Study On Robust Image Watermarking Algorithms Based On Artificial Neural Networks

The rapid development of information technology has greatly enriched people's lives.Meanwhile more and more digital works in the form of multimedia presented on the Internet to facilitate people to communicate and learn,which brought a lot of information security issues.Some piracy issues and copyright disputes are becoming increasingly serious.Digital watermarking is an effective digital product protection and data security maintenance technology,it will hide the copyright information into the relevant digital works in a specific form to prove the creators' ownership of works,and become effective means of intellectual property protection and digital multimedia security.This paper mainly studies the application of artificial neural network in digital image watermarking.Artificial neural network(ANN)is a data processing model inspired by biological neural networks,which is used to model the input data by adjusting the weights between neurons.Artificial neural network has a high degree of non-linear global fitting ability and fault tolerance,and a good adaptive self-learning function.Based on above,two blind watermarking algorithms for digital images are proposed in this paper.One is a spatial domain watermarking algorithm,which e mbeds watermarking information into the local region with complex texture of the image to improve theimperceptibilityofthe watermark on thebasis of characteristic value of gray-level co-occurrence matrix.Besides,the relation model between pixel dots in the original image is constructed by using artificial neural network so that the blind extraction of watermark information can be realized.The other watermarking algorithm is based on the discrete wavelet transform.It researchs image by Multi-resolution Analysis and embeds watermark information into the singular value of low frequency domain of the image to enhance the robustness of the watermark.The experimental results and datas show that these two algorithms have good imperceptibility.The spatial watermark algorithm has good resistance to JPEG2000 compression in middle and high quality,noise and Gauss filtering attacks,but is completely invalid to the median filter and mean filter.By contrast,the wavelet domain algorithm proposed in this paperhas stronger anti-interference ability,and is robust to JPEG2000 compression,noise,filtering and cropping attacks which are tested in this paper.