An Image Watermarking Scheme Resilient to Geometric Distortions

In this article, a robust watermarking scheme is proposed to lighten the problem of geometric distortion using Lifting Wavelet Transform (LWT). Initially, the host image is normalized to make it as invariant to affine distortion. Then LWT is applied and LD subband is divided into 8 × 8 non overlapping blocks. The first element of each block is retrieved to form a matrix to apply Singular Value Decomposition (SVD). The watermark is embedded in these singular values after applying Arnold scrambling to enhance the security of the watermark. Robustness of the proposed scheme are estimated using the metric Peak Signal to Noise Ratio (PSNR) and Structural Similarity (SSIM). The experimental results are compared and it exemplifies that the proposed method generates good robustness for geometric distortion.