CN115278266B - Method for embedding, transmitting and extracting secret information in digital image - Google Patents

Abstract

The present invention relates to a method of embedding, transmitting and extracting secret information in a digital image. At the sender, the digital image is decomposed into a single bit matrix, then whether the lowest bit information of the corresponding position in the original image is replaced by the data information to be embedded is judged according to the value of the evaluation function, after all the data information to be embedded is embedded into the lowest bit matrix, the data information to be embedded is synthesized with the high bit matrix and is transmitted to the receiver through a public channel, and meanwhile, the content of the evaluation function is transmitted to the receiver through a secret channel. At the receiving side, the receiving side extracts secret information from the received image by the evaluation function. The technology is simple in implementation mode, can realize that the image embedded with secret information is mixed with common image transmission, and has good application potential.

Description

Method for embedding, transmitting and extracting secret information in digital image

Technical Field

The invention relates to the field of information security, in particular to a method for embedding, transmitting and extracting secret information in a digital image.

Background

In the present internet age, digital images are an important carrier for transmitting information, and an unlimited number of information can be transmitted through the image content. Meanwhile, in many application scenarios, there is a large amount of secret information that needs to be transmitted. One straightforward way of transmission is to encrypt secret information in some form directly and then transmit it through public channels, but the disadvantage of encrypted transmission is that often the encrypted data file has obvious statistical features, for example, the gray histograms of the encrypted images are often uniformly distributed, so that the detection program can identify and intercept these ciphertext information through various statistical features and then crack it. Alternatively, the secret information is embedded in some form into another common and public data file for transmission, which has the advantage that the carrier file embedded with the secret information is similar to the common file in statistics, is difficult to identify in a large number of files, and avoids the risk of being cracked. The two methods play respective advantages in different application scenes.

The common information recording and transmission can be regarded as "white paper communication", i.e. the information to be transmitted is written into a blank file and then transmitted. Secret information embedding technology can be regarded as "dirty paper communication", i.e., writing information to be transmitted into a file in which the information has been recorded, and thus the written paper can be regarded as "dirty".

Dirty paper communication technology is one of important issues in the field of digital information security, and how to improve the concealment of information embedding, improve the calculation speed and the information extraction method is the key point of related research.

Disclosure of Invention

The invention aims to solve the problem of providing a method for embedding, transmitting and extracting secret information in a digital image, which has the advantages of high processing speed, high concealment of embedded information, larger embedded information load and smaller image distortion degree.

The technical scheme adopted for solving the technical problems is as follows: a method of embedding, transmitting and extracting secret information in a digital image, comprising the steps of:

(1) The sender converts the secret information into a binary bit stream S, the information length of which is T.

(2) The sender selects a gray image I (x, y) as the embedding carrier, x and y being pixel coordinates, the pixel size being M x N, the gray value bit depth being B.

(3) Taking the lowest bit matrix H1 (x, y) of the image I (x, y), taking the high bit matrix Hr (x, y) of the image I (x, y) except the lowest bit matrix.

(4) Defining a local evaluation function F (p, q),

，（1）

wherein E represents the pixel mean of the image part, defining x0= [2, (M-1) ], y0= [2, (N-1) ], defining p0= [ (X-1), (x+1) ], q0= [ (Y-1), (y+1) ].

(5) Traversing all pixel coordinates in < X, Y > ∈x0×y0, calculating the value of the local evaluation function F (P, Q), where < P, Q > ∈p0×q0; if F (p, q) > = 0, the value of S (t) is used to replace the value of H1 (x, y), and t+1; if F (p, q) <0, the value of H1 (x, y) is retained.

(6) Until all values in S are replaced with H1 (x, y).

(7) The Hr (x, y) matrix and the H1 (x, y) matrix are recombined into the gray image Ir (x, y).

(8) The sender transmits the content of the evaluation function F (p, q) to the receiver through the secret channel; and Ir (x, y) is transmitted to the receiver through the common channel.

(9) The receiver decomposes Ir (x, y) into a lowest bit matrix HH1 (x, y) and a high bit matrix HHr (x, y).

(10) The secret information bit stream Sr is fetched from HH1 (x, y) according to the local evaluation function F (p, q).

