Fuzzy-GA hybridization in M-band wavelets for collusion resilient optimized SS watermarking
Seba Maity, Jaya Sil, Santi P. Maity, Claude Delpha · 2011
An optimized high payload spread spectrum (SS) image watermarking scheme robust against fading-like collusion attack is proposed using genetic algorithms (GA) in multiband (M-band) wavelets. M-band decomposition of host offers advantages of better scale-space tiling and good energy compactness that make watermarking robust against frequency selective fading. On the other hand, GA would determine threshold value for the host coefficients selection along with the respective embedding strengths. Performance against collusion operation is improved in multiple stage detection through cancelation of multiple bit interference (MBI) effect experienced by the individual embedded bit. Fuzzy logic is used to classify decision magnitudes for multiple group combined interference cancelation (MGCIC) in the intermediate stage(s). Simulation results show convergence of GA and validate relative performance gain achieved in this algorithm compared to the existing works.