Semi-Blind Watermarking of Digital Images using Evolutionary Algorithms

Abstract

In this paper, the evolutionary techniques- Firefly and Bat algorithmsareexercised to optimize the proposed semi-blind watermarking scheme of digital images to attain better imperceptibility and robustness. The insertion of the watermark is performed by normalized block processing (NBP) and hybrid transforms such as IWT, DCT and SVD. The embedding and extraction processes have been performed with the parameters obtained from the evolutionary algorithms. The experimental results prove that this scheme exhibits very good perceptual quality of the watermarked image and better robustness of the watermark for various image attacks. The proposed approach is also superior to the other existing methods in terms of PSNR.