Improved watermarking scheme based on best color channel selection using discrete slantlet transform

Abstract

Digital watermarking is a process to embed the secret information into digital data for verifying identity of the owners by making assertion about the data and image authentication applications that provide security to watermark, W which is converted to a sequence of random binary R of size n adopted to encrypt the watermark. The adaptation process uses a pseudo-random number generator to determine the pixel to be used on a given key. The digital watermarking is created as a method to solve this kind of problems. There are two issues which are embedded watermark image in the host image without causing any kind of degradation, achieve and improve both imperceptibility and robustness of watermarked image before and after attacks. In this thesis, The RGB colour image watermarking is proposed using by Discrete Slantlet Transform (DST) to generate higher degree of robustness and imperceptibility of watermarked image. After applying 2-level DST on the host image to divided Red, Green and Blue select the best channel to embedding. The experimental results show that the proposed approach provides extra imperceptibility, robustness and security against JPEG compression and different noises attacks compared to the previous methods. The robustness of the proposed image is evaluated by calculating the Normalized Cross Correlation (NCC) value of watermarked before and after the image process. After applying the proposed approach the results proved that the way The Peak Signal-to-Noise Ratio (PSNR) and NCC values were greater than 30 db and 0.6, respectively