Extended Capabilities of Feature-Extraction for Digital Image Sharing by Diverse Image Media

In accustomed visual secret sharing algorithm issue like hide secret images in shares that may be printed on transparencies or are encoded and stored in a digital form exist. The shares can be either meaningful images or noise-like pixels, which will definitely arouse suspicion and increase interception threat during transmission of the shares. Therefore, VSS schemes have limitation that it suffers from a transmission risk problem for the secret itself and for the participants who are involved in the VSS scheme. To overcome the problem of transmission risk, we proposed a natural image based VSS scheme (NVSS scheme) which can share secret images via different carrier media to protect the secret and the participants during the transmission phase. We are introducing new technique i. e. k-means clustering for extracting the features of the natural shares and with that there will be an additional permutation and normalization technique on halftone visual cryptography which will replace the random selection of pixel which we were using for generating the noise -like share before the shares moves to pixel swapping. The unaltered natural shares are various and inoffensive, thus to great extent reducing the transmission risk problem. Experimental results indicate that the proposed approach is an excellent solution for solving the transmission risk problem for the VSS schemes