Efficient Cognitive Visual Cryptography for Secure Image Transmission

Abstract

Visual Cryptography (VC) is a technique used to protect images by dividing them into multiple meaningless shares, where the original image becomes visible only when the correct shares are combined. Traditional Cognitive Visual Cryptography (CVC) approaches mainly depend on stacking these shares for reconstruction, but they often result in larger image sizes, reduced visual quality, and limited practical usability. To overcome these challenges, this work introduces an enhanced CVC framework that makes effective use of human visual perception during the reconstruction process. Instead of relying on complex computations, the human eye itself plays a key role in interpreting the hidden image, making the recovery process simple and intuitive. The proposed method applies adaptive encoding and decoding based on image characteristics such as contrast and complexity, ensuring that important visual details are preserved. Perceptual modelling further improves robustness against noise and lighting variations. As a result, the reconstructed images exhibit better clarity and reliability. In addition, shifting much of the decoding responsibility to human perception reduces computational effort and energy consumption. The proposed approach therefore offers a practical and user-friendly solution for secure image sharing, with potential applications in areas such as real-time authentication and secure digital document transmission.