Multiple Secret Sharing Using Natural Language Letter Based Visual Cryptography Scheme

Visual cryptography (VC) is a method used for encrypting visual informations in such a way that it can be easily decrypted by human visual system. Existing visual cryptographic schemes focuses only on using transparencies as shares and the shares appear as noise like pixels. These meaningless shares are not user friendly and create transmission risk problems. And schemes generating meaningful shares are limited to single secret sharing. This paper proposes a Letter-based Visual Cryptography Scheme (LVCS), which creates meaningful shares for sharing multiple secrets that can be used for both binary and grayscale images. In this method we use natural language letters for representing pixels of secret images. Here we are generating meaningful text files instead of share images. So here the text files contain meaningful data and an attacker will not recognize them as containing secrets. This method satisfies security conditions since the secret information can be reconstructed by any k shares but with less than k shares reveal nothing.