A Reversible Information Hiding Method for Encrypted 3D Mesh Based on Coordinate Axis Segmentation
By using a reversible information hiding method for encrypted 3D meshes based on coordinate axis segmentation, the problem of balancing data privacy and integrity in existing 3D models is solved, achieving high-capacity information hiding and lossless recovery of 3D mesh models.
Patent Information
- Application Number
- CN202310293174.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-03-23
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-03-23
AI Technical Summary
Existing 3D model protection technologies struggle to increase information hiding capacity while ensuring data privacy and integrity, and to recover 3D mesh models without loss when extracting secret data.
A reversible information hiding method for encrypted 3D meshes based on coordinate axis segmentation is adopted. By preprocessing and encrypting the vertex coordinates of the 3D mesh model, auxiliary information and secret data are embedded using quadrant grouping and stream cipher functions, and encryption and decryption operations are performed using a key to ensure the integrity and privacy of the data.
It enables lossless recovery of 3D mesh models on the cloud platform, improves information hiding capacity, and can accurately extract secret data, protecting data integrity and privacy.
Smart Images

Figure CN116467725B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of information hiding technology, specifically relating to a reversible information hiding method for encrypted 3D meshes based on coordinate axis segmentation. Background Technology
[0002] With the widespread application of 3D digital media, such as 3D printing, virtual reality, and computer-aided design, the protection and security of 3D models are receiving increasing attention. Traditional digital watermarking technology can embed information into 3D models, but it often cannot guarantee the privacy and integrity of the data. Encryption-based technologies can guarantee the privacy and integrity of the data, but they compromise readability and have a low information hiding capacity. Summary of the Invention
[0003] The purpose of this invention is to provide a reversible information hiding method for encrypted 3D meshes based on coordinate axis segmentation, which can improve the information hiding capacity and ensure accuracy when extracting secret data, and can recover the 3D mesh model without loss.
[0004] The technical solution adopted in this invention is a reversible information hiding method for encrypted 3D meshes based on coordinate axis segmentation, which is implemented according to the following steps:
[0005] Step 1: Read the original 3D mesh file, obtain the floating-point coordinate information of each vertex in the original 3D mesh file, preprocess the floating-point coordinate information of each vertex, and obtain the long binary bit stream of the x, y, z axis coordinates of each vertex respectively;
[0006] Step 2: Based on the long binary bit stream of x, y, z axis coordinates of each vertex obtained in Step 1, determine the quadrant to which the current vertex belongs. Divide the vertices in the same quadrant into a group, and obtain the binary bit stream of x, y, z axis coordinates of each vertex in each group. Take the first vertex in each group as the reference center point and do not make any changes. Compare the binary bit streams of the first vertex in each group with the binary bit streams of other vertices to obtain auxiliary information. The second vertex in each group is the auxiliary information point. The 2nd to 6th bits of the auxiliary information point are used as auxiliary information bits to embed auxiliary information.
[0007] Step 3: Set the key K, use a stream cipher function to generate a pseudo-random bit stream, encrypt the vertices processed in Step 2, and then convert the binary information of the encrypted vertices back to floating-point information.
[0008] Step 4: Send the floating-point information obtained in Step 3 to the data embedder. The data embedder obtains the length of the secret data that can be embedded based on the obtained auxiliary information, embeds the secret data into each vertex, and converts the binary vertex information back to floating-point numbers after the embedding is completed.
[0009] Step 5: After receiving the floating-point information obtained in step 4, repeat the embedding operation in step 3 to complete the extraction of secret data;
[0010] Step 6: Decrypt the mesh and restore it without loss.
[0011] The invention is further characterized in that,
[0012] In step 1, the specific preprocessing process is as follows:
[0013] Step 1.1: Convert the floating-point coordinates of each vertex to 0 to 2. k Integers between -1, expressed as:
[0014]
[0015] In equation (1), x, y, and z represent the x, y, and z axes of the three-dimensional coordinate system, respectively; v i,j Represents the x, y, and z coordinates of each vertex;
[0016] Step 1.2, take each integer v′ obtained in Step 1.1 i,j Convert to a 32-bit binary bit stream;
[0017] Step 1.3: Concatenate the binary bit streams of the x, y, and z axis coordinates of each vertex obtained in Step 1.2 into a long binary bit stream according to the vertex index.
