An optimized image watermarking algorithm based on SVD and IWT
Ting Zhu,Wen Qu,Wenliang Cao
DOI: https://doi.org/10.1007/s11227-021-03886-2
IF: 3.3
2021-05-25
The Journal of Supercomputing
Abstract:With the wide spread of image information, it is an urgent problem to protect image property rights and crack down on piracy. Watermarking algorithm is an effective means to solve the problem of copyright protection. In this paper, we propose an optimized image watermarking algorithm based on singular value decomposition (SVD) and integer wavelet transform (IWT). First, the carrier images are divided into blocks. Then, the block-based integer wavelet transform is performed, and the singular value decomposition is performed in the low frequency part. Finally, the first singular value is used to extract energy effectively, so as to improve the robustness of digital watermarking. At the same time, genetic algorithm is used to optimize the robustness and imperceptibility of image watermarking. Four classic gray images, including Lena, baboon, peppers, and Goldhill, are used as carrier images to test, the test results show that the watermarking algorithm has good imperceptibility, and robustness. Compared with other methods, the experimental results show that the algorithm has good PSNR(peak signal-to-noise ratio) and NC(normalized correlation coefficient) values. In the attacks including Gaussian noise, low-pass filtering, changing the size, straight square error equalization, image blur, image contrast, JPEG compression, and gamma correction, the proposed method shows good performance. The NC value of this method is better than that of the contrast method, especially in Gaussian noise attack.
computer science, theory & methods,engineering, electrical & electronic, hardware & architecture