Dual Watermarking Based on Human Visual Characteristics with IWT-SVD

Abstract

The security and protection of digital content have become crucial for protecting ownership. Watermarking is an essential technique for protecting copyright and preserving the integrity of digital images. This study suggested a robust and secure image watermarking scheme that employs a combination of IWT-SVD to embed dual watermarks. The watermark logo is embedded by examining the orthogonal U with a set of conditions. This study also introduces threshold values in the red and blue channels for two watermarks embedded. Threshold values of 0.06 and 0.13 are proposed for red and blue channels, respectively, as the strength of watermark embedding. The threshold is revealed by incrementing a threshold one at a time against JPEG compression. The results indicate method exhibits a smaller reconstruction error of about 1.2415 and an SSIM value of 0.9158. The proposed watermarking method achieved an NC value of 0.9246 and a BER value of 0.0742. This indicates that the method exhibits better robustness than the existing method against various attacks. The suggested method shows good robustness against Gaussian noise, median filter, salt and pepper noise, and JPEG compression.