基于特征点检测的MPEG-2视频盲水印算法  

Watermarking Algorithm Based on Feature Points Detected in MPEG-2 Video

作　　者：李瑶[1] 朱桂斌[1] 暴晋飞[1] 

机构地区：[1]重庆通信学院,重庆400035

出　　处：《微计算机信息》2009年第30期88-90,共3页Control & Automation

基　　金：基金申请人:朱桂斌;项目名称:视频信息隐藏及其应用技术研究;基金颁发部门:重庆市科技委员会(CSTC.2007BB2105)

摘　　要：为了使数字水印技术更加有效地应用于数字媒体知识产权的保护,提出了一种基于特征点检测的MPEG-2视频盲水印算法。首先利用Harris检测器得到视频图像的稳定特征点来判断水印的嵌入位置,然后结合MPEG-2压缩标准,通过调整相应的宏块内8*8分块DCT变换直流分量来嵌入水印。实验结果表明,水印嵌入前后视频内容改变小,信噪比损失很小,在保证不可感知性的前提下,对平移,缩放,帧删除和MPEG-2压缩码率变换攻击都具有很好的鲁棒性。With the fast development of computer and network techniques, digital watermarking is becoming an efficient method for protecting the intellectual property rights of digital products. This paper proposes a novel blind watermarking algorithm based on feature points detected in MPEG-2 video. In this algorithm, feature points of video are found by using Harris Detectors firstly, and then combined with MPEG-2, embedding the watermarking by adjusting the 8＊8 block DCT＇s DC coefficient. The experimental results in dicate that compared with the original video, the watermarked video is changed slightly; the PSNR of watermarked frames loses a little. Our proposed algorithm is easy to perform and can obtain the imperceptibility performance, meanwhile this algorithm has strong robustness towards translation, zoom, frame deleting and MPEG-2 compressing.

关 键 词：视频水印 特征点 检测器 离散余弦变换 直流分量 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]