ExamVault: Online Examination System with Live Proctoring

International Journal of Computer Techniques–IJCT Volume 12 Issue 6, November 2025 Open Access and Peer Review Journal ISSN 2394-2231 https://ijctjournal.org/ ISSN :2394-2231 http://www.ijctjournal.org Page 15 EXAM VAULT: ONLINE EXAMINTATION SYSTEMS WITH LIVE PROTORING Udhayasankar R, Siddharth AJ, Poovarsan k,Sivamuthu E,Buvana M BACHELORE OF TECHNOLOGY – FINALYEAR DEPARTMENT OF INFORMATION TECHNOLOGY SRI SHAKTHI INSTITUTE OF ENGINEERING AND TECHNOLOGY (AUTONOMOUS) COIMBATORE – 641062 Abstract: Exam Vault: Online Examination System with Live Proctoring is an advanced digital assessment platform designed to streamline the conduct of secure online exams for educational institutions and organizations. The system integrates robust features such as AI-driven proctoring, automated candidate registration, flexible question bank management, and instant result analytics. The live proctoring module uses real-time monitoring and intelligent algorithms to detect and prevent malpractices, ensuring exam integrity. The platform supports multiple question formats and automates processes from scheduling to results generation, reducing administrative workload while enhancing efficiency and security. Exam Vault's intuitive interface provides a seamless experience for stakeholders, scalable exam delivery across geographies, and detailed insights for continuous improvement. By leveraging adaptive technology and enterprise-grade security, Exam Vault promotes transparent, reliable, and accessible online assessments For modern education needs Introduction: In recent years, the growth of digital technology and online education platforms has revolutionized the traditional methods of learning and assessment. The concept of conducting examinations through the internet has gained massive popularity, particularly after the rise of remote learning environments. However, one of the major challenges faced by educational institutions and organizations is ensuring the authenticity, transparency, and fairness of online examinations. Traditional online exams often suffer from security issues such as impersonation, cheating, and use of unauthorized materials, which directly affect the credibility of the evaluation process. The Online Examination System with Live Proctoring is designed to overcome these limitations by integrating artificial intelligence (AI) and real-time video surveillance. The system offers a secure, scalable, and user-friendly platform for conducting exams remotely while maintaining the same level of supervision as physical examination halls. It provides functionalities for student authentication, exam scheduling, question randomization, time tracking, and automatic result evaluation, all within a webbased environment. A key innovation of this system lies in its AI-based proctoring mechanism. During an examination, the candidate’s webcam and microphone are continuously active to monitor facial movements, multiple presences, tab switching, and other suspicious behaviors. The system uses machine learning algorithms and image recognition techniques to detect anomalies in real time, ensuring the integrity of the examination. In case of any irregularities, the system alerts the administrator or automatically flags the session for review. The proposed system not only ensures fairness but also simplifies the entire examination process for institutions. It eliminates the need for physical presence, reduces operational costs, and enables the conduction of large-scale assessments across diverse geographical International Journal of Computer Techniques–IJCT Volume 12 Issue 6, November 2025 Open Access and Peer Review Journal ISSN 2394-2231 https://ijctjournal.org/ ISSN :2394-2231 http://www.ijctjournal.org Page 16 locations. Furthermore, the platform maintains data security through encryption and secure authentication, ensuring that all personal and exam-related information remains protected. In summary, this project introduces a next-generation solution that bridges the gap between convenience and security in online assessments. By combining automation, artificial intelligence, and real-time monitoring, the Online Examination System with Live Proctoring ensures credibility, efficiency, and reliability in modern digital examination environments. Literature Review: The transformation of education into a digital medium has led to the development of various online examination systems that aim to simplify, automate, and digitize the process of conducting assessments. Over the past decade, researchers and developers have focused on improving accessibility, automation, and scalability in online testing platforms. However, maintaining the integrity and security of these examinations has remained a persistent challenge, particularly in remote environments. This section reviews several existing approaches and research studies that have contributed to the evolution of online examination and proctoring technologies. Early studies in online testing systems primarily focused on exam management and automation. Systems developed during the early 2000s emphasized the storage of questions in databases, random question generation, and