Neural Network-Based Digital Image Watermarking Algorithm
Huang Xing · Journal of Jiangxi Normal University · 2007
A novel digital image watermarking algorithm based on the association memory ability of neural network is proposed in this paper.The watermark is stored in a neural network firstly;then embed into the significant and detail wavelet coefficients of original image.The inverse embedding process is first used to extract watermark,then the extracted watermark is processed using the association memory ability of neural network.Experimental results show that the algorithm can improve the watermark image quality which is extracted from the watermarked image attacked by JPEG compression,noise adding,filtering,cropping etc.