Robust watermarking in multiwavelet domain based on average subimage of low frequency

In this paper, a simple and robust watermarking algorithm using balanced multiwavelet transform is proposed. The latter transform we use achieves simultaneous orthogonality and symmetry without requiring any input prefiltering or border extending. This reduction in computational complexity makes the transform a good candidate for real-time watermarking implementations. Novelty of the system lies in 1) the use of a new multiwavelet suitable for watermarking techniques, 2) the proposition of the average subimage of low frequency in multiwavelet domain, 3) watermarks generated from the above subimage during the embedding procedure which are not predetermined. The original image is transformed using the new multiwavelet and flags are created with the secret key. Then the extracted flags using the same key are compared with the original ones to confirm the copyright of digital product without the original one. Theoretical analysis and simulation results show that the proposed watermarking system has excellent robustness against various watermark attacks including printing and scanning distortions with the high quality of the watermarked image.