基于MPEG的实时视频水印技术  

MPEG-based Real-time Watermarking for Compressed Video

作　　者：凌贺飞[1] 卢正鼎[1] 邹复好[1] 

机构地区：[1]华中科技大学计算机学院,湖北武汉430074

出　　处：《小型微型计算机系统》2005年第12期2181-2185,共5页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(60502024)资助;信息产业部电子信息产业发展基金项目(2004-2006)资助;科技部科技型中小企业创新基金项目(04C26214201284;2004-2006)资助.

摘　　要：作为数字视频版权保护的重要手段,视频水印技术的研究逐渐成为当前数字水印技术研究方向中的一个热点和难点.与静态图像水印相比,视频水印需要更高的要求,如实时性.已存在的能量差水印(DEW)算法本身具有低计算复杂性优点,但是给出的参数选择的优化算法计算复杂性过高,同时鲁棒性比较差,需要进一步地提高.本文提出两个有效措施来改进算法的性能:一是用能量差和整体能量的比率RD代替能量差D,作为一比特水印信号的嵌入模式;二是嵌入过程的截断索引c的挑选不再要求两个子区域的能量都大于能量差D.结果表明改进的能量差水印算法(IDEW)在水印的视觉失真、容量和鲁棒性方面都比原始能量差算法(DEW)性能更好.As a main method for copyright protection of digital video data, video watermarking has gradually become a focus. However, unlike still image, video watermarking technology must meet the real-time requirement. The DEW algorithm itself has been shown to have relatively low complexity, but the optimization approach for parameters setting is too complex to be practical, and the robustness is still low, so these need to be improved. In this paper, two measures are proposed to improve the algorithm＇s performance. One is using the ratio of energy difference to total energy RD to replace energy difference D as pattern to embed label bits. The other is that the selection of embedding cut-off index c doesn＇t need that both subregions A and B have enough energy. The experimental results show that the improved algorithm performs better on watermarkrs visual quality impact, capacity and robustness than the original DEW algorithm.

关 键 词：视频水印 实时性 差分能量水印 MPEG 离散余弦变换 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]