一种基于视觉掩盖的数字图像水印技术  

An Image Watermarking Technology Based on Visual Concealment

作　　者：曹岭[1] 杨小帆[1] 柏森[2] 

机构地区：[1]重庆大学计算机学院,重庆400030 [2]重庆通信学院研究生管理大队,重庆400035

出　　处：《重庆大学学报（自然科学版）》2005年第6期64-67,共4页Journal of Chongqing University

基　　金：受到重庆市应用基础研究基金项目资助(8028)

摘　　要：随着计算机通信技术的迅速发展,数字水印技术在数字产品的版权保护领域发挥着越来越重要的作用,数字水印技术的研究也成为了一个重要的研究领域.笔者提出了一种基于原图像的基本特征的视觉掩盖水印嵌入及提取检测算法.将原图像进行8×8的分块,再将分块进行DCT变换.对每分块的DCT系数进行图像特征量的计算,对图像变化剧烈的地方加重水印强度,反之减轻水印的加入强度.最后根据HVS空域掩盖特性,对加入水印后的中间图像进行自适应掩盖,得到最终的水印图像.实验表明该算法具有良好的水印不可见性和鲁棒性,尤其是本算法对抗JPEG压缩攻击有很好的效果.A new image watermarking scheme is proposed, which is based on the visual concealment technique. This scheme consists of three stages: at the first stage, the original image is divided into 8×8 blocks and then the DCT transform is conducted on each block; at the second stage, the features of the original image are extracted by considering the distribution of the obtained DCT coefficients; while at the third stage, the watermark is embedded into the original image with the watermark strength adaptive to the textures. Experimental results show that this method is robust against the common geometrical attacks and signal processing attacks.

关 键 词：数字水印算法 人类视觉系统 DCT变化 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]