automatic grading. For instance, multiple-choice–based assessment systems were introduced to reduce manual effort and human error in evaluation. However, these systems lacked security measures to prevent impersonation or cheating, making them unreliable for large-scale academic use. Later, researchers began incorporating authentication mechanisms to ensure that only legitimate users could access exams. Methods such as password-based logins, one-time passwords (OTPs), and biometric authentication (face and fingerprint recognition) were implemented to verify candidates. Though these approaches improved security, they did not address realtime monitoring, which is essential for preventing malpractice during the examination itself. The emergence of AI and machine learning technologies paved the way for intelligent proctoring systems. Modern systems started integrating webcambased surveillance, where a student’s behavior could be monitored through continuous video capture. Studies by various educational technology researchers introduced AI-based face detection, gaze tracking, and head movement analysis to identify abnormal behaviors such as looking away from the screen or the presence of multiple individuals. Some systems also utilized audio analysis through microphones to detect background noises and potential conversations, further strengthening exam supervision. Recent works have explored hybrid proctoring models that combine AI-based automated detection with human supervision. This dual-layer approach ensures higher accuracy, as the AI system flags suspicious activities while human proctors verify them. Cloudbased architectures have also gained attention for offering scalability and ease of deployment, allowing institutions to conduct exams for thousands of users simultaneously. Despite these advancements, many existing systems face limitations such as high resource consumption, inconsistent AI accuracy, and privacy concerns regarding continuous video recording. Some opensource platforms, though efficient, lack real-time fraud detection or integration with institutional databases. Moreover, the effectiveness of AI-based monitoring is influenced by environmental factors like International Journal of Computer Techniques–IJCT Volume 12 Issue 6, November 2025 Open Access and Peer Review Journal ISSN 2394-2231 https://ijctjournal.org/ ISSN :2394-2231 http://www.ijctjournal.org Page 17 lighting conditions, internet connectivity, and camera quality, which can impact the accuracy of anomaly detection. To address these shortcomings, the proposed Online Examination System with Live Proctoring integrates multiple levels of authentication, intelligent behavioral monitoring, and secure data management. The system uses advanced algorithms for real-time face detection, tab-switch detection, and continuous session validation to ensure the authenticity of the examination. Additionally, the architecture is designed to balance performance and security, enabling efficient monitoring without excessive computational overhead. Methodology: The methodology of the Online Examination System with Live Proctoring is designed to create a secure, intelligent, and reliable platform that ensures the authenticity of remote examinations while maintaining user convenience and system efficiency. The system follows the principles of the Software Development Life Cycle (SDLC), consisting of five key stages: requirement analysis, system design, implementation, testing, and deployment. Each phase plays a vital role in ensuring that the platform meets the objectives of reliability, scalability, and security. In the requirement analysis phase, the system’s functional and nonfunctional needs are clearly defined. The primary functional requirements include user registration, authentication, exam scheduling, real-time proctoring, and automated result evaluation. The nonfunctional requirements focus on aspects like data privacy, accessibility, responsiveness, and scalability to handle multiple users simultaneously. This phase also identifies the main actors of the system—administrators, students, and proctors—and defines their roles and permissions. Hardware requirements include a computer or mobile device with a working webcam and microphone, along with stable internet connectivity, while the software requirements consist of a modern browser, Node.js runtime, MongoDB database, and cloud hosting services. After gathering the requirements, the next stage involves system design. The architecture of the system is based on a three-tier structure comprising the presentation layer, application layer, and data layer. The presentation layer is responsible for user interaction and is built using web technologies such as HTML, CSS, JavaScript, and React.js, providing an intuitive and responsive interface for both students and administrators. The application layer, implemented using Node.js and Express.js, handles all business logic, including user authentication, data validation, proctoring controls, and communication with the database. The data layer, powered by MongoDB, securely stores all essential data, such as user credentials, exam questions, logs, and live proctoring evidence. The system