An asymmetric public detection watermarking technique

Abstract

The new watermarking technique1 presented in this paper is an example of an asymmetric public detection scheme. The detection process does not need the original picture nor the secret key used in the embedding process. It is the translation, in the watermarking domain, of a public key pair cryptosystem. The key idea is to filter the pseudo-noise sequence before embedding it in the cover-content. Contrary to classical techniques, the heart of the detection algorithm is not a correlation measure but a consistent statistical test hypothesis in spectral analysis. Some signal based considerations show that knowing the public key used in the detection process is no use for pirate who wants to discover the secret key. An example of a copyright protection system for digital content using this technique is briefly described.