Published Paper Details:

Authentication has always been a critical challenge in examination settings, where ensuring the legitimacy of candidates is essential to maintaining the integrity of the assessment process. Traditional manual verification methods, such as paper-based identity checks and physical clearance cards, are highly vulnerable to security threats, including impersonation, fraudulent identity documents, and administrative errors. These weaknesses not only compromise the credibility of examinations but also place an additional burden on exam administrators.To address these challenges, this research proposes an advanced, automated authentication system that leverages facial biometrics and Convolutional Neural Networks (CNNs) to provide a robust and secure solution for student verification during exams. By integrating deep learning models, the system is capable of detecting, analyzing, and recognizing candidates' faces in real-time, ensuring a highly accurate and tamper-proof authentication process.The proposed approach significantly reduces the risks associated with impersonation by using image-processing techniques to match each candidate's face against a pre-registered database, preventing unauthorized individuals from gaining access to the examination hall. Furthermore, the system enhances operational efficiency by automating the verification process, eliminating the need for manual document checks, and reducing the administrative workload. By implementing this biometric authentication system, institutions can reinforce examination security, enhance transparency, and ensure a fair evaluation process. The integration of deep learning and facial recognition technologies represents a significant step toward modernizing exam security protocols, ultimately leading to a more reliable and fraud-resistant examination framework.