CN114900700A - A motion vector histogram correction method and system for video steganography - Google Patents

Abstract

The invention provides a motion vector histogram correction method for video steganography, comprising the following steps: selecting a preset specific frame, and dividing the specific frame into m square areas; The vector direction angle variance, the balance coefficient and the number m of divided areas in a specific frame are obtained, and the correction function is obtained; the above steps are repeated until all the correction functions of the specific frame are obtained; the correction function is divided into n parts according to the preset ratio, and according to the preset Sort order; obtain correction parameters according to the sorting of the correction function; take the horizontal and vertical components of the motion vector as an embedded pair; divide the embedded pairs into multiple sets according to the value of the embedded pair and the correction parameters; The set where the motion vector is located, the corresponding information is extracted, and the value of the motion vector is corrected. This method can effectively improve the embedding capacity and invisibility of video-based steganography algorithms.

Description

A motion vector histogram correction method and system for video steganography

technical field

The invention relates to the technical field of data encryption, in particular to a motion vector histogram correction method and system for video steganography.

Background technique

At present, information has become an important resource for people, and the transmission of information is more important to people's lives. With the development of network technology and the popularization of various multimedia data, only relying on encryption technology can not fully meet the requirements of current data security communication. In order to further enrich the technical means of confidential communication, information hiding technology emerges as the times require. Information hiding technology can embed secret information into a carrier that can be disclosed, so as to hide the behavior of secret communication from third parties. The embedded carrier can be video, image, audio, etc. Among them, digital video has a large amount of data and a complex coding structure, and it is the absolute amount of information that can be embedded in the carrier, so steganalysis is relatively difficult. At the same time, with the popularity of short videos, the information hiding technology based on video has a good performance in practical applications.

Because the video structure is relatively complex, the encoded motion vectors, DCT coefficients, inter-frame and intra-frame prediction modes, and data block modes can all be used as information embedding points. Among them, the use of motion vectors and DCT coefficients to embed information is the current mainstream method. Motion vector data has the characteristics of strong independence, large absolute number and relatively simple structure, so it has the characteristics of controllable security, large embedding capacity and fast embedding speed after embedding information. However, it is prone to error accumulation and motion offset after modification, which greatly affects the visual quality of the video.

Information embedding will reduce video visual quality and video statistical characteristics, especially when the amount of information embedding is too large, it will have a serious impact on video visual quality, and video statistical characteristics will change significantly, affecting the security of covert communication. The reason is that the original coded data and structure will be greatly changed when the information is embedded.

When the steganographer chooses the motion vector-based histogram correction algorithm in covert communication, the large-capacity histogram is designed because the number of small-valued motion vectors is large, and the small-valued motion vectors are often selected for information embedding, resulting in Video visual quality is poor. When the amount of communication is small, and it is no longer necessary to embed too many small-valued motion vectors, the histogram cannot be adjusted to improve the visual quality of the video.

Therefore, it is an urgent problem for those skilled in the art to provide a motion vector histogram correction method and system for video steganography that can effectively improve the embedding capacity and invisibility of the video-based steganography algorithm.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a motion vector histogram correction method and system for video steganography, the method is clear in logic, safe, effective and reliable, and the system has a simple structure and simple operation, and can effectively improve the use of video as a carrier. The embedded capacity and invisibility of the steganography algorithm further improve the efficiency and concealment of steganography.

Based on the above purpose, the technical scheme provided by the present invention is as follows:

A motion vector histogram correction method for video steganography, comprising the following steps:

Selecting a preset specific frame, and dividing the specific frame into m square areas;

Obtain a correction function according to the average value of the motion vector value in the square area, the variance of the direction angle of the motion vector, the balance coefficient and the number m of divided areas in the specific frame;

Repeat the above steps until all correction functions of the specific frame are obtained;

Divide all the correction functions of the specific frame into n parts according to a preset ratio, and sort them according to a preset order;

Obtain the correction parameters according to the order of all correction functions in a specific frame;

Using the horizontal component and the vertical component of the motion vector as an embedded pair;

dividing the embedded pairs into multiple sets according to the embedded pair value and the correction parameter;

According to the set of motion vectors obtained after video decoding, corresponding information is extracted, and the value of the motion vector is corrected.

