Computer Engineering and Applications ›› 2018, Vol. 54 ›› Issue (14): 115-119.DOI: 10.3778/j.issn.1002-8331.1712-0003

Previous Articles     Next Articles

High fidelity reversible data hiding algorithm based on SVD compression

LI Tianxue, WANG Jianping, ZHANG Minqing, KONG Yongjun   

  1. Key Laboratory of Network and Information Security under the Armed Police Force, Department of Electronic Technology, Engineering University of the Chinese Armed Police Force, Xi’an 710086, China
  • Online:2018-07-15 Published:2018-08-06



  1. 武警工程大学 电子技术系 网络与信息安全武警部队重点实验室,西安 710086

Abstract: In order to solve the problem of distortion caused by embedding the secret data on the original carrier image, high-fidelity reversible information hiding algorithm using Singular Value Decomposition(SVD) prediction pixel is proposed in this paper. Firstly, the original carrier is divided into gray and white layers. Pixels in the gray layer are selected as the target pixels, and the white pixels in the field are used as the reference pixels. Then, the neighborhood matrix is constructed using these reference pixels, the matrix is processed using SVD, and is restored with a large singular value, the target pixel is predicted by its average. Finally, the secret data is embedded by the extended prediction error. Extensive experiments have shown that the proposed algorithm effectively reduces the embedded distortion of the carrier.

Key words: reversible data hiding, high fidelity, Singular Value Decomposition(SVD)

摘要: 针对嵌入秘密数据对原始图像造成失真明显的问题,提出一种利用奇异值分解(Singular Value Decomposition,SVD)进行像素预测的可逆信息隐藏算法。首先将原始载体分成灰和白两层,选取灰色层中的像素作为目标像素,其领域上的白色层像素作为参考像素;而后利用这些参考像素构成邻域矩阵,再对其SVD压缩处理,利用压缩结果预测目标像素;最后通过扩展预测误差嵌入秘密数据。实验数据显示,该算法有效降低了携密载体的嵌入失真。

关键词: 可逆信息隐藏, 高保真, 奇异值分解(SVD)