[0018] In step 2, the specific process of comparing the binary bitstreams of the first vertex with those of the other vertices is as follows:
[0019] The binary bit stream of the x-axis coordinate of the first vertex in each group is compared with the binary bit stream of the x-axis coordinate of other vertices starting from the most significant bit. The comparison is stopped when a position with mismatched bit information is encountered, and the length of that position is recorded as auxiliary information. The binary bit streams of the y-axis coordinate and z-axis coordinate are processed in the same way.
[0020] In step 2, the auxiliary information embedding needs to be converted from decimal length information into a binary bit stream of length 5, expressed as:
[0021]
[0022] In equation (2), b represents a binary bit stream and m represents a decimal integer.
[0023] In step 3, the specific encryption process is as follows:
[0024] The first bit of each vertex and the first 6 bits of auxiliary information points within each group are not encrypted. All other vertices within each group are encrypted. The vertices to be encrypted are in binary bitstream form (b). i,j,0 b i,j,1 …b i,j,k Where 1≤i≤N and j∈{x,y,z}, that is:
[0025]
[0026] Using a key K, a pseudo-random bitstream is generated using a stream cipher function, and then the corresponding vertices are encrypted.
[0027]
[0028] In equation (4), k i,j,u It is a key bit stream, e i,j,u It is the generated ciphertext, and ⊕ represents the XOR operation.
[0029] The specific process of step 4 is as follows:
[0030] The 3D network file processed in step 3 is sent to the data embedding party. After obtaining the 3D network file, the data embedding party traverses the floating-point vertex information according to the vertex index, converts the floating-point number into a binary bit stream. Since the first bit of the binary information of each vertex is not encrypted, the quadrant to which the vertex belongs can be obtained based on the first bit of the binary information of each vertex. Then, vertices belonging to the same quadrant are grouped into the same quadrant group. Auxiliary information is obtained based on the second to sixth bits of the auxiliary information points in each quadrant group, thereby obtaining the length of information that can be embedded in the current quadrant group. The auxiliary information points in each quadrant group start embedding secret data from the seventh bit. The first bit of each vertex remains unchanged, and the other vertices start embedding secret data from the second bit. After the embedding is completed, the binary vertex information is converted back to floating-point number.
[0031] Since the auxiliary information is a binary bit stream of length 5, it needs to be converted into decimal to obtain the length of information that can be embedded in the current quadrant group. The expression is:
[0032] m = b1 × 2 L-1 +b2×2 L-2 +…b n-1 ×2 L-n+1 (5)
[0033] In equation (5), m represents a decimal integer, b represents a binary bit, and L represents the binary bit length.
[0034] The specific process of step 6 is as follows: Since the first vertex in each quadrant group obtained in step 5 has not changed, according to the auxiliary information, it can be found that the first few bits of the other vertices in each quadrant group are the same as the first vertex. Then, the first few bits of the first vertex are directly overwritten to the other vertices. At this time, the other vertices are restored to the original information. Then, the decrypted bit stream is generated according to the key K and XORed with the vertices with embedded secret information to obtain the decrypted grid data and completely restore the grid.
[0035] The beneficial effects of this invention are:
[0036] (1) The method of the present invention firstly encrypts the grid file, reserving space for third-party cloud administrators to embed data so as to facilitate their management of the file. However, the cloud cannot see the original file. The receiving end can recover the file without loss through the encryption key, thus achieving the purpose of protecting the file. By encrypting the grid file in advance through the encryption system, the problem of cloud platform data leakage can be solved well.
[0037] (2) According to the quadrant grouping method of the present invention, all mesh models can be divided into 8 groups, that is, there are only 8 auxiliary information. More or less information can be embedded in the same group, so a large amount of information can be embedded, thus having a higher embedding capacity.
[0038] (3) The method of the present invention can restore the original information by using the first vertex in each quadrant group of the embedded information points, thus enabling the complete recovery of the original grid. Attached Figure Description
[0039] Figure 1 This is a framework diagram of the encrypted 3D mesh reversible information hiding method based on coordinate axis segmentation according to the present invention;
[0040] Figure 2 The original model of the 3D mesh model "elephant";
[0041] Figure 3 A demonstration of embedding the 3D mesh model "elephant" into quadrant groups;
[0042] Figure 4 It is the encrypted version of the 3D mesh model "elephant";
[0043] Figure 5 It is the model after extracting data from the 3D mesh model "elephant";
[0044] Figure 6 It is the decrypted and restored model of the 3D mesh model "elephant". Detailed Implementation
[0045] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.
