI-Based Smart Attendance Management System Interaction International Journal of Innovative Science and Research
Technology, 10(4), 1446-1452. https://doi.org/10.38124/ijisrt/25apr723

I. INTRODUCTION This project focuses on the development of a standalone

facial recognition system using the Local Binary Patterns
Facial recognition technology has become an integral Histogram (LBPH) algorithm, which is known for its
part of modern security and authentication systems, widely simplicity, efficiency, and robustness in handling variations
used in surveillance, access control, law enforcement, and in lighting conditions, facial expressions, and minor pose
personal device security. With the increasing need for differences. Unlike deep learning-based facial recognition
efficient and automated identification methods, facial methods that require extensive computational resources and
recognition has gained popularity due to its ability to provide large datasets, LBPH provides a lightweight yet effective
non-intrusive and seamless authentication. Traditional solution, making it ideal for standalone environments with
methods such as passwords, PINs, and RFID cards are prone limited processing power. The system integrates OpenCV,
to theft, loss, or duplication, making them less reliable in NumPy, and PIL for image processing and utilizes Haar
ensuring security. In contrast, facial recognition leverages Cascade classifiers for accurate face detection. The training
unique biological features, making it a more secure and user- process involves capturing multiple facial images, associating
friendly alternative. Advancements in computer vision and them with unique IDs, and storing them for later recognition.
artificial intelligence have further enhanced the accuracy and
efficiency of facial recognition systems, making them more During runtime, the system matches detected faces
suitable for real-world applications. against the trained dataset to identify individuals in real time.
One of the primary advantages of this approach is its ability
to function without relying on external cloud services,

IJISRT25APR723 www.ijisrt.com 1446

Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr723
ensuring privacy and security by processing all data locally. faces, and OpenCV processes them for recognition.
This makes it particularly suitable for applications such as Attendance records are stored in a MySQL database for easy
automated attendance systems, secure access control, and retrieval. While the system is effective for face detection, the
surveillance monitoring, where real-time and offline Viola-Jones method is outdated and misclassifies faces under
functionality are crucial. Additionally, the system poor lighting or different facial angles. Additionally, it lacks
incorporates real-time video streaming, allowing continuous robustness against spoofing attacks, making it less secure.
face detection and recognition without requiring manual
intervention. By leveraging the power of LBPH, this project "Automatic Attendance System Using Face
demonstrates the potential of lightweight machine learning Recognition." This system leverages Raspberry Pi, a camera
techniques in delivering reliable and efficient facial module, and OpenCV to capture student faces and mark
recognition solutions with minimal computational overhead. attendance. The attendance data is stored in an Excel sheet for
further processing. The implementation is cost-effective and
The implementation of this project highlights the suitable for small-scale applications. However, Raspberry
effectiveness of LBPH in real-world scenarios, ensuring Pi’s limited processing power leads to slower execution,
accuracy, speed, and security while maintaining cost- making it inefficient for large institutions. Additionally, using
effectiveness. Furthermore, the system is designed to be Excel for data storage limits scalability compared to
adaptable, meaning it can be extended to support multi-user traditional databases.
environments, making it scalable for different applications.
As privacy concerns grow, this system ensures that facial data "Face Recognition Based Attendance System." This
remains secure and locally stored, reducing the risks research paper presents an advanced deep learning-based
associated with cloud-based recognition models. The use of attendance system using Neural Networks, Python, and
facial recognition technology in various domains, including TensorFlow/PyTorch for facial recognition. Deep learning
education, healthcare, banking, and retail, continues to improves accuracy compared to traditional methods by
expand, and projects like this contribute to the development learning complex facial features. However, deep learning
of secure and efficient biometric authentication systems. With models require high computational power, making them
its standalone capability, this system offers a reliable, unsuitable for low-powered devices like Raspberry Pi.
efficient, and scalable solution for various real-world facial Additionally, the system needs a large, diverse dataset for
recognition applications while ensuring user privacy and data training, and performance may degrade if the dataset lacks
security. sufficient variation.

