Digital Watermark Embedding and Extraction Method for Real Scene 3D Model Data

Abstract

The present invention provides a method for digital watermark embedding and extraction of real - scene three - dimensional model data, belonging to the technical field of digital watermarking. The embedding method includes: determining watermark information in binary data format and determining the total number of bits of the watermark information; determining at least one carrier triangular face in the real - scene three - dimensional model data, where the carrier triangular face includes three edges; adding vertices to each of the edges to obtain new vertices, and the new vertices divide the edge into a first partial edge and a second partial edge; constructing a correspondence relationship between the ratio of the first partial edge and the edge and the stored bits; segmenting the watermark information based on the stored bits and the total number of bits to generate multiple watermark information segments corresponding to the new vertices, and embedding the watermark information segments into the real - scene three - dimensional model data based on the correspondence relationship. The present invention improves the applicability and robustness of the digital watermark embedding method.

Description

Technical Field

[0001] The present invention relates to the technical field of digital watermarking, and particularly relates to a method for embedding and extracting digital watermarks in real - scene three - dimensional model data. Background Art

[0002] As an important geographical data, real - scene three - dimensional model data has commercial value and confidentiality. The leakage, illegal use or dissemination of model data will not only damage the interests of enterprises, but also pose a threat to national security. Therefore, it is particularly important to take effective protection measures to prevent data theft and illegal dissemination. As a common information security means, digital watermarking technology is widely used in fields such as audio, video, vector graphics, and raster data. This technology embeds copyright information into digital data through specific methods, making it an inseparable part of the data, thus endowing the data with an "identity". This "identity" can not only effectively prevent unauthorized copying and use, but also provide evidence in case of infringement to help copyright owners safeguard their legitimate rights and interests.

[0003] Existing three - dimensional model digital watermarking algorithms can be divided into the following two types: point - based watermarking algorithms and texture - based watermarking algorithms. Point - based watermarking algorithms hide information by modifying vertex coordinates, which to a certain extent destroys the accuracy of the original data. At the same time, point - based watermarking algorithms are relatively vulnerable and are not easy to resist conventional mesh processing means such as rotation, translation, scaling, and cropping; Texture - based watermarking algorithms are inherently ineffective for models lacking textures and lack universality.

[0004] Therefore, there is an urgent need to provide a method for embedding and extracting digital watermarks in real - scene three - dimensional model data to improve the applicability and robustness of the digital watermark embedding method. Summary of the Invention

[0005] In view of this, it is necessary to provide a method for embedding and extracting digital watermarks in real - scene three - dimensional model data to solve the technical problem of the low applicability and robustness of digital watermarking algorithms in the prior art.

[0006] On the one hand, to solve the above - mentioned technical problems, the present invention provides a method for embedding digital watermarks in real - scene three - dimensional model data, including:

[0007] Determine the watermark information in binary data format and determine the total number of bits of the watermark information;

[0008] Determine at least one carrier triangular face in the real - scene three - dimensional model data, and the carrier triangular face includes three edges;

[0009] Add vertices to each of the edges to obtain new vertices, and the new vertices divide the edge into a first - part edge and a second - part edge;

[0010] Construct the correspondence between the ratio of the first part of the edge and the edge and the stored bit positions;

[0011] Segment the watermark information based on the stored bit positions and the total bit positions to generate multiple watermark information segments corresponding to the new vertices, and embed the watermark information segments into the real-scene three-dimensional model data based on the correspondence.

[0012] In a possible implementation, the determining of at least one carrier triangular face in the real-scene three-dimensional model data includes:

[0013] Obtain multiple target triangular faces in the real-scene three-dimensional model data;

[0014] Determine the face storage bit positions of each of the target triangular faces, and determine the first ratio of the total bit positions to the face storage bit positions;

[0015] Determine the at least one carrier triangular face among the multiple target triangular faces based on the first ratio; the number of the at least one carrier triangular face is the same as the first ratio.

[0016] In a possible implementation, the determining of at least one carrier triangular face in the real-scene three-dimensional model data includes:

[0017] Obtain multiple target triangular faces in the real-scene three-dimensional model data;

[0018] Obtain at least one set of target triangular faces based on the multiple target triangular faces; the set of target triangular faces includes a characteristic triangular face and three neighbor triangular faces adjacent to the characteristic triangular face; both the characteristic triangular face and the neighbor triangular faces are the carrier triangular faces;

[0019] Determine the set storage bit positions of each of the sets of target triangular faces, and determine the second ratio of the total bit positions to the set storage bit positions;

[0020] Determine at least one set of carrier triangular faces among the at least one set of target triangular faces based on the second ratio; the number of the at least one set of carrier triangular faces is the same as the second ratio.

[0021] In a possible implementation, the real-scene three-dimensional model data includes multiple initial triangular faces; then the obtaining of multiple target triangular faces in the real-scene three-dimensional model data includes:

[0022] Determine each interior angle of each of the initial triangular faces;

[0023] Judge whether the interior angle is within a preset interior angle range, and if so, the initial triangular face is the target triangular face.

