Image watermarking on Gaussian low pass components ignoring edges and its neighbor pixels

Abstract

This paper proposes a method used to improve the performance of the digital watermarking based on the modification in the blue channel of the image pixels. The main objective is to reduce the error prediction caused by the high different values of neighbor pixels. We use 5x5 Gaussian Low Pass Filter in the embedding side and then the edge detection technique takes place in order to find out the high different values of neighbor pixels, so that on the embedding side, the watermark bits are embedded on the low frequency components of the blue channel ignoring the edges and neighbor of 3×3 pixel positions, where in the extraction side, only the edge and its neighbor ignorance are applied. The experimental results show the improved performance in terms of average NC, compared to the previous methods. The improved robustness of our method against different types of attacks at various strengths is also presented and compared.