An intelligent watermarking method based on particle swarm optimization

Meerwald, Koidl, and Uhl (2009) pointed out that the method proposed in Lin et al. (2008) exists potential insecurity. This paper proposes an intelligent watermarking by invoking particle swarm optimization (PSO) technique in wavelet domain to overcome the revealed insecurity issue, furthermore resolve the conflict between imperceptibility and robustness of watermarking. In the proposed method, PSO is fused with the method proposed in Lin et al. (2008) (denoted SDWCQ) to avoid potentially insecurity in Lin et al. (2008). That is, the method of using the fixed block size in one subband and the permutation is unable to disguise which coefficients make up a block. The attacker can utilize the insecure property and analyze the significant difference between bipolar watermarks in Lin et al. (2008) to detect the embedded blocks, furthermore modify the significant difference, and result in unable to extract the watermark. In this paper, coefficients are randomly selected from different subbands to make up a block to promote the disguise. Performance analysis shows that the proposed algorithm obviously outperforms SDWCQ which does not use PSO.

Research highlights
► This paper proposes an intelligent watermarking by invoking particle swarm optimization (PSO) technique in wavelet domain.
► The coefficients are randomly selected from different subbands and are grouped into a block to promote the perceptual transparency and robustness.
► Performance analysis shows that the proposed algorithm obviously outperforms the method proposed in (Lin et al., 2008).