design emphasizes modularity, ensuring that individual components can be updated or replaced without affecting the entire application. To enhance security, the system employs JWT (JSON Web Token) authentication for session management and bcrypt hashing for password encryption, protecting sensitive information from unauthorized access. A major innovation of the proposed system lies in its AIbased proctoring mechanism, which acts as the core component ensuring the fairness and integrity of examinations. When a student begins an exam, the system activates the device’s webcam and microphone to monitor the test environment in real time. The AI model, developed using computer vision and deep learning techniques, continuously analyzes the candidate’s face, gaze, and background. It verifies the student’s identity at login using face recognition technology and monitors during the exam for irregularities. The algorithm is trained to detect suspicious behaviors such as multiple faces in the frame, the absence of a user for an extended period, or frequent eye movements that indicate cheating. The system also monitors tab-switching events or attempts International Journal of Computer Techniques–IJCT Volume 12 Issue 6, November 2025 Open Access and Peer Review Journal ISSN 2394-2231 https://ijctjournal.org/ ISSN :2394-2231 http://www.ijctjournal.org Page 18 to open other applications. Any detected violation triggers an automatic flag in the system log, and the incident is reported to the examiner for review. The AI operates seamlessly in the background, minimizing manual supervision while maximizing accuracy and integrity. The implementation phase translates the design into functional components. The front end provides users with a clean interface for login, exam participation, and viewing results. Administrators can access a dashboard that allows them to create exams, manage question banks, assign time limits, and review flagged incidents from the proctoring module. The system’s backend is implemented using RESTful APIs that enable smooth communication between the front end and the database. MongoDB is used for storing structured and unstructured data efficiently, while the use of asynchronous processing in Node.js ensures smooth performance even under heavy load. The application also supports randomized question generation, ensuring that each student receives a unique set of questions, thereby minimizing the chances of collaboration or cheating. Time tracking features ensure that exams automatically submit once the allotted duration expires, preventing students from exceeding their limits. Once the system is developed, testing and evaluation are carried out to ensure stability and reliability. Multiple testing methods are applied, including unit testing to verify individual modules, integration testing to check interactions between modules, and system testing to assess overall functionality. The proctoring system undergoes intensive testing under simulated conditions to validate its accuracy in detecting cheating behaviors. Performance and load testing are also conducted to ensure the platform can handle hundreds of concurrent users without compromising response time. Furthermore, user acceptance testing (UAT) is performed by students and administrators to ensure that the system is easy to navigate and functions as expected. Security testing, including vulnerability assessments and penetration testing, is performed to verify that all user data and video streams are safely handled. Once all errors are resolved, the system moves to deployment. The deployment of the online examination system is done using cloud-based infrastructure, ensuring accessibility from any location. Hosting the application on cloud services such as AWS, Azure, or Google Cloud provides scalability, high availability, and secure data storage. The system architecture supports both web and mobile devices, allowing users to attend exams conveniently. The deployment phase includes configuring the database connections, setting up SSL certificates for secure communication, and optimizing the application for performance. Post-deployment, the system undergoes continuous monitoring and maintenance to ensure stability, address bugs, and implement periodic updates to improve AI accuracy and user experience. The maintenance phase also includes regular data backups, software patching, and monitoring of server performance to prevent downtime during critical exam sessions. Data security and privacy are given top priority in this methodology. All data—such as user details, exam content, and proctoring logs—are encrypted both during transmission and in storage using SSL/TLS protocols and modern encryption algorithms. Access to sensitive data is restricted to authorized personnel through role-based authentication. The proctoring video feeds are processed in real time and are not stored permanently unless flagged for review, ensuring compliance with privacy regulations like GDPR. The system also includes detailed audit trails and logging mechanisms that record every significant action, from login attempts to exam submissions, ensuring transparency and accountability. These features together create a secure