Attendify: A Cloud-Based Multi-Factor Attendance System using QR Code, Facial Recognition and Geofencing

Attendify: A Cloud-Based Multi-Factor Attendance System using QR Code, Facial Recognition and Geofencing

Revanth. D, Sathwik. K, Akshitha. N

Students, Department of Computer Science and Engineering, Geethanjali College of Engineering and Technology

Kamakshiah Kolli

Professor, Department of Computer Science and Engineering, Geethanjali College of Engineering and Technology

Abstract – Traditional attendance systems are often inefficient, time-consuming, and prone to manipulation such as proxy attendance. These limitations reduce the reliability and accuracy of attendance records in educational and organizational environments. To address these challenges, a cloud-based multi-factor attendance management system named Attendify is proposed. The system integrates multiple layers of verification to ensure secure and accurate attendance marking. Initially, students are required to scan a dynamically generated QR code or enter a One-Time Password (OTP) that refreshes at regular intervals to prevent sharing. This is followed by facial recognition using computer vision techniques, including Haar Cascade and Convolutional Neural Networks (CNN), to verify the identity of the student. Additionally, Global Positioning System (GPS)-based geofencing is used to confirm the physical presence of the student within a predefined location. The system also incorporates an offline mode, where attendance data is securely stored on the device and synchronized with the central server once connectivity is restored. A real-time dashboard is provided for teachers to monitor attendance and generate analytical reports. The proposed system significantly improves accuracy, reduces fraudulent practices, and enhances overall efficiency, making it suitable for educational institutions and corporate environments.

Keywords: Attendance System, Facial Recognition, QR Code, One-Time Password (OTP), Geofencing, Offline Synchronization, Anti- Spoofing, Cloud-Based System

1. INTRODUCTION

   Attendance management plays a critical role in academic institutions and organizations, as it directly reflects participation, discipline, and performance of individuals. Traditional attendance systems, such as manual roll calls and paper-based registers, are widely used but suffer from several limitations. These methods are time-consuming, prone to human errors, and highly vulnerable to fraudulent practices such as proxy attendance, where one individual marks attendance on behalf of another. Such issues reduce the reliability and authenticity of attendance records.

   With the advancement of technology, various automated attendance systems have been introduced, including Radio Frequency Identification (RFID), biometric fingerprint scanners, and QR code-based solutions. Although these systems provide partial automation, they often rely on a single method of verification, making them susceptible to misuse. For example, QR codes can be easily shared among users, and biometric systems may fail due to environmental conditions, device limitations, or spoofing attempts. As a result, these systems do not completely eliminate the problem of proxy attendance.

   To overcome these challenges, a more robust and secure solution is required that combines multiple verification techniques. In this context, this paper proposes Attendify, a cloud-based multi-factor attendance management system designed to ensure high accuracy, security, and reliability. The system integrates three key verification mechanisms: time-sensitive QR code or One-Time Password (OTP), facial recognition using advanced computer vision techniques, and Global Positioning System (GPS)based geofencing. This layered approach ensures that attendance is recorded only when all verification conditions are satisfied, thereby minimizing the possibility of fraudulent activities.

   In addition to security, reliability is also a major concern in real-world environments where internet connectivity may be unstable or unavailable. To address this issue, the proposed system incorporates an offline mode that securely stores attendance data on the device and automatically synchronizes it with the central server once connectivity is restored. This ensures uninterrupted functionality and data integrity.

   Furthermore, the system provides a comprehensive dashboard for teachers, enabling real-time monitoring of attendance, detection of suspicious activities, and generation of detailed analytical reports. The architecture of the system is designed to be scalable and adaptable, making it suitable for deployment in educational institutions, corporate training environments, and large-scale events.

   The primary objective of this research is to develop a secure, efficient, and scalable attendance management system that eliminates proxy attendance, improves accuracy, and enhances user experience through the integration of modern technologies.

2. Literature Review

   The development of automated attendance systems has gained significant attention in recent years due to the limitations of traditional methods. Various technologies such as biometric systems, Radio Frequency Identification (RFID), and QR code-based solutions have been explored to improve efficiency and accuracy. This section reviews existing approaches and highlights their limitations, which motivated the proposed system.

