Performance Comparison of Various Particle Swarm Optimizers in DWT-SVD Watermarking for RGB Images

In this paper we evaluate a non-blind watermarking technique for colored digital image based on singular value decomposition (SVD) in the DWT (Discrete Wavelet Transformation) domain under different PSO variants. PSO is incorporated in the scheme to identify the most suited single value scaling factor to achieve high robustness and similarity ratio, existing PSO variants are compared and the best variant is then applied on the scheme. Experimental results are provided along with statistical analysis to exemplify that proposed scheme using best PSO variant is able to absorb various image processing attacks to a great extent and provide copyright protection.