A Multiple-Level DCT Based Robust DWT-SVD Watermark Method

• Published in: 2014 Tenth International Conference on Computational Intelligence and Security pp. 479 - 483
• Main Authors: Xueyi Ye, Xueting Chen, Meng Deng, Shuyun Hui, Yunlu Wang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2014
• Subjects: Discrete cosine transforms; discrete wavelet transform-singular value decomposition; Discrete wavelet transforms; image watermark; large-capacity; Matrix decomposition; multiple-level discrete cosine transform; Noise; Robustness; Watermarking
• DOI: 10.1109/CIS.2014.28

The contradiction between the capacity and the robustness of image watermark has not been solved in the digital watermark research. In this paper, a robust Discrete Wavelet Transform (DWT) - Singular Value Decomposition (SVD) watermark method based on multiple-level Discrete Cosine Transform (DCT) is proposed by the advantage of DCT's energy concentration. Firstly, the carrier image is done discrete wavelet transform and its low frequency sub-band is obtained. Secondly, the coefficient matrix of the low frequency sub-band is partitioned. Then each block is done multiple-level DCT. Thirdly, according to the capacity of the embedded watermark, the appropriate DCT coefficients are selected in each block to form a new matrix, and then the encrypted watermark is embedded in the singular value matrix of the new matrix. Finally, the image containing watermark is obtained by SVD synthesis, multiple-level inverse DCT and multiple-level inverse DWT. Experimental results show that compared with other similar watermark methods, this paper's method not just can embed the large-capacity watermark by dynamically selecting the appropriate DCT coefficients, but also has robustness to different image attacks.