利用自适应有序抖动块截断编码的鲁棒水印  

Robust watermarking utilizing adaptive order dither block truncation coding

作　　者：谢琨[1] 曾平[1,2] 郑海红[1] 郭涛[1] 

机构地区：[1]西安电子科技大学计算机学院,陕西西安710071 [2]西安石油大学计算机学院,陕西西安710065

出　　处：《西安电子科技大学学报》2015年第3期198-204,共7页Journal of Xidian University

基　　金：国家自然科学基金资助项目(61100156);国家部委基金资助项目(51XXXX702);基本科研业务费资助项目(JB150317;K5051303014;JB140311)

摘　　要：为克服现有压缩域水印算法图像质量或鲁棒性较差的缺点,提出了一种基于自适应有序抖动块截断编码的鲁棒水印算法.首先,利用图像的信息熵和方差自适应地将图像按四叉树分解为平滑块、纹理块和边缘块,再对不同大小的分块进行编码;然后,利用自适应有序抖动块截断编码得到的位平面,通过调整位平面水印嵌入窗口内元素的奇偶和,实现水印嵌入;最后,通过位平面中元素的奇偶和与投票策略来实现水印的盲提取.实验结果表明,与同类算法相比,该算法能够在取得较低压缩率的同时,获得良好的图像质量;并且有较高且灵活的水印嵌入容量;同时能够抵抗各种常见攻击,尤其对椒盐噪声、亮度调整、对比度调整、旋转、剪切攻击有很强的鲁棒性.To overcome the shortcomings of bad image quality or poor robustness of the current methods, a robust watermarking method based on adaptive order dither block truncation coding is proposed. Firstly, the image is adaptively divided into smooth regions, texture regions and edge regions using a quadtree segmentation, according to the information entropy and the variance. After that, these blocks are coded separately. Secondly, we exploit the bit plane obtained from adaptive order dither block truncation coding, and embed the watermark by changing the parity sum of the pixels within a window in the bit plane. Lastly, the watermark is retrieved with the parity sum and the majority voting strategy, not referring to the original image. The experimental results indicate that compared with the similar algorithm, the proposed method can achieve good image quality with a low compression rate, a high watermark rate and watermark rate flexibility. Moreover, it is robust to common attacks, especially to Salt ＆ Pepper noising, brightness adjustment, contrast adjustment, rotating and cropping attacks.

关 键 词：块截断编码 有序抖动 数字水印 图像压缩 

分 类 号：TN911.7[电子电信—通信与信息系统]