A Digital Blind Multipurpose Error Correction Coding And HVS-Based Image Watermarking

The paper primarily carries on the systemic research of the algorithm of digital image watermarking which is based on error correction coding.At first, the paper introduces the background and the actuality of digital watermarking, and dissertates the digital watermarking about the basic concept and characteristics, then summarizes the generic model of the digital blind watermark and the typical algorithms. The attack analysis and evaluation of performance of the watermark are discussed.Secondly, after the theory and application of the digital watermark are discussed, the concept of fragile watermark, the robust watermark and multipurpose watermark are proposed and simulated. Through a series of experiments, the characteristics of all above three are validated. The multipurpose watermark has the strongpoint of both the fragile watermark and the robust watermark. It account for that the multipurpose watermark not only can resist some image processing operations, but also being effective in image authentication. So its applications are more abroad. This paper is mainly discussing the multipurpose watermark, and it respectively chooses the gray image and the color image which being embedded the watermarks. Through the data of the simulation, supportive evidence was provided to demonstrate the proposed method being effective in image authentication and resisting image processing attacks.Compare with the fragile watermark or the robust watermark, the multipurpose watermark has the obvious strongpoint. In order to combine the copyright protection with the content authentication commendably in the digital watermark, the multipurpose watermarks embed the fragile watermark and the robust watermark into one image synchronously. The robust watermark embedded can resist image processing attacks, protecting the integrality of the watermark itself, and the fragile watermark embedded can validate the reliability of the image. Once the image is attacked, the fragile watermark can not be extracted, accordingly the image can not be trusted anymore, but it can locate the area which is attacked maliciously. In order to strengthen the security, the watermarking algorithm still utilizes the RSA public-key cryptosystem,the error correction coding and scrambling. The advantage of public-key cryptosystem is that no one can destroy the watermark or get the original unwatermarked image without the owner's private key. In the pretreatment, the error correction coding is more important. This paper adopts two coding mode, one is (7, 4) Hamming code, the other is (15, 7) BCH code. These two coding mode will be introduced and simulated respectively in the paper, and the bit error rate will be analyzed. The data of the simulation can prove that the error correction coding improve the robustness of the watermark.At last, from three aspect of chosen of the watermark information, chosen of watermark transform domain and transform method, and watermark hiding and detecting algorithm, the digital blind watermark technology is studied in detail. In former methods, the watermark is usually the pseudo-random sequence, and the original image is needed when extracting the watermark. There are two defaults with the algorithm. One is that it can only carry little information and have the poor ability of keeping secrecy. And another is that because of the need of the original image, it brings the inconvenience when extracting the watermark. Aiming at the problem, the meaningful blind watermark is generally studied and analyzed. This paper simulates a multipurpose error correction coding based image watermarking.The algorithm chooses two-value images which have special meaning as watermark information. First of all, the algorithm codes, encrypts and scrambles the robust watermark to a disorderly image, then embeds the pretreated robust watermark into the DCT middle frequency coefficients and the authentication watermark into the DCT high frequency coefficients based on the HVS. The experiments prove the method proposed is robust against some image processing operations, such as low- pass filtering, Gaussian noise, JPEG lossy compression and cropping. While in the aspect of authentication, the fragile watermark can not be extracted after signal processing, but it can locate the area which is attacked maliciously.Based on the analysis of the research in existence of the digital blind watermark algorithm in the DCT domain, a colorful multipurpose digital watermarking was put forward. The algorithm transforms the original color image into the YUV color space, embedding the multipurpose watermarks into the Y ponderance, through the inverse transform, and then the watermarked colorful image is obtained. Through the simulation, the algorithm proposed is also good for colorful image.The paper mainly carried on three aspects work of the following:1. The multipurpose watermarking has the advantage both of the robust watermark and the fragile watermark, so this paper simulates these two classical watermark algorithm—fragile watermarking and robust watermarking. The fragile watermarking simulates the embedding algorithm in the spatial domain and the DCT domain respectively. Both two have good performance that the fragile watermark can not be extracted after the attack, and it can locate the area which is attacked maliciously too. While in the robust watermarking, the watermark is embedded into the DCT domain. The simulation shows that the robust algorithm can resist image processing attacks, and be extracted clearly.2. In the multipurpose watermarking proposed, two-value images were chosen to be watermarks, and gray image to be original image. Before the watermarks embedded, the algorithm pretreated the robust watermark such as error correction coding, public-key cryptosystem, and Arnold scrambling to get a disorderly watermark which can disperse concentrative mistakes into the entire image. Afterward, it embedded the watermarks in the DCT domain based on the HVS theoretic. Consequently, the security and the imperceptibility were enhanced.3. Moreover, the algorithm also chooses the valuable colorful image as an original image. The paper analyses the capability of the copyright protection and the content authentication detailed. The ascendants of the multipurpose watermark make it more effective in some especial instance.Through a series of experiment, supportive evidence was provided to demonstrate the proposed method being effective in imperceptibility, image authentication and resisting image processing attacks.