最优像素调整耦合基因算法的高容量图像隐写研究

为了解决当前图像隐写方案存在阶梯效应,使其不可感知能力差,且其信息隐藏容量小(≤50%)等不足,本文设计了最优像素调整耦合基因算法的高容量图像隐写术。基于HDWT(Haara Discrete Wavelet Transform) 机制,构造隐藏信息长度计算模型,找出图像分块的频域表示,以改善隐写鲁棒性；根据载体图像与隐写图像之间的绝对误差,设计适应度函数,借助基因算法,获取最优映射函数,将秘密信息嵌入到HDWT系数中；并设计最优像素变换方案,降低载秘图像与载体图像之间的嵌入误差,显著增大隐写容量；再设计该算法的提取机制,获取信息图像；以PSNR(Peak Signal to Noise Ratio)构建反馈机制,优化提取质量。仿真结果显示：与其他隐写机制相比,本文算法具备更大的隐写容量和更强的不可感知性能；拥有更高的检测精度,可有效区分载体与隐写图像特征值。

The High Capacity Image Steganography Optimize Algorithm Based on Best Pixel Adjustment and Genetic Algorithm

In order to sovle these defects such as small capacity of infimation hiding and poor imperceptible performance in current image steganography mechanism, the high capacity image steganography based on optimal pixel adjustment and genetic algorithm was proposed in this paper. The frequency domain representation of blocks were founde by HDWT mechanism to improve the robustness of steganography. Then the optimal mapping function was obtained by introducing the genetic algorithm to embed the sceret message into HDWT coeffeints ; the optimal pixel adjustment algorithm was designed to reduce the erroe between cover image and stego-image and increase the hiding capavity. Finanly, the resrotation image was got by the extraction mechaonism. And the simualtion results showed that: comparision with current image steganography mechansim, this algorithm had larger hiding capacity and grearer imperceptible performance.