摘要
本文提出利用小波变换技术的变换域水印方法。灰度级宿主图像划分为四个子带:LL、LH、HH、HL。其中的低频分量LL包含最大信息量实验图像的平均成分。由于图像被划分四分带,应用离散小波变换技术插入大小相等水印至LL子带,并使用XOR运算方法来嵌入水印。经过实验,如宿主图像和水印图像嵌入对比,嵌入前后对比,以及峰值信噪比进行了分析。结果表明该水印算法,具有良好的隐秘性和鲁棒性。
we propose the transformation of wavelet transform domain watermarking method. The gray level host image is divided into four sub bands : LL, LH, HH, HL. The low - frequency component LL contains the average composition of the maximum information experimental image. Because the image is divided into four bands, the discrete wavelet transform technology is applied to insert the same size watermark into the LL subband, and the XOR operation method is used to embed watermark. Results the comparison between the image of host image and watermark image, the comparison between embedding and comparison, and the results of peak signal - to - noise ratio were analyzed. The results show that the watermark algorithm has good secrecy and strong robustness.
作者
梁艳招
LIANG Yanzhao(Shanwei Polytechnic,Guangdong Shanwei 516600,China)
出处
《电声技术》
2018年第5期25-27,共3页
Audio Engineering
基金
汕尾职业技术学院青年课题(SWKT16-006)资助
广东省高职高专云计算与大数据专业委员会教育科研课题(GDYJSKT16-13)资助
关键词
低频分量
离散小波变换
嵌入
low frequency
discrete wavelet transform
embedding