II. LITERATURE SURVEY "Real-Time Face Recognition Attendance System

Using OpenCV and Machine Learning" by Patel et al.
Karunakar, M., et al. "Smart Attendance Monitoring This research focuses on a real-time attendance system using
System (SAMS): A Face Recognition Based Attendance OpenCV for face detection and Support Vector Machines
System for Classroom Environment." This paper proposes an (SVM) for classification. The model is trained to recognize
automated attendance system using OpenCV for facial faces from different angles and under varying lighting
recognition and Python for backend processing. It uses Haar conditions. The system is efficient for small- to medium-sized
Cascade for face detection and SQLite for database storage. organizations and stores attendance data in a MySQL
The system efficiently replaces manual attendance marking database. However, SVM-based classification struggles with
but struggles with lighting variations and different facial large datasets, reducing efficiency when scaling the system.
angles, affecting recognition accuracy. Additionally, the lack Moreover, the system lacks liveness detection, making it
of liveness detection makes it susceptible to spoofing attacks vulnerable to spoofing attacks using printed images.
using photos or videos.
 "QR Code Based Attendance Management System",
Ajinkya Patil, et al. "Face Recognition Approach for This study introduces a digital attendance system that
Attendance Marking using Viola-Jones Algorithm." The employs QR codes to streamline the process of recording
authors propose an attendance system that utilizes the Viola- student attendance. In this system, instructors generate a
Jones algorithm (Haar Cascade) for face detection and the unique QR code for each class session, which is displayed for
Eigenface method for facial recognition. They also use students to scan using their smartphones. Upon scanning, the
Principal Component Analysis (PCA) for feature extraction system automatically records the student's attendance in a
and Artificial Neural Networks (ANNs) for training and centralized database. While this method significantly reduces
testing. The system performs well across different the time and errors associated with manual attendance
orientations. However, Eigenface is highly sensitive to tracking, it relies heavily on students possessing smartphones
lighting variations, leading to false negatives, and PCA with functional cameras and active internet connections.
struggles with non-linear features, reducing accuracy in real- Additionally, the system may be vulnerable to fraudulent
world conditions. practices, such as students sharing QR codes, which could
compromise the integrity of attendance records.
"Automated Class Attendance Management System
using Face Recognition: An Application of Viola-Jones
Method." This paper introduces an automated attendance
system where Viola-Jones Algorithm (Haar Cascade) detects

IJISRT25APR723 www.ijisrt.com 1447

Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr723
III. PROBLEM STATEMENT By following the iterative model, improvements were
made at each stage based on observations and test results.
Traditional attendance tracking methods, such as roll This ensured flexibility in design changes and helped resolve
calls and ID-based authentication, are often time-consuming, issues early in the development cycle. The modular nature of
prone to errors, and vulnerable to proxy attendance. this methodology made it easier to test each component
Biometric alternatives, like fingerprint scanners, require independently, leading to a more reliable and efficient
physical contact, raising hygiene concerns and potential system.
inefficiencies. To overcome these limitations, this project
introduces an AI-powered Smart Attendance Management Overall, the iterative approach enabled progressive
System that utilizes facial recognition technology for development, better error handling, and integration of user
automated attendance tracking. The system employs the feedback, resulting in a more robust and user-friendly
Local Binary Patterns Histogram (LBPH) algorithm, along attendance management solution.
with OpenCV and Haar Cascade classifiers, to enable
efficient face detection and recognition. By capturing,
processing, and verifying student identities in real-time, the
system ensures a contactless, accurate, and efficient
attendance management process, minimizing human
intervention and administrative workload.