[0046] This invention relates to a reversible information hiding method for encrypted 3D meshes based on coordinate axis segmentation, such as... Figure 1 As shown, please follow these steps:
[0047] Step 1, read as follows Figure 2 The original 3D mesh file is shown. The 32-bit floating-point coordinate information of each vertex in the original 3D mesh file is obtained. The floating-point coordinate information of each vertex is preprocessed to obtain the long binary bit stream of the x, y, and z axis coordinates of each vertex.
[0048] The specific process of preprocessing is as follows:
[0049] Step 1.1: Convert the floating-point coordinates of each vertex to 0 to 2. k Integers between -1, expressed as:
[0050]
[0051] In equation (1), x, y, and z represent the x, y, and z axes of the three-dimensional coordinate system, respectively; v i,j Represents the x, y, and z coordinates of each vertex;
[0052] Step 1.2, take each integer v′ obtained in Step 1.1 i,j Convert to a 32-bit binary bit stream;
[0053] Step 1.3: Concatenate the binary bit streams of the x, y, and z axis coordinates of each vertex obtained in Step 1.2 into a long binary bit stream according to the vertex index.
[0054] Step 2: Based on the long binary bit stream of x, y, z axis coordinates of each vertex obtained in Step 1, determine the quadrant to which the current vertex belongs. Divide the vertices in the same quadrant into a group, and obtain the binary bit stream of x, y, z axis coordinates of each vertex in each group. Take the first vertex in each group as the reference center point and do not make any changes. Compare the binary bit streams of the first vertex in each group with the binary bit streams of other vertices to obtain auxiliary information. The second vertex in each group is the auxiliary information point. The 2nd to 6th bits of the auxiliary information point are used as auxiliary information bits to embed auxiliary information.
[0055] The specific process of comparing the binary bitstream of the first vertex with that of the other vertices is as follows:
[0056] The binary bit stream of the x-axis coordinate of the first vertex in each group is compared with the binary bit stream of the x-axis coordinate of other vertices starting from the most significant bit. The comparison is stopped when a position with mismatched bit information is encountered, and the length of that position is recorded as auxiliary information. The binary bit streams of the y-axis coordinate and z-axis coordinate are processed in the same way.
[0057] The auxiliary information embedding needs to be converted from decimal length information into a binary bit stream of length 5, as shown in the expression:
[0058]
[0059] In equation (2), b represents a binary bit stream and m represents a decimal integer;
[0060] like Figure 3 As shown, taking the x-coordinate of each vertex within a group as an example, the leftmost bit is a floating-point number, which is then converted to an integer, and finally becomes 32 bits of binary information on the right. Comparing from the highest bit, an inconsistency is found up to the 16th bit. Therefore, the first 15 bits of the x-coordinate of this group are consistent, and the number 15 is recorded. This number 15 is auxiliary information, used to determine where to embed the data. The dotted ellipse in the diagram indicates where the auxiliary information 15 is embedded. However, 15 must be converted to binary before it can be embedded. The first bit of each vertex cannot be manipulated because this bit is needed to determine the sign of the vertex's x-coordinate, which is used to determine which coordinate system the vertex belongs to. The positions of the second vertex, excluding the first bit and the auxiliary information bits, can be used to embed secret data. The positions of other vertices, excluding the first bit, can also be used to embed secret data. The same logic applies to the binary bit streams of the y-axis and z-axis coordinates.
[0061] Step 3: Set the key K, use a stream cipher function to generate a pseudo-random bitstream, encrypt the vertices processed in Step 2, and then convert the binary information of the encrypted vertices back to floating-point information, such as... Figure 4 As shown, the encrypted model was obtained;
[0062] The specific encryption process is as follows:
[0063] The first bit of each vertex and the first 6 bits of auxiliary information points within each group are not encrypted. All other vertices within each group are encrypted. The vertices to be encrypted are in binary bitstream form (b). i,j,0 b i,j,1 …b i,j,k Where 1≤i≤N and j∈{x,y,z}, that is:
[0064]
[0065] Using a key K, a pseudo-random bitstream is generated using a stream cipher function, and then the corresponding vertices are encrypted.
