A Watermark Embedding and Extraction Method Combining Non-overlapping Blocks and Equal-Band Fusion

Abstract

The invention provides non-overlapping block and frequency band equivalent fusion combined watermark embedding and extracting methods. The initial values and parameters of a watermark encryption link and a watermark embedding link are generated through the SHA-1 value of a cover image and a secret key given by a user, chaotic mapping is driven to generate a 2-value pseudorandom matrix to encrypt a 2-value watermark and generate random coordinate positions of non-overlapping blocks for an embedded carrier image, 1-level wavelet coefficient Pyramid decomposition is carried out on each random non-overlapping block, frequency band coefficients HL1 and LH1 are fused in an equal weighed manner to form a fusion coefficient, and the quantified value of the fusion coefficient is adjusted to embed a watermark. Watermark embedding and extracting strategies can be disclosed completely; and compared with a traditional disclosed sensitive watermark, the methods of the invention avoid a cast effect caused by that different levels of wavelet coefficients are fused, prevent twice adjustment for watermark extraction via an isolated point filtering strategy, but ensure the sufficiently-high embedded safety level.

Description

technical field [0001] The invention belongs to the cross research field of information security and digital image signal processing, relates to a watermark embedding and extraction method, in particular to a watermark embedding and extraction method combining non-overlapping block and equal frequency band fusion. Background technique [0002] The development of multimedia and network technology enables people to obtain various digital image information more conveniently. Digital images have the advantages of easy storage, easy transmission, and easy operation, but at the same time, pirates can make unauthorized copies and dissemination of them; malicious attackers can easily tamper or forge them, which is easy to cause serious consequences and lead to major economic losses. [0003] For image information security, people have proposed a variety of methods, such as: ① digital image encryption technology that converts plaintext images into ciphertext images; ② digital image ...

AI Technical Summary

Problems solved by technology

Combining multi-level wavelet coefficients to embed watermarks, while obtaining better performance, there are also the following problems: when the image is transformed, the impact on different levels of wavelet coefficients is different, and when the changes in different levels of wavelet coefficients are the same, the impact on the amount of change in the image matrix It is also different. Although we balance the influence of wavelet coefficients at all levels by weighting and adjusting the weighted mean of wavelet coefficients, and balance the invisibility and robustness of the watermarking algorithm, the performance differences of wavelet coefficients at all levels lead to short-board effects, making the extraction The quality of the watermark is degraded, and the anti-attack ability is weakened

Images