WO2019242254A1 - Image area copy detection method supporting privacy protection function - Google Patents

Abstract

An image area copy detection method supporting a privacy protection function, comprising: a user side and N independent cloud server sides. Part of the server sides is used for ciphertext storage, and the other part is used for ciphertext calculation; the user side splits an image into a plurality of ciphertext by using an image encryption algorithm, and the ciphertext is respectively transmitted to the server sides for ciphertext storage; the server sides for ciphertext storage and the server sides for ciphertext calculation complete detection and positioning of a copy tampering operation of an image area under a ciphertext space by means of interaction and calculation; finally, each server side for ciphertext storage obtains a suspected tampering area, and the server side for ciphertext storage sends the result to the user side. The cloud server side provides an effective area copy evidence obtaining service on the premise that the image content is not known, and detection and positioning of the tampering operation are realized. With respect to a conventional area copy evidence obtaining tool or service, the method can provide privacy protection for the image content of a user.

Description

Image area copy detection method supporting privacy protection function Technical field

The present invention relates to the field of digital image forensics, and in particular, to a method for detecting copying of an image area supporting a privacy protection function.

Background technique

Area copy is the most common image tampering technique to copy part of the image to other locations to cover some objects in the image or emphasize some content. Forensics of this type of image tampering usually requires finding multiple highly similar areas in the image, and the similarity of these areas will obviously exceed the normal natural image as evidence of tampering and locate the tampered area. Outsourcing regional forensics to the cloud, and cloud servers to provide forensic services is a cost-effective solution, but it also brings the risk of privacy leakage. The image data that users need to forensic often contain sensitive content, and there is often an interest relationship between users and image content. This information is usually not intended to be leaked to digital forensics cloud service providers. On the other hand, the sharing of computing resources on cloud platforms has exacerbated the security risks of data leakage. However, the existing image area copying, tampering and forensics operations can only be performed in plain text, and therefore face the risk of image content leakage. How to protect the security and privacy of image data in regional copy forensics outsourcing services has become an important factor in determining whether image forensics outsourcing can be practically applied.

Summary of the Invention

The purpose of the present invention is to overcome the shortcomings and shortcomings of the prior art, and to provide an image area copy detection method that supports the privacy protection function. The user terminal can send the image in cipher text to the cloud server. The cloud server does not know the image. Under the premise of the content, it provides effective regional copying and forensics services to detect and locate tampering operations. Compared with the traditional area copying forensics tool or service, the present invention can provide privacy protection of user image content.

The object of the present invention is achieved by the following technical solution: a method for detecting the copy of an image area supporting a privacy protection function, comprising: a user terminal, N mutually independent cloud server terminals, some of which are used for ciphertext storage, and some services The end is used for ciphertext calculation; the user end uses the image encryption algorithm to split the image into multiple ciphertexts and hand them over to the server for ciphertext storage. These servers interact with the server for ciphertext calculation through interaction. The detection and positioning of the image tampering operation in the ciphertext space are completed with calculation. Finally, each server for ciphertext storage obtains the suspected tampering area, and the server for ciphertext storage sends the result to the user.

Preferably, includes 1 client

3 independent cloud servers, 2 of them,

with

For ciphertext storage, 1 server

For ciphertext calculations;

Use image encryption algorithm to split the image into two ciphertexts

with

By these two servers and

Detecting and locating the tampering operation of the image area in ciphertext space through interaction and calculation; finally

with

All obtained suspected tampered areas,

with

Send results to

Specifically, the detection method includes the following steps:

S1, client

Image ciphertext generation step;

S2, cloud server

with

Detection and location steps of tampered area in ciphertext space;

S2-1. Calculate Harris corner;

S2-2, extracting a point of interest descriptor;

S2-3. Points of interest match;

S2-4. Tampering with the positioning of the area;

S3, client

Obtain detection results and suspected tampered areas.

Preferably, in step S1, the client

The image ciphertext generation steps include:

Before calculation,

Have an image to be inspected X with a size of l _m × l _m ;

step 1:

Select a random number matrix R of size l _m × l _m , each element of which is a random number between 0 and 2 ¹⁰ ; generate 2 copies of the image ciphertext:

I ⁽¹⁾ = X + R, I ⁽²⁾ = R

Step 2:

Send I ⁽¹⁾ to the secure channel

Send I ⁽²⁾ over the secure channel to

After calculation,

Owns the ciphertext I ⁽¹⁾ ,

Owns the ciphertext I ⁽²⁾ .

Preferably, in step S2-1, calculating the Harris corner specifically includes:

Before calculation,

Has image ciphertext I ⁽¹⁾ ,

Has image ciphertext I ⁽²⁾ ,

with

Shared key set

step 1:

Calculation

Calculation

Step 2:

with

Interactive computing

After calculation

have

have

Step 3:

with

Interactive computing

After calculation

have

have

Step 4:

with

Interactive computing

After calculation

have

have

Step 5:

with

Interactive computing

After calculation

have

have

Step 6:

Calculation

Where h is the Gaussian filter kernel;

Calculation

Step 7:

with

Interactive computing

After calculation

have

have

Step 8:

with

Interactive computing

After calculation

have

have

Step 9:

with

Interactive computing

After calculation

have

have

Step 10:

with

Interactive computing

After calculation

have

have

Step 11:

with

Interactive computing

After calculation

Owns A ⁽¹⁾

Owns A ⁽²⁾ ;

Step 12:

Calculation

Where t is the Harris corner check constant;

Calculation

Step 13:

with

Generate an all-zero matrix H of size l _m × l _m for storing Harris corners; for all i and j,

Calculate coefficient blocks U ⁽¹⁾ (8i + 1: 8i + 8,8j + 1: 8j + 8) and U ⁽²⁾ (8i + 1: 8i + 8,8j + 1: 8j + 8) in ciphertext form The corresponding local minimum of the plaintext U (8i + 1: 8i + 8,8j + 1: 8j + 8), the method is:

Step 13-1:

with

Let θ _m = 8i + 1, θ _n = 8j + 1 both;

Step 13-2: For all 8i + 1 <m≤8i + 8, 8j + 1 <n≤8j + 8

with

Interactive calculation b = SCP (U ⁽¹⁾ (θ _m , θ _n ) -U ⁽¹⁾ (m, n), U ⁽²⁾ (θ _m , θ _n ) -U ⁽²⁾ (m, n)) After calculation

with

Both have b; if b = 0,

with

Let θ _m = m and θ _n = n both;

Step 13-3:

with

Let H (θ _m , θ _n ) = 1;

After calculation,

with

Each has a detected Harris corner H.

