Digital Watermarking Approach Based on Edge Based Sorted Pixel Value Difference (ESPVD)

Abstract

Objectives: The major challenge in Imaging applications, particularly Digital Multimedia data transfer over the internet is data authentication with copyright protection to ensure ownership of identification. This paper aims to propose a new digital watermarking approach based on Edgebased sorted pixel value difference (ESPVD) to protect copyrights. Method: The hypothesis behind the proposed method is mainly composed of two phases: Watermark inserting phase, detection and extraction phase. In the inserting phase, first of all motif pattern approach is used to generate the mixed Image and then, identify the edge pixel locations using morphological edge (ME) operator to embed the watermark. Now, sorted pixel value difference method is used to insert the watermark. The watermark extraction phase also uses the same procedure which is stated earlier to retrieve the watermark image. To evaluate and match the ground truth images with the proposed watermarked images, peak signal to noise ratio (PSNR) and normalized cross correlation (NCC) are employed as quantitative measures. Findings: The proposed algorithm is tested with two logo images inserted over 30 different popular images and the outcomes indicates that the proposed approach is consistently performing well in terms of PSNR and NCC. Application/Improvements: The present approach provides high level of authentication, robustness, security and copyright protection against several attacks by using motif patterns while changing the order of pattern sequence in the formation of mixed image.