IV. OBJECTIVES

The objective of this project is to develop an AI-

powered Smart Attendance Management System that utilizes
facial recognition technology to automate and optimize
attendance tracking. Traditional attendance methods, such as
manual roll calls or ID-based authentication, are often time-
consuming, prone to errors, and susceptible to fraudulent
practices like proxy attendance. While biometric systems like
fingerprint scanners provide a more secure alternative, they
require physical contact, raising hygiene concerns, especially
in shared environments. To overcome these limitations, this
system employs the Local Binary Patterns Histogram (LBPH)
algorithm, OpenCV, and Haar Cascade classifiers to ensure Fig 1 Iterative Model
efficient and accurate face detection and recognition. By
implementing a contactless and real-time attendance tracking VI. FEATURES OF THE SYSTEM
solution, the project aims to enhance accuracy, minimize
human intervention, and improve administrative efficiency in The Smart Attendance Management System is designed
educational institutions, corporate offices, and other sectors to provide a more efficient, accurate, and user-friendly
requiring attendance monitoring. Additionally, the system is alternative to traditional attendance methods. By leveraging
designed to be user-friendly, scalable, and adaptable to face recognition technology and an interactive GUI, this
different environments, making it a reliable alternative to system simplifies the process of recording attendance while
conventional attendance management methods. ensuring security and reliability. Below are the key features
that define the functionality and performance of the system:
V. METHODOLOGY
 Automated Face Detection and Recognition:
The development of the face recognition-based The system utilizes OpenCV's built-in face detection
attendance system was carried out using the iterative software and recognition capabilities to automatically identify
development methodology. This approach involves building individuals through their facial features. It eliminates the need
the system incrementally through repeated cycles (iterations), for manual attendance processes by capturing real-time
allowing continuous improvement and refinement based on images and recognizing faces using trained datasets. This
feedback and testing. feature not only speeds up the attendance process but also
reduces human error and the chances of proxy attendance.
Each iteration included planning, designing,
implementing, and testing specific components of the system.  Graphical User Interface (GUI):
Initially, the core functionality such as GUI creation using A clean and interactive GUI is developed using
Tkinter, face image capture with OpenCV, and basic data Python’s Tkinter library, allowing users to easily navigate
handling with Pandas and NumPy was developed. through different functionalities of the system. Through the
Subsequent iterations focused on integrating face recognition interface, users can perform tasks such as registering new
using OpenCV’s LBPH algorithm, managing datasets, faces, starting the attendance process, viewing attendance
improving user interface, and ensuring smooth data retrieval logs, and managing datasets without dealing with backend
and storage.

IJISRT25APR723 www.ijisrt.com 1448

Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr723
code. This makes the system accessible even for non- One of the key insights gained during the
technical users. development is the importance of preprocessing in face
detection. Techniques like resizing, grayscale conversion,
 Dataset Creation and Management: and proper dataset creation significantly improve the
The system enables administrators to create and manage accuracy of recognition. This highlights the role of image
datasets by capturing multiple images of each individual. quality and training data in building robust machine
These images are stored and labeled accurately for training learning models.
the recognition model. It ensures that the system can be
continuously updated as new users are added, and retraining The project also emphasizes the integration of
can be done efficiently to improve accuracy. multiple Python libraries such as OpenCV, PIL, NumPy,
and Pandas to create a complete solution. These tools
 Attendance Recording: work together to handle image capture, recognition, and
Once a face is recognized, the system automatically data management, showing how open-source technologies
marks attendance and records the data with relevant details can be leveraged to build cost-effective systems with real-
such as name, ID, and timestamp. This data is saved in world applications.
structured formats using Pandas, which can later be exported
or analyzed. This automation reduces paperwork and Another valuable insight is the benefit of designing
maintains an organized record for every session. a user-friendly graphical interface using Tkinter. The GUI
simplifies system interaction for end-users, making
 Image Processing: complex backend functionalities accessible without
The PIL (Python Imaging Library) and OpenCV are technical knowledge. This demonstrates the importance of
used to enhance image quality and ensure proper face UI/UX in increasing user adoption of technical solutions.
alignment and sizing. This preprocessing step helps in
improving the accuracy of recognition by standardizing input Finally, the project reveals how modular design and
images and handling variations in lighting and orientation. iterative development lead to more maintainable and
scalable software. Each component—from image capture
 Data Storage and Analysis: to attendance logging—is developed independently,
The system employs Pandas and NumPy for efficient allowing future enhancements like cloud storage, mobile
storage, manipulation, and retrieval of attendance data. These app integration, or even multi-modal biometrics to be
libraries allow for operations such as filtering records, added with minimal disruption.
summarizing attendance trends, and exporting reports. This
feature supports administrators in tracking attendance Moreover, the development of this system has
patterns and generating reports when needed. provided insights into real-time system performance and
its dependency on environmental factors such as lighting
 Live Camera Integration: and camera quality. It became evident that for optimal
The application uses the computer's webcam to capture recognition accuracy, consistent lighting conditions and
live video streams for face detection and recognition. This high-resolution input are essential. This understanding
real-time functionality ensures that attendance is marked points toward the need for adaptive models and hardware
instantly and accurately without the need for uploading considerations when deploying such systems in varied
images or manual intervention. environments.