Preferably, in step S2-2, extracting the point of interest descriptor specifically includes:

Before calculation,

Have image ciphertext I ⁽¹⁾ , Harris corner H detected,

Own image ciphertext I ⁽²⁾ , Harris corner H detected

step 1:

Generate a three-dimensional all-zeros matrix D ^{(1) of} size l _m × l _m × 49,

Generate a three-dimensional all-zero matrix D ^{(2) of the} same size, which is used to store the descriptors of points of interest;

Step 2: For all i and j that satisfy H (i, j) = 1, calculate the point of interest descriptor in ciphertext form:

Step 2-1:

Calculation

Where FFT () represents fast Fourier transform and LPM () represents log-polar mapping

Represents a 7 × 7 pixel block with (i, j) as the center in I ⁽¹⁾ ;

Calculation

among them

Represents a 7 × 7 pixel block centered at (i, j ⁾ in I ⁽²⁾ ;

Step 2-2:

Let D ⁽¹⁾ (i, j, :) = [F ⁽¹⁾ (1,1), ..., F ⁽¹⁾ (1,7), ..., F ⁽¹⁾ (7,1), ..., F ⁽¹⁾ (7,7)];

Let D ⁽²⁾ (i, j, :) = [F ⁽²⁾ (1,1), ..., F ⁽²⁾ (1,7), ..., F ⁽²⁾ (7,1), ..., F ⁽²⁾ (7,7)].

After calculation,

Has a point of interest descriptor D ^{(1) in} cipher text,

The point-of-interest descriptor D ^{(2) in} ciphertext form.

Preferably, in step S2-3, the interest point matching specifically includes:

Before calculation,

Has a point of interest descriptor D ^{(1) in} cipher text,

Has a point of interest descriptor D ^{(2) in} ciphertext,

with

Both have control parameters α and β, shared key set

step 1:

Generate a l _D × 6 matrix T ⁽¹⁾ , where l _D is the number of all (i, j) with D ⁽¹⁾ (i, j, :) ≠ 0,

Generate a matrix T ^{(2) of the} same size. Note that (i, j ⁾ with D ⁽¹⁾ (i, j, :) ≠ 0 must also satisfy D ⁽²⁾ (i, j, :) ≠ 0;

Step 2: Use p to represent the index of the currently matched feature points,

with

Both start from the first feature point, that is, let p = 1, find the smallest i and corresponding j that satisfy D ⁽¹⁾ (i, j, :) ≠ 0;

Step 3: In cipher text form, find the descriptor closest to the descriptor at (i, j) and its location. The method is:

Step 3-1:

Let T ⁽¹⁾ (p, 1) = i, T ⁽¹⁾ (p, 2) = j, if q <p satisfies T ⁽¹⁾ (q, 3) = i, T ⁽¹⁾ (q, 4) = j, then let T ⁽¹⁾ (p, 3) = T ⁽¹⁾ (q, 1), T ⁽¹⁾ (p, 4) = T ⁽¹⁾ (q, 2), T ^{(1 )} (p, 5) = T ⁽¹⁾ (q, 5), otherwise let T ⁽¹⁾ (p, 5) = inf;

Let T ⁽²⁾ (p, 1) = i, T ⁽²⁾ (p, 2) = j, if q <p satisfies T ⁽²⁾ (q, 3) = i, T ⁽²⁾ (q, 4) = j, then let T ⁽²⁾ (p, 3) = T ⁽²⁾ (q, 1), T ⁽²⁾ (p, 4) = T ⁽²⁾ (q, 2), T ^{(2 )} (p, 5) = T ⁽²⁾ (q, 5), otherwise let T ⁽¹⁾ (p, 5) = 0;

Step 3-2: For any (i ′, j ′), if D ⁽¹⁾ (i ′, j ′, :) ≠ 0 && (i ′> i || j ′> j) && ((i′- i) ² + (j′-j) ² )> α, to determine whether the descriptor at this position is closest to the descriptor at (i, j) in cipher text, the method is:

Step 3-2-1:

with

Interactive computing

After calculation

have

have

Step 3-2-2:

with

Interactive computing

After calculation

have

have

Step 3-2-3:

with

Interactive computing

After calculation

have

have

Step 3-2-4:

with

Interactive computing

After calculation

have

have

Step 3-2-5:

Calculation

Calculation

Step 3-2-6:

Calculate u ⁽¹⁾ = (U ⁽¹⁾ (1) + ... + U ⁽¹⁾ (49));

Calculate u ⁽²⁾ = (U ⁽²⁾ (1) + ... + U ⁽²⁾ (49));

Step 3-2-7:

with

Interactive calculation b = SCP (T ⁽¹⁾ (p, 5) -u ⁽¹⁾ , T ⁽²⁾ (p, 5) -u ⁽²⁾ ), after calculation

with

Both have b;

Step 3-2-8: If b = 1,

Let T ⁽¹⁾ (p, 3) =, T ⁽¹⁾ (p, 4) = j ', T ⁽¹⁾ (p, 6) = T ⁽¹⁾ (p, 5), T ⁽¹⁾ ( p, 5) = u ⁽¹⁾ ,

Let T ⁽²⁾ (p, 3) = i ', T ⁽²⁾ (p, 4) = j', T ⁽²⁾ (p, 6) = T ⁽²⁾ (p, 5), T ^{(2 )} (p, 5) = u ⁽²⁾ .

Step 3-3:

with

Interactive calculation b = SCP (T ⁽¹⁾ (p, 5) × β-T ⁽¹⁾ (p, 6), T ⁽²⁾ (p, 5) × β-T ⁽²⁾ (p, 6)) After calculation

with

Both have b;

Step 3-4: If b = 1,

Let T ⁽¹⁾ (p, 3) = 0, T ⁽¹⁾ (p, 4) = 0, T ⁽¹⁾ (p, 5) = inf,

Let T ⁽²⁾ (p, 3) = 0, T ⁽²⁾ (p, 4) = 0, T ⁽²⁾ (p, 5) = 0;

Step 4: Find the smallest i ′ and corresponding j ′ that satisfy D ⁽¹⁾ (i ′, j ′, :) ≠ 0 && (i ′> i || j ′> j), if they exist,

with

Let p = p + 1, i = i ′, j = j ′, return to step 3 and execute again, otherwise go to step 5;

Step 5: The plaintext T corresponding to T ⁽¹⁾ and T ^{(2) is} reordered in the cipher text form according to the size of T (:, 5);

Step 6: For all l _D ≥i> 1, if (T ⁽¹⁾ (i, 1)> T ⁽¹⁾ (i, 3) || T ⁽¹⁾ (i, 2)> T ⁽¹⁾ ( i, 4)), then

Interacting T ⁽¹⁾ (i, 1) and T ⁽¹⁾ (i, 3), T ⁽¹⁾ (i, 2) and T ⁽¹⁾ (i, 4);

If (T ⁽²⁾ (i, 1)> T ⁽²⁾ (i, 3) || T ⁽²⁾ (i, 2)> T ⁽²⁾ (i, 4)), then

Interacting T ⁽²⁾ (i, 1) and T ⁽²⁾ (i, 3), T ⁽²⁾ (i, 2) and T ⁽²⁾ (i, 4);

Step 7: If l _D ≥i> j satisfies T ⁽¹⁾ (i,:) = T ⁽¹⁾ (j, :), then

Delete T ⁽¹⁾ (i, :);

If l _D ≥i> j satisfies T ⁽²⁾ (i, :) = T ⁽²⁾ (j, :), then

Delete T ⁽²⁾ (i, :);

After calculation,

Ranked p matching feature points T ^{(1) in} ciphertext form,

Ranked p matching feature points T ^{(2) in} ciphertext form.

