AI-Based Fake Certificate Verification System Using QR Code
G Deepadharshini
Abstract
Purpose The growing use of digital certificates has increased concerns about certificate forgery in education, government, and corporate sectors. This study proposes an AI-based certificate verification system integrated with QR codes to enable secure, rapid, and reliable authentication. Each genuine certificate receives a unique QR code linked to verified database records, reducing manual intervention, preventing fraud, minimizing errors, and supporting scalable digital verification. Design/Methodology/Approach The proposed system generates a unique, cryptographically hashed QR code for every certificate at the time of issuance, embedding key credential attributes and a digital signature linked to a secure institutional database. A verification module built on optical character recognition (OCR) and a convolutional neural network (CNN) based image-forgery detection model cross-checks scanned certificates against the database, flags layout or seal tampering, and validates the QR payload in real time. The system was tested on a dataset of genuine and artificially forged certificates to assess detection accuracy, processing speed, and robustness against common forgery techniques such as image splicing and font manipulation. Findings Results indicate that the integrated AI and QR-based approach achieves substantially higher forgery-detection accuracy and verification speed than manual or QR-only methods. The CNN-based forgery detector reliably identified tampered seals, signatures, and altered text fields, while the QR-hash validation instantly rejected certificates with mismatched or absent database records, reducing verification time from days to seconds. Practical Implication Educational institutions and employers can adopt the proposed system to automate credential checks, reduce recruitment and admission fraud, and lower the administrative burden associated with manual verification requests. Originality: The originality lies in integrating Artificial Intelligence, encrypted QR codes, and database validation for automated certificate authentication. Unlike conventional manual or QR-based verification, the proposed framework improves fraud detection, security, efficiency, and accuracy. Using Python, Django, and SQL, it provides a scalable, cost-effective solution for educational, corporate, and government certificate verification.