Near optimum local image watermarking using Krawtchouk moments

A novel methodology that deals with the optimum watermarking of image media by using Krawtchouk moments, in terms of locality, is presented in this paper. The proposed technique uses a simple genetic algorithm as optimization procedure, in order to find the best parameters set (p₁, p₂) defining the local behaviour of the Krawtchouk polynomials. The resulted near optimum parameters are used to construct the corresponding image´s moments where the watermark information is inserted. These parameters are selected regarding the PSNR (Peak Signal to Noise Ratio) of the resulted watermarked image and the NC (Normalized Correlation) between the original and the reconstructed watermark, under several attacking conditions. Appropriate experiments have taken place and the results are promising, in the sense that the selection of the region, where the watermark is inserted can improve the quality of the watermarked image, but most of all it is robust to signal processing attacks.