A novel watermarking for images using neural networks

Abstract

Watermarking, which can be applied to the copyright protection and the integrity of multimedia products, has recently become a very active area of research. This paper proposes a novel watermarking scheme for an image. The image is firstly decomposed by multiwavelet transformation, and then the relation among subblocks in the coarsest level of the multiwavelet domain is learned by a back-propagation neural network, which is trained using coefficients in three subblocks as input vectors and corresponding coefficients in another sub-block as output values. Finally a logo watermark is embedded into the multiwavelet domain by adjusting the relation among these subblocks. Experimental results show that the proposed method is superior to the similar one in the literature.