关键词: 数字隐写, 通用盲检测, 纹理复杂度, 边缘区域, LSB匹配

Abstract: According to the analysis on the effect of the texture complexity on steganalysis, the authors proposed a data hiding algorithm based on texture complexity and pixel-value difference. Firstly, the image was decomposed into blocks. Then, the texture complexity of each block was calculated and blocks with high complexity were selected. Then the pixel-value difference on two directions of the selected blocks was computed to determine the edge areas. The secret messages were only embedded in the pixel-value difference above the threshold. When compared with several steganographic methods on different image databases, the experimental results show that the proposed method outperforms previous steganographic methods such as LSB matching and adaptive data hiding in edge areas of images on the capability of resisting universal blind detection, such as wavelet high-order statistics analysis.

Key words: steganography, universal blind detection, texture complexity, edge area, LSB matching