一种高容量的自适应图像可逆信息隐藏算法

An Adaptive Reversible Data Hiding Algorithm for Digital Images with High Embedding Capacity

基于传统的预测误差扩张算法,创新性地提出一种根据不同像素局部复杂度适应性嵌入1bit或3bits的可逆数据隐藏算法,有效地避免对有较大预测误差像素的嵌入,提高了水印图像质量并且增加了可嵌入容量。同时结合像素选择策略,在被改变的图像像素总数量中加大嵌入像素数量比重,并降低转移像素数量比重,以达到优化水印图像质量的目标。实验结果表明,该算法相比于传统的预测误差扩张算法与相似理念适应性算法,在嵌入量和水印质量方面都有一定提升。

In this paper,an efficient reversible data hiding scheme is proposed which adaptive embedding 1 bit or 3 bits data according to the local complexity of pixels based on the traditional prediction error expansion algorithm.It effectively avoids choosing pixels with large prediction error,improves the quality of watermark images and increases the embedding capacity.At the same time,the pixel selection strategy is adopted to increase the proportion of embedded pixels and reduce the number of shifted pixels.In this way,the quality of watermarked images is optimized.Experiments showed that our method outperforms traditional prediction error expansion and other similar method.