Abstract:

This paper proposes a new image watermarking scheme based on Generalized Singular Value Decomposition (GSVD). This image watermarking method embeds the principal components of watermark image into the host image using a specific scaling factor to control the imperceptibility and robustness aspects. Using this strategy, the GSVD-based image watermarking has ability to avoid the false positive problem and ambiguity situation which often occur in the former existing Singular Value Decomposition (SVD)-based image watermarking schemes. As documented in the experimental results, GSVD-based image watermarking method offers a promising result and reliable performances compared to the former schemes.