Title: Digital image watermarking technique for big size watermarks and high-resolution images using discrete cosine transform
Authors: Ahmed Al-Gindy
Addresses: School of Engineering and Information Technology, AlDar University College, Dubai, UAE
Abstract: The capability of any watermarking system to hide large or small amounts of data while maintaining the robustness and quality is an important challenge in watermarking. The proposed algorithm can store hidden watermarking information up to 25% of the host image size. This means that it can also be an extremely robust technique when working with high-resolution images. The green channel of the RGB model has been used to host the watermark by modifying the very low-frequency coefficients of the discrete cosine transform transformation. The used frequency coefficients are predefined after a process of zigzag patterns. The new technique can resist classical attacks such as JPEG compression, low-pass filtering, median filtering and geometric attacks such as, cropping, rotation, affine and scaling. The method has been applied to high-resolution images that have been found to be extremely robust against attacks. The recovery method is blind because it does not need the original host image for extraction. The simulation results show that the proposed watermarking technique outperforms many other existing techniques.
Keywords: blind; discrete cosine transform (DCT); high capacity; high resolution; RGB; watermarking.
International Journal of Signal and Imaging Systems Engineering, 2017 Vol.10 No.6, pp.306 - 315
Received: 21 Jul 2016
Accepted: 26 Jun 2017
Published online: 13 Nov 2017 *