为了提高信息隐藏的不可感知性和鲁棒性, 结合人眼视觉系统(HVS)的相关知识, 提出了一种新的小波域信息隐藏算法。将载体图像的中等亮度区域设为不可用区域, 不在该区域隐藏信息, 同时选择在载体图像小波变换的高频系数细节子图隐藏信息, 并优先选择在对角高频系数细节子图中隐藏。计算局部子块的方差值作为隐藏信息的嵌入系数, 并通过计算高频子图的噪声敏感模型ωθ(i,j)决定隐藏信息的位置。充分考虑了人眼的各种视觉掩蔽效应, 使得隐藏信息后的结果图像相比隐藏之前没有出现明显的降质。实验结果表明该方法能够有效提高信息隐藏的不可知性, 达到了很好的隐藏效果, 同时能够有效地抗击噪声、裁剪和压缩等攻击影响, 具有很好的鲁棒性, 在信息安全领域有较好的应用前景。

To improve the imperceptibility and robustness of information hiding, a new method in wavelet domain is proposed in combination with correlative knowledge of HVS. Medium luminance districts of carrier image are supposed to unavailable districts without hiding information. High-frequency coefficients approaching subgraph of discrete wavelet transform of carrier image are chosen to hide information and diagonal high-frequency coefficients approaching subgraph is prior choice. The variance of partial subgraph is defined as imbedding coefficient. The positions to hide information are determined by noise sensitivity model ωθ(i,j) of high-frequency subgraph. The quality decline of the result image is unapparent compared with carrier image owing to sufficiently considering the visual masking effect. The proposed method is proved to be a valid approach for improving the imperceptibility of information hiding method via computer simulation and leading to a good hiding result. Meanwhile the method can effectively resist the influence of noise, cutting, compression and other attacks. And the good robustness is of high practical value in the field of information security.