基于双重变换域的鲁棒水印算法研究

发布时间：2018-07-13 15:24

【摘要】：随着互联网的迅猛发展和日益普及,越来越多的人们开始利用网络发布自己的作品、传递数据、共享信息,多媒体数据的存储、复制与传播变得越来越容易。网络在方便人们交流的同时,也对多媒体作品的版权保护提出了新的挑战。随意篡改、肆意传播数字作品,甚至侵犯版权行为,已经引起人们的高度重视,数字作品的版权问题亟待解决。数字水印技术,作为保护数字作品版权的有效手段,为人们打击盗版,保护多媒体知识产权提供了新思路。鲁棒数字水印技术在数字内容保护的相关领域中有很好的应用价值,如盗版追踪、拷贝控制和版权保护等。但目前鲁棒水印技术方面的研究还不是很充分,还有很多技术方面的难题需要解决。本学位论文将粒子群优化算法应用于数字水印的嵌入算法中,以提高水印算法的鲁棒性。首先,利用小波变换的多分辨分析特性,对二维载体信号做一层小波分解,提取中低频系数,并进行分块处理,对每一块系数进行离散余弦变换。其次,设计自适应量化算法。利用粒子群优化算法,在载体信号的二维空间中动态搜索水印的最优嵌入位置,并取出最优嵌入位置上的系数,根据这些系数的特点,在载体信号的三维空间中动态地确定水印算法的最优量化步长。最后,利用得到的最优嵌入位置与最优量化步长,将水印信息自适应地嵌入到水印载体中,得到含有水印的载体信号。实验结果表明,本学位论文提出的基于双重变换域的鲁棒水印算法与传统水印算法相比,不仅消除了变换过程中可能产生的块效应,而且充分利用粒子群优化算法确定了水印嵌入的最优位置与最优量化步长,实现了自适应地嵌入,提高了水印的不可感知性与鲁棒性。通过一系列的攻击实验与对比实验证明,本学位论文中提出的算法构造的水印系统性能优于传统算法。
[Abstract]:With the rapid development and popularity of the Internet, more and more people begin to use the network to publish their works, transfer data, share information, multimedia data storage, replication and dissemination become more and more easy. Network not only facilitates communication, but also poses new challenges to copyright protection of multimedia works. Arbitrary tampering, wanton dissemination of digital works, and even copyright infringement have attracted great attention, and the copyright problems of digital works need to be solved urgently. Digital watermarking technology, as an effective means to protect copyright of digital works, provides a new way for people to crack down on piracy and protect multimedia intellectual property. Robust digital watermarking technology has good application value in the field of digital content protection, such as piracy tracking, copy control and copyright protection. However, the research on robust watermarking is not enough, and there are still many technical problems to be solved. In this thesis, particle swarm optimization algorithm is applied to the embedding algorithm of digital watermarking to improve the robustness of the watermarking algorithm. Firstly, using the multi-resolution analysis characteristic of wavelet transform, the two-dimensional carrier signal is decomposed by a layer of wavelet, the middle and low frequency coefficients are extracted and processed in blocks, and each coefficient is transformed by discrete cosine transform. Secondly, an adaptive quantization algorithm is designed. The particle swarm optimization algorithm is used to dynamically search the optimal embedding position of the watermark in the two-dimensional space of the carrier signal, and the coefficients on the optimal embedding position are extracted. According to the characteristics of these coefficients, The optimal quantization step size of the watermarking algorithm is dynamically determined in the three dimensional space of the carrier signal. Finally, using the optimal embedding position and the optimal quantization step, the watermark information is adaptively embedded into the watermark carrier, and the carrier signal containing the watermark is obtained. Experimental results show that the proposed robust watermarking algorithm based on dual transform domain not only eliminates the block effect in the transformation process, but also compares with the traditional watermarking algorithm. Moreover, the particle swarm optimization (PSO) algorithm is used to determine the optimal position and the optimal quantization step size of the watermark embedding, which realizes adaptive embedding and improves the inperceptibility and robustness of the watermark. Through a series of attack experiments and contrast experiments, it is proved that the performance of the proposed algorithm is superior to that of the traditional algorithm.
【学位授予单位】：延边大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP309.7

【参考文献】