Constructive information hiding method based on automatic contour generation

Abstract

The invention discloses a constructive information hiding method based on automatic contour generation. The method specifically comprises the following steps: segmenting a binary sequence S of secret information with the length of L into n binary sequences with the lengths of l; 2, constructing a target contour image; 3, constructing a contour-image reversible transformation model; and 4, inputting the target contour image into the trained contour-image reversible transformation model, generating a secret-containing image by the generation model according to the target contour image, and extracting the contour of the secret-containing image by the extraction model so as to hide the binary sequence of the secret information in the contour extracted by the extraction model. The generated secret-containing image is high in quality, and the visual effect is better.

Description

technical field [0001] The invention belongs to the field of information security. Background technique [0002] For a long time, information security has been an issue that people pay close attention to. The traditional method based on cryptography encrypts and retransmits information, but the encrypted information is disordered and therefore vulnerable to attacks by monitors. In contrast, steganography is an information hiding technology that embeds secret information in multimedia carriers (such as text, images, audio, video, etc.) for covert communication. security. Steganography has important application value in information transmission and other fields. [0003] In the previous research, researchers often adopt the traditional carrier modification information hiding method, design a reasonable loss function, and modify the carrier image according to certain rules to realize the embedding of secret information. To mitigate the impact of modification on image qualit...

AI Technical Summary

Problems solved by technology

[0005] 1. The existing carrier-constructed information hiding method uses GAN to convert secret information into a dense image, but since the essence of GAN is not a reversible Markov model, it is difficult to extract secret information

[0006] 2. In order to avoid mode collapse, the existing carrier-based information hiding methods usually construct a mapping relationship between one or a type of picture and one bit of secret information, which has the disadvantage of small hidden capacity

Images