Robust and Imperceptible Image Watermarking using Full Counter Propagation Neural Networks
Summrina Kanwal, M. Arfan Jaffar, Wajid Rasul, Anwar Majeed Mirza, Saudia Arabia · 2011
Digital Media is the main source of communication now a days. A drastic change emerged with the access of internet to everyone, which made it possible to copy and distribute the digital content illegally. Digital Watermarking/Steganography has been used for digital media protection. Many techniques have been developed for effective watermarking. However, there is always a tradeoff between robustness and imperceptibility features of watermarking offered by these techniques. In this paper, we have proposed a method based on Full Counter Propagation Neural Network (FCPN) to train multiple grey or color cover images to produce a desired watermark image. We have used grey, binary and color images as watermarks. The results have proved the higher imperceptibility and robustness of this scheme when compared with the other available watermarking techniques.