信息隐藏
加密
嵌入
有效载荷(计算)
计算机科学
图像(数学)
按位运算
像素
块(置换群论)
锯齿波
分拆(数论)
排列(音乐)
算法
密码学
理论计算机科学
编码(内存)
人工智能
数学
计算机视觉
方案(数学)
数字水印
最低有效位
数据提取
模式识别(心理学)
彩色图像
图像处理
作者
Xianming Wang,Li Li,Mingjing Wang,Chin-Chen Chang
标识
DOI:10.1007/s44443-025-00238-5
摘要
Reversible data hiding in encrypted images (RDHEI) is a popular research topic in data hiding, and embedding capacity is the most important evaluation metric. To obtain high embedding capacity, many block-based data hiding methods—which usually exploit spatial correlations between adjacent pixels in the same block—have been proposed and have achieved good results. Currently, images are typically partitioned into non-overlapping squares, rectangles, rhombuses, or line-shaped regions. In this paper, we propose a novel reversible data hiding method for encrypted images based on a Most Significant Bits sharing encoding (MSBSE) scheme and an overlapping sawtooth partition (OSP) strategy, named MSBSE-OSP. First, by using the MSBSE-OSP scheme, embeddable space in the image is reserved for auxiliary data and payload data. Then, the image is encrypted through block permutation and bitwise eXclusive OR(XOR) operations. Finally, auxiliary data and secret data are embedded into the encrypted image. The recipient can extract secret data or reconstruct the original image with one key, or perform both operations with two keys. Experimental results show that our method achieves separable extraction and decryption operations, with higher capacity compared to state-of-the-art methods.
科研通智能强力驱动
Strongly Powered by AbleSci AI