A robust watermarking scheme based on singular value decomposition and quantization technique

Abstract

In this paper, a robust watermarking scheme using singular value decomposition (SVD) and quantization is proposed. The SVD technique is used to extract important features from a host protected image; the important features are preserved in the singular values which are resistant to a rotation attack. Due to the sensitivity of singular values, the quantization technique is applied to find an appropriate modification range for singular values during watermark embedding. The proposed scheme does not need an original host protected image and an original watermark image for watermark extraction. The experimental results clearly present a robust watermarking effectively resistant to various attacks.