摘要
为了提高嵌入水印图像的视觉效果和解决自适应调节参数简单的问题,借鉴了离散小波变换下纹理特征清晰度分析评价标准的思想,对文献[1]提出的分块自适应水印算法做出了改进,并提出了一种改进的水印算法。对图像离散小波变换的近似子带进行分块和纹理特征计算处理,对纹理特征计算值进行分类和确定对应分类的嵌入强度,将水印信息嵌入到近似子带分块SVD(奇异值分解)的S矩阵中。实验结果表明,通过纹理清晰度分析出来的自适应效果好于文献[1]中提到的方法,PSNR=70,NC=0.9926,而且水印信息具有良好的鲁棒性和不可见性。
In order to improve the visual effect of the watermarked image and solve the problem that adaptive adjusting parameters are too simple,the evaluation standard idea which in the discrete wavelet transform analysis of texture feature is drawn,then the literature[1] is proposed to make the improvement and an improved watermarkig algorithm is put forword.The idea is dividing the approximate sub-band of discrete wavelet transform image into many blocks and calculating texture feature of blocks,so the features need to be classified and thier embedding strength is distributed,at the same time,the S matrix of the sub-band block will be embedded to some watermark information.Proved by experiments,compared the literature[1],the adaptive effect of proposed algorithm is better and the watermark has good robustness invisibility for PSNR=70,NC=0.9926.
出处
《计算机工程与设计》
CSCD
北大核心
2013年第8期2695-2699,共5页
Computer Engineering and Design
基金
国家自然科学基金项目(60972077)
北京市自然科学基金B类项目(KZ2010009008)
中央高校基本科研业务费专项基金项目(BUPT 2011RC0210)
关键词
离散小波变换
奇异值分解
Arnold置乱处理
纹理清晰度
自适应
discrete wavelet transform
singular value decomposition
arnold scrambling processing
texture resolution
adaptive