ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 7. Vol. 25. 2019

DOI: 10.17587/it.25.405-413

A. N. Shniperov, Assistant Professor, e-mail: ashniperov@sfu-kras.ru, Siberian Federal University, M. S. Sosnovskiy, Software Development Engineer, e-mail: maximsosnovskiy@gmail.com, P. M. Shipulin, Software Development Engineer, e-mail: pshipulin@gmail.com, Academician M. F. Reshetnev Information Satellite Systems

The Robust Image Digital Watermark Labeling Method Based on Orthogonal Zernike Moments

Nowadays visible watermarks are usually used for the copyright protection of images. But this method does not give the best result in many cases. The best practices lie in the digital steganography field. In this article authors suggest a digital watermark labeling method for full-color images based on orthogonal Zernike moments calculated for certain image points (and their neighborhoods). Using only certain points makes the method applicable for any image sizes with good robustness features and low computation time (orthogonal Zernike moments calculation for big image is quite a difficult computation task). To find certain points the Shi-Tomasi corner detector algorithm is used. These points are invariant under affine transformation. The suggested method uses only blue channel of the RGB-model, because distortions in this channel are less visible for the human eye. For performance evaluation of the method some popular metrics were used: BER (Bit Error Rate), PSNR (Peak Signal-To-Noise Ratio), MSE (Mean Square Error), SSIM (Structural SIMilarity). A series of experiments illustrates high values of these metrics not only in case of simple digital watermark embedding but in case of the most popular active attacks on the labeling method: JPEG compression, image reflexing, scaling, rotating, cropping, noise adding. The group of tests proves the labeling method immunity to the majority of popular active attacks. So the suggested method can be regarded as an effective method for labeling images by invisible watermarking for copyright protection purposes.
Keywords: digital watermarks, steganography, image labeling, information security, copyright protection, image moments, Zernike moments

P. 405–413

To the contents