【作者】 贾柳燕；

【导师】 王珂；

【作者基本信息】 吉林大学 ， 信号与信息处理， 2007， 硕士

【摘要】 论文主要对基于提升小波变换下的有意义数字盲水印算法及其应用展开研究,提出了一种提升框架下基于量化补偿的有意义的数字盲水印算法。算法选择彩色图像的YIQ空间中的Y分量嵌入水印,利用人眼视觉特性,采用量化补偿的方法,将置乱变换后的二值水印图像预处理后重复的嵌入到Y分量的提升小波变换系数中去。以提升小波变换为理论依据,利用小波的多分辨率分析特性,在提升小波域内进行水印的嵌入和提取,并且嵌入时采用带有补偿系数的量化理论,不但更好的兼顾了水印不可见性与鲁棒性之间的矛盾,而且实现了水印的盲提取。采用提升小波变换方法有效的避免了运算过程中引入量化误差的缺陷,从而达到精度上的要求,满足图像的精确分解与重构。另外,提升小波变换是JPEG2000的核心技术,在此变换域嵌入水印,可以有效的防止由于JPEG2000有损压缩而造成的水印消除。通过仿真实验以及与其它算法的对比,证明了算法在保证水印不可见性的同时,对常见的图像处理操作如加噪、压缩、剪切等,均具有较好的鲁棒性。更多还原

【Abstract】 The paper primarily carries on the systemic research of the algorithm of digital image watermarking which is based on wavelet transform.At first the background and the actuality of digital watermarking is introduced, dissertate the digital watermarking about the basic concept and characteristics, the generic model of the digital blind watermark, and then summarize the typical algorithms. The attack analysis and evaluation of performance of the watermark are discussed.Then the theory and application of wavelet transform, the principle of image permutation are introduced. Wavelets have found wide applications in image proceeding due to the multi-scale analysis property. In recent years, the study of wavelet transform urged its usage in image coding. The multi-resolution analysis of the wavelet transformation matches with the character of the human vision. It can give sub-bands for different levels and different directions of an image. The second generation wavelet transform (the lifting wavelet) is a new method for constructing wavelets and performing wavelet transform. The application of the second generation wavelet transform in the watermark algorithm is talked over based on the research of the wavelet transform theory.Lifting wavelet is different from the traditional wavelet transform and analyzes the items in the time region (space region). It not only holds the features of the traditional wavelet and also contains traditional wavelet, so that all the traditional wavelet can be formed through the lifting method. The lifting scheme can offers a method of the multi-resolution analysis. Lifting framework technique can effectively accomplish the wavelet decompose and restoration with a series lifting (split, predict and update). Lifting wavelet also has the feature of multi-resolution analysis. Lifting wavelet calculates in the original location. It is simple and understandable and its reverse transformation is easy to achieve. And also, the lifting wavelet is the core arithmetic of the new generation compression technology–JPEG2000. Then embedding the watermark in the lifting wavelet domain can avoid the elimination of the watermark which is arisen by the JPEG2000 compression. So apply the lifting wavelet transform which has the feature of multi-resolution analysis to digital image watermarking, it not only reserves the superiority of multi-resolution based on wavelet, and also improve the computing speed, It is a simpler, more understandable method and have the better robustness.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.For one thing, A blind watermark algorithm is simulated. It chooses the gray image as the watermark which is needed to be embedded. After the Arnold transform and the bit-plane decomposition, obtain eight bit-planes four of which is chose to be embedded. Then using the quantization method, the watermark imformation is embedded into the intermediate frequency of the wavelet coefficient. The experiment shows that this algorithm has the ability to resist the JPEG compression and the cutting. But the ability to resist the filter and the noise is not good.Based on the analysis of the research in existence of the digital blind watermark algorithm in the wavelet domain, a blind watermark technique using lifting wavelet transform was put forward. To increase the robustness, the original color image is transformed into the YIQ color space. Exploiting the excellent properties of the wavelet transform, encrypted watermark is embedded into the Y luminance component of the YIQ in Lifting Wavelet Transform domain using quantization compensation. This method does not require the original image in the proposal detection algorithm. Through Arnold transformation, the watermarking image was hashed to eliminate the spatial correlation among pixels. The ability to withstand attacks was enhanced. The security of watermarking was enhanced too.The paper mainly carried on three aspects work of the following:(1) Embed and extract the watermark in the lifting wavelet transform domain and choose the colorful image which has the practicality as the watermark to be embedded. The paper analyzes the superiority which is between the lifting wavelet and the traditional wavelet in detail. So the watermark will have the better performance when handled in the lifting wavelet.(2) Put forward using the two-value image which has special meaning as Watermark information, and pretreatment the watermark with image permutation before it be embedded. Make it become disorderly information and then embed the watermark, and in this way the watermark’s security and imperceptibility was strengthened.(3) Use the quantization compensation. Although the blind watermark is carried out with the general quantization, the robustness will be reduced. We employ the method of the quantization compensation to amend the default of the quantization so as to improve the robustness.(4) For strengthening the steadiness of the watermark algorithm, the paper put forward to the thought of the watermark which was repeated embedding. After the original image was disassembled with the lifting wavelet, according to the HVS, the right frequency bands were selected where the watermark image was repeated embedding. So the algorithm not only ensures the watermark’s imperceptibility, but also tremendous strengthen the watermark’s robustness.Plentiful emulate experiments prove the algorithm assures the watermark’s imperceptibility and robustness of the algorithm to familiar image process such as noise contamination, compression, cropping, zoom, etc. Finally, review entire work, and make the expectation of the direction for the future research of watermarking.更多还原