节点文献

基于Tchebichef不变矩的数字图像被动认证算法

Passive digital image authentication algorithm based on Tchebichef moment invariants

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 谷宗运吕皖丽罗斌

【Author】 GU Zong-yun1,2,Lü Wan-li1,LUO Bin1(1.School of Computer Science and Technology,Anhui University,Hefei 230039,China;2.Information Engineering College of Traditional Chinese Medicine,Anhui University of Traditional Chinese Medicine,Hefei 230031,China)

【机构】 安徽大学计算机科学与技术学院安徽中医学院医药信息工程学院

【摘要】 文章提出了一种新的数字图像被动认证算法,用于检测区域复制粘贴的伪造图像。算法首先利用小波变换提取图像的低频分量,再对低频分量进行分块并提取每一块的Tchebichef不变矩特征向量,然后将特征向量进行字典排序,比较相邻2组特征向量的相似性,最后利用阈值判别实现篡改伪造区域的检测与定位。实验表明,该算法不但能够较精确地定位出复制和粘贴的图像伪造区域,还能有效抵抗噪声污染、有损JPEG压缩以及旋转等攻击,并有效地减少了运算量,提高了检测效率。

【Abstract】 This paper presents a new passive digital image authentication algorithm to detect and identify the location of image copy-move forgery.The low-frequency images are firstly extracted by discrete wavelet transform(DWT) and split into blocks.Then the eigenvector of the Tchebichef moment invariants of each block is extracted and lexicographically sorted.The similarity of two sets of neighboring eigenvectors is compared and similar eigenvectors are matched by a certain threshold.Finally,the forgery regions are detected and identified by the threshold.The experimental results show that the proposed method can localize the copy-move forgery regions accurately,and check some attacks like random noise contamination,lossy JPEG compression,and rotation transformation effectively.Moreover,it can reduce the amount of computation and improve the efficiency of detection.

【基金】 国家自然科学基金资助项目(60772122);安徽高校省级自然科学研究重点资助项目(KJ2007A045;KJ2008A033)
  • 【文献出处】 合肥工业大学学报(自然科学版) ,Journal of Hefei University of Technology(Natural Science) , 编辑部邮箱 ,2011年06期
  • 【分类号】TP391.41
  • 【被引频次】7
  • 【下载频次】124
节点文献中: 

本文链接的文献网络图示:

本文的引文网络