摘要
水印的鲁棒性受背景、纹理、边缘等条件影响较大。为此,提出基于离散余弦变换(DCT)域和纹理复杂度的图像水印算法。对水印信息进行预处理,计算每个图像块的方差并将其按大小排序,对各子块进行离散余弦变换,计算变换后每个图像块的方差并将其按大小排序,根据中频系数将经过方差值排序的水印图像块嵌入载体图像中。测试结果表明,该算法比原始基于DCT的水印算法具有更好的鲁棒性和图像可见性。
The robustness of watermark is largely influenced by texture of image background,texture,edge and so on.This paper proposes an image watermark algorithm based on Discrete Cosine Transformation(DCT) domain and texture complexity,including preprocessing the watermark information,and calculating each image block's variance and sort the value,then transferred by DCT.It also calculates each varied image block's variance and sorts the value.It insets the watermark image block which goes through the variance yields sort into the carrier image.Experimental results show that the algorithm compared with the primitive DCT-based transformed watermark algorithm has great robustness and invisibility.
出处
《计算机工程》
CAS
CSCD
北大核心
2011年第18期148-150,共3页
Computer Engineering
基金
陕西省科技攻关计划基金资助项目(200405-G25)
关键词
离散余弦变换
纹理复杂度
水印图像
鲁棒性
Discrete Cosine Transformation(DCT)
texture complexity
watermark image
robustness