   1. QR Code-Based Attendance Systems

      QR code-based attendance systems are widely adopted due to their simplicity and ease of implementation. In these systems, a QR code is generated and scanned by users to mark attendance. While this method reduces manual effort and speeds up the process, it has significant security limitations. QR codes can be easily shared among users, enabling proxy attendance. Additionally, such systems typically lack identity verification, making them unreliable in ensuring authentic attendance.

   2. Biometric and Facial Recognition Systems

      Biometric systems, particularly fingerprint and facial recognition technologies, have been introduced to enhance security in attendance management. Facial recognition systems use computer vision techniques to identify individuals based on their facial features. Recent studies have implemented algorithms such as Haar Cascade, Local Binary Patterns (LBP), and Convolutional Neural Networks (CNN) for improved accuracy. Although these systems reduce proxy attendance, they often face challenges such as sensitivity to lighting conditions, variations in facial expressions, and spoofing attacks using photos or videos. Moreover, standalone facial recognition systems may not verify the actual location of the user.

   3. GPS and Location-Based Attendance Systems

      Location-based attendance systems utilize Global Positioning System (GPS) technology to verify whether a user is present within a specified geographic boundary. These systems introduce geofencing techniques to ensure that attendance can only be marked from authorized locations. While this approach adds an additional layer of validation, it is not entirely secure, as GPS data can be manipulated using spoofing tools or Virtual Private Networks (VPNs). Therefore, relying solely on location-based verification does not guarantee authenticity.

   4. Limitations of Existing Systems

      Despite advancements in attendance technologies, most existing systems rely on a single method of verification, making them vulnerable to manipulation. QR-based systems lack identity verification, biometric systems may fail under certain conditions, and GPS-based systems can be spoofed. Furthermore, many existing solutions do not support offline functionality, making them unreliable in environments with poor internet connectivity. These limitations highlight the need for a more comprehensive and secure approach.

   5. Motivation for Proposed System

      To overcome the shortcomings of existing systems, a multi-factor approach is required that combines multiple verification techniques into a single framework. The proposed system, Attendify, integrates QR/OTP verification, facial recognition, and GPS- based geofencing to ensure a higher level of security and accuracy. Additionally, the inclusion of offline synchronization and anti- spoofing mechanisms makes the system more robust and reliable compared to existing solutions.

3. PROBLEM STATEMENT

   Attendance management is an essential component in educational institutions and organizations, as it directly impacts performance evaluation and discipline monitoring. However, existing attendance systems suffer from several limitations that reduce their effectiveness and reliability. Traditional methods such as manual roll calls and paper-based registers are time-consuming, prone to human errors, and inefficient for large groups. These methods also fail to ensure authenticity, making them unsuitable for modern environments.

   One of the major challenges in attendance systems is proxy attendance, where one individual marks attendance on behalf of another. This issue is prevalent not only in manual systems but also in semiautomated systems such as QR code-based attendance. Since QR codes can be easily shared through digital platforms, these systems remain vulnerable to misuse and fail to guarantee genuine participation.

   Biometric systems, including fingerprint and facial recognition technologies, were introduced to improve security. While these systems provide identity verification, they also have limitations. Fingerprint systems require physical interaction and may face hygiene concerns, whereas facial recognition systems can be affected by lighting conditions, camera quality, and facial obstructions such as masks. Additionally, they may be susceptible to spoofing attacks using images or videos.

   Another limitation is the lack of reliable location verification. Users can sometimes mark attendance remotely without being physically present. Although GPS-based systems attempt to solve this problem through geofencing, they are still vulnerable to manipulation using spoofing tools or Virtual Private Networks (VPNs), which compromises their effectiveness.

   Furthermore, many existing systems rely heavily on continuous internet connectivity. In real-world scenarios, network issues can interrupt attendance marking and lead to data inconsistency. The absence of a secure offline mechanism reduces system reliability, especially in environments with unstable connectivity.

   These limitations highlight the need for a more robust and secure attendance system that ensures authenticity, prevents fraudulent practices, supports offline functionality, and provides accurate and reliable results. The proposed system addresses these challenges by integrating multiple verification techniques into a unified and efficient framework.

