基于纹理特性的能量差调制视频水印算法  

Energy Difference Modulated Video Watermarking Algorithm Based on Texture Feature

作　　者：刘慧超 王志君[1] 梁利平[1] LIU Huichao;WANG Zhijun;LIANG Liping(Institute of Microelectronics,Chinese Academy of Sciences,Beijing 100029,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院微电子研究所,北京100029 [2]中国科学院大学,北京100049

出　　处：《湖南大学学报（自然科学版）》2020年第10期116-123,共8页Journal of Hunan University:Natural Sciences

基　　金：中国科学院知识创新项目(KGCX2-YW-134)。

摘　　要：针对现有视频水印算法复杂度高、鲁棒性差的问题,提出一种基于残差纹理特征的能量差调制视频水印算法.该算法首先利用视频编码过程得到的残差DCT系数进行区域纹理强度建模,然后依据纹理强度对DCT中高频系数的能量进行自适应调制实现水印的嵌入.算法在不引入额外复杂计算过程的条件下,针对不同区域自适应地嵌入不同强度的水印信息,保证视频质量的同时提高水印的鲁棒性.算法以H.264/AVC视频编解码器为测试实例,实验结果表明,本文提出的水印算法在保证视频质量和编码码率的情况下,对重编码、滤波和噪声等攻击具有很好的鲁棒性,水印提取过程满足盲检测应用需求.Aiming at the problems such as high complexity and poor robustness for current video watermarking algorithms,a video watermarking algorithm based on residual texture feature is proposed by adopting an energy difference modulation.Firstly,the residual DCT coefficients obtained from the video coding process are used to model the regional texture intensity,and then the energy of the DCT coefficients on the middle and high frequency are adjusted adaptively according to the texture intensity to realize the watermark embedding.Under the condition that no additional complicated computational process is introduced,the algorithm embeds different intensity watermark information adaptively for different regions to ensure the video quality and improve the watermark robustness.The scheme takes the H.264/AVC video codec as a verification example.The experimental results show that,under the premise of guaranteeing video quality and encoding rate,the watermarking algorithm proposed in this paper has good robustness to resist attacks such as re-encoding,filter and noise.Meanwhile,the watermark extraction process meets the demand of blind detection for practical applications.

关 键 词：纹理特性 离散余弦变换 能量差 数字水印 

分 类 号：TN919.3[电子电信—通信与信息系统]