An Adaptive Image Watermarking Algorithm Based on Neural Network

摘要

According to the characteristic of Human Visual System and the association memory ability of neural network, an adaptive image watermarking algorithm based on neural network is proposed. The watermarking signal is embedded in higher frequency, which is in the lower frequency of original image by DWT joined with DCT. The ability of attracting is improved by pretreatment and re-treatment of image scrambling and Hop field network. Experimental results show that the embedded digital watermarking is invisible and robust enough against common image processing such as JPEG compression, noises, filter and shearing.