and trustworthy examination environment that upholds academic integrity. Overall, the methodology ensures the creation of an intelligent, efficient, and secure online examination International Journal of Computer Techniques–IJCT Volume 12 Issue 6, November 2025 Open Access and Peer Review Journal ISSN 2394-2231 https://ijctjournal.org/ ISSN :2394-2231 http://www.ijctjournal.org Page 19 system that integrates advanced web technologies with artificial intelligence to deliver a next-generation assessment platform. Through real-time monitoring, automated evaluation, and encrypted data management, the system offers a complete digital solution that mirrors the fairness of physical examinations while offering the flexibility of remote accessibility. The systematic development process, combined with rigorous testing and AI-based surveillance, ensures that the Online Examination System with Live Proctoring achieves its objectives of maintaining integrity, enhancing convenience, and setting new standards for secure digital assessments in modern education. Platform Features and Functionality: The Online Examination System with Live Proctoring is a comprehensive and secure platform that integrates digital examination management with artificial intelligence–based monitoring. The system is designed to simplify the process of creating, conducting, and evaluating exams while ensuring fairness, authenticity, and transparency through live proctoring mechanisms. The platform operates through three key user roles: Administrator, Faculty, and Student, each with distinct functionalities that interact seamlessly within a centralized database and secure server environment. Key Features User Authentication and Role Management: The system provides secure login for Admins, Faculty, and Students using encrypted credentials. Multi-level authentication ensures that only authorized users gain access. Additional verification using webcam-based facial recognition or OTP can be integrated for added security. Exam Creation and Question Management: Faculty members can design exams with flexible question types such as multiple-choice, descriptive, and true/false. Questions are stored in a secure database and can be randomized automatically to prevent cheating. Each exam includes timer functionality, question weightage, and automatic scoring parameters. Live Proctoring and AI Surveillance: One of the core features of this system is real-time AIbased proctoring. The platform activates the candidate’s webcam and microphone during the exam to continuously record and analyze behavior. The system detects face absence, multiple faces, background voices, tab switching, and unusual movements using AI-driven image and audio processing. Suspicious activities are instantly flagged and reported to the admin or stored for review. Secure Exam Environment: The application disables unauthorized actions such as copy-paste, screen capture, or opening new tabs. Browser activity is monitored to ensure that candidates remain focused on the exam window. Encrypted communication protocols protect data exchange between client and server, ensuring exam integrity. Result Evaluation and Report Generation: Once an exam is completed, the system automatically evaluates objective questions and allows manual review for subjective answers. Results are generated instantly and stored in the database, accessible to faculty and students through their respective dashboards. Detailed analytics, such as score distribution and performance insights, can be viewed by administrators. Admin Control and Monitoring Dashboard: The administrator has complete control over user registration, exam scheduling, and proctoring supervision. The live dashboard displays ongoing exams, proctoring alerts, and student activity logs in real time. This enables the admin to take immediate action if any rule violations are detected. Scalability and Accessibility: The system is built using scalable web technologies, allowing multiple users to access the platform simultaneously without performance issues. It supports International Journal of Computer Techniques–IJCT Volume 12 Issue 6, November 2025 Open Access and Peer Review Journal ISSN 2394-2231 https://ijctjournal.org/ ISSN :2394-2231 http://www.ijctjournal.org Page 20 multiple devices and browsers, ensuring accessibility across laptops, desktops, and tablets. Functionality Overview The platform follows a client-server architecture. Students log in to the web interface, take exams, and stream webcam data to the central proctoring system. The AI module processes these inputs to identify anomalies, while the backend handles exam logic, scoring, and storage. Faculty and admin modules interact with the same database to create, monitor, and review examinations in real time. Overall, the Online Examination System with Live Proctoring combines automation, artificial intelligence, and cybersecurity principles to deliver a robust digital examination platform that enhances trust and efficiency in modern education. Conclusion and Future: The Online Examination System with Live Proctoring successfully addresses the key challenges faced in conducting remote examinations, such as authentication, cheating