Preferably, obtaining the average value of the motion vector value includes the following steps:

Obtain the motion vector value of the i-th macroblock according to the horizontal component and the vertical component of the motion vector;

According to the motion vector value and the total number of motion vectors, the average value of the motion vector value is obtained.

Preferably, obtaining the direction angle variance of the motion vector includes the following steps:

According to the total number of the motion vectors and the motion vector direction angle of the i-th macroblock, obtain the average value of the motion vector direction angle;

The variance of the direction angle of the motion vector is obtained according to the direction angle of the motion vector of the i-th macroblock and the average value of the direction angle of the motion vector.

Preferably,

The balance coefficient is used to balance the proportion of the velocity and direction characteristics of the motion vector in the correction function.

Preferably, the specific formula for obtaining the correction function is:

Among them, H(p) is the correction function of the p-th specific frame, n is the balance coefficient, and m is the number of divided regions in the reference frame.

Preferably,

The sorting according to the preset order is specifically sorting from large to small.

Preferably, all the correction functions of the specific frame are divided into n parts according to a preset ratio, and they are sorted in descending order. According to the sorting of all the correction functions of the specific frame, the correction parameters obtained are specifically:

When n=3, all the correction functions of the specific frame are divided into three parts according to the preset ratio, and they are sorted in descending order, which are h(1), h(2), and h(3);

The specific formula of the correction parameter (k, j) is:

Among them, k is the horizontal component of the motion vector, and j is the vertical component of the motion vector.

A motion vector histogram correction system for video steganography, comprising,

The selection module is used to select a specific frame;

a first dividing module, configured to divide the specific frame into m square areas;

The first calculation module is used to calculate and obtain the correction function according to the average value of the motion vector value in the square area, the method of the direction angle of the motion vector, the balance coefficient and the number m of the specific frame division area;

an arrangement module, configured to divide all the correction functions of the specific frame into n parts according to a preset ratio, and sort them according to a preset order;

The second calculation module is used to calculate and obtain correction parameters according to the sorting of all correction functions of the specific frame;

The second division module is used to divide the embedded pairs into multiple sets according to the embedded pair value and the correction parameter;

an extraction module, used for extracting corresponding information according to the set where the motion vector obtained after video decoding is located;

The correction module is used for correcting the motion vector according to the set of the motion vector obtained after the video is decoded.

In the motion vector histogram correction method for video steganography provided by the present invention, the sender first selects a video as an embedded carrier, and selects a preset specific frame in the video, and after selecting the specific frame, the specific frame is selected. Divide into m square areas; obtain the average value of the motion vector value and the variance of the direction angle of the motion vector in a square area, and combine the balance coefficient and the number m of spending areas in a specific frame to calculate and obtain the correction function; repeat the above two step, until all the correction functions of the specific frame are obtained; all the correction functions of the specific frame are divided into n parts according to the preset ratio, and are sorted according to the preset order; according to the corrected functions after the arrangement, calculate and obtain the correction parameters; then Taking the horizontal component and vertical component in the motion vector as an embedded pair, and combining the obtained correction parameters, divide the embedded pair into multiple non-overlapping sets; The set fits together, extracts the corresponding information, and recovers the motion vector. Since in the video, the data of several adjacent frames have great correlation, it can be considered that the current frame can represent certain characteristics of several subsequent frames. The speed and direction characteristics of the motion vector are used to design the correction parameters for the subsequent frames. In order to expand the amount of information embedding, a two-dimensional histogram is used for embedding, and the two components of the motion vector, that is, the horizontal component and the vertical component, are regarded as an embedding pair, and these embedding pairs are divided into multiple non-overlapping pairs according to the embedding pair value and the correction parameter. Different sets correspond to different embedding algorithms, and the value of the embedding pair is modified according to the corresponding algorithm to realize the embedding of information in the motion vector, which can effectively improve the embedding capacity and invisibility of the video-based steganography algorithm.

