Author(s)

Nan Lin

Corresponding Author

Nan Lin

Abstract

In view of the low robustness and poor concealment of traditional digital image watermarking algorithms, this paper proposes an improved digital image watermarking algorithm based on discrete wavelet transform and singular value decomposition. It first uses discrete wavelet transform to transform the carrier image into the wavelet domain to obtain multiple different frequency bands; then it uses singular value decomposition to train the coefficients of different frequency bands so that it can embed the watermark into the singular value of the carrier image according to the scale factor. In the watermark embedding and extraction process, in order to get a balanced state of the concealment and robustness, it searches the optimal scale factor according to particle swarm optimization algorithm. Experiments show that compared with similar algorithms, the proposed algorithm has stronger robustness and better concealment.

Keywords

Watermark; Discrete wavelet transform; Particle swarm optimization; Singular value decomposition; Scale factor