High spatial domain concealment method for adversarial sample pair analysis

Abstract

High-level airspace hiding method for resistive sample pair analysis. Information hiding technology is a new research direction in the field of information security in recent years. This technology hides secret information in other carriers so that people cannot detect it, thus ensuring the security of Safety. In the carrier grayscale image with a size of N1×N2, find out the pixels that can be hidden at a high level, convert the hidden secret image with a size of M1×M2 into a binary information flow, and use the logistic chaotic map Judging which bits of these binary information streams are hidden in the carrier image, the Logistic chaotic mapping formula is Zn+1=4Zn(1-Zn), where Zn∈(0,1), if the secret information is the same as the bit information, no Replacement, if not the same, replacement, and the necessary compensation algorithm to control the range of pixel changes within 4 pixels, repeat this process until all information is hidden. The invention is used for information hiding technology.

Description

Technical field: [0001] The invention relates to a method for concealing high-level space, in particular to a method for hiding high-level space that is resistant to sample pair analysis. Background technique: [0002] Information hiding technology is a new research direction in the field of information security in recent years. This technology hides secret information in other carriers so that people can't detect it, thus ensuring the security of secret information. In the image hiding algorithm, according to the classification of embedded domain, it can be divided into spatial domain and transform domain hiding. In the image information hiding technology based on the space domain, the least significant bit (LSB) algorithm has a large hiding capacity and is easy to implement, but the hiding position is unreliable and the robustness is poor, so that many analysis methods for LSB hiding have appeared in recent years. , such as classic SPA (sample pair analysis) and RS analys...

AI Technical Summary

Problems solved by technology

In the image information hiding technology based on the space domain, the least significant bit (LSB) algorithm has a large hiding capacity and is easy to implement, but the hiding position is unreliable and the robustness is poor, so that many analysis methods for LSB hiding have appeared in recent years. , such as classic SPA (sample pair analysis) and RS analysis methods

Then people proposed the HB algorithm based on the most significant bit of the image. Compared with the LSB algorithm, its robustness has been greatly improved, but the hidden amount can only reach 30% of the carrier image; in 2005, Zhu Congxu proposed that each watermark bit It is randomly embedded into a middle bit of the pixel, and the optimization strategy of minimizing the amount of pixel change is adopted, so that the robustness has been greatly improved, but even the optimized pixel value still has a large difference from the original image. Changes will have a certain impact on the vision

Images