Atlantis Press Paper Details: Blind Wavelet-based Image Watermarking Based on HVS and Neural Networks

This work proposes a Blind Wavelet-based Image Watermarking (BWIW) technique, based on the human visible system (HVS) model and neural networks, for image copyright protection. A characteristic of the HVS, the Just Noticeable Difference (JND) profile, is employed in the watermark embedding so that the BWIW technique can make the watermark further imperceptible. The technique is developed in the wavelet domain. While embedding a watermark in an image, it adds the adaptive strengths to the wavelet coefficients of the image according to the JND thresholds. Moreover, an artificial neural network (ANN) is used to memorize the relationships between the wavelet version of an original image and its watermarked image. An advantage of the BWIW technique is that it utilizes the trained ANN to estimate the watermark without the original image to be applied in the calculation of the JND profile of the image. Finally, computer simulations demonstrate that both the transparency and the robustness of the BWIW technique are better than that of other proposed methods.

		back to author index

title:		Blind Wavelet-based Image Watermarking Based on HVS and Neural Networks
publication:		JCIS-2006 Proceedings
part of series:		Advances in Intelligent Systems Research
ISBN:		978-90-78677-01-7
ISSN:		1951-6851
DOI:		doi:10.2991/jcis.2006.40 (how to use a DOI)
author(s):		Hung-Hsu Tsai, Chi-Chih Liu, Kuo-Chun Wang
corresponding author:		Hung-Hsu Tsai
publication date:		October 2006
keywords:		Image Watermarking, Human Visual System, Neural Networks, Wavelets, Image Authorization
abstract:		This work proposes a Blind Wavelet-based Image Watermarking (BWIW) technique, based on the human visible system (HVS) model and neural networks, for image copyright protection. A characteristic of the HVS, the Just Noticeable Difference (JND) profile, is employed in the watermark embedding so that the BWIW technique can make the watermark further imperceptible. The technique is developed in the wavelet domain. While embedding a watermark in an image, it adds the adaptive strengths to the wavelet coefficients of the image according to the JND thresholds. Moreover, an artificial neural network (ANN) is used to memorize the relationships between the wavelet version of an original image and its watermarked image. An advantage of the BWIW technique is that it utilizes the trained ANN to estimate the watermark without the original image to be applied in the calculation of the JND profile of the image. Finally, computer simulations demonstrate that both the transparency and the robustness of the BWIW technique are better than that of other proposed methods.
copyright:		© Atlantis Press. This article is distributed under the terms of the Creative Commons Attribution License, which permits non-commercial use, distribution and reproduction in any medium, provided the original work is properly cited.
full text:		JCIS06-CVPRIP-53.pdf (180 K)