科研之友
微信
新浪微博
Facebook
分享链接
摘要
Visual cryptographic (VC) schemes have been widely used in secure multimedia systems for data transmission and data storage. It divides a secret image into n random-seemingly share images printed on the transparencies. Superimposing the designed shares will display the recovered secret image which can be recognized by human visual system. It is very convenient to decode the secret since it requires no cryptographic knowledge and computation. However, there is a high chance for dishonest shareholders to present faked shares in the secret reconstruction phase, which would result in a huge damage to the honest shareholders. In this article, a secure approach to verify the cheating shares has been proposed to achieve fair reconstruction of the image secret. It is designed to share a verification image among the original shares of the XOR based VC scheme. It only increases pixel expansion by one to achieve the verification function. Cheating detection ability is attained by pairwise superimposing the shares so that any cheating behavior can be detected by the honest participant. The secret image is recovered and its recovered contrast becomes 1times of the original contrast where m denotes the pixel expansion of the original scheme. The verification image is probabilistically recovered and its recovered contrast is 12(m+1) Compared with traditional verifiable (k,n)-VC schemes against cheating, it overcomes the drawbacks such as requiring additional shares, additional large pixel expansion, or lower contrast. The experimental results show that the visual quality of the recovered secret image is as good as expected. The security analysis and comparative results based on various aspects of VC schemes demonstrate the better efficiency of the proposed approach over existing schemes.