Preferably, the plaintext T corresponding to T ⁽¹⁾ and T ^{(2) in the} cipher text form is reordered according to the size of T (:, 5) by the method:

Step 5-1:

with

Let i = 1;

Step 5-2: Find the position of the minimum value of T (i: l _D , 5) in ciphertext form, the method is:

Step 5-2-1:

with

Let θ = i;

Step 5-2-2: For all l _D ≥i ′> i,

with

Interactive calculation b = SCP (T ⁽¹⁾ (θ, 5) -T ⁽¹⁾ (i ′, 5), T ⁽²⁾ (θ, 5) -T ⁽²⁾ (i ′, 5)), calculate Rear

with

Both have b; if b = 1,

with

Let θ = i ′;

Step 5-3:

Exchange T ⁽¹⁾ (θ, :) and T ⁽¹⁾ (i, :);

Exchange T ⁽²⁾ (θ, :) and T ⁽²⁾ (i, :);

Step 5-4:

with

Let i = i + 1, if i ≤ l _D , return to step 5-2 and execute again, otherwise go to step 6.

Preferably, in step S2-4, the location of the tampering area specifically includes:

Before calculation,

p = 1,2, with sorted matching feature points T ^{(1) in} ciphertext form,

Have control parameters γ and δ;

step 1:

Generate an all-zero matrix E of size l _m × l _m , and let i = 1;

Step 2:

Find the number of points that satisfy the distance less than γ from the T ⁽¹⁾ (i, :) position by the method:

Step 2-1:

Let n = 0, j = i + 1, and set ε = {i};

Step 2-2:

Calculate a = (T ⁽¹⁾ (i, 1) -T ⁽¹⁾ (j, 1)) ² + (T ⁽¹⁾ (i, 2) -T ⁽¹⁾ (j, 2)) ² , b = (T ⁽¹⁾ (i, 3) -T ⁽¹⁾ (j, 3)) ² + (T ⁽¹⁾ (i, 4) -T ⁽¹⁾ (j, 4)) ² ;

Step 2-3: If a> γ &&b> r / 2 && b <2r is satisfied, then

Let n = n + 1, ε = ε + {j};

Step 2-4: Let j = j + 1, if j≤l _D , return to step 2-2 and execute again, otherwise go to step 3;

Step 3: If n≥δ, then for all j∈ε ₁ ,

Let E (T ⁽¹⁾ (j, 1), T ⁽¹⁾ (j, 2)) = i, E (T ⁽¹⁾ (j, 3), T ⁽¹⁾ (j, 4)) =- i;

Step 4:

Find the smallest circle that includes all points (j, k) where E (j, k) = i, and get all points (j ′, k ′) located inside the circle

Let E (j ′, k ′) = i;

Step 5:

Find the smallest circle that includes all points (j, k) where E (j, k) = -i, and obtain all points (j ′, k ′) located inside the circle

Let E (j ′, k ′) =-i;

Step 6: Let i = i + 1, if i≤5, return to step 2-3 and execute again, otherwise end execution;

After calculation,

p = 1, 2 and has a suspected tampering area E.

Preferably, in step S3, the client

Obtaining detection results and suspected tampering areas include:

Before calculation,

With the suspected tampering area E, the two areas of the stitching are labeled as i and -i, i ∈ {1, ..., 5}

step 1:

Send E over a secure channel to

After calculation,

Possess suspected tampering area E.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

1. The present invention provides a method for forensic copying of image areas with privacy protection capabilities. The server can detect the copy operation of the image area and locate the suspected tampering area without knowing the content of the image. Image encryption technology and the proposed forensic algorithm based on multi-party secure computing.

2. The forensic method provided by the present invention has higher calculation efficiency and better accuracy, and the effect is brought about by the image forensics flow, the secure multi-party multiplication protocol, and the secure multi-party comparison protocol, which are proposed by the present invention, and have a relatively low calculation amount. Since multiple regions are located simultaneously in the region positioning operation, the present invention can be used to detect a tampered image with multiple copied regions.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is an overall flowchart of the method of the embodiment.

FIG. 2 is a flowchart of tampering positioning calculation of a stitching region.

detailed description

The present invention is described in further detail below with reference to the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.

Example 1

The invention combines multi-party secure computing, image interest point descriptor extraction, and nearest neighbor search technologies to realize the detection and location of image region replication on the premise of protecting the privacy of image content. In order to achieve the confidentiality of image content, the present invention uses image encryption to split the image into two image ciphertexts. In order to achieve image forensics under the premise of privacy protection, the present invention designs a forensic algorithm based on secure multi-party calculations. Through the interaction of three servers, it can extract the interest point descriptor without knowing the ciphertext of the other party, and find Best match descriptor. In order to ensure the efficiency and accuracy of algorithm execution, the present invention designs a simple image forensics process and a secure multi-party computing protocol.

There are 4 individuals in this embodiment: 1 client

3 independent cloud servers, 2 of them,

with

Used for ciphertext storage, 1 server,

Used for ciphertext calculation, as shown in Figure 1.

Use the image encryption algorithm to split the image into two ciphertexts and hand them over to

with

By these two servers and

The detection and location of the image tampering operation in the ciphertext space are completed through interaction and calculation. At last

with

All obtained suspected tampered areas,

with

Send results to

In this embodiment, uppercase letters (such as I, J) represent matrices, bold lowercase letters (such as i, j) represent vectors, and italic lowercase letters (such as i, j) represent numbers. All multiplications are performed according to Element multiplication, if C = AB, then each element in C C (i, j) = A (i, j) × B (i, j). A * B represents the convolution of A and B. The specific implementation process is as follows:

First, the client

Image ciphertext generation process

Before calculation,

Have an image X to be detected with a size of l _m × l _m .

step 1:

L select a random number R _m × l _m matrix size, each element of a random number between 0 and ^210. Generate 2 image ciphertexts:

I ⁽¹⁾ = X + R, I ⁽²⁾ = R

Step 2:

Send I ⁽¹⁾ to the secure channel

Send I ⁽²⁾ over the secure channel to

After calculation,

Owns the ciphertext I ⁽¹⁾ ,

Owns the ciphertext I ⁽²⁾ .

