The good statistical character of image blocks was used to present a novel digital reversible data hiding algorithm based on block-classification and difference-expansion. The host image was divided into data embedded area and auxiliary information area. In the former area, the types of image block were determined by the statistical relationship between the surrounding image blocks and the target image blocks, thus different amount of data information was embedded according to the different types. At the same time, the data were embedded in the less impact direction by the direction determined criterion, and the problem of image distortion due to single embed direction was solved. In auxiliary information embedded area, modified prediction-error was used to embed the location map and other auxiliary information for its zero overflow performance, and the new location map generated by auxiliary information embedded was avoided. Simulation of the algorithm was performed on different type images, and the results were compared with those of the exiting algorithms. The results show that the proposed algorithm has good performance, which can not only increase the amount of the embedded data, but also improve the embedded images' quality.%利用图像块的统计特性,描述一种新的基于差值扩展和块分类技术的可逆数据隐藏算法.首先,将图像分为数据嵌入区域和辅助信息嵌入区域.在数据嵌入区域,将分块后的图像利用周围图像块与目标图像块间的统计关系判断图像块的类型,从而根据类型差异嵌入不同数据量的信息,实现图像内容复杂度对数据嵌入量的控制.同时,利用方向判断准则选择对图像影响较小的方向嵌入,解决了单一嵌入方向对图像造成较大失真的问题.在辅助信息嵌入区域,采用MPE零溢出的特性嵌入位图辅助信息,避免嵌入辅助信息带来新的位图信息.对不同纹理图像进行实验测试以及与其他算法进行比较.研究结果表明:该算法具有良好的综合性能;在提高算法嵌入容量的同时,有效地降低了嵌入数据后对原始图像质量的影响.
展开▼