4. PROPOSED SYSTEM

   The proposed system, Attendify, is a cloud-based multi-factor attendance management solution designed to ensure secure, accurate, and reliable attendance recording. The system addresses the limitations of existing methods by integrating multiple verification techniques, including QR code or One-Time Password (OTP) validation, facial recognition, and Global Positioning System (GPS)- based geofencing.

   This layered approach ensures that attendance is marked only when all verification criteria are successfully satisfied.

   The system operates through three main user roles: student, teacher, and administrator, each with distinct responsibilities. The teacher initiates an attendance session, during which a dynamically generated QR code and OTP are displayed. Students are required to scan the QR code or enter the OTP within a limited time frame, ensuring real-time participation. Following this, the system captures a live image of the student and performs facial recognition using advanced computer vision algorithms to verify identity. Finally, the system validates the student's location using GPS to ensure that attendance is marked only within the designated area.

   To enhance reliability, the system incorporates an offline mode that allows attendance data to be securely stored on the device when internet connectivity is unavailable. Once the connection is restored, the data is automatically synchronized with the central server. Additionally, anti-spoofing mechanisms are integrated to detect and prevent fraudulent attempts such as GPS manipulation or use of VPNs.

   The architecture of the system is designed to be scalable and modular, enabling seamless integration with various platforms and environments. A centralized backend manages authentication, data storage, and verification processes, while dedicated applications are provided for students, teachers, and administrators. 4.1 Multi-Factor Verification Mechanism

   The core strength of the proposed system lies in its multi-factor verification mechanism, which combines three independent layers of validation to ensure authenticity. The first layer involves scanning a dynamically generated QR code or entering a time-sensitive OTP. This prevents unauthorized access and ensures that only users present during the session can attempt attendance marking.

   The second layer involves facial recognition, where the system captures a live image and compares it with the registered facial data using machine learning algorithms such as Convolutional Neural Networks (CNN). This step ensures identity verification and reduces the possibility of impersonation.

   The third layer involves GPS-based geofencing, which verifies whether the user is physically present within a predefined location. This prevents remote attendance marking and adds an additional level of security. The combination of these three layers significantly enhances the reliability and robustness of the system.

   1. QR Code and OTP Generation

      The system generates a unique QR code and OTP for each attendance session initiated by the teacher. These codes are dynamically refreshed at regular intervals, typically every 60 to 90 seconds, to prevent sharing and reuse. The QR code can be scanned using the mobile application, while the OTP can be manually entered as an alternative method.

      This mechanism ensures flexibility while maintaining security, as the time-bound nature of the codes restricts unauthorized usage. It also supports environments where camera access may be limited, allowing OTP-based attendance marking as a fallback option.

   2. Facial Recognition System

      Facial recognition is implemented as a key component of identity verification. The system uses computer vision techniques such as Haar Cascade for face detection and Convolutional Neural Networks (CNN) for feature extraction and classification. This combination enables accurate recognition even under varying conditions such as changes in lighting or partial facial occlusion.

      The system captures a live image during attendance marking and compares it with the stored facial data collected during user registration. This ensures that only the registered individual can successfully complete the verification process. Additional measures are incorporated to prevent spoofing attempts using static images or pre-recorded videos.

   3. GPS-Based Geofencing

      To ensure physical presence, the system incorporates GPS-based geofencing. A predefined geographic boundary is set for each attendance session, and users must be within this boundary to successfully mark attendance. The system retrieves real-time location data from the users device and validates it against the defined geofence.

      Anti-spoofing mechanisms are also implemented to detect suspicious location patterns and prevent manipulation using external tools such as VPNs or GPS spoofing applications. This ensures that attendance is marked only by users who are physically present at the designated location.

   4. Offline Mode and Data Synchronization

   One of the key features of the proposed system is its ability to function in offline environments. When internet connectivity is unavailable, attendance data is securely stored on the users device using local storage mechanisms. The data is encrypted to ensure security and prevent unauthorized access.

   Once the device regains connectivity, the stored data is automatically synchronized with the central server without requiring manual intervention. This ensures data consistency and prevents loss of attendance records, making the system reliable even in low- connectivity scenarios.