[0066]
[0067] In equation (4), k i,j,u It is a key bit stream, e i,j,u This is the generated ciphertext, and ⊕ represents the XOR operation;
[0068] Step 4: Send the floating-point information obtained in Step 3 to the data embedder. The data embedder embeds the secret data into each vertex. After embedding, the binary vertex information is converted back to floating-point numbers.
[0069] The specific process is as follows:
[0070] The 3D network file processed in step 3 is sent to the data embedding party. After receiving the 3D network file, the data embedding party traverses the floating-point vertex information according to the vertex index, converts the floating-point number into a binary bit stream. Since the first bit of the binary information of each vertex is not encrypted, the quadrant to which the vertex belongs can be obtained based on the first bit of the binary information of each vertex. Then, vertices belonging to the same quadrant are grouped into the same quadrant group. Auxiliary information is obtained based on the second to sixth bits of the second vertex (auxiliary information point) in each quadrant group, thereby obtaining the length of information that can be embedded in the current quadrant group. The second vertex (auxiliary information point) in each quadrant group embeds secret data starting from the seventh bit. The first bit of each vertex remains unchanged, and the other vertices embed secret data starting from the second bit. After the embedding is completed, the binary vertex information is converted back to floating-point number.
[0071] Since the auxiliary information is a binary bit stream of length 5, it needs to be converted into decimal to obtain the length of information that can be embedded in the current quadrant group. The expression is:
[0072] m = b1 × 2 L-1 +b2×2 L-2 +…b n-1 ×2 L-n+1 (5)
[0073] In equation (5), m represents a decimal integer, b represents a binary bit, and L represents the binary bit length;
[0074] Step 5: After receiving the floating-point information obtained in Step 4, repeat the operation in Step 4 to extract the secret data embedded in Step 5, such as... Figure 5 As shown, the model after data extraction was obtained;
[0075] The specific process is as follows:
[0076] Convert the floating-point number into a binary bit stream. Since the first bit of the binary information of each vertex is not encrypted, the quadrant to which the vertex belongs can be obtained based on the first bit of the binary information of each vertex. Then, vertices belonging to the same quadrant are grouped into the same quadrant group. Auxiliary information is obtained based on the second to sixth bits of the second vertex (auxiliary information point) in each quadrant group, thereby obtaining the length of information that can be embedded in the current quadrant group. The secret data of the second vertex (auxiliary information point) in each quadrant group is extracted from the seventh bit. The first bit of each vertex remains unchanged, and the secret data of other vertices is extracted from the second bit.
[0077] Since the auxiliary information is a binary bit stream of length 5, it is necessary to convert the auxiliary information into decimal according to formula (5) in order to obtain the length of information that can be embedded in the current quadrant group.
[0078] Step 6: Decrypt the mesh and restore it without loss, such as... Figure 6 The decrypted and recovered model is shown below;
[0079] Since the first vertex in each quadrant group obtained in step 5 has not changed, according to the auxiliary information, it can be found that the first few bits of the other vertices in each quadrant group are the same as the first vertex. Therefore, the first few bits of the first vertex are directly overwritten to the other vertices. At this time, the other vertices are restored to the original information. Then, the decrypted bit stream generated according to the key K is XORed with the vertices with embedded secret information to obtain the decrypted grid data and completely restore the grid.
