NVF and CSF Based Image Watermarking Using SVD and Contourlet Transform

This article proposes a human visual system (HVS) with stable transform image watermarking scheme. This scheme uses the Noise Visibility Function (NVF) and Contrast Sensitivity Function (CSF) to apply the HVS, making it image adaptive. A hybrid operation is the stable transform domain hiding scheme. This is due to the contourlet transform (CT) centered on the pyramidal and directional filter bank (PDFB) as well as the singular value decomposition being used (SVD). The scheme presented in this article is visually transparent, robust, and with high capacity due to the usage of the HVS, and the unique blend of the two transforms. The analysis of imperceptibility has been done with respect to a number of previous standard works, along with checking for robustness and results after applying the popular attacks.