   Figure 1: Proposed Multi-Factor Attendance System

5. SYSTEM ARCHITECTURE

   The system architecture of Attendify is designed to provide a scalable, secure, and modular framework for managing attendance across multiple platforms. The system follows a centralized architecture in which all client applications communicate with a backend server that acts as the core processing unit. This design ensures consistency, security, and efficient data management.

   The architecture consists of three primary client applications: the student mobile application, the teacher mobile application, and the admin web application. Each application is designed to perform specific functions based on user roles while interacting with the

   backend through secure APIs. The backend system handles authentication, attendance processing, verification mechanisms, and data storage.

   The student mobile application is responsible for enabling students to mark attendance using the multifactor verification process. It integrates device components such as the camera for QR code scanning and facial capture, as well as GPS for location tracking. The application also supports offline functionality by storing attendance data locally and synchronizing it with the server when connectivity is restored.

   The teacher mobile application is used to initiate and manage attendance sessions. It generates dynamic QR codes and OTPs, monitors real-time attendance status, and allows teachers to supervise and flag suspicious entries. This application ensures that attendance sessions are conducted securely and efficiently.

   The admin web application provides centralized control and management of the system. It allows administrators to manage users, assign roles, monitor attendance records, and generate reports. It also supports audit mechanisms and manual attendance adjustments when required.

   The backend server acts as the core component of the system, handling all critical operations. It manages user authentication using secure protocols, processes attendance requests, performs facial recognition and geolocation validation, and stores data in the database. The backend also ensures role-based access control and maintains system integrity.

   The database layer is responsible for storing user details, attendance records, facial data, and session information. Both relational and non-relational databases can be used to handle structured and unstructured data efficiently. The system ensures data security through encryption and secure communication protocols.

   The overall architecture is designed to be modular, allowing independent development and integration of different components. This improves maintainability and enables future enhancements without affecting the entire system.

   Figure 2: System Architecture of Attendify

6. MODULES DESCRIPTION

   The proposed system is divided into multiple functional modules to ensure clear separation of responsibilities, efficient operation, and ease of maintenance. Each module is designed to handle specific tasks within the system while interacting with the centralized backend. The major modules of the system include the Student Module, Teacher Module, and Admin Module.

   1. Student Module

      The Student Module is responsible for enabling students to securely mark their attendance using the multifactor verification process. The module provides a user-friendly interface through a mobile application, allowing students to log in and access their assigned subjects.

      To mark attendance, the student is required to scan a dynamically generated QR code or enter a One-Time Password (OTP) displayed on the teachers device. The system ensures that QR codes cannot be uploaded from the gallery, enforcing real-time scanning through the device camera. This prevents unauthorized usage and sharing of QR codes.

      Once the QR/OTP verification is completed, the system captures a live image of the student using the device camera. This image is processed using facial recognition algorithms and compared with the stored facial data collected during registration. This step ensures that only the authorized student can mark attendance.

      After successful identity verification, the system validates the students location using GPS. The attendance is marked only if the student is within the predefined geofence area. The module also supports offline functionality, where attendance data is securely stored on the device and automatically synchronized with the server once connectivity is restored. Additionally, students can view their attendance history and status through the application.

   2. Teacher Module

      The Teacher Module is designed to manage attendance sessions and monitor student participation in real time. Through the mobile application, teachers can log in and access their assigned subjects and sections.

      The teacher initiates an attendance session for a selected class, upon which the system generates a dynamic QR code and OTP. These codes are refreshed at regular intervals, typically every 60 seconds, to prevent misuse and ensure real-time attendance marking.

      During the session, the teacher can monitor live attendance updates and view the verification status of each student. The module provides the ability to identify and flag suspicious or potentially fraudulent attendance entries. After the session is completed, the teacher can end the session and review a summary of attendance records.

      The Teacher Module ensures controlled and supervised attendance marking while maintaining the integrity of the system.

   3. Admin Module

      The Admin Module provides centralized control and management of the entire system. It is typically implemented as a web-based application that allows administrators to manage users, academic structures, and attendance records.

      Administrators can create, update, and deactivate student and teacher accounts. They can also manage sections, assign subjects, and allocate teachers to specific classes. During the registration process, the admin is responsible for storing the facial data of students, which is later used for identity verification.

