Video steganography method, system and device based on macroblock complexity

Abstract

The invention discloses a video steganography method, system and device based on the complexity of macroblocks. The method is specifically to perform normal compression on the video, calculate the complexity of the reconstructed block, and determine the embedding threshold of the steganographic carrier based on the statistical length of embedded secret information. ;According to the embedding threshold, the macroblock motion vector is steganographically, and then the secret information is embedded, and the macroblock is reconstructed according to the steganographic motion vector to obtain the steganographic frame; the reconstruction block of the steganographic frame is looped through, and the reconstruction of the steganographic frame is calculated The complexity of the block is compared with the embedding threshold. If it is less than the embedding threshold, the secret information is extracted from the least significant bit of the horizontal component and the vertical component of the steganography macroblock motion vector; the sender repeats the steganography to obtain the steganography video; the receiver Secret information can be extracted by repeatedly extracting the steganography video received by the terminal. The invention can obviously maintain the local optimality of the motion vector, and it is pointed out through analysis that it is easier to maintain the local optimality by modifying the motion vector of a smooth macroblock.

Description

technical field [0001] The invention belongs to the technical field of image information hiding, and in particular relates to a video steganography method, system and device based on macroblock complexity. Background technique [0002] The video steganography algorithm based on the motion vector has been extensively studied. The idea of ​​using the motion vector as the steganographic carrier was proposed by Kutter et al. A video watermarking method was implemented by directly modifying the motion vector. In 2006, Xu et al. embed information by embedding information in the least significant bits of motion vector components with larger magnitudes. Aly found through research that the selection strategy based on the magnitude of the motion vector cannot guarantee smaller distortion, so a selection strategy directly related to the macroblock prediction error is designed, and the motion vector of the macroblock with a higher prediction error is selected and Information embedding ...

AI Technical Summary

Problems solved by technology

In order to ensure the local optimality of motion vectors, Zhang et al. proposed a local optimality preservation algorithm in 2015, which uses candidate motion vectors that simultaneously satisfy local optimality and are opposite to the parity of the optimal motion vector as the post-steganographic motion vector, since the replaced motion vector may not satisfy the global optimum within the search range, the increase in the residual error will lead to a significant increase in bit

[0003] Chinese patent CN 105704489 A measures the complexity of macroblocks, but it cannot completely describe the difference between adjacent pixels of the reconstruction block. In addition, there are certain defects in the selection method of the threshold, which may lead to incomplete embedding of secret information

[0004] In the existing steganographic algorithm based on motion vector, modifying the motion vector may cause the reconstructed block to not satisfy the local optimality at the decoding end, thus being detected by the steganalysis method based on local optimality hiding, thus exposing the concealed communication fact The security of the algorithm will be greatly reduced, so how to select the appropriate motion vector for information embedding and ensure that the reconstruction block can still maintain the local optimality of the motion vector at the decoding end has a great impact on the security of the steganography algorithm. plays a vital role

Images