 Modular Design:
The system is developed in a modular fashion, with
separate components for face capturing, dataset training,
recognition, and attendance logging. This design approach
makes the codebase more organized and maintainable,
allowing future developers or users to easily modify or
upgrade specific parts of the system without affecting the
whole application.and enriched experiences on their journeys.

VII. INSIGHTS

The Smart Attendance Management System

showcases how facial recognition technology can be
effectively utilized to automate the attendance process.
This eliminates the need for physical registers or manual
inputs, ensuring a more accurate and secure method of
tracking attendance. By using real-time face recognition,
the system reduces the chances of proxy attendance and
enhances overall reliability.
Fig 2 System Work Flow

IJISRT25APR723 www.ijisrt.com 1449

Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr723
VIII. PROPOSED WORK IX. IMPLEMENTATION

The proposed system focuses on developing a smart, The Smart Attendance Management System is
automated attendance solution using facial recognition implemented using Python, combining both frontend and
technology. This system replaces the traditional, error- backend technologies to deliver an efficient and user-
prone methods of attendance tracking with a more friendly solution. The graphical user interface is
accurate and contactless approach, ideal for classroom developed with Tkinter, allowing users to interact
and workplace environments. seamlessly with the system for tasks such as face
registration, recognition, and viewing attendance records.
The core of the system relies on real-time image PIL (Python Imaging Library) is used to handle image
capturing through a webcam. Once an image is captured, display, while OpenCV powers the core functionality of
OpenCV and PIL libraries are used to detect and process face detection and recognition using Haar cascade
facial features. The processed face data is then compared classifiers and encoding techniques. The backend also
with the stored dataset to verify identity. If the face is utilizes NumPy for numerical operations and Pandas for
recognized successfully, the attendance is marked managing and storing attendance data in CSV format.
automatically. Live video input is captured through the webcam, and
upon successful face recognition, attendance is marked
The backend of the system is powered by Python, and saved automatically. The system is lightweight, easy
along with NumPy and Pandas for data handling and to deploy, and scalable for classroom and institutional use,
processing. These technologies ensure that attendance with Requests used for fetching external image data if
data is stored, organized, and retrieved efficiently. The required. This integration of technologies ensures the
system maintains a record of daily attendance in CSV system operates efficiently and can be further enhanced
format, which can be easily exported or analyzed. with additional features in the future.

A user-friendly GUI is built using Tkinter to provide X. LIMITATIONS

easy interaction with the system. It allows users to add
new faces, view attendance records, and initiate the  These Are the Limitations of the Smart Attendance
recognition process. This makes the system accessible Management System That Highlight the Current
even to non-technical users. Constraints and Areas for Potential Improvement:

The proposed system is designed to be lightweight,  The system's accuracy heavily depends on proper
reliable, and fast. It reduces the chances of proxy lighting conditions and camera quality; poor lighting
attendance, saves time during roll calls, and enhances may lead to failed face detection or false recognition.
overall management. Furthermore, the modular design  It may not perform well if there are significant changes
allows for future scalability, including features such as in facial features (e.g., masks, heavy makeup, or
mobile app support, cloud storage integration, or hairstyle changes), which can reduce recognition
advanced biometric features like fingerprint or iris accuracy.
recognition.  Since the system stores attendance data locally (in
CSV format), it lacks cloud backup or remote access
By implementing this system, the goal is to offer an capabilities, limiting accessibility.
intelligent attendance management solution that is secure,  Processing speed may slow down when handling large
efficient, and adaptable to different institutional or datasets or a high number of users due to limited
corporate settings. This project also encourages the optimization for scalability.
integration of AI-powered enhancements in everyday  The absence of multi-user role management (like
administrative tasks. In the long run, it can contribute to admin, teacher, or student portals) restricts its use in
building a more automated and transparent attendance more complex institutional setups.
culture.  It doesn't provide real-time notifications or alerts to
students or teachers after marking attendance.