Claims
1. A method for reversible information hiding in encrypted 3D meshes based on coordinate axis segmentation, characterized in that, The specific steps are as follows: Step 1: Read the original 3D mesh file, obtain the floating-point coordinate information of each vertex in the original 3D mesh file, preprocess the floating-point coordinate information of each vertex, and obtain the long binary bit stream of the x, y, z axis coordinates of each vertex respectively; Step 2: Based on the long binary bit stream of x, y, z axis coordinates of each vertex obtained in Step 1, determine the quadrant to which the current vertex belongs. Divide the vertices in the same quadrant into a group, and obtain the binary bit stream of x, y, z axis coordinates of each vertex in each group. Take the first vertex in each group as the reference center point and do not make any changes. Compare the binary bit streams of the first vertex in each group with the binary bit streams of other vertices to obtain auxiliary information. The second vertex in each group is the auxiliary information point. The 2nd to 6th bits of the auxiliary information point are used as auxiliary information bits to embed auxiliary information. In step 2, the specific process of comparing the binary bitstreams of the first vertex with those of the other vertices is as follows: The binary bit stream of the x-axis coordinate of the first vertex in each group is compared with the binary bit stream of the x-axis coordinate of other vertices starting from the most significant bit. The comparison is stopped when a position with mismatched bit information is encountered, and the length of that position is recorded as auxiliary information. The binary bit streams of the y-axis coordinate and z-axis coordinate are processed in the same way. Step 3: Set the key K, use a stream cipher function to generate a pseudo-random bit stream, encrypt the vertices processed in Step 2, and then convert the binary information of the encrypted vertices back to floating-point information. In step 3, the specific encryption process is as follows: The first bit of each vertex and the first 6 bits of auxiliary information points within each group are not encrypted. All other vertices within each group are encrypted, and the vertices to be encrypted are in binary bitstream form. , of which 1 And j {x,y,z}, that is: Using a key K, a pseudo-random bitstream is generated using a stream cipher function, and then the corresponding vertices are encrypted. In the formula, It is a key bitstream. It is the generated ciphertext. Indicates the XOR operation; Step 4: Send the floating-point information obtained in Step 3 to the data embedding party. The data embedding party obtains the length of the secret data that can be embedded based on the obtained auxiliary information, embeds the secret data into each vertex, and converts the binary vertex information back to floating-point numbers after the embedding is completed. The specific process of step 4 is as follows: The 3D mesh file processed in step 3 is sent to the data embedding party. After obtaining the 3D mesh file, the data embedding party traverses the floating-point vertex information according to the vertex index, converts the floating-point number into a binary bit stream. Since the first bit of the binary information of each vertex is not encrypted, the quadrant to which the vertex belongs can be obtained based on the first bit of the binary information of each vertex. Then, vertices belonging to the same quadrant are grouped into the same quadrant group. Auxiliary information is obtained based on the second to sixth bits of the auxiliary information points in each quadrant group, thereby obtaining the length of information that can be embedded in the current quadrant group. The auxiliary information points in each quadrant group start embedding secret data from the seventh bit. The first bit of each vertex remains unchanged, and the other vertices start embedding secret data from the second bit. After the embedding is completed, the binary vertex information is converted back to floating-point number. Since the auxiliary information is a binary bit stream of length 5, it needs to be converted into decimal to obtain the length of information that can be embedded in the current quadrant group. The expression is: In the formula, m represents a decimal integer, b represents a binary bit stream, and L represents the binary bit length; Step 5: After receiving the floating-point information obtained in Step 4, repeat the encryption operation in Step 3 to complete the extraction of the secret data; Step 6: Decrypt the mesh and restore it without loss; The specific process of step 6 is as follows: Since the first vertex in each quadrant group obtained in step 5 has not changed, according to the auxiliary information, it can be found that the first few bits of the other vertices in each quadrant group are the same as the first vertex. Then, the first few bits of the first vertex are directly overwritten to the other vertices. At this time, the other vertices are restored to the original information. Then, according to the key K, the decrypted bit stream is generated and XORed with the vertices with embedded secret information to obtain the decrypted grid data and completely restore the grid.
2. The method for reversible information hiding in encrypted 3D meshes based on coordinate axis segmentation according to claim 1, characterized in that, In step 1, the specific preprocessing process is as follows: Step 1.1: Convert the floating-point coordinates of each vertex to 0 to 1. Integers between, the expression is: = ,j=x,y,z In the formula, x, y, and z represent the x, y, and z axes of the three-dimensional coordinate system, respectively; Represents the x, y, and z coordinates of each vertex; Step 1.2, take each integer obtained in Step 1.1 Convert to a 32-bit binary bit stream; Step 1.3: Concatenate the binary bit streams of the x, y, and z axis coordinates of each vertex obtained in Step 1.2 into a long binary bit stream according to the vertex index.
3. The method for reversible information hiding in encrypted 3D meshes based on coordinate axis segmentation according to claim 1, characterized in that, In step 2, the auxiliary information embedding needs to be converted from decimal length information into a binary bit stream of length 5, expressed as: In the formula, b represents a binary bit stream and m represents a decimal integer.
Citation Information
Patent Citations
3D mesh reversible information hiding algorithm based on partitioning and multi-MSB prediction
CN115150621A
Reversible data hiding method based on Huffman coding rule and position graph compression
CN115297218A
