首页 | 本学科首页   官方微博 | 高级检索  
     

抗拼贴攻击的二值图像认证算法
引用本文:秦瑶, 和红杰, 陈帆. 抗拼贴攻击的二值图像认证算法. 自动化学报, 2014, 40(6): 1184-1190. doi: 10.3724/SP.J.1004.2014.01184
作者姓名:秦瑶  和红杰  陈帆
作者单位:1.西南交通大学信息科学与技术学院 成都 610031
基金项目:国家自然科学基金(61373180,60970122),中央高校基本科研业务专项基金资助(SWJTU09CX039,SWJTU10CX09),西藏自治区科技计划项目资助
摘    要:为提高算法抵抗拼贴攻击的能力,本文结合块相关思想提出一种二值图像认证算法.该算法首先将每个图像块中的像素点分为“可翻转”和“不可翻转”两类,将“不可翻转”像素点 Hash 生成水印信息,用其替代相应映射块的“可翻转”像素点实现水印嵌入.通过比较图像块重构水印及提取水印的一致性判断该图像块的真实性,并结合各图像块邻域篡改特征进一步提高篡改检测性能.实验结果表明,该算法不仅提高了在替换、添加、删除攻击下的篡改检测性能,而且能够有效抵抗拼贴攻击.

关 键 词:二值图像   拼贴攻击   块相关   邻域篡改特征
收稿时间:2013-01-25
修稿时间:2013-05-29

Binary Image Authentication Watermarking Scheme Against Collage Attack
QIN Yao, HE Hong-Jie, CHEN Fan. Binary Image Authentication Watermarking Scheme Against Collage Attack. ACTA AUTOMATICA SINICA, 2014, 40(6): 1184-1190. doi: 10.3724/SP.J.1004.2014.01184
Authors:QIN Yao  HE Hong-Jie  CHEN Fan
Affiliation:1. School of Information Science and Technology, Southwest Jiaotong University, Chengdu 610031
Abstract:To improve the ability against collage attack, a watermarking scheme for binary image authentication is proposed by introducing dependence of blocks in this paper. In the proposed algorithm, all pixels in each block are divided into the flipped pixels and the unflipped ones. The watermark information of each block is generated by hashing the unflipped pixels of it, and embedded in the flipped pixels of the corresponding mapping block. The authenticity of each block is detected by comparing the consistency between the reconstructed watermark and the extracted one. The neighborhood tampering feature is used to improve the tamper detection performance. Experimental results show that the proposed algorithm can not only improve the performance of tamper detection in the general tampering such as replace, add, delete, etc., but also effectively resist collage attack.
Keywords:Binary image  collage attack  block correlation  neighborhood tampering feature
本文献已被 CNKI 等数据库收录!
点击此处可从《自动化学报》浏览原始摘要信息
点击此处可从《自动化学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号