Direction changing difference expansion and synchronous embedding reversible watermark embedding and extraction method

Abstract

The invention provides a direction changing difference expansion and synchronous embedding reversible watermark embedding and extraction method. Embedding capacity and embedding data are enabled to be unrelated by adopting direction changing difference expansion and the threshold is enabled to select the maximum embedding capacity; and the classification accuracy and the maximum embedding capacity are enhanced by adopting gradient and direct classification for avoiding normalized gradient and reduction of the classification accuracy. A compressed location graph is applied so that large consumption of the embedding capacity and excessive additional data caused by recording overflow pixel location of a generic difference expansion and local complexity reversible watermark method can be avoided. The embedding data are ensured to be completely reversible by giving the embedding strategy of synchronous addition of backup data for avoiding irreversibility caused by direct embedding of the additional data after embedding of load data, and an embedding parameter selection method based on ordering and enumeration is also given so as to reduce computational complexity. Compared with the reversible watermark method based on generic difference expansion and local complexity, the method is completely reversible and parameter selection time is greatly reduced so that the method has larger maximum embedding capacity.

Description

technical field [0001] The invention belongs to the cross field of image information security and digital image signal processing, relates to a reversible watermark embedding and extraction method, in particular to a reversible watermark embedding and extraction method of variable direction difference expansion and synchronous embedding. Background technique [0002] Reversible watermark refers to a special type of watermark that can be completely recovered after the watermark is extracted. Compared with traditional watermarking, reversible watermarking has strict requirements for lossless recovery of embedded carriers, and is generally used for distortion-free protection of important images, and has important application value in military images, medical images and remote sensing images. [0003] The difference-extended reversible watermarking method proposed by Tian et al. is a typical method of image reversible watermarking. This method performs Haar integer wavelet trans...

AI Technical Summary

Problems solved by technology

[0042] ① The expected value of the capacity does not represent the actual capacity. According to formula (26), if t p +1 is not a power of 2, for any e, only if can be embedded data, otherwise only embedded data, but the position where the data is embedded cannot be determined before the capacity is determined, so for any embeddable pixel, its actual capacity is instead of log 2 (t p +1);

[0043] ② In formula (26), the preset threshold T is related to the size of the difference expansion area and the translation area. Since there is no linear relationship between T and the maximum embedding capacity, in order to obtain a larger embedding capacity, T should be determined by the parameter selection method instead of Preset, while the selection of T is related to all categories of pixels, it cannot guarantee that all categories of pixels provide the maximum embedding capacity;

[0044] ③According to Equation (27) and Equation (28), the normalized local complexity determines the affected pixel category, but the local complexity calculated according to Equation (28) is the gradient sum after normalization, and different gradient sums may be the same after normalization, so the classification The accuracy is lower than the case of directly using the gradient sum as the local complexity, which will affect the maximum capacity and the visual quality of the embedding carrier;

[0045] ④ From formula (29) and formula (30), the process of determining the ideal parameter set does not consider the influence of overflow pixels, and it is difficult to ensure that the selected ideal embedding parameters still have a large real capacity;

[0046] ⑤Record overflow pixels according to the recording position, each overflow pixel will consume Bit capacity, when there are many overflow pixels, it will have a greater impact on the maximum embedding capacity;

[0047] ⑥In order to obtain a larger embedding capacity, L usually takes 3, from 0≤s L ≤...≤s 1 ≤256, 01 L L , in order to determine Φ, a large number of parameter sets need to be enumerated, and equations (29) and (30) need to be calculated multiple times. The algorithm complexity of direct brute force enumeration is too high and requires a long running time;

[0048] ⑦ After embedding of payload data β is completed and backup data ξ is obtained, if new overflow pixels are encountered during the process of embedding backup data ξ, these overflows cannot be recorded, resulting in the existence of unbacked least significant bits, so this method is irreversible;

Images