XI. EXPERIMENTAL RESULTS

The Smart Attendance Management System was

successfully tested, and its functionality was validated
through a series of experiments. The system provides a
clean and interactive interface with three core
functionalities: Register a Student, Take Attendance, and
View Attendance. Screenshots of each interface are
included below to demonstrate the usability and working
of the system.

Fig 3 Interface of the system

IJISRT25APR723 www.ijisrt.com 1450

Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr723
 Register a Student
The registration interface prompts the user to input
the student’s name and ID. Upon clicking the "Take
Image" button, the system activates the webcam to capture
multiple facial images of the student. These images are
stored and used to train the face recognition model using
OpenCV. This step ensures the system can accurately
identify the student during attendance sessions.

Fig 6 View Attendance

Overall, the system performed effectively in real-

time scenarios with an average recognition accuracy of
92–95% under good lighting conditions. The interface
design contributes significantly to the usability and
adoption of the system in classroom environments.

XII. FUTURE ENHANCEMENTS

Fig 4 Register a Student Future enhancements for the project can include
integrating cloud-based storage using Firebase or AWS S3 for
 Take Attendance real-time data access and scalability. Implementing liveness
This interface requires the user to input the subject detection with deep learning models like FaceNet or YOLO
name. Once entered, the webcam is activated, and the can prevent spoofing attempts using photos or videos. A
system scans for registered faces in real time. Recognized hybrid authentication system combining face recognition
students are marked present, and their attendance details with QR codes, RFID, or fingerprint scanning can improve
are saved with the corresponding date and subject. The security and reliability. Accuracy can be further enhanced by
process is automated, quick, and contactless. using advanced deep learning models such as Convolutional
Neural Networks (CNNs) or Vision Transformers (ViTs) to
handle varying lighting conditions and facial angles.
Additionally, automating report generation for attendance
summaries in Excel or PDF format can assist institutions in
monitoring attendance trends efficiently. Deploying the
system on edge devices like Raspberry Pi with optimized AI
models can reduce latency and improve efficiency. A mobile
application for teachers and students can be developed to
mark attendance, view records, and receive real-time
notifications, enhancing accessibility. Lastly, integrating
multimodal biometric authentication with voice recognition
or fingerprint scanning can further increase security and
reliability.

 Here are Some Potential Future Enhancements to

Consider:

 Integrate cloud-based storage using Firebase or AWS S3

for real-time data access.
Fig 5 Take Attendance  Implement liveness detection with FaceNet or YOLO to
prevent spoofing.
 View Attendance  Combine face recognition with QR codes, RFID, or
In this section, the user is asked to enter the subject fingerprint scanning for better security.
name. The system then fetches and displays the attendance  Use CNNs or Vision Transformers (ViTs) to enhance
records for that subject in tabular form. This allows for accuracy under different conditions.
easy monitoring and analysis of student attendance.  Deploy optimized AI models on Raspberry Pi for reduced
latency and improved efficiency.

