Affiliations: [a] Department of Computer Engineering, Sharif University of Technology, Tehran, Iran | [b] Department of Computer Engineering, University of Guilan, Rasht, Iran
Correspondence:
[*]
Corresponding author: Mohammadreza Tahzibi, Department of Computer Engineering Sharif University of Technology Azadi Ave., Tehran, Iran. E-mail:tahzibi@ce.sharif.edu
Abstract: The growth of data communication
networks has made digital watermarking an important issue for copyright and
content protection. Achieving high level of robustness and good transparency
are the main objectives of developing every digital watermarking algorithm.
From among transform domains as the basis of watermark hiding place,
Discrete Cosine Transform (DCT), Discrete Wavelet Transform (DWT) and
Singular Value Decomposition (SVD) are the most commonly used transforms in
literature. In this paper we propose a new hybrid DWT-SVD domain
watermarking scheme taking into account the energy content of every chosen
block of the selected DWT sub-band coefficients. Before embedding, we append
a special class of forward error correcting codes into watermark sequence to
improve robustness against all kinds of attacks. In our proposed method,
watermark embedding is performed by manipulating a new
defined metric based on singular values of each block and watermark
extraction is done blindly i.e., without using the original cover image.
Experimental results show a good level of imperceptibility and better
resistance against different kinds of attacks specially JPEG compression.
Keywords: Digital watermarking, discrete wavelet transform (DWT), energy, singular value decomposition (SVD)
