基于凸优化的图像隐藏信息重构方法

摘要

Aiming at shortcomings of weak attack resistance of image information hiding least significant bits (LSB) algorithm,a reconstructing method which applies matrix reconstructing theory based on convex optimization to extracted hiding image is proposed.The method makes discrete cosine transform (DCT) on extracted hiding image,to make transformed data spare,singular value iteration algorithm is used to reconstruct data and hiding image is obtained through inverse operation,feasibility of the method is verified by experimental result.%针对图像信息隐藏最不重要位(LSB)算法抗攻击性弱的缺点,提出了一种将凸优化的矩阵重建理论运用于所提取的隐密图像中的重建方法.对所提取的隐密图像作离散余弦变换(DCT),使变换后的数据稀疏化,利用奇异值迭代算法重构数据,通过逆运算得到精度较高的隐秘图像,实验结果验证了方法的可行性.