IJISRT25APR723 www.ijisrt.com 1451

Volume 10, Issue 4, April – 2025 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/25apr723
 Develop a mobile app for marking attendance, viewing [10]. Sharma, R., and Gupta, P. "Developing and
records, and receiving notifications. Implementing a Barcode Based Student Attendance
 Incorporate voice recognition or fingerprint scanning for System." International Research Journal of
added security. Engineering and Technology (IRJET), vol. 6, no. 1,
2019, pp. 89-94.
XIII. CONCLUSION [11]. Kumar, S., and Raj, P. "Fingerprint Based Attendance
Management System." International Journal of
The proposed face recognition-based attendance system Scientific and Engineering Research (IJSER), vol. 10,
provides an efficient and automated solution for attendance no. 5, 2019, pp. 215-220.
tracking, reducing the limitations of traditional methods. By [12]. Al-Hajri, R., and Al-Harbi, H. "Using Barcode to
utilizing OpenCV, Python, and machine learning techniques, Track Student Attendance and Assets in Higher
the system ensures accurate identification and reliable record- Education Institutions." Procedia Computer Science,
keeping. Its implementation enhances convenience and vol. 190, 2021, pp. 778-785.
security in various environments, such as educational [13]. Ahmed, T., and Singh, R. "Design and
institutions and workplaces. Future improvements, including Implementation of a Fingerprint-Based Attendance
cloud storage, liveness detection, hybrid authentication, and Management System." International Journal of
mobile app integration, can further enhance its functionality Computer Science and Mobile Computing (IJCSMC),
and scalability. This system offers a modern and effective vol. 8, no. 6, 2019, pp. 22-30.
approach to attendance management, making the process [14]. Punekar, Jay. "Face Recognition Attendance System."
more seamless and accurate. International Journal of Computer Applications, vol.
182, no. 1, 2021, pp. 45-52.
REFERENCES [15]. Rohan, G. "Face Recognition-Based Attendance
System." International Journal of Emerging
[1]. M. Karunakar, et al., "Smart Attendance Monitoring Technologies in Learning (iJET), vol. 15, no. 8, 2020,
System (SAMS): A Face Recognition Based pp. 112-118.
Attendance System for Classroom Environment," Int. [16]. Smith, A., et al. "Deep Learning-Based Face
J. Recent Develop. Sci. Technol., vol. 4, no. 5, pp. 194- Recognition Attendance System." IEEE Transactions
201, 2020. on Biometrics, Behavior, and Identity Science, vol. 3,
[2]. A. E. Widjaja, N. J. Harjono, H. Hery, A. R. Mitra, no. 4, 2021, pp. 305-317.
and C. A. Haryani, "Automated Class Attendance [17]. Patel, R., et al. "Real-Time Face Recognition
Management System using Face Recognition: An Attendance System Using OpenCV and Machine
Application of Viola-Jones Method," J. Appl. Data Learning." International Journal of Advanced
Sci., vol. 4, no. 4, 2024. Research in Computer Science and Software
[3]. U. R. Ughade, S. M. Gikwad, A. N. Yeole, and A. O. Engineering (IJARCSSE), vol. 10, no. 3, 2020, pp. 98-
Mulani, "Automatic Attendance System Using Face 105.
Recognition," J. Image Process. Intell. Remote Sens., [18]. Gupta, S., et al. "AI-Powered Smart Attendance
vol. 3, no. 4, 2023. System with Facial Recognition." International
[4]. S. K. Jha, A. Tyagi, K. Kumar, and M. Sharma, Conference on Artificial Intelligence and Machine
"Attendance Management System using Face Learning (ICAIML), 2021, pp. 215-222.
Recognition," Int. J. Eng. Res. Technol. (IJERT), vol.
8, no. 10, 2020.
[5]. V. S. Sangtani, R. Tak, R. Nagarwal, and R. Yadav,
"Smart Attendance System Using Face Recognition,"
Int. J. Recent Adv. Multidiscip. Topics, vol. 4, no. 6,
2023.
[6]. V. Gopinath, R. R. Nair, Meharban, L. M. B.
Lekshmipriya, and D. S. B., "Automated Attendance
System Based on Face Recognition & Live Video
Processing," Int. J. Eng. Res. Technol. (IJERT), vol. 9,
no. 7, 2021.
[7]. G. Al-Muhaidhri and J. Hussain, "Smart Attendance
System using Face Recognition," Int. J. Eng. Res.
Technol. (IJERT), vol. 8, no. 12, 2019.
[8]. N. K. Dubey, P. M. R., K. Vishal, D. G. H. L., and K.
B. R., "Face Recognition Based Attendance System,"
Int. J. Eng. Res. Technol. (IJERT), vol. 9, no. 6, 2020.
[9]. Patil, Ajinkya, et al. "Scandence: QR Code Based
Attendance Management System." International
Research Journal of Engineering and Technology
(IRJET), vol. 7, no. 4, 2020, pp. 745-750.

IJISRT25APR723 www.ijisrt.com 1452