      The module also provides access to detailed attendance reports, including student-wise, subject-wise, and date-wise analytics. Administrators can review flagged attendance entries and take appropriate actions when necessary. Additionally, the system allows manual attendance marking for special cases such as official events, with proper audit tracking.

      The Admin Module ensures governance, transparency, and accountability within the system.

   4. Backend Module

      The Backend Module acts as the central core of the system, handling all processing, validation, and data maagement operations. It provides secure APIs for communication between client applications and ensures proper authentication and authorization using mechanisms such as JSON Web Tokens (JWT).

      The backend manages attendance sessions, generates QR codes and OTPs, performs facial recognition and GPS validation, and stores attendance data in the database. It also handles offline data synchronization and ensures consistency across the system.

      This modular architecture allows independent development and scalability, making the system efficient and adaptable for future enhancements.

7. METHODOLOGY

   The methodology of the proposed system describes the step-by-step process involved in securely marking attendance using a multi- factor verification approach. The system is designed to ensure that attendance is recorded only when all verification stages are successfully completed. This structured workflow enhances accuracy, prevents fraudulent activities, and ensures reliability in different operational conditions.

   The process begins with the teacher initiating an attendance session through the teacher application. Upon starting the session, the system generates a time-bound QR code and One-Time Password (OTP), which are displayed to students. These codes are dynamically refreshed at regular intervals to prevent sharing and unauthorized access.

   Students then access the system through the student application and scan the QR code using their device camera or manually enter the OTP. This step ensures that the student is actively present during the session. The system validates the entered or scanned data before proceeding to the next stage.

   Following successful QR/OTP verification, the system captures a live image of the student using the device camera. The captured image is processed using facial recognition algorithms and compared with the pre-registered facial data stored in the system. This step verifies the identity of the student and ensures that the attendance is not marked by another individual.

   After identity verification, the system performs location validation using Global Positioning System (GPS) data. The students real- time location is checked against a predefined geofence corresponding to the class location. Attendance is marked only if the student is within the allowed geographic boundary.

   Once all verification steps are successfully completed, the attendance record is generated and stored in the system. In cases where internet connectivity is unavailable, the data is securely stored on the device and automatically synchronized with the server once connectivity is restored. This ensures uninterrupted functionality and prevents data loss.

   The system also includes monitoring and validation mechanisms, allowing teachers to view attendance status in real time and identify any suspicious activities. This complete workflow ensures a secure, efficient, and reliable attendance management process.

   Let V$ be the final validation. V = Q ^ F ^ G, where: Q = Dynamic QR/OTP Match

   F = Face Embedding Distance $<$ Threshold

   G = GPS Coordinate within Radius $R$

   Figure 3: Workflow of Attendance Marking Process

8. TECHNOLOGY STACK

   The proposed system utilizes a combination of modern technologies to ensure efficient performance, scalability, and security. The technology stack is divided into components based on their functionality within the system.

   The frontend of the system is developed using technologies such as HTML5, CSS3, and JavaScript to create an interactive and user- friendly interface. For mobile application development, Android-based technologies are used to enable features such as camera access for QR code scanning, facial image capture, and GPS-based location tracking. Frameworks such as React.js can be used to develop responsive web interfaces for administrative dashboards.

   The backend of the system is implemented using Node.js with Express.js, which provides a scalable environment for handling API requests and managing system operations. The backend acts as the core of the system, handling authentication, attendance processing, and communication between different modules.

   For facial recognition and image processing, Python is used along with libraries such as OpenCV. Machine learning models such as Convolutional Neural Networks (CNN) are used to ensure accurate face detection and verification under varying conditions.

   MongoDB is used as the primary database for storing unstructured data such as attendance logs and session data. Additionally, relational databases such as PostgreSQL can be used for structured data management, including user records and academic information.

   For real-time communication and updates, technologies such as WebSockets or Socket.IO can be used to enable instant synchronization of attendance data during active sessions. Authentication and security are implemented using JSON Web Tokens (JWT), along with HTTPS protocols to ensure secure data transmission.

   The system integrates geolocation services using the HTML5 Geolocation API to implement GPS-based validation. Mapping services such as Google Maps or OpenStreetMap can be used to support geofencing functionality.

