Self-embedded full-blind color image robust watermarking method based on multiple transform domains

Abstract

The invention discloses a self-embedded full-blind color image robust watermarking method based on multiple transform domains. The method comprises the following steps of: performing fast finite shearwave transformation on a blue part of an original color image, and generating an original characteristic watermark for copyright protection based on QR decomposition and one-dimensional discrete sinetransformation in a transformed low-frequency sub-band by virtue of a'zero watermark 'thought; and then performing dual-density dual-tree complex wavelet transform on the green part of the original color image, and completing embedding of the original feature watermark in the obtained low-frequency sub-band based on two-dimensional discrete sine transform and a pre-generated random number pair. The watermark embedding end does not need to select additional information as a watermark, and the watermark detection end can carry out copyright recognition on a color image in a'full-blind 'manner only by means of an attacked image and a related secret key. The method has good watermark invisibility, and the generation and self-embedding scheme of the original feature watermark can effectively resist various common attacks.

Description

technical field [0001] The invention belongs to the fields of information security and image processing, and in particular relates to a self-embedded robust watermarking method for blind color images based on multi-transform domains. Background technique [0002] With the rapid development of computing technology and network technology, operations such as illegal duplication and malicious tampering of digital image products, especially color images, have become relatively easy. As an effective means, digital watermarking technology can solve the problems of copyright protection and content authentication of color digital images, so it has become a research hotspot in recent years. [0003] Digital watermarking algorithms are divided according to their functions, and can be divided into robust watermarking algorithms, fragile watermarking algorithms and semi-fragile watermarking algorithms. Among them, the robust watermarking algorithm can only be used for image copyright pr...

AI Technical Summary

Problems solved by technology

In 2015, Han Shaocheng and others applied the idea of ​​"self-embedded blindness" to the copyright protection of color images for the first time, and proposed a color image self-embedded blindness robust watermarking algorithm based on QR decomposition, which is resistant to adding noise, JPEG compression and scaling, etc. The attack has certain robustness, but the self-embedding method of feature watermark based on quantization method is almost ineffective for brightness and contrast adjustment attacks

The space domain algorithm is simple in operation and high in timeliness, but has the disadvantage of poor robustness; although the transform domain algorithm is relatively complex, it has high robustness, so it has attracted the attention of researchers.

Most of the existing robust watermarking algorithms for digital images use "extra" information such as binary images as watermarks and embed them into the carrier image. The watermark detection end can identify the copyright of the image by extracting the watermark information, but it cannot achieve "full blindness". detection

At present, the "full-blind" robust watermarking algorithm for color images is still rare, and the robustness of the existing algorithms is also very poor

Images