A Robust Zero-Watermark Copyright Protection Scheme Based on DWT and Image Normalization

Abstract

Abstract. Recently, protecting the copyright of digital media has become an imperative issue due to the growing illegal reproduction and modification of digital media. A large number of digital watermarking algorithms have been proposed to protect the integrity and copyright of images. Traditional watermarking schemes protect image copyright by embedding a watermark in the spatial or frequency domain of an image. However, these methods degrade the quality of the original image in some extend. In recent years, a new approach called zero-watermarking algorithms is introduced. In these methods, the watermark does not require to be embedded into the protected image but is used to generate a verification map which is registered to a trusted authority for further protection. In this paper a robust copyright proving scheme based on discrete wavelet transform is proposed. It uses a normalization procedure to provide robustness against geometric distortions and a cellular automaton for noise robustness. Experimental results on images with different complexity demonstrate that our proposed scheme is robust against common geometric and non geometric attacks including blurring, JPEG compression, noise addition, sharpening, scaling, rotation, and cropping. In addition, our experimental results obtained on images with different complexities showed that our method could outperform the related methods in most cases.

Robust image zero-watermarking based on Complex Zernike moment

“…When encoding quality of the image's ROI gained higher decoding quality than that in the background region, the algorithm possesses higher robustness compared with other watermarking algorithm not targeted at ROI. Shakeri and Jamzad presented a robust copyright proving scheme based on discrete wavelet transform, firstly conducted normalization of the image and then conducted wavelet transformation of the normalized image [8]. Next, cellular automaton is employed to conduct noise infiltration processing of the transformed lowfrequency subgraph.…”