The invention has the technical advantages that the secret information to be transmitted can be embedded into the common digital image, thereby realizing the function of secret transmission or digital watermarking, the secret image is very close to the common non-secret image, so that the secret image is not easy to be recognized by human sense organs, other non-secret images can be mixed through a public channel for transmission, and meanwhile, an unauthorized third party can not correctly acquire the secret information even if the image is suspected to contain the secret information due to the fact that the evaluation function is transmitted through the secret channel, and the secret information can not be acquired under the condition that the evaluation function is not acquired.

Drawings

The invention will be further described with reference to the drawings and examples.

Fig. 1 is a flowchart of secret information embedding, transmission and extraction according to the method of the present invention.

Fig. 2 shows a gray-scale image I of a carrier selected by a sender according to the invention.

Fig. 3 shows a least significant bit image H1 extracted from image I according to the present invention.

Fig. 4 is a lowest bit image H1 after embedding secret information according to the present invention.

Fig. 5 shows a gray scale image Ir recombined after embedding secret information according to the present invention.

Detailed Description

So that the manner in which the above recited objects, features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized below, may be had by reference to the appended drawings.

Examples:

(1) The sender secret information binary bit stream is S, and the information length T is 90000 bits.

(2) The carrier gray image I (x, y) selected by the sender, as shown in fig. 1, x and y are pixel coordinates, the pixel size is 500×480, and the gray value bit depth b=8.

(3) The lowest bit matrix H1 (x, y) of the image I (x, y) is taken, and as shown in fig. 3, the high bit matrix Hr (x, y) of the image I (x, y) other than the lowest bit matrix is taken.

(4) The local evaluation function isWhere E represents the image pixel mean, defining x0= [2,499 ]]，Y0=[2,479]Definition p0= [ (x-1), (x+1)]，Q0=[(y-1),(y+1)]。

(5) Traversing all pixel coordinates in < X, Y > ∈x0×y0, calculating the value of the local evaluation function F (P, Q), where < P, Q > ∈p0×q0; if F (p, q) > = 0, the value of S (t) is used to replace the value of H1 (x, y), and t+1; if F (p, q) <0, the value of H1 (x, y) is retained.

(6) Until all values in S are replaced with H1 (x, y).

(7) The Hr (x, y) matrix and the H1 (x, y) matrix are recombined into a gray image Ir (x, y), as shown in fig. 5.

(8) The sender transmits the content of the evaluation function F (p, q) to the receiver through the secret channel; and Ir (x, y) is transmitted to the receiver through the common channel.

(9) The receiver decomposes Ir (x, y) into a lowest bit matrix HH1 (x, y) and a high bit matrix HHr (x, y).

(10) The secret information bit stream Sr is fetched from HH1 (x, y) according to the local evaluation function F (p, q).

Claims (1)

1. A method of embedding, transmitting and extracting secret information in a digital image, characterized by: the method comprises the following steps:

(1) The sender converts the secret information into a binary bit stream S (T), the information length of which is T,

t＝1,2,…,T；

(2) The sender selects a gray image I (x, y) as an embedding carrier, wherein x and y are pixel coordinates, the pixel size is M x N, and the gray value bit depth is B;

(3) Taking the lowest bit matrix H1 (x, y) of the image I (x, y), and taking the high bit matrix Hr (x, y) of the image I (x, y) except the lowest bit matrix;

(4) Defining a local evaluation function F (p, q),

wherein E represents the pixel mean of the image part, defining x0= [2, (M-1) ], y0= [2, (N-1) ], defining p0= [ (X-1), (x+1) ], q0= [ (Y-1), (y+1) ], < P, Q > ∈p0×q0;

(5) Traversing all pixel coordinates in < X, Y > -X0X Y0, calculating the value of the local evaluation function F (p, q); if F (p, q) > = 0, the value of S (t) is used to replace the value of H1 (x, y) and t is added to 1; if F (p, q) <0, the value of H1 (x, y) is retained;

(6) Until all values in S (t) are replaced with H1 (x, y);

(7) Recombining the Hr (x, y) matrix and the H1 (x, y) matrix into a gray image Ir (x, y);

(8) The sender transmits the content of the evaluation function F (p, q) to the receiver through the secret channel; and transmitting Ir (x, y) to the receiver through the common channel;

(9) The receiver decomposes Ir (x, y) into a lowest bit matrix HH1 (x, y) and a high bit matrix HHr (x, y);

(10) The secret information bit stream Sr is fetched from HH1 (x, y) according to the local evaluation function F (p, q).