   QR code generation and OTP mechanisms are implemented using libraries such as QRCode and Speakeasy, ensuring secure and time-bound access to attendance sessions.

   To support offline functionality, technologies such as Service Workers and IndexedDB are used to store data locally on the device. The Background Sync API ensures that data is automatically synchronized with the server once connectivity is restored.

   For deployment and scalability, cloud platforms such as Amazon Web Services (AWS), Google Cloud, or Microsoft Azure can be used. Containerization tools such as Docker can also be used to simplify deployment and ensure consistency across different environments.

   Table 1: Comparison of Attendance Systems

   System Face Recognition QR Authentication GPS Verification Offline Support

   Traditional Attendance	No	No	No	No
   QR Based System	No	Yes	No	No
   Face Recognition System	Yes	No	No	No
   Attendify (Proposed)	Yes	Yes	Yes	Yes

9. Advantages and Limitations

   The proposed system offers several advantages over traditional and existing attendance systems by incorporating multiple verification mechanisms and modern technologies. However, like any system, it also has certain limitations that can be addressed in future improvements.

   1. Advantages

      The primary advantage of the proposed system is its multi-factor verification approach, which significantly enhances the security and reliability of attendance marking. By combining QR code or OneTime Password (OTP), facial recognition, and GPS-based geofencing, the system effectively eliminates proxy attendance and ensures that only authorized users can mark their presence.

      The system is highly accurate due to the use of machine learning-based facial recognition techniques, which can identify users even under varying conditions such as changes in lighting or partial facial occlusion. The integration of time-bound QR codes and OTPs further prevents unauthorized access and sharing.

      Another major advantage is the inclusion of offline functionality. The system allows attendance data to be stored locally when internet connectivity is unavailable and automatically synchronizes it once the connection is restored. This ensures uninterrupted operation and data reliability.

      The system is also scalable and adaptable, making it suitabl for various environments such as educational institutions, corporate training programs, and large-scale events. Additionally, the user-friendly interface of the mobile applications and admin dashboard enhances usability for students, teachers, and administrators.

      Real-time monitoring and analytical reporting features provide valuable insights into attendance patterns, enabling better decision- making and management.

   2. Limitations

      Despite its advantages, the proposed system has certain limitations. The performance of the facial recognition module may be affected by poor lighting conditions, low-quality cameras, or extreme facial obstructions. Although advanced algorithms are used, accuracy may vary depending on environmental factors.

      The system relies on GPS for location verification, which may sometimes produce inaccurate results due to signal limitations, especially in indoor environments or densely populated areas. Additionally, although anti-spoofing measures are implemented, completely eliminating GPS manipulation remains a challenge.

      Another limitation is the dependency on device hardware capabilities. The system requires access to camera and GPS features, which may not perform consistently across all devices. This could impact user experience in some cases.

      Furthermore, initial setup, including user registration and facial data collection, may require additional effort and administrative support. The system also requires proper data security measures to ensure privacy and protection of sensitive user information.

10. FUTURE SCOPE

    While the Attendify system demonstrates strong performance in controlled classroom environments, several avenues exist for further research and development to enhance its capabilities and broaden its applicability.

    1. Advanced Liveness Detection

       The current facial recognition module relies on deep learning-based embeddings for identity verification. Future improvements can include dedicated liveness detection techniques such as 3D depth analysis and blink detection to distinguish between a live face and spoofing attempts using photographs or screen replays. This enhancement will significantly improve resistance to facial spoofing attacks.

    2. Cross-Platform Support

       The current implementation primarily targets Android devices. Extending support to iOS platforms will broaden the usability of the system, especially in institutions with diverse device ecosystems. Crossplatform frameworks such as Flutter or React Native can be explored to maintain a unified and efficient codebase.

    3. Cloud-Based Microservices Architecture

       The current backend architecture is designed for single-institution deployment. For large-scale adoption, the system can be migrated to a cloud-based microservices architecture using containerization technologies such as Docker and orchestration tools like Kubernetes. This approach enables horizontal scaling, improved fault tolerance, and independent deployment of system components.