Cloud server

with

Detection and localization process of tampered area in ciphertext space

(1) Calculate Harris Corner:

Before calculation,

Has image ciphertext I ⁽¹⁾ ,

Has image ciphertext I ⁽²⁾ ,

with

Shared key set

step 1:

Calculation

Calculation

Step 2:

with

Interactive computing

After calculation

have

have

Step 3:

with

Interactive computing

After calculation

have

have

Step 4:

with

Interactive computing

After calculation

have

have

Step 5:

with

Interactive computing

After calculation

have

have

Step 6:

Calculation

Where h is the Gaussian filter kernel;

Calculation

Step 7:

with

Interactive computing

After calculation

have

have

Step 8:

with

Interactive computing

After calculation

have

have

Step 9:

with

Interactive computing

After calculation

have

have

Step 10:

with

Interactive computing

After calculation

have

have

Step 11:

with

Interactive computing

After calculation

Owns A ⁽¹⁾

Owns A ⁽²⁾ ;

Step 12:

Calculation

Where t is the Harris corner check constant, which is generally taken between 0.04-0.06;

Calculation

Step 13:

with

An all-zero matrix H of size l _m × l _{m is} generated for storing Harris corner points. For all i and j,

Calculate coefficient blocks U ⁽¹⁾ (8i + 1: 8i + 8,8j + 1: 8j + 8) and U ⁽²⁾ (8i + 1: 8i + 8,8j + 1: 8j + 8) in ciphertext form The corresponding local minimum of plaintext U (8i + 1: 8i + 8,8j + 1: 8j + 8).

Step 13-1:

with

Let θ _m = 8i + 1, θ _n = 8j + 1 both;

Step 13-2: For all 8i + 1 <m≤8i + 8, 8j + 1 <n≤8j + 8,

with

Interactive calculation b = SCP (U ⁽¹⁾ (θ _m , θ _n ) -U ⁽¹⁾ (m, n), U ⁽²⁾ (θ _m , θ _n ) -U ⁽²⁾ (m, n)) After calculation

with

Both have b; if b = 0,

with

Let θ _m = m and θ _n = n both;

Step 13-3:

with

Let H (θ _m , θ _n ) = 1.

After calculation,

with

Each has a detected Harris corner H.

(2) Extract the point of interest descriptor:

Before calculation,

Have image ciphertext I ⁽¹⁾ , Harris corner H detected,

Have image ciphertext I ⁽²⁾ , Harris corner H detected.

step 1:

Generate a three-dimensional all-zeros matrix D ^{(1) of} size l _m × l _m × 49,

Generate a three-dimensional all-zero matrix D ^{(2) of the} same size, which is used to store the descriptors of points of interest;

Step 2: For all i and j that satisfy H (i, j) = 1, calculate the point of interest descriptor in ciphertext form. The method is:

Step 2-1:

Calculation

Where FFT () represents a fast Fourier transform, and LPM () represents a log-polar mapping,

Represents a 7 × 7 pixel block with (i, j) as the center in I ⁽¹⁾ ;

Calculation

among them

Represents a 7 × 7 pixel block centered at (i, j ⁾ in I ⁽²⁾ ;

Step 2-2:

Let D ⁽¹⁾ (i, j, :) = [F ⁽¹⁾ (1,1), ..., F ⁽¹⁾ (1,7), ..., F ⁽¹⁾ (7,1), ..., F ⁽¹⁾ (7,7)];

Let D ⁽²⁾ (i, j, :) = [F ⁽²⁾ (1,1), ..., F ⁽²⁾ (1,7), ..., F ⁽²⁾ (7,1), ..., F ⁽²⁾ (7,7)].

After calculation,

Has a point of interest descriptor D ^{(1) in} cipher text,

The point-of-interest descriptor D ^{(2) in} ciphertext form.

(3) Points of interest match:

Before calculation,

Has a point of interest descriptor D ^{(1) in} cipher text,

Has a point of interest descriptor D ^{(2) in} ciphertext,

with

Both have control parameters α and β, shared key set

step 1:

Generate a l _D × 6 matrix T ⁽¹⁾ , where l _D is the number of all (i, j) with D ⁽¹⁾ (i, j, :) ≠ 0,

Generate a matrix T ^{(2) of the} same size. Note that (i, j ⁾ with D ⁽¹⁾ (i, j, :) ≠ 0 must also satisfy D ⁽²⁾ (i, j, :) ≠ 0;

Step 2: Use p to represent the index of the currently matched feature points,

with

Both start from the first feature point, that is, let p = 1, find the smallest i and corresponding j that satisfy D ⁽¹⁾ (i, j, :) ≠ 0;

Step 3: In cipher text form, find the descriptor closest to the descriptor at (i, j) and its location. The method is:

Step 3-1:

Let T ⁽¹⁾ (p, 1) = i, T ⁽¹⁾ (p, 2) = j, if q <p satisfies T ⁽¹⁾ (q, 3) = i, T ⁽¹⁾ (q, 4) = j, then let T ⁽¹⁾ (p, 3) = T ⁽¹⁾ (q, 1), T ⁽¹⁾ (p, 4) = T ⁽¹⁾ (q, 2), T ^{(1 )} (p, 5) = T ⁽¹⁾ (q, 5), otherwise let T ⁽¹⁾ (p, 5) = inf;

Let T ⁽²⁾ (p, 1) = i, T ⁽²⁾ (p, 2) = j, if q <p satisfies T ⁽²⁾ (q, 3) = i, T ⁽²⁾ (q, 4) = j, then let T ⁽²⁾ (p, 3) = T ⁽²⁾ (q, 1), T ⁽²⁾ (p, 4) = T ⁽²⁾ (q, 2), T ^{(2 )} (p, 5) = T ⁽²⁾ (q, 5), otherwise let T ⁽¹⁾ (p, 5) = 0;

Step 3-2: For any (i ′, j ′), if D ⁽¹⁾ (i ′, j ′, :) ≠ 0 && (i ′> i || j ′> j) && ((i′- i) ² + (j′-j) ² )> α, to determine whether the descriptor at this position is closest to the descriptor at (i, j) in cipher text, the method is:

Step 3-2-1:

with

Interactive computing

After calculation

have

have

Step 3-2-2:

with

Interactive computing

After calculation

have

have

Step 3-2-3:

with

Interactive computing

After calculation

have

have

Step 3-2-4:

with

Interactive computing

After calculation

have

have

Step 3-2-5:

Calculation

Calculation

Step 3-2-6:

