Adaptive Robust Watermarking Algorithm Based on DCT and SVD in Curvelet Domain

Abstract

In order to realize the copyright protection and proof of ownership of the digital multimedia products, this paper proposes an adaptive robust watermarking algorithm based on DCT and SVD in Curvelet domain. Firstly, the color image is converted from RGB space to Lab space and the L component is extracted using Curvelet transform. Then the sixth layer high frequency sub-band of Curvelet transform is divided into nonoverlapping sub-blocks of size 8 × 8 by using DCT and Singular Value Decomposition (SVD) is performed on each sub-block. Finally, the encrypted watermark image is adaptively embedded in the maximum singular value of each sub-block according to the texture complexity based on L component and the energy distribution of the DCT coefficients. The experimental results show that the proposed algorithm has better performance in terms of imperceptibility, and also has strong robustness against different kinds of attacks, such as cropping, median filtering, motion blur, salt & pepper noise and JPEG compression attacks in comparison with other related existing methods.