prevention, and data security. By integrating artificial intelligence, web-based automation, and real-time monitoring, the system ensures that online assessments maintain the same integrity and credibility as traditional in-person exams. The platform’s design enables smooth interaction between students, faculty, and administrators through a secure and intuitive interface. It provides end-to-end functionalities — from question paper generation to live proctoring, result evaluation, and report generation. The incorporation of AI-driven proctoring techniques such as face detection, tab-switch detection, and audio analysis enables continuous supervision throughout the examination process. This intelligent monitoring significantly reduces the chances of malpractice and ensures transparency. Moreover, the system’s modular architecture enhances scalability and performance, allowing multiple users to take part in exams simultaneously without compromising response time or security. The use of encrypted data communication and secure authentication methods further ensures the confidentiality of user information and examination data. By automating administrative tasks and minimizing human intervention, the platform saves time and resources while maintaining high accuracy in result processing. In essence, the Online Examination System with Live Proctoring bridges the gap between flexibility and fairness in digital education. It not only provides convenience for institutions and learners but also contributes to the growing global need for reliable, technology-driven educational tools. The project demonstrates how AI and software engineering can collaboratively enhance the trustworthiness and accessibility of online assessments in academic and professional environments. While the current system provides a robust and secure solution, there is considerable potential for enhancement through emerging technologies. Future developments may include: Advanced AI Models: Implementing deep learning–based behavioral analysis to improve detection accuracy for cheating, emotional state, and focus level. Voice Recognition and Natural Language Processing (NLP): Integrating speech recognition to identify unauthorized verbal communication during exams and to enable voice-controlled navigation for differently abled users. Browser Lockdown and Screen Activity Capture: Enhancing exam security by restricting access to external websites, screen sharing, or secondary devices during the examination. Cloud-based Scalability: Hosting the platform on distributed cloud servers to support a large number of simultaneous users globally, ensuring uninterrupted performance. International Journal of Computer Techniques–IJCT Volume 12 Issue 6, November 2025 Open Access and Peer Review Journal ISSN 2394-2231 https://ijctjournal.org/ ISSN :2394-2231 http://www.ijctjournal.org Page 21 Blockchain for Data Integrity: Using blockchain technology to store exam data and proctoring logs, providing an immutable and transparent record of examination activities. Mobile Application Integration: Developing Android and iOS applications to make the system more accessible for students using mobile devices. By integrating these advancements, the system can evolve into a next-generation digital examination ecosystem that offers complete automation, intelligent monitoring, and universal accessibility. The project thus holds significant promise for shaping the future of secure online assessments in educational institutions, recruitment processes, and certification programs. Reference: 1. Alotaibi, M., & Alqahtani, A. (2022). AIBased Online Examination Proctoring Systems: A Review of Methods and Challenges. International Journal of Advanced Computer Science and Applications, 13(5), 123–131. https://doi.org/10.14569/IJACSA.2022.01305 16 2. Bawarith, R., Basuhail, A., Fattouh, A., & Gamalel-Din, S. (2017). E-exam system based on facial recognition and detection techniques. Procedia Computer Science, 112, 1839–1848. https://doi.org/10.1016/j.procs.2017.08.246 3. Dutta, S., & Saha, S. (2021). Artificial Intelligence in E-Proctoring: Enhancing Integrity in Online Assessments. IEEE Access, 9, 146321–146333. https://doi.org/10.1109/ACCESS.2021.31178 49 4. Jha, S., & Kaur, P. (2020). A Secure Online Examination System Using Web Technologies. International Journal of Scientific & Engineering Research, 11(9), 1505–1512. 5. Okada, A., Whitelock, D., Holmes, W., & Edwards, C. (2019). E-Assessment with Secure and Trustworthy AI: Strategies for Academic Integrity in Online Learning. British Journal of Educational Technology, 50(6), 2971–2986. https://doi.org/10.1111/bjet.12842 6. Pande, M., & Raut, S. (2021). Design and Development of an Online Examination System with Live Proctoring. International Journal of Computer Applications, 183(22), 10–16. https://doi.org/10.5120/ijca2021921595 7. Patel, A., & Shah, N. (2020). Remote Proctored Examination: A Review on Challenges and Solutions. International Research Journal of Engineering and Technology (IRJET), 7(6), 2300–2304. International Journal of Computer Techniques–IJCT Volume 12 Issue 6, November 2025 Open Access and Peer Review Journal ISSN 2394-2231 https://ijctjournal.org/