摘要
针对一些用于标识和篡改定位的双水印在抗几何攻击方面表现不好的问题,提出了一种基于Contourlet变换的彩色图像双水印算法。该算法首先在红绿蓝(red green blue,RGB)颜色空间,对G分量进行Contourlet变换和QR分解后嵌入鲁棒水印;其次,提取R、G、B三个通道的特征信息,构成可对该图像进行篡改定位的脆弱水印,在B分量上嵌入该脆弱水印,得到含双水印的彩色图像;最后,采用盲提取水印算法提取脆弱水印和鲁棒水印。选取不同刑侦图像进行实验,实验结果表明,脆弱水印能够准确检测篡改位置;鲁棒水印在抵抗剪切、旋转、缩放、随机打马赛克、噪声等几何攻击方面鲁棒性较好。
For the problem that dual watermarking algorithms for identification and tampering are not well in anti-geometric attacks,a color image dual watermarking algorithm based on Contourlet transform is proposed.After Contourlet transform and QR decomposition,the robust watermark is embedded in the G component of the RGB color space.A fragile watermark with the image feature information and used for tampering is formed.A color image with dual watermark is formed,after embedding the fragile watermark in the B component.Fragile watermark and robust watermark are extracted using blind extraction method.For the experiment,different criminal investigation images are chosen.The experimental results show that tamper location can be accurately detected with fragile watermark and the robust watermark is robust in resisting geometric attacks such as shearing,rotation,scaling,random mosaic,and noise.
作者
苗鑫梅
朱婷鸽
刘颖
来毅
刘卫华
MIAO Xinmei;ZHU Tingge;LIU Ying;LAI Yi;LIU Weihua(School of Communication and Information Engineering,Xi'an University of Posts and Telecommunications,Xi'an 710121,China;Key Laboratory of Electronic Information Application Technology for Crime Scene Investigation,Ministry of Public Security,Xi'an University of Posts and Telecommunications,Xi'an 710121,China;International Joint Research Center for Wireless Communication and Information Processing Technology of Shaanxi Province,Xi'an University of Posts and Telecommunications,Xi'an 710121,China;Center for Image and Information Processing,Xi'an University of Posts and Telecommunications,Xi'an 710121,China)
出处
《西安邮电大学学报》
2018年第5期77-84,共8页
Journal of Xi’an University of Posts and Telecommunications
基金
公安部科技强警专项项目(2016GABJC51)
陕西省国际科技合作与交流计划项目(2017KW-016)
陕西省教育厅科学研究计划项目(17JK0692
17JK0707)
作者简介
苗鑫梅(1993-),女,硕士研究生,研究方向为信号与信息处理。E-mail:miaoxm93@163.com;朱婷鸽(1976-),女,硕士,讲师,从事图像安全及其应用研究。E-mail:tgzhu114@163.com
