A Dwt-Svd Based Watermarking Techniques For Digital Images

Robustness geometric against distortion is one of the crucial important issues in watermarking. In this paper, a new discrete wavelet transform- singular value decomposition (DWTSVD) image watermarking algorithm that is robust against affine transformation and ordinary image processing is presented. We use DWT transform to obtain four different frequency subbands. Watermarking is embedded in high frequency subbands by singular value decomposition (SVD). This is unlike traditional viewpoint that assumes watermarking should be embedded in low or middle frequency to have good robustness. Peak signal to Noise Ratio(PSNR) and Normal Cross Correlation are computed to measure image quality and template matching. In addition, the competency of the proposed scheme is verified under common image processing operations and a comparative study is made against our previous technique [6].