摘要
提出了一种基于彩色图像色差分量DCT域下的数字图像水印算法。该算法在埋植水印信息时利用了彩色图像两色差分量DCT系数相关值的统计特点，通过修改两色差分量的DCT系数，将水印信号埋植到图像中，而在水印信息检测时，不需要原始图像以及其它额外的数据。经实验测试，提出的算法可以抵抗诸如JPEG压缩、低通滤波、降采样、放大和缩小以及裁剪等大多数图像处理方法的攻击，具有很高的鲁棒性和不可见性。
Abstract; To be practical, color image watermarking method should not require resorting to original image or to any other data sets. The method we present is, to our best knowledge, a new method based on DCT-domains of color channels. As is well known, digital watermarking is used to discourage illicit copying and distribution of copyrighted material. Our method embeds digital watermarking information into DCT-domains of color channels. We observe that the correlation values between the color-channel DCT (Discrete Cosine Transform) coefficients are satisfied through utilizing certain statistical characteristics; by means of these characteristics we can modify the selected sets of DCT coefficients to make the correlation coefficients show such specific distribution that we can embed a binary string into the image. Experimental results, shown in Fig. 4, demonstrate preliminarily that the watermark produced by our method is robust against attacks by most of the image and signal processing techniques and that our method provides very good results in terms of both image transparency and robustness.
出处
《西北工业大学学报》
EI
CAS
CSCD
北大核心
2002年第1期70-74,共5页
Journal of Northwestern Polytechnical University
关键词
版权保护
数字水印
DCT
色差分量
彩色图像
彩色分量
数字图像
Key words: copyright protection, digital watermarking, DCT(Discrete Cosine Transform), color image
