通过像素相关性改进的基于均值的密文域可逆信息隐藏

The Mean-Based Ciphertext Domain Reversible Information Hiding Is Improved by Pixel Correlation

摘要: 这篇文献改进了图像加密算法，提高了图像的嵌入容量。原文献中将图像按列分为256 * 1大小的块，嵌入的信息和块的多少有关。原文中分块太大而导致嵌入的信息相对较少，所以，本文将块尽可能地分小，并修改了嵌入方法并根据均值像素的相邻像素完全提取信息，使得嵌入的信息增加。本分主要讨论的是基于加密域的可逆信息隐藏的算法中的加密后腾出空间的方法。实验结果表明，修改后的算法嵌入的数据比原文多了十多倍，提出的方案同样简单易操作并具有安全性和可逆性、隐私保护性。

Abstract: This paper improves the image encryption algorithm and the image embedding capacity. The document divided image into the block columned by 256 * 1. The number of blocks determines the number of bits embedded. The original block is too large, resulting in relatively few embedded data. Therefore, this paper divides the block as small as possible and modifies the embedding method and extracts the information completely according to the neighboring pixels of the average pixel. Using this method, the embedded information increases. Experimental results show that the em-bedding capacity obtained by this method is almost ten times of that. In this part, we mainly discuss the method of making space after encryption in the algorithm of reversible information hiding based on encryption domain. The proposed scheme is also simple and easy to operate with security and reversibility and privacy protection.