We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
基于 BEMD、DCT 和 SVD 的混合图像水印算法.
- Authors
谭晓东; 赵奇; 文明珠; 王小超
- Abstract
The invisibility of watermark and robustness of watermarking algorithms are important issues of concern in the field of image copyright protection, however, most algorithms cannot balance well the relationship between the two. Therefore, a hybrid image watermarking algorithm with high invisibility and robustness based on bi-dimensional empirical mode decomposition (BEMD), discrete cosine transform (DCT) and singular value decomposition (SVD) is proposed in this paper. The watermark image is firstly decomposed by Arnold scrambling and 2D-DCT, and then the host image is decomposed by BEMD to obtain a finite number of different scale intrinsic mode functions and residue. The first intrinsic modal function (IMF1) with low correlation with the host image is selected to perform 2D-DCT, and it is divided into non-overlapping blocks according to the size of the watermark. Then, SVD is performed on the image of each block and the watermarked image after DCT. Finally, the watermark embedding strength is determined and the watermark is repeatedly embedded into each chunk according to the adaptive optimal embedding criterion, which effectively enhances the fault tolerance of the algorithm. Extensive experiments and comparisons with existing algorithms show that the proposed watermarking algorithm not only ensures the robustness of the algorithm against large scale attacks, but also improves the invisibility of the algorithm.
- Subjects
DIGITAL image watermarking; DISCRETE cosine transforms; DIGITAL watermarking; SINGULAR value decomposition; COPYRIGHT; HILBERT-Huang transform; HOUGH transforms; FAULT-tolerant computing
- Publication
Journal of Zhejiang University (Science Edition), 2023, Vol 50, Issue 4, p442
- ISSN
1008-9497
- Publication type
Article
- DOI
10.3785/j.issn.1008-9497.2023.04.008