ABSTRACT
Assessing and comparing the performance of watermarking algorithms is difficult. The visibility of the watermark is an
important aspect in this process. The ease in which digital data can be manipulated has created a need for techniques that
determine the credibility of digital information. In the recent development of digital multimedia in an entire range of our
everyday life has brought forth two active areas of research, namely, retrieval systems and watermarking technology. The
problem of associating messages to multimedia content can be addressed by a watermarking system which embeds the
associated messages into the multimedia content (also called Works). A drawback of watermarking is that the content will be
distorted during embedding. On the other hand, if we assume that the database is available, the problem can be addressed by a
retrieval system. Although no undesirable distortion is created, searching in large databases is fundamentally difficult (also
known as the dimensionality curse). In the present study a novel framework is presented which strikes a trade off between
watermarking and retrieval systems. The framework avoids the dimensionality curse by introducing small distortions
(watermark) into the multimedia content. From another perspective, the framework improves the watermarking performance,
marked by significant reduction in distortion, by introducing searching ability in the message detection stage. To prove the
concept, we give an algorithm based on the proposed notion of “clustering by watermarking”.
Keywords: Algorithm, clustering, multimedia messages, watermark.