节点文献

基于图像插值和参考矩阵的可逆信息隐藏算法

Reversible Data Hiding Based on Image Interpolation and Reference Matrix

  • 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】 孙容海施林甫黄丽艳唐振军俞春强

【Author】 SUN Ronghai;SHI Linfu;HUANG Liyan;TANG Zhenjun;YU Chunqiang;Guangxi Key Lab of Multi-source Information Mining and Security,Guangxi Normal University;Guangxi Normal University Press(Group);Network Information Center,Guangxi Normal University;

【通讯作者】 俞春强;

【机构】 广西师范大学广西多源信息挖掘与安全重点实验室广西师范大学出版社集团广西师范大学网络信息中心

【摘要】 本文提出一种基于插值技术和参考矩阵的可逆信息隐藏算法。该算法首先用一种改进的线性插值方法对载体图像进行插值,生成一幅插值图像;然后对插值图像进行不重叠分块,分块大小为2×2,在每个分块中以左上角的像素值作为平面坐标点的横坐标,其他像素值作为纵坐标构造3个坐标点并将其映射到参考矩阵中;最后根据秘密信息的十进制值和参考矩阵中相应坐标点的值来修改纵坐标以实现信息隐藏。在提取秘密信息时,通过信息隐藏时相同方法构造每个分块的3个坐标点并映射到参考矩阵中获取相应坐标点处的值完成秘密信息的提取。由于信息隐藏过程仅修改插值像素,原始像素保持不变,因此可无损还原载体图像。大量实验结果表明,该算法具有较大的信息隐藏容量和较好的视觉效果。

【Abstract】 A reversible data hiding algorithm based on interpolation technique and reference matrix is proposed in this paper.Firstly,the cover image is interpolated to generate an interpolated image using an improved linear interpolation method.Then,the interpolated image is divided into non-overlapped blocks with size 2×2.And the pixel in top-left corner of each block is taken as horizontal ordinate of the point in plane.Moreover,other pixels are taken as vertical ordinates to construct three coordinate points,which are mapped into reference matrix.Finally,the vertical ordinates are modified in accordance with the decimal value of secret data and the value of corresponding coordinate points in reference matrix to realize data hiding.During data extraction,three coordinate points of each block are constructed via the same method in data hiding and mapped into the reference matrix to obtain the values of corresponding coordinate points to achieve data extraction.Since only interpolated pixels are modified and the original pixels keep unchanged,the cover image can be recovered without loss.Experimental results demonstrate that the proposed algorithm reaches high embedding capacity and has good visual quality.

【基金】 国家自然科学基金(61562007,61762017);广西自然科学基金(2017GXNSFAA198222,2015GXNSFDA139040)
  • 【文献出处】 广西师范大学学报(自然科学版) ,Journal of Guangxi Normal University(Natural Science Edition) , 编辑部邮箱 ,2019年02期
  • 【分类号】TP391.41;TP309
  • 【被引频次】7
  • 【下载频次】198
节点文献中: 

本文链接的文献网络图示:

本文的引文网络