Shearlet变换和奇异值分解的图像隐藏方法

为了更好地平衡Shearlet域图像隐藏不可见性、鲁棒性和算法时间复杂度之间的关系,提出了一种基于Shearlet变换和奇异值分解的图像隐藏方法。利用Shearlet变换的能量聚集性、小波包分解低频子带抗攻击性强和矩阵奇异值良好的稳定性,载体图像先进行Shearlet分解,得到的低频子带再进行二级小波包分解。将秘密图像的重要信息位平面隐藏到小波包分解低频系数的奇异值矩阵中,次要信息嵌入Shearlet高频子带中。实验表明,该算法对高斯噪声、滤波和剪切等攻击都有较好的鲁棒性,同时,不可见性较好,时间复杂度较低。

A image hiding method based on Shearlet transform and SVD

In order to balance the relationship between image hiding imperceptibility, robustness and time complexity of the algorithm in the shearlet, a kind of image hiding method is proposed based on shearlet combination and Singular Value Decomposition (SVD), which uses the energy aggregation of shearlet combination, the strong anti-jamming of low frequency coefficient matrix of wavelet packers transform and the stability of singular value. The carrier image is firstly decomposed by the shearlet transformation, and then the low-frequency sub-band from the shearlet transformation is decomposed by the two level wavelet packets. The key bit-planes of secret image are hidden into the singular values of the wavelet low-frequency sub-band, and the bit-planes that contain less information are hidden into the high-frequency sub-band of the shearlet. The results show that, the algorithm is effectively against the normal image attacks, such as cutting processing, filtering processing and adding noise. Meanwhile, the invisibility and the time complexity are good.