Calculate u ⁽¹⁾ = (U ⁽¹⁾ (1) + ... + U ⁽¹⁾ (49));

Calculate u ⁽²⁾ = (U ⁽²⁾ (1) + ... + U ⁽²⁾ (49));

Step 3-2-7:

with

Interactive calculation b = SCP (T ⁽¹⁾ (p, 5) -u ⁽¹⁾ , T ⁽²⁾ (p, 5) -u ⁽²⁾ ), after calculation

with

Both have b;

Step 3-2-8: If b = 1,

Let T ⁽¹⁾ (p, 3) =, T ⁽¹⁾ (p, 4) = j ', T ⁽¹⁾ (p, 6) = T ⁽¹⁾ (p, 5), T ⁽¹⁾ ( p, 5) = u ⁽¹⁾ ,

Let T ⁽²⁾ (p, 3) = i ', T ⁽²⁾ (p, 4) = j', T ⁽²⁾ (p, 6) = T ⁽²⁾ (p, 5), T ^{(2 )} (p, 5) = u ⁽²⁾ .

Step 3-3:

with

Interactive calculation b = SCP (T ⁽¹⁾ (p, 5) × β-T ⁽¹⁾ (p, 6), T ⁽²⁾ (p, 5) × β-T ⁽²⁾ (p, 6)) After calculation

with

Both have b;

Step 3-4: If b = 1,

Let T ⁽¹⁾ (p, 3) = 0, T ⁽¹⁾ (p, 4) = 0, T ⁽¹⁾ (p, 5) = inf,

Let T ⁽²⁾ (p, 3) = 0, T ⁽²⁾ (p, 4) = 0, and T ⁽²⁾ (p, 5) = 0.

Step 4: Find the smallest i ′ and corresponding j ′ that satisfy D ⁽¹⁾ (i ′, j ′, :) ≠ 0 && (i ′> i || j ′> j), if they exist,

with

Let p = p + 1, i = i ′, j = j ′, return to step 3 and execute again, otherwise go to step 5;

Step 5: The plaintext T corresponding to T ⁽¹⁾ and T ^{(2) in} cipher text form is reordered according to the size of T (:, 5).

Step 5-1:

with

Let i = 1;

Step 5-2: Find the position of the minimum value of T (i: l _D , 5) in ciphertext form, the method is:

Step 5-2-1:

with

Let θ = i;

Step 5-2-2: For all l _D ≥i ′> i,

with

Interactive calculation b = SCP (T ⁽¹⁾ (θ, 5) -T ⁽¹⁾ (i ′, 5), T ⁽²⁾ (θ, 5) -T ⁽²⁾ (i ′, 5)), calculate Rear

with

Both have b; if b = 1,

with

Let θ = i ′;

Step 5-3:

Exchange T ⁽¹⁾ (θ, :) and T ⁽¹⁾ (i, :);

Exchange T ⁽²⁾ (θ, :) and T ⁽²⁾ (i, :);

Step 5-4:

with

Let i = i + 1, if i ≤ l _D , return to step 5-2 and execute again, otherwise go to step 6;

Step 6: For all l _D ≥i> 1, if (T ⁽¹⁾ (i, 1)> T ⁽¹⁾ (i, 3) || T ⁽¹⁾ (i, 2)> T ⁽¹⁾ ( i, 4)), then

Interacting T ⁽¹⁾ (i, 1) and T ⁽¹⁾ (i, 3), T ⁽¹⁾ (i, 2) and T ⁽¹⁾ (i, 4);

If (T ⁽²⁾ (i, 1)> T ⁽²⁾ (i, 3) || T ⁽²⁾ (i, 2)> T ⁽²⁾ (i, 4)), then

Interacting T ⁽²⁾ (i, 1) and T ⁽²⁾ (i, 3), T ⁽²⁾ (i, 2) and T ⁽²⁾ (i, 4);

Step 7: If l _D ≥i> j satisfies T ⁽¹⁾ (i,:) = T ⁽¹⁾ (j, :), then

Delete T ⁽¹⁾ (i, :);

If l _D ≥i> j satisfies T ⁽²⁾ (i, :) = T ⁽²⁾ (j, :), then

Delete T ⁽²⁾ (i, :).

After calculation,

Ranked p matching feature points T ^{(1) in} ciphertext form,

Ranked p matching feature points T ^{(2) in} ciphertext form.

(4) Tampering area positioning:

Before calculation,

p = 1,2, has the sorted matching feature point T ^{(1) in} ciphertext form.

Have control parameters γ and δ.

step 1:

Generate an all-zero matrix E of size l _m × l _m , and let i = 1;

Step 2:

Find the number of points that satisfy the distance less than γ from the T ⁽¹⁾ (i, :) position by the method:

Step 2-1:

Let n = 0, j = i + 1, and set ε = {i};

Step 2-2:

Calculate a = (T ⁽¹⁾ (i, 1) -T ⁽¹⁾ (j, 1)) ² + (T ⁽¹⁾ (i, 2) -T ⁽¹⁾ (j, 2)) ² , b = (T ⁽¹⁾ (i, 3) -T ⁽¹⁾ (j, 3)) ² + (T ⁽¹⁾ (i, 4) -T ⁽¹⁾ (j, 4)) ² ;

Step 2-3: If a> γ &&b> r / 2 && b <2r is satisfied, then

Let n = n + 1, ε = ε + {j};

Step 2-4: Let j = j + 1, if j≤l _D , return to step 2-2 and execute again, otherwise go to step 3;

Step 3: If n≥δ, then for all j∈ε ₁ ,

Let E (T ⁽¹⁾ (j, 1), T ⁽¹⁾ (j, 2)) = i, E (T ⁽¹⁾ (j, 3), T ⁽¹⁾ (j, 4)) =- i;

Step 4:

Find the smallest circle that includes all points (j, k) where E (j, k) = i, and get all points (j ′, k ′) located inside the circle

Let E (j ′, k ′) = i;

Step 5:

Find the smallest circle that includes all points (j, k) where E (j, k) = -i, and obtain all points (j ′, k ′) located inside the circle

Let E (j ′, k ′) =-i;

Step 6: Let i = i + 1, if i≤5, return to step 2-3 and execute again, otherwise end execution;

After calculation,

p = 1, 2 and has a suspected tampering area E.

Third, the client

Obtain detection results and suspected tampered areas

Before calculation,

With the suspected tampering area E, the two regions of the stitching are labeled as i and -i, i ∈ {1, ..., 5}.

step 1:

Send E over a secure channel to

After calculation,

Possess suspected tampering area E.