Description of drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without any creative effort.

1 is a flowchart of a method for correcting a motion vector histogram for video steganography according to an embodiment of the present invention;

FIG. 2 is a flowchart of obtaining an average value of a motion vector value provided by an embodiment of the present invention;

Fig. 3 is the flow chart of obtaining the direction angle variance of motion vector provided by the embodiment of the present invention;

FIG. 4 is a structural diagram of a motion vector histogram correction system for video steganography according to an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

The embodiments of the present invention are written in a progressive manner.

The embodiment of the present invention provides a motion vector histogram correction method for video steganography. It mainly solves the problem that in the prior art, the small value of the motion vector is used for information embedding, resulting in poor video visual quality. When the communication volume is small and information embedding of too many small-valued motion vectors is no longer necessary, it is also a technical problem that the histogram cannot be adjusted to improve the visual quality of the video.

A motion vector histogram correction method for video steganography, comprising the following steps:

S1. Select a preset specific frame, and divide the specific frame into m square areas;

S2. Obtain a correction function according to the average value of the motion vector value in the square area, the variance of the direction angle of the motion vector, the balance coefficient and the number m of divided areas in a specific frame;

S3. Repeat the above steps until all correction functions of a specific frame are obtained;

S4. Divide all the correction functions of a specific frame into n parts according to a preset ratio, and sort them according to the preset order;

S5. According to the sorting of all correction functions of a specific frame, the correction parameters are obtained;

S6. Use the horizontal component and the vertical component of the motion vector as an embedded pair;

S7. According to the embedded pair value and the correction parameter, divide the embedded pair into multiple sets;

S8. According to the set where the motion vector obtained after the video is decoded, the corresponding information is extracted, and the value of the motion vector is corrected.

In step S1, a specific frame of the input video is selected by referring to a frame transformation technique, and the frame is divided into M×M regions.

In step S2, the correction function H(p) is calculated and obtained according to the average value of the motion vector value, the variance of the direction angle of the motion vector, the balance coefficient and the number m of divided regions in a specific frame in one of the square regions.

In step S3, the above steps are repeated until all correction functions H(p) of a specific frame are obtained.

In step S4, all the correction functions H(p) of a specific frame are divided into n parts according to a preset ratio, and are sorted according to the preset order. In this embodiment, n=3.

In step S5, after sorting, the correction parameters (k, j) are obtained.

In step S6, the horizontal component and the vertical component of the motion vector are used as an embedded pair.

In step S7, these embedded pairs are divided into multiple non-overlapping sets A according to the embedded pair value and the correction parameter (k, j).

In step S8, according to the set where the motion vector obtained after decoding the video is located, the corresponding information is extracted and the value of the embedded pair (that is, the value of the motion vector) is corrected.

The present invention designs the correction parameters of several subsequent frames by counting the speed and direction characteristics of the motion vector in a specific frame. In order to expand the amount of information embedding, a two-dimensional histogram is used for embedding, and the two components of the motion vector, that is, the horizontal component and the vertical component, are regarded as an embedding pair, and these embedding pairs are divided into multiple non-overlapping pairs according to the embedding pair value and the correction parameter. Different sets correspond to different embedding algorithms, and the value of the embedding pair is modified according to the corresponding algorithm to realize the embedding of information in the motion vector, which can effectively improve the embedding capacity and invisibility of the video-based steganography algorithm.

Preferably, obtaining the average value of the motion vector value includes the following steps:

A1. According to the horizontal component and the vertical component of the motion vector, obtain the motion vector value of the i-th macroblock;

A2. Obtain the average value of the motion vector values according to the motion vector value and the total number of motion vectors.

In step A1, the specific formula for obtaining the motion vector value of the i-th macroblock is:

where L(i) represents the size of the motion vector of the i-th macroblock in the region, and mv _x and mv _y represent the horizontal and vertical components of the motion vector, respectively.

