Research On Digital Watermarking Algorithm For Image Authentication And Certificate Anti-Counterfeiting

At present, as a representative of computer network, the Internet technology is developing rapidly and making the communication and transmission of the multimedia more simple and quick.However, convenience and insecurity are exist, in the process of the transmission, the information of multimedia is tampered and forged easily by attackers, which causes serious consequences because people cannot judge the information whether true or not. For these attacks, the traditional cryptosecurity based on cryptography was ability not equal to its ambition; therefore, a new and effective confidentiality technology-digital watermarking technique is developping and becoming a hot issue in the field of information security.The significance of this thesis is to give the corresponding strategies according to the image suffered malicious tampering and forge using digital watermarking technique. This article is mainly as follows:Firstly, aiming at positioning accuracy and security of existing spatial fragile watermarking algorithms, a new fragile watermarking algorithm based on a chaos of the fragile for image tamper detection and accurate location is presented. The algorithm encrypts watermarking and confirms the position bits of watermarking embedded using Logistic binary sequence Experimental results show that, the algorithm has a higher security and has a capacity of effective detection and accurate positioning to the Localized tamper.Secondly, aiming at the watermarking embedded into its low frequency coefficient be easy to cause a quality drop of images, this paper puts forward a new kind of adaptive watermarking algorithm. This method is based on the multiresolution property of the wavelet transform and the property of Human Visual System. The energy of subbloek and the texture of the carrier image are utilized to determine where the watermark will be embedded. And the embedding intensity will be adaptively determined by the noise visibility funetion.At last, some test results are provided which indicate this method has a good imperceptivity and robustness. Finally, when extracting watermark, the algorithm does not need original image and watermarking, and it realizes blind extraction by independent component analysis technique.