基于独立分量分析的盲视频水印方案

A Blind Video Watermarking Scheme Based on Independent Component Analysis

本文提出一种新的盲视频水印方案 .该方案利用独立分量分析方法根据相邻的视频帧提取相应的运动分量 ,并且把这些运动分量进行小波变换 ,采用邻近特征值平均法 ,将水印嵌入到运动分量中 .运动分量是视频的重要特征 ,压缩的处理对于它的影响较小 ,因此可以提高鲁棒性 .基于帧的运动分量的提取避免了基于块的重复性操作 ,小波域的水印嵌入可以将水印带来的影响分散到整个帧上 ,具有良好的不可见性 .计算机仿真结果表明 ,该方法能够保证很好的视频质量 ,对于常见的视频处理有较好的鲁棒性 ,能够进行水印的盲检测 ,虚警概率小 .

A novel blind video watermarking scheme is proposed.In this scheme,independent component analysis (ICA) is used to extract the motion component between the consecutive frames and the watermark is embedded into the wavelet coefficients of the motion component according to the mean of the neighboring wavelet coefficients.Motion component is important to the video,but it is not obviously affected by compression.So,the robustness is improved by embedding the watermark into the motion components.Frame-based motion extraction avoids the repeat operations of block-based extraction and reduces the complexity.Embedding watermark in wavelet domain disperses the effects caused by watermark to the whole frame.Better invisibility can be achieved.The simulation results show that the proposed scheme can preserve high video quality.The watermark can be extracted blindly and is robust to common video processing with little probability of false positive.