In step A2, the specific formula for obtaining the average value of the motion vector value is:

表示该帧内第q个区域的运动矢量大小的平均值，n表示该区域内运动矢量的总个数。i的取值范围从1到n。in,

Represents the average value of the motion vector size of the qth region in the frame, and n represents the total number of motion vectors in the region. The value of i ranges from 1 to n.

Preferably, obtaining the direction angle variance of the motion vector includes the following steps:

B1. According to the total number of motion vectors and the motion vector direction angle of the i-th macroblock, obtain the average value of the motion vector direction angle;

B2. Obtain the variance of the direction angle of the motion vector according to the direction angle of the motion vector of the ith macroblock and the average value of the direction angle of the motion vector.

In step B1, the specific formula for obtaining the average value of the direction angle of the motion vector is:

表示该帧内第q个区域的运动矢量方向角的平均值，θ(i)表示该区域内第i个宏块的运动矢量方向角。in

represents the average value of the direction angle of the motion vector of the qth region in the frame, and θ(i) represents the direction angle of the motion vector of the ith macroblock in the region.

In step B2, the specific formula for obtaining the variance of the direction angle of the motion vector is:

where D(q) represents the variance of the direction angle of the motion vector of the qth region in the frame.

Preferably,

The balance factor is used to balance the proportion of the velocity and direction characteristics of the motion vector in the correction function.

In the actual application process, the balance coefficient η is used to balance the proportion of the velocity and direction characteristics of the motion vector in the correction function

Preferably, the specific formula for obtaining the correction function is:

Among them, H(p) is the correction function of the p-th specific frame, n is the balance coefficient, and m is the number of divided regions in the reference frame.

Preferably,

Sorting according to the preset order is in descending order.

In the actual application process, the preset order can also be selected from various orders. In this embodiment, the order from large to small is selected for the convenience of calculation.

Preferably, all the correction functions of a specific frame are divided into n parts according to a preset ratio, and they are sorted in descending order. According to the sorting of all the correction functions of the specific frame, the correction parameters obtained are specifically:

When n=3, divide all the correction functions of a specific frame into 3 parts according to the preset ratio, and sort them in descending order, which are h(1), h(2), h(3);

The specific formula of the correction parameter (k, j) is:

Among them, k is the horizontal component of the motion vector, and j is the vertical component of the motion vector.

In the actual application process, in step S7, according to the embedded pair value and the correction parameter (k, j), these embedded pairs are divided into a plurality of non-overlapping sets A. The set A is as follows:

Wherein, B ₁ and B ₂ are 4 times of the horizontal and vertical components of the motion vector, respectively.

In step S8, according to the set where the motion vector obtained after decoding the video is located, the value of the embedded pair is corrected as follows:

(1) When (B ₁ , B ₂ )∈A ₁ :

Wherein, B' ₁ and B' ₂ represent the modified B ₁ and B ₂ respectively, and m _t and m _t+1 represent the secret information of the t-th bit and the t+1-th bit, respectively.

(2) When (B ₁ , B ₂ )∈A ₂ :

(3) When (B ₁ , B ₂ )∈A ₃ :

(4) When (B ₁ , B ₂ )∈A ₄ :

(5) When (B ₁ , B ₂ )∈A ₅ :

(6) When (B ₁ , B ₂ )∈A ₆ :

(7) When (B ₁ , B ₂ )∈A ₇ :

(8) When (B ₁ , B ₂ )∈A ₈ :

(9) When (B ₁ , B ₂ )∈A ₉ :

(10) When (B ₁ ,B ₂ )∈A ₁₀ ∪A ₁₁ ∪A ₁₂ ∪A ₁₃ ∪A ₁₄ ∪A ₁₅ ∪A ₁₆ ∪A ₁₇ ∪A ₁₈ , only make corrections:

When the corresponding information needs to be extracted, the formed set A is specifically:

Wherein, B ₁ and B ₂ are 4 times of the horizontal and vertical components of the motion vector, respectively.

