摘要
提出了一种基于Hopfield神经网络的盲检测数字水印算法。基于噪声可见函数实现了水印的自适应嵌入,利用Hopfield神经网络记忆宿主图像以及原始水印信息。在水印检测时,通过神经网络从嵌入水印的图像中联想出宿主图像和水印嵌入信息,再利用嵌水印图像和联想出的宿主图像提取出水印,实现了水印的盲检测。
A blind digital watermarking algorithm based on Hopfield neural network is proposed. The shost image and original watermark are stored by the Hopfield neural network. In the watermarking embedding, the noise visibility function (NVF) is used for adaptive watermark embedding. In the watermarking extracting, the host image and the original watermark are retrieved by the neural network.
出处
《计算机与数字工程》
2006年第10期149-153,共5页
Computer & Digital Engineering