Decoding and coding network steganography based on improved attention and loss function

Abstract

The invention relates to a computer vision and image processing technology, in particular to a decoding and coding network steganography based on an improved attention and loss function, which comprises the following steps of: extracting a two-dimensional image feature of a secret image and an attention mask of a container image; splicing the extracted two-dimensional image features and attention masks at a channel layer and a container image, and inputting the spliced two-dimensional image features and attention masks into a coding network to obtain a secret-carrying image; inputting the secret-carrying image into a decoding network, and recovering a reconstructed secret image; inputting the container image into a decoding network to obtain a generated secret image; through a composite function constructed based on a pixel value mean square error and image multi-scale structure similarity, constructing an overall loss function considering the similarity between a container image and a secret carrying image, the similarity between a secret image and a reconstructed secret image and the difference between the reconstructed secret image and a generated secret image, and training a network model; and generating a secret carrying image and a reconstructed secret image. The steganography improves the imperceptibility and robustness of the secret-carrying image.

Description

technical field [0001] The invention belongs to the technical field of computer vision and image processing, in particular to a decoding and encoding network steganography based on improved attention and loss functions. Background technique [0002] In the information age, it is necessary for both individuals and nations to transmit and receive confidential information securely over the Internet. In the field of information security, there are two main research directions, namely cryptography and steganography. Cryptography protects the content of information through the incomprehensibility of ciphertext. Its main task is to allow only the sender and target receiver of the message to view its transmission content, and to encode information to achieve information concealment, but the incomprehensibility of cryptography It also exposes the importance of information. Steganography is to protect the content of information through the imperceptibility of ciphertext, by embeddin...

AI Technical Summary

Problems solved by technology

First, because its loss function is only the mean square error loss function of calculating the distance pixel by pixel, the generated image has differences in overall brightness, contrast and resolution compared with the original image; second, the reconstructed secret image The secret information will be disturbed by the container image information; third, the location of the secret hiding will not be selected according to the characteristics of the container image itself, thus leading to the most fatal problem in steganography: the secret information is basically evenly embedded in the channel of the container image corresponding to Once the stealer obtains the original container image, the approximate shape and basic information of the secret image can be obtained by calculating the residual value between the secret image and the container image

Images