Hybrid watermarking algorithm based on Singular Value Decomposition and Radon transform

A hybrid robust digital watermarking algorithm based on finite Radon transform (FRAT) and Singular Value Decomposition (SVD) is proposed and simulated. The Discrete Wavelet Transform (DWT) is popularly used as it possesses properties similar to the human visual system (HVS). We will show that, the watermark embedding procedure comes to get better normalized correlation (NC) factor by modifying singular values. Trade off robustness and transparence of digital watermarking is done by using middle frequencies of HL3 and LH3. Experimental results demonstrate that this method is robust against several attacks such as: noise addition, filtering, histogram equalization, gamma correction, JPEG compression, scaling and rotation.

Digital watermarking ,Wavelet transform, Singular Value Decomposition, Human visual system