摘要
针对传统视频水印算法鲁棒性和安全性较差的问题,提出了一种新的结合双树复小波变换(DT-CWT)和奇异值分解(SVD)的视频半盲水印方法。算法首先从视频中提取I帧,其数量和位置均由用户决定,具有良好的保密性。I帧的亮度分量经DT-CWT变换的系数有较强的鲁棒性、更佳的重构效果、灵活的水印嵌入位置以及更多的方向选择性,SVD分解后的奇异值具有良好的抗几何失真不变性。同时还生成了一个随机密钥,有效提高了视频水印的安全性。实验结果表明,该算法对视频质量的影响较小,并且能有效抵御高斯噪声、椒盐噪声、JPEG压缩、直方图均衡化、帧交换、帧删除及帧平均等多种攻击,具有较强的鲁棒性。
Aimed at some drawbacks of traditional video watermarking algorithms,such as bad robustness and low security,a new video semi-blind watermarking algorithm is proposed based on the combination of the dual-tree complex wavelet transform （DT-CWT） and singular value decomposition （SVD）.The algorithm firstly selects i frames from the video.The number and position are determined by the user,it leads to high confidentiality.The coefficients of the luminance component transformed by DT-CWT have stronger robustness,better reconstruction results,more flexible watermark embedding position and direction selectivity,the singular values are robust to geometric distortion.Meanwhile,a random secret key is produced to enhance watermark security.The simulation results show the algorithm has little influence on the video quality,and effectively resists multiple attacks,such as Gaussian noise,salt and pepper noise,JPEG compression,histogram equalization,frame exchange,frame deletion and frame averaging.In a word,the new algorithm has better robustness.
出处
《实验室研究与探索》
CAS
北大核心
2014年第7期23-28,共6页
Research and Exploration In Laboratory
基金
国家自然科学基金项目(61301126)
重庆市高等学校教学改革研究项目(133009)
重庆市高等教育教学改革研究项目(112025)
重庆市研究生教育教学改革研究项目(yjg132001)
关键词
双树复小波变换
视频水印
奇异值分解
鲁棒性
dual-tree complex wavelet transform
video watermarking
singular value decomposition
robustness