In the above steps, secure multiparty multiplication protocol ^{(Y (1), Y (} 2)) = SMP (X 1, X 2) for calculating

Ciphertext matrix X ₁ and

Multiplication of the ciphertext matrix X ₂ and during calculation

Will not know X ₂ ,

Will not know X ₁ ,

X ₁ or X ₂ will not be known, and X ₁ X ₂ will not be known to all three servers. The protocol flow is:

Before calculation,

Has a ciphertext matrix X ₁ ,

Has a ciphertext matrix X ₂ ,

with

Shared key set

step 1:

with

Select two key matrices K _{1 of the} same size as X ₁ (or X ₂ ) according to pre-negotiation,

Step 2:

Calculate U ₁ = (X ₁ + K ₂ ) / K ₁ and send U ₁ to

Calculate U ₂ = X ₂ K ₁ and send U ₂ to

Step 3:

Randomly select a random number matrix R of the same size as U ₁ , each element of which is between 0 and 2 ¹⁰ , and calculate V = U ₁ U ₂ + R;

Step 4:

Send V to

Send R to

Step 5:

Let Y ⁽¹⁾ = V;

Let Y ⁽²⁾ = X ₂ K ₂ + R;

After calculation,

Has a ciphertext matrix Y ⁽¹⁾ ,

Have a ciphertext matrix Y ⁽²⁾ , which satisfies Y ⁽¹⁾ -Y ⁽²⁾ = X ₁ X ₂ .

In the above steps, the secure multiparty comparison protocol b = SCP (X ₁ , X ₂ ) is used for comparison

Ciphertext matrix X ₁ and

The size of the ciphertext matrix X ₂ and the calculation process

Will not know X ₂ ,

Will not know X ₁ ,

Will not know X ₁ or X ₂ , the protocol flow is:

Before calculation,

Has a ciphertext matrix X ₁ ,

Has a ciphertext matrix X ₂ ,

with

Shared key set

step 1:

with

Select two key matrices K _{1 of the} same size as X ₁ (or X ₂ ) according to pre-negotiation,

Step 2:

Calculate U ₁ = (X ₁ + K ₁ ) / K ₂ and send U ₁ to

Calculate U ₂ = (X ₂ + K ₁ ) / K ₂ and send U ₂ to

Step 3:

Calculate U ₁ / U _2. If the result is greater than 1, let b = 1, otherwise let b = 0.

Step 4:

Send b to

with

After calculation,

with

Each has a comparison result b, which satisfies X ₁ > X ₂ if b = 1, otherwise X ₁ ≤ X ₂ .

The image encryption technology used in this solution can be implemented with similar technologies, such as replacing the addition operation with the subtraction operation to achieve the same effect.

The multiplication and comparison protocols based on multi-party secure calculations used in this solution can be implemented by using various other multi-party security protocols to achieve similar results, but the computing efficiency or the number of cloud servers used varies.

The sorting algorithm in this solution can be replaced with any other sorting algorithm to achieve the same effect.

Existing image area copying and forensics methods can only be performed in plain text, which will leak image content. In the method of the present invention, the user terminal can send the image in the form of cipher text to the cloud server. The cloud server provides an effective area copy forensics service without realizing the image content, so as to detect and locate the tampering operation. Compared with the traditional area copying forensics tool or service, the present invention can provide privacy protection of user image content.

This method is suitable for image digital forensics cloud services, and provides individuals, enterprises, government departments, etc. with a safe and reliable regional copy forensics and forensics service. This solution is suitable for unreliable cloud service environments.

The above embodiment is a preferred embodiment of the present invention, but the embodiment of the present invention is not limited by the above embodiment. Any other changes, modifications, substitutions, combinations, and modifications made without departing from the spirit and principle of the present invention, Simplified, all should be equivalent replacement methods, and all are included in the protection scope of the present invention.

Claims (10)

1. An image region copy detection method supporting privacy protection function, comprising: a user terminal and N mutually independent cloud servers, some of which are used for ciphertext storage, and some of which are used for ciphertext calculation; The user terminal uses the image encryption algorithm to split the image into multiple ciphertexts and hand them over to the server for ciphertext storage. These servers and the server for ciphertext calculation complete the ciphertext space through interaction and calculation. Detection and location of image tampering operations; finally, each server for ciphertext storage obtains a suspected tampering area, and the server for ciphertext storage sends the results to the user.
2. The method for detecting a copy of an image area supporting privacy protection according to claim 1, further comprising: a client

   

   3 independent cloud servers, 2 of them,

   

   with

   

   For ciphertext storage, 1 server

   

   For ciphertext calculations;

   

   Use the image encryption algorithm to split the image into two ciphertexts and hand them over to

   

   with

   

   By these two servers and

   

   Detecting and locating the tampering operation of the image area in ciphertext space through interaction and calculation; finally

   

   with

   

   All obtained suspected tampered areas,

   

   with

   

   Send results to

   
3. The method for detecting a copy of an image area supporting the privacy protection function according to claim 2, wherein the detection method comprises the following steps:

   S1, client

   

   Image ciphertext generation step;

   S2, cloud server

   

   with

   

   Detection and location steps of tampered area in ciphertext space;

   S2-1. Calculate Harris corner;

   S2-2, extracting a point of interest descriptor;

   S2-3. Points of interest match;

   S2-4. Tampering with the positioning of the area;

   S3, client

   

   Obtain detection results and suspected tampered areas.
4. The method for detecting a copy of an image area supporting privacy protection according to claim 3, wherein in step S1, the user

   

   The image ciphertext generation steps include:

   Before calculation,

   

   Have an image to be inspected X with a size of l _m × l _m ;

   step 1:

   

   Select a random number matrix R of size l _m × l _m , each element of which is a random number between 0 and 2 ¹⁰ ; generate 2 copies of the image ciphertext:

   I ⁽¹⁾ = X + R, I ⁽²⁾ = R

   Step 2:

   

   Send I ⁽¹⁾ to the secure channel

   

   Send I ⁽²⁾ over the secure channel to

   

   After calculation,

   

   Owns the ciphertext I ⁽¹⁾ ,

   

   Owns the ciphertext I ⁽²⁾ .
5. The method for detecting a copy of an image area supporting privacy protection according to claim 4, wherein, in step S2-1, calculating the Harris corner specifically comprises:

   Before calculation,

   

   Has image ciphertext I ⁽¹⁾ ,

   

   Has image ciphertext I ⁽²⁾ ,

   

   with

   

   Shared key set

   

   step 1:

   

   Calculation

   

   

   Calculation

   

   Step 2:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 3:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 4:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 5:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 6:

   

   Calculation

   

   

   Where h is the Gaussian filter kernel;

   

   Calculation

   

   

   Step 7:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 8:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 9:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 10:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 11:

   

   with

   

   Interactive computing

   

   After calculation

   

   Owns A ⁽¹⁾

   

   Owns A ⁽²⁾ ;

   Step 12:

   

   Calculation

   

   

   Where t is the Harris corner check constant;

   

   Calculation

   

   

   Step 13:

   

   with

   

   Generate an all-zero matrix H of size l _m × l _m for storing Harris corners; for all i and j,

   

   Calculate coefficient blocks U ⁽¹⁾ (8i + 1: 8i + 8,8j + 1: 8j + 8) and U ⁽²⁾ (8i + 1: 8i + 8,8j + 1: 8j + 8) in ciphertext form The corresponding local minimum of plaintext U (8i + 1: 8i + 8,8j + 1: 8j + 8).

