Enhancing the effectiveness of CAPTCHA using an improved visual cryptography scheme

Abstract

Traditional CAPTCHA systems, designed to distinguish humans from bots, are increasingly ineffective due to advancements in artificial intelligence (AI), particularly deep learning and optical character recognition (OCR) technologies, which enable bots to bypass these systems. This paper proposes a new CAPTCHA authentication method that combines enhanced visual cryptography with traditional techniques to improve security. Visual cryptography divides information into visually distinct shares, reinforcing CAPTCHA’s defenses against automated attacks, especially those using deep learning. This approach not only strengthens security but also improves user experience by adjusting the time required to complete CAPTCHA challenges, addressing usability concerns associated with traditional systems. Overall, the proposed method offers a more secure, efficient, and user friendly solution for online authentication.