Abstract: Aiming at the problems of traditional reversible information hiding algorithms such as declining quality of the host image, low security, and extra overhead for information recovery, a reversible information hiding algorithm combining chaos theory and integer transform is proposed. The carrier image is divided into sub-blocks of size 2×2 and their pixels are divided into labeled pixels and hidden pixels. The sub-block pixels are evened and expanded to obtain the difference values with the base pixels. If the extended pixel and the difference value satisfy a given threshold condition, the chaotic encrypted hidden information is transformed into the least significant bit of the concealed position pixel of the carrier sub-block by integer transform. Otherwise, the mark position pixel is odd-numbered and the least significant bit of the original pixel is recorded. The simulation test shows that the proposed hiding method can achieve the lossless restoration of the carrier image and the original image, which not only reduces the transmission load effectively, but also has high security.

Key words: image encryption, information hiding, chaos encryption, differential expansion, integer transformation

摘要： 针对传统可逆信息隐藏算法存在宿主图像质量下降明显、安全性低、信息恢复需要较多额外负载等问题，提出一种混沌理论和整数变换相结合的可逆信息隐藏算法。将载体图像分为大小为2×2的子块并对其像素区分为标记像素和隐藏像素；将子块像素偶数化并将其扩展再获取与基像素的差异值；若扩展像素和差异值满足给定门限条件时，将混沌加密隐藏信息经整数变换再嵌入载体子块隐藏位置像素的最低有效位；否则，将标记位置像素奇数化并记录该原始像素的最低有效位。仿真测试表明，所提出的隐藏方法能够实现载体图像与原始图像无损恢复，不仅有效地减少了传输负载，且具有较高的安全性。

关键词: 图像加密, 信息隐藏, 混沌理论, 差异扩展, 整数变换