[0024] In a possible implementation, determining the watermark information in binary data format includes:

[0025] Obtain the initial watermark information, perform a cyclic redundancy check on the initial watermark information, and generate a cyclic redundancy check code;

[0026] Merge the cyclic redundancy check code to the end of the initial watermark information to obtain the initial verified watermark information;

[0027] Convert the initial verified watermark information into binary data format to obtain the watermark information.

[0028] In a possible implementation, before storing the watermark information segment to the new vertex, it further includes:

[0029] Convert the number of segments of the multiple watermark information segments into binary data format to obtain the binary number of segments;

[0030] Merge the binary number of segments to the head of each watermark information segment to obtain the merged watermark information segment;

[0031] Verify the merged watermark information segment to obtain a check code, and merge the check code to the end of the merged watermark information segment to obtain the verified watermark information segment.

[0032] In a possible implementation, embedding the watermark information segment into the real - scene three - dimensional model data based on the corresponding relationship includes:

[0033] Determine the vertex position of the new vertex based on the corresponding relationship and the watermark information segment;

[0034] Based on the vertex position, divide the carrier triangular face into four triangular faces, and use the triangular face composed of the new vertex among the four triangular faces as the replacement triangular face;

[0035] Replace the carrier triangular face with the replacement triangular face to obtain the real - scene three - dimensional model data embedded with the watermark information.

[0036] In a possible implementation, embedding the watermark information segment into the real - scene three - dimensional model data based on the corresponding relationship includes:

[0037] Obtain the interior angles of each triangular face of the carrier triangular face;

[0038] Determine the embedding order of multiple new vertices based on the interior angles of the triangular face;

[0039] Embed the watermark information segment into the real - scene three - dimensional model data based on the embedding order and the corresponding relationship.

[0040] In a possible implementation, the method further includes:

[0041] Determine the new texture coordinates of the new vertex based on the texture coordinates of each vertex in the carrier triangular face, and assign the new texture coordinates to the new vertex.

[0042] On the other hand, the present invention also provides a method for extracting digital watermark from real-scene three-dimensional model data, including:

[0043] Determine new vertices from the real-scene three-dimensional model data;

[0044] Determine the first partial edge of the edge where the new vertex is located and the ratio of the edge;

[0045] Determine the stored bit positions based on the ratio and the correspondence between the ratio and the stored bit positions, where the stored bit positions store watermark information segments;

[0046] Stitch together multiple watermark information segments to obtain watermark information.

[0047] The beneficial effects of the present invention are as follows: In the digital watermark embedding method for real-scene three-dimensional model data provided by the present invention, after adding vertices to each edge of the carrier triangular face, by constructing the correspondence between the first partial edge of the edge where the new vertex is located and the ratio of the edge and the stored bit positions, the stored bit positions can be converted into the position coordinates of the new vertex according to the correspondence, thereby realizing the embedding of watermark information. Moreover, for the watermark in the form of a ratio, the geometric ratio does not change during the rotation, translation, and a certain degree of scaling of the real-scene three-dimensional model data, that is, it satisfies anti-rotation, scale, and zoom, improving the robustness of the digital watermark embedding method. And the digital watermark embedding in the present invention is applicable to both real-scene three-dimensional model data including texture information and those without texture information, improving the applicability of the digital watermark embedding method.

[0048] Furthermore, the watermark information in the present invention is divided into multiple watermark information segments corresponding to the new vertices, with a relatively high embedding ratio. When the real-scene three-dimensional model data is cropped, it is easy to include complete watermark information segments, having a certain degree of cropping robustness. And by adopting the segmented embedding method, the watermark information can be completely embedded into the real-scene three-dimensional model data without setting up a database to store the complete watermark information, realizing the traceability of the source and copyright information of the real-scene three-dimensional model data without relying on a database.

[0049] Even further, the present invention realizes the embedding of watermark information based on the ratio relationship. When extracting the watermark information, as long as the position offset of the new vertex containing the watermark information does not exceed 1 / 20 of the length of the edge where it is located, it can ensure the correct extraction of the watermark information at this point, having good robustness to data compression with significantly reduced precision, and further improving the robustness of the digital watermark embedding method. Description of the Drawings

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0051] Figure 1 It is a schematic flowchart of an embodiment of the digital watermark embedding method for real-scene three-dimensional model data provided by the present invention;

[0052] Figure 2 It is a schematic structural diagram of an embodiment of adding a new vertex provided by the present invention;

[0053] Figure 3 It is a schematic diagram of the first embodiment of the correspondence between the ratio and decimal value of the new vertex provided by the present invention;

[0054] Figure 4 It is a schematic flowchart of an embodiment of determining the carrier triangular face in step S102 of the present invention;

[0055] Figure 5 It is a schematic flowchart of another embodiment of determining the carrier triangular face in step S102 of the present invention;

[0056] Figure 6 It is a schematic structural diagram of an embodiment of the new vertex in the carrier triangular face set provided by the present invention;

[0057] Figure 7 It is a schematic flowchart of an embodiment of determining the target triangular face provided by the present invention;

[0058] Figure 8 It is a schematic flowchart of an embodiment of determining the watermark information in step S101 of the present invention;