In step S8, according to the set where the obtained motion vector is located after decoding the video, the corresponding algorithm is used to extract information, and the motion vector is restored to before the modification, specifically:

(1). When (B ₁ ,B ₂ )∈A ₉ ∪A ₁₀ ∪A ₁₃ ∪A ₂₂ , extract information m _t =0:

(2). When (B ₁ ,B ₂ )∈A ₁₁ ∪A ₁₂ ∪A ₂₅ , extract information m _t =1:

(3). When (B ₁ ,B ₂ )∈A ₁ ∪A ₂ ∪A ₃ ∪A ₄ ∪A ₂₁ , extract information m _t m _t+1 =00:

(4). When (B ₁ ,B ₂ )∈A ₅ ∪A ₆ ∪A ₇ ∪A ₈ ∪A ₁₉ , extract the information m _t m _t+1 =01:

(5). When (B ₁ ,B ₂ )∈A ₁₄ ∪A ₁₆ ∪A ₂₀ ∪A ₂₄ , extract information m _t m _t+1 =10:

(6). When (B ₁ ,B ₂ )∈A ₁₅ ∪A ₁₇ ∪A ₁₈ ∪A ₂₃ , extract information m _t m _t+1 =11:

(7). When (B ₁ ,B ₂ )∈A ₂₆ ∪A ₂₇ ∪A ₂₈ ∪A ₂₉ ∪A ₃₀ ∪A ₃₁ ∪A ₃₂ ∪A ₃₃ ∪A ₃₄ , only make corrections:

A motion vector histogram correction system for video steganography, comprising,

The selection module is used to select a specific frame;

a first division module, for dividing a specific frame into m square areas;

The first calculation module is used to calculate and obtain the correction function according to the average value of the motion vector value in the square area, the method of the direction angle of the motion vector, the balance coefficient and the number m of the specific frame division area;

The arrangement module is used to divide all the correction functions of a specific frame into n parts according to a preset ratio, and sort them according to the preset order;

The second calculation module is used to calculate and obtain the correction parameters according to the sorting of all correction functions of a specific frame;

The second division module is used to divide the embedded pairs into multiple sets according to the embedded pair value and the correction parameter;

an extraction module, used for extracting corresponding information according to the set where the motion vector obtained after video decoding is located;

The correction module is used for correcting the motion vector according to the set of the motion vector obtained after the video is decoded.

In the actual application process, the selection module selects a specific frame from the video, and the specific frame is divided into m square areas by the first division module; The angle method, the balance coefficient and the number m of divided areas of a specific frame are calculated to obtain the correction function; the first calculation module obtains all the correction functions of the specific frame after multiple calculations; the arrangement module divides all the correction functions of the specific frame into n parts , and sort according to the preset order; according to all the correction functions of the sorted specific frame, the second calculation module calculates and obtains the correction parameters; the second division module divides the embedded pairs into multiple sets according to the embedded pair values and the correction parameters; video After decoding, the set where the motion vector is located is obtained, and the extraction module extracts corresponding information according to the set; the correction module corrects the motion vector according to the set.

In the embodiments provided in this application, it should be understood that the disclosed method and apparatus may be implemented in other manners. The device embodiments described above are only illustrative. For example, the division of modules is only a logical function division. In actual implementation, there may be other division methods, for example, multiple modules or components may be combined or integrated. to another system, or some features can be ignored, or not implemented. In addition, the coupling, or direct coupling, or communication connection between the various components shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or modules may be electrical, mechanical or other forms. of.

In addition, each functional module in each embodiment of the present invention may all be integrated into one processor, or each module may be used as a separate device, or two or more modules may be integrated into one device; the present invention Each functional module in each embodiment may be implemented in the form of hardware, or may be implemented in the form of hardware plus software functional units.

Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be accomplished through program instructions and related hardware, and the aforementioned program instructions can be stored in a computer-readable storage medium, and when the program instructions are executed , perform the steps including the above method embodiments; and the aforementioned storage medium includes: a removable storage device, a read only memory (Read Only Memory, ROM), a magnetic disk or an optical disk and other media that can store program codes.

It should also be noted that in this document, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply those entities or operations There is no such actual relationship or order between them. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device that includes a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The method and system for correcting a motion vector histogram for video steganography provided by the present invention have been described in detail above. The above description of the disclosed embodiments enables any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (8)

1. A motion vector histogram modification method for video steganography, comprising the steps of:

selecting a preset specific frame, and dividing the specific frame into m square areas;

obtaining a correction function according to the motion vector value average value in the square area, the motion vector direction angle variance, the balance coefficient and the number m of the divided areas in the specific frame;

repeating the steps until all correction functions of the specific frame are obtained;

dividing all correction functions of the specific frame into n parts according to a preset proportion, and sequencing according to a preset sequence;

acquiring correction parameters according to the sequence of all correction functions of a specific frame;

taking the horizontal component and the vertical component of the motion vector as an embedding pair;

dividing the embedded pairs into a plurality of sets according to embedded pair values and the correction parameters;

and extracting corresponding information according to a set of motion vectors acquired after video decoding, and correcting the values of the motion vectors.

2. The motion vector histogram modification method for video steganography as described in claim 1, wherein said obtaining said motion vector value mean value comprises the steps of:

acquiring a motion vector value of the ith macro block according to the horizontal component and the vertical component of the motion vector;

and obtaining the average value of the motion vector values according to the motion vector values and the total number of the motion vectors.

3. The method of motion vector histogram modification for video steganography as described in claim 2, wherein said obtaining said motion vector direction angle variance comprises the steps of:

acquiring a motion vector direction angle average value according to the total number of the motion vectors and the motion vector direction angle of the ith macro block;

and obtaining the variance of the direction angle of the motion vector according to the direction angle of the motion vector of the ith macro block and the average value of the direction angle of the motion vector.

4. The method of motion vector histogram modification for video steganography as claimed in claim 2,

the balance coefficients are used to balance the ratio of the speed and direction characteristics of the motion vectors in the correction function.

5. The method of claim 4, wherein the specific formula of the obtained modification function is:

where h (p) is the modification function of the pth specific frame, η is the balance coefficient, and m is the number of divided regions in the reference frame.

6. The motion vector histogram modification method for video steganography as set forth in claim 5,

the sorting according to the preset sequence is specifically sorting from big to small.

7. The method according to claim 6, wherein the step of dividing all the modification functions of the specific frame into n parts according to a predetermined ratio, and sorting the n parts from large to small, and obtaining the modification parameters according to the sorting of all the modification functions of the specific frame specifically comprises:

when n is 3, dividing all correction functions of the specific frame into 3 parts according to a preset proportion, and sequencing the correction functions from large to small, wherein the correction functions are h (1), h (2) and h (3);

the specific formula of the correction parameter (k, j) is as follows:

where k is the horizontal component of the motion vector and j is the vertical component of the motion vector.

8. A motion vector histogram modification system for video steganography, comprising,

the selection module is used for selecting a specific frame;

a first dividing module for dividing the specific frame into m square regions;

the first calculation module is used for calculating and obtaining a correction function according to the motion vector value average value in the square area, the motion vector direction angle method, the balance coefficient and the number m of the specific frame division areas;

the arrangement module is used for dividing all correction functions of the specific frame into n parts according to a preset proportion and sequencing the n parts according to a preset sequence;

the second calculation module is used for calculating and acquiring correction parameters according to the sequence of all correction functions of the specific frame;

the second dividing module is used for dividing the embedding pair into a plurality of sets according to the embedding pair value and the correction parameter;

the extraction module is used for extracting corresponding information according to a set where the motion vector obtained after video decoding is located;

and the correction module is used for correcting the motion vector according to the set of the motion vector acquired after the video decoding.

Images