An SVD-based watermarking scheme using SA algorithm and compared to GA based method

摘要

Watermarking image, the process of embedding data into a cover image and extract watermark under different attacks, can be used primarily for copyright protection and other purposes. Two key components in watermarking are perceptual transparency and robustness of the watermark. A tradeoff between these two features exists, which can be introduced as an optimization problem. Simulated annealing (SA) and genetic algorithms (GA) are two stochastic methods currently in wide use for difficult optimization problems. Simulated annealing as optimization algorithm has not been considered in image watermarking and was remain as an open work. So we apply simulated annealing and genetic algorithms on watermarking image using singular value decomposition (SVD) and discuss their empirical comparison. Simulation results show that the SA algorithm method has better results from the other algorithm in some attacks.