Cryptanalysis on majority-voting based self-recovery watermarking scheme

In 2007, Wang-Chen proposed a majority-voting based self-recovery watermarking scheme, in which the additional authentication watermark of each block was embedded in the less significant bits and used to examine the authenticity of the same block. In this paper we examine the security of the Wang-Chen self-recovery watermarking scheme and show that it is vulnerable to the proposed XOR-equivalent attack. Specifically, given a watermarked image, one can forge the authentication watermark it contains into another image by the proposed XOR-equivalent attack. Experimental results are given to support our conclusions and demonstrate that our attack is successful in forging watermarked image. This implies that it is insecure to detect the validity of image block only by the additive authentication watermark embedded in the same block in self-recovery watermarking techniques.