Decision-based hybrid image watermarking in wavelet domain using HVS and neural networks

1Citations
Citations of this article
1Readers
Mendeley users who have this article in their library.
Get full text

Abstract

This paper presents a Decision-based Hybrid Image Watermarking (DHIW) technique, based on the Human Visible System (HVS) and an Artificial Neural Network (ANN), for image copyright protection in wavelet domain. In [1], an image watermarking technique, called the IWNN technique, utilizes an ANN to extract watermarks without original images. However, the IWNN technique performs poorly for highly complicated image textures because the generalization capability of neural networks is powerfully effective in dealing with smooth image textures. Therefore, the PAIW method is proposed to enhance the IWNN technique, which uses the spatial information associated with wavelet-transformed images. The DHIW technique takes advantages of these two techniques by using a decision preprocessor. Experimental results prove that the DHIW technique remarkably outperforms other existing schemes. © Springer-Verlag Berlin Heidelberg 2007.

Cite

CITATION STYLE

APA

Tsai, H. H. (2007). Decision-based hybrid image watermarking in wavelet domain using HVS and neural networks. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4493 LNCS, pp. 904–913). Springer Verlag. https://doi.org/10.1007/978-3-540-72395-0_111

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free