   Step 13-1:

   

   with

   

   Let θ _m = 8i + 1, θ _n = 8j + 1 both;

   Step 13-2: For all 8i + 1 <m≤8i + 8, 8j + 1 <n≤8j + 8,

   

   with

   

   Interactive calculation b = SCP (U ⁽¹⁾ (θ _m , θ _n ) -U ⁽¹⁾ (m, n), U ⁽²⁾ (θ _m , θ _n ) -U ⁽²⁾ (m, n)) After calculation

   

   with

   

   Both have b; if b = 0,

   

   with

   

   Let θ _m = m and θ _n = n both;

   Step 13-3:

   

   with

   

   Let H (θ _m , θ _n ) = 1;

   After calculation,

   

   with

   

   Each has a detected Harris corner H.
6. The method for detecting a copy of an image area supporting privacy protection according to claim 5, wherein, in step S2-2, extracting a point of interest descriptor comprises:

   Before calculation,

   

   Have image ciphertext I ⁽¹⁾ , Harris corner H detected,

   

   Own image ciphertext I ⁽²⁾ , Harris corner H detected

   step 1:

   

   Generate a three-dimensional all-zeros matrix D ^{(1) of} size l _m × l _m × 49,

   

   Generate a three-dimensional all-zero matrix D ^{(2) of the} same size, which is used to store the descriptors of points of interest;

   Step 2: For all i and j that satisfy H (i, j) = 1, calculate the point of interest descriptor in ciphertext form:

   Step 2-1:

   

   Calculation

   

   Where FFT () represents fast Fourier transform, and LPM () represents log-polar mapping,

   

   Represents a 7 × 7 pixel block with (i, j) as the center in I ⁽¹⁾ ;

   

   Calculation

   

   among them

   

   Represents a 7 × 7 pixel block centered at (i, j ⁾ in I ⁽²⁾ ;

   Step 2-2:

   

   Let D ⁽¹⁾ (i, j, :) = [F ⁽¹⁾ (1,1), ..., F ⁽¹⁾ (1,7), ..., F ⁽¹⁾ (7,1), ..., F ⁽¹⁾ (7,7)];

   

   Let D ⁽²⁾ (i, j, :) = [F ⁽²⁾ (1,1), ..., F ⁽²⁾ (1,7), ..., F ⁽²⁾ (7,1), ..., F ⁽²⁾ (7,7)];

   After calculation,

   

   Has a point of interest descriptor D ^{(1) in} cipher text,

   

   The point-of-interest descriptor D ^{(2) in} ciphertext form.
7. The method for detecting a copy of an image area supporting privacy protection according to claim 6, wherein, in step S2-3, the matching of the points of interest specifically includes:

   Before calculation,

   

   Has a point of interest descriptor D ^{(1) in} cipher text,

   

   Has a point of interest descriptor D ^{(2) in} ciphertext,

   

   with

   

   Both have control parameters α and β, shared key set

   

   step 1:

   

   Generate a l _D × 6 matrix T ⁽¹⁾ , where l _D is the number of all (i, j) with D ⁽¹⁾ (i, j, :) ≠ 0,

   

   Generate a matrix T ^{(2) of the} same size. Note that (i, j ⁾ with D ⁽¹⁾ (i, j, :) ≠ 0 must also satisfy D ⁽²⁾ (i, j, :) ≠ 0;

   Step 2: Use p to represent the index of the currently matched feature points,

   

   with

   

   Both start from the first feature point, that is, let p = 1, find the smallest i and corresponding j that satisfy D ⁽¹⁾ (i, j, :) ≠ 0;

   Step 3: In cipher text form, find the descriptor closest to the descriptor at (i, j) and its location. The method is:

   Step 3-1:

   

   Let T ⁽¹⁾ (p, 1) = i, T ⁽¹⁾ (p, 2) = j, if q <p satisfies T ⁽¹⁾ (q, 3) = i, T ⁽¹⁾ (q, 4) = j, then let T ⁽¹⁾ (p, 3) = T ⁽¹⁾ (q, 1), T ⁽¹⁾ (p, 4) = T ⁽¹⁾ (q, 2), T ^{(1 )} (p, 5) = T ⁽¹⁾ (q, 5), otherwise let T ⁽¹⁾ (p, 5) = inf;

   

   Let T ⁽²⁾ (p, 1) = i, T ⁽²⁾ (p, 2) = j, if q <p satisfies T ⁽²⁾ (q, 3) = i, T ⁽²⁾ (q, 4) = j, then let T ⁽²⁾ (p, 3) = T ⁽²⁾ (q, 1), T ⁽²⁾ (p, 4) = T ⁽²⁾ (q, 2), T ^{(2 )} (p, 5) = T ⁽²⁾ (q, 5), otherwise let T ⁽¹⁾ (p, 5) = 0;

   Step 3-2: For any (i ′, j ′), if D ⁽¹⁾ (i ′, j ′, :) ≠ 0 && (i ′> i || j ′> j) && ((i′- i) ² + (j′-j) ² )> α, to determine whether the descriptor at this position is closest to the descriptor at (i, j) in cipher text, the method is:

   Step 3-2-1:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 3-2-2:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 3-2-3:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 3-2-4:

   

   with

   

   Interactive computing

   

   After calculation

   

   have

   

   have

   

   Step 3-2-5:

   

   Calculation

   

   

   

   Calculation

   

   

   Step 3-2-6:

   

   Calculate u ⁽¹⁾ = (U ⁽¹⁾ (1) + ... + U ⁽¹⁾ (49));

   

   Calculate u ⁽²⁾ = (U ⁽²⁾ (1) + ... + U ⁽²⁾ (49));

   Step 3-2-7:

   

   with

   

   Interactive calculation b = SCP (T ⁽¹⁾ (p, 5) -u ⁽¹⁾ , T ⁽²⁾ (p, 5) -u ⁽²⁾ ), after calculation

   

   with

   

   Both have b;

   Step 3-2-8: If b = 1,

   

   Let T ⁽¹⁾ (p, 3) =, T ⁽¹⁾ (p, 4) = j ', T ⁽¹⁾ (p, 6) = T ⁽¹⁾ (p, 5), T ⁽¹⁾ ( p, 5) = u ⁽¹⁾ ,