    4. Integration with Learning Management Systems

       The system can be integrated with Learning Management Systems (LMS) such as Moodle, Google Classroom, or institutional Enterprise Resource Planning (ERP) platforms. This integration would allow seamless synchronization of attendance data with academic records, improving administrative efficiency and reducing manual effort.

    5. AI-Driven Attendance Analytics

       Future enhancements can include the use of machine learning algorithms to analyze attendance patterns and predict student performance. Techniques such as clustering and anomaly detection can be used to identify irregular attendance behavior, enabling early intervention and better academic monitoring.

    6. Multi-Campus and Multi-Institution Deployment

       The system can be extended to support multi-tenant architecture, allowing a single deployment to serve multiple campuses or institutions. Each entity can have isolated data, independent administrative control, and customizable configurations. This enhancement would enable the system to function as a scalable Software-as-a-Service (SaaS) solution.

11. CONCLUSION

    The proposed system, Attendify, presents a comprehensive and secure solution to the challenges associated with traditional attendance management methods. By integrating multiple verification techniques such as QR code or One-Time Password (OTP), facial recognition, and Global Positioning System (GPS)-based geofencing, the system effectively eliminates proxy attendance and ensures high accuracy in recording attendance.

    The implementation of a multi-factor authentication approach significantly enhances the reliability and security of the system compared to existing single-method solutions. The inclusion of time-bound QR codes and OTPs prevents unauthorized access, while facial recognition ensures identity verification. Additionally, GPS-based validation confirms the physical presence of users, further strengthening the authenticity of attendance records.

    One of the key strengths of the system is its ability to function in environments with limited or no internet connectivity through the implementation of offline data storage and automatic synchronization. This feature ensures uninterrupted operation and makes the system practical for real-world deployment.

    The system also provides a user-friendly interface for students, teachers, and administrators, along with real-time monitoring and analytical reporting capabilities. These features contribute to improved efficiency, better decision-making, and enhanced overall user experience.

    In conclusion, the proposed system successfully addresses the limitations of existing attendance systems by providing a secure, efficient, and scalable solution. The integration of modern technologies and multilayered verification mechanisms makes it suitable for deployment in educational institutions, corporate environments, and large-scale applications. Future enhancements can further improve the systems capabilities and expand its applicability across diverse domains.

12. ACKNOWLEDGEMENT

    The authors would like to express their sincere gratitude to their internal guide, Professor Kamakshiah

    Kolli, for continuous guidance, support, and valuable suggestions throughout the development of this research work. The authors also extend their thanks to the Department of Computer Science and Engineering, Geethanjali College of Engineering and Technology, for providing the necessary resources and a conducive environment for completing this project successfully. The authors are also grateful to all faculty members and peers who contributed directly or indirectly to the successful completion of this work.

13. REFERENCES

1. Singh R., Kumar A., Smart Attendance System Using Face Recognition, Traitement du Signal, 2023, 40 (2), 345356.

2. Hassan M., El-Sayed A., Intelligent Attendance Monitoring System Using Deep Learning Techniques, Journal of Engineering Science and Information Technology, 2025, 12 (1), 88104.

3. Zhang K., Zhang Z., Li Z., Qiao Y., Joint Face Detection and Alignment Using Multi-task Cascaded Convolutional Networks, IEEE Signal Processing Letters, 2016, 23 (10), 14991503.

4. Schroff F., Kalenichenko D., Philbin J., FaceNet: A Unified Embedding for Face Recognition and Clustering, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, 1 (1), 815823.

5. Viola P., Jones M., Rapid Object Detection using a Boosted Cascade of Simple Features, IEEE onference on Computer Vision and Pattern Recognition, 2001, 1 (2), 511518.

6. Ahonen T., Hadid A., Pietikäinen M., Face Description with Local Binary Patterns: Application to Face Recognition, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28

7. (12), 20372041.

8. Dey N., Ashour A., Smart Attendance System Using QR Code and Mobile Application, International Journal of Engineering and Advanced Technology, 2018, 8 (1), 4550.

9. Jain A., Ross A., Prabhakar S., An Introduction to Biometric Recognition, IEEE Transactions on Circuits and Systems for Video Technology, 2004, 14 (1), 420.