Novel Image Watermarking Approach against Noise and RST Attacks

Digital watermarking is used for copyright protection of intellectual properties and hence it is important to protect and recover data from attacks. Existing systems work on and recover the image from single attacks. This paper aims at proposing a technique for retrieval of cover image and watermarked data from dual attacks like rotation and noise attacks. The proposed system takes as input, Color images and then embeds data using alpha blending method. And also tries to recover from dual attacks using median filtering and pseudo Zernike moment. Proposed approach selects LH band from the 4 sub bands image divided into using DWT and for reducing noise from the image SVD is used. Proposed approach is giving better results in terms of PSNR and MSE as compared to other existing systems using QWT. (C) 2020 The Authors. Published by Elsevier B.V.