   

   Let T ⁽²⁾ (p, 3) = i ', T ⁽²⁾ (p, 4) = j', T ⁽²⁾ (p, 6) = T ⁽²⁾ (p, 5), T ^{(2 )} (p, 5) = u ⁽²⁾ ;

   Step 3-3:

   

   with

   

   Interactive calculation b = SCP (T ⁽¹⁾ (p, 5) × β-T ⁽¹⁾ (p, 6), T ⁽²⁾ (p, 5) × β-T ⁽²⁾ (p, 6)) After calculation

   

   with

   

   Both have b;

   Step 3-4: If b = 1,

   

   Let T ⁽¹⁾ (p, 3) = 0, T ⁽¹⁾ (p, 4) = 0, T ⁽¹⁾ (p, 5) = inf,

   

   Let T ⁽²⁾ (p, 3) = 0, T ⁽²⁾ (p, 4) = 0, T ⁽²⁾ (p, 5) = 0;

   Step 4: Find the smallest i ′ and corresponding j ′ that satisfy D ⁽¹⁾ (i ′, j ′, :) ≠ 0 && (i ′> i || j ′> j), if they exist,

   

   with

   

   Let p = p + 1, i = i ′, j = j ′, return to step 3 and execute again, otherwise go to step 5;

   Step 5: The plaintext T corresponding to T ⁽¹⁾ and T ^{(2) is} reordered in the cipher text form according to the size of T (:, 5);

   Step 6: For all l _D ≥i> 1, if (T ⁽¹⁾ (i, 1)> T ⁽¹⁾ (i, 3) || T ⁽¹⁾ (i, 2)> T ⁽¹⁾ ( i, 4)), then

   

   Interacting T ⁽¹⁾ (i, 1) and T ⁽¹⁾ (i, 3), T ⁽¹⁾ (i, 2) and T ⁽¹⁾ (i, 4);

   If (T ⁽²⁾ (i, 1)> T ⁽²⁾ (i, 3) || T ⁽²⁾ (i, 2)> T ⁽²⁾ (i, 4)), then

   

   Interacting T ⁽²⁾ (i, 1) and T ⁽²⁾ (i, 3), T ⁽²⁾ (i, 2) and T ⁽²⁾ (i, 4);

   Step 7: If l _D ≥i> j satisfies T ⁽¹⁾ (i,:) = T ⁽¹⁾ (j, :), then

   

   Delete T ⁽¹⁾ (i, :);

   If l _D ≥i> j satisfies T ⁽²⁾ (i, :) = T ⁽²⁾ (j, :), then

   

   Delete T ⁽²⁾ (i, :);

   After calculation,

   

   Ranked p matching feature points T ^{(1) in} ciphertext form,

   

   Ranked p matching feature points T ^{(2) in} ciphertext form.
8. The method for detecting the copy of an image area supporting privacy protection according to claim 7, wherein the plaintext T corresponding to T ⁽¹⁾ and T ⁽²⁾ in the ciphertext form is weighted according to the size of T (:, 5) Sort by:

   Step 5-1:

   

   with

   

   Let i = 1;

   Step 5-2: Find the position of the minimum value of T (i: l _D , 5) in ciphertext form, the method is:

   Step 5-2-1:

   

   with

   

   Let θ = i;

   Step 5-2-2: For all l _D ≥i ′> i,

   

   with

   

   Interactive calculation b = SCP (T ⁽¹⁾ (θ, 5) -T ⁽¹⁾ (i ′, 5), T ⁽²⁾ (θ, 5) -T ⁽²⁾ (i ′, 5)), calculate Rear

   

   with

   

   Both have b; if b = 1,

   

   with

   

   Let θ = i ′;

   Step 5-3:

   

   Exchange T ⁽¹⁾ (θ, :) and T ⁽¹⁾ (i, :);

   

   Exchange T ⁽²⁾ (θ, :) and T ⁽²⁾ (i, :);

   Step 5-4:

   

   with

   

   Let i = i + 1, if i ≤ l _D , return to step 5-2 and execute again, otherwise go to step 6.
9. The method for detecting a copy of an image area supporting privacy protection according to claim 7, wherein, in step S2-4, the location of the tampered area specifically comprises:

   Before calculation,

   

   p = 1,2, with sorted matching feature points T ^{(1) in} ciphertext form,

   

   Have control parameters γ and δ;

   step 1:

   

   Generate an all-zero matrix E of size l _m × l _m , and let i = 1;

   Step 2:

   

   Find the number of points that satisfy the distance less than γ from the T ⁽¹⁾ (i, :) position by the method:

   Step 2-1:

   

   Let n = 0, j = i + 1, and set ε = {i};

   Step 2-2:

   

   Calculate a = (T ⁽¹⁾ (i, 1) -T ⁽¹⁾ (j, 1)) ² + (T ⁽¹⁾ (i, 2) -T ⁽¹⁾ (j, 2)) ² , b = (T ⁽¹⁾ (i, 3) -T ⁽¹⁾ (j, 3)) ² + (T ⁽¹⁾ (i, 4) -T ⁽¹⁾ (j, 4)) ² ;

   Step 2-3: If a> γ &&b> r / 2 && b <2r is satisfied, then

   

   Let n = n + 1, ε = ε + {j};

   Step 2-4: Let j = j + 1, if j≤l _D , return to step 2-2 and execute again, otherwise go to step 3;

   Step 3: If n≥δ, then for all j∈ε ₁ ,

   

   Let E (T ⁽¹⁾ (j, 1), T ⁽¹⁾ (j, 2)) = i, E (T ⁽¹⁾ (j, 3), T ⁽¹⁾ (j, 4)) =- i;

   Step 4:

   

   Find the smallest circle that includes all points (j, k) where E (j, k) = i, and get all points (j ′, k ′) located inside the circle

   

   Let E (j ′, k ′) = i;

   Step 5:

   

   Find the smallest circle that includes all points (j, k) where E (j, k) = -i, and obtain all points (j ′, k ′) located inside the circle

   

   Let E (j ′, k ′) =-i;

   Step 6: Let i = i + 1, if i≤5, return to step 2-3 and execute again, otherwise end execution;

   After calculation,

   

   p = 1, 2 and has a suspected tampering area E.
10. The method for detecting a copy of an image area supporting privacy protection according to claim 9, characterized in that, in step S3, the user terminal

    

    Obtaining detection results and suspected tampering areas include:

    Before calculation,

    

    With the suspected tampering area E, the two areas of the stitching are labeled as i and -i, i ∈ {1, ..., 5}

    step 1:

    

    Send E over a secure channel to

    

    After calculation,

    

    Possess suspected tampering area E.

Images