[0059] Figure 9 It is a schematic flowchart of an embodiment of verifying the watermark information segment provided by the present invention;

[0060] Figure 10 It is a schematic flowchart of an embodiment of embedding the watermark information segment into the real-scene three-dimensional model data based on the correspondence in step S105 of the present invention;

[0061] Figure 11 It is a schematic flowchart of another embodiment of embedding the watermark information segment into the real-scene three-dimensional model data based on the correspondence in step S105 of the present invention;

[0062] Figure 12Schematic flow chart of an embodiment of the digital watermark extraction method for the real - scene three - dimensional model data provided by the present invention;

[0063] Figure 13 Schematic diagram of an embodiment of the recoverable original large triangle provided by the present invention;

[0064] Figure 14 Schematic diagram of an embodiment of the non - recoverable original large triangle provided by the present invention;

[0065] Figure 15 Schematic diagram of an embodiment of the extraction of the carrier triangular face set provided by the present invention. Detailed implementation manners

[0066] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with 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, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0067] It should be understood that the schematic drawings are not drawn to scale. The flowcharts used in the present invention illustrate the operations implemented according to some embodiments of the present invention. It should be understood that the operations in the flowchart may not be implemented in sequence, and steps without logical context relationships may be reversed or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of the present invention. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor systems and / or microcontroller systems.

[0068] Referring to "embodiments" herein means that the specific features, structures, or characteristics described in conjunction with the embodiments can be included in at least one embodiment of the present invention. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0069] The present invention provides a digital watermark embedding and extraction method for real - scene three - dimensional model data, which will be described separately below.

[0070] Before presenting the embodiments, the watermark information in the embodiments of the present invention will be introduced first.

[0071] The watermark information in the embodiment of the present invention refers to an implicit watermark, which embeds copyright information or identification marks into the original data through a specific algorithm, making the copyright information almost imperceptible visually or aurally, while being able to be extracted when needed. This watermark information has the following characteristics:

[0072] 1. Concealment, which is not easy for users to detect during use;

[0073] 2. Data integrity: data is not destroyed and users can still use the data normally;

[0074] 3. Robustness. The watermark can resist various attacks and processing, such as compression, cropping, noise addition, etc.

[0075] 4. Security: Unauthorized users cannot access or tamper with watermark information. Security can usually be enhanced through encryption technology.

[0076] Figure 1 A schematic flow chart of an embodiment of a method for embedding a digital watermark into real-scene 3D model data provided by the present invention is shown in FIG. Figure 1 As shown, the digital watermark embedding method of real scene 3D model data includes:

[0077] S101, determining watermark information in binary data format, and determining the total bits of the watermark information;

[0078] S102, determining at least one carrier triangle face in the real-scene 3D model data, where the carrier triangle face includes three edges;

[0079] S103, adding vertices to each edge to obtain newly added vertices, wherein the newly added vertices divide the edge into a first part of edges and a second part of edges;

[0080] S104, constructing a correspondence between the first part of edges, the ratio of edges and the storage bits;

[0081] S105, segmenting the watermark information based on the storage bits and the total bits, generating a plurality of watermark information segments corresponding to the newly added vertices, and embedding the watermark information segments into the real-scene 3D model data based on the corresponding relationship.

[0082] The corresponding relationship represents the corresponding relationship between the position of the newly added vertex and the binary information, that is, the newly added vertex falling at different positions on the edge represents different binary information, thereby achieving the purpose of embedding the watermark information in the real-scene three-dimensional model data.

[0083] It should be understood that the number of newly added vertices and the corresponding relationship can be reasonably set according to actual needs. Figure 2As shown in the figure, each side of the carrier triangular surface is divided into 9 parts, that is, 8 positions are set on each side of the carrier triangular surface for adding new vertices. Specifically, the new vertex on side AB is F, the new vertex on side AC is E, and the new vertex on side BC is D.

[0084] Based on the above new vertices, the corresponding relationship is as Figure 3 shown. Taking side AB as an example, when adding new vertex F, the ratios of AF to AB can be 0.15, 0.25, 0.35, 0.45, 0.55, 0.65, 0.75, and 0.85 respectively. The decimal values corresponding to the stored bit positions are 0, 1, 2, 3, 4, 5, 6, 7 respectively. The above decimal values can be represented by three binary bit positions.

[0085] In other words, the number of bits of the stored bit position is 3. Then the size of the watermark information that can be stored in a carrier triangular surface is 3 bit × 3 = 9 bit.

[0086] Compared with the prior art, the digital watermark embedding method for real - scene three - dimensional model data provided by the embodiments of the present invention adds vertices to each side of the carrier triangular surface, constructs the corresponding relationship between the ratio of the first part of the side where the new vertex is located and the side and the stored bit position, and then the stored bit position can be converted into the position coordinates of the new vertex according to the corresponding relationship to realize the embedding of watermark information. Moreover, the watermark in the form of ratio does not change the geometric ratio during the rotation, translation, and a certain degree of scaling of the real - scene three - dimensional model data, that is, it satisfies anti - rotation, ratio, and scaling, improving the robustness of the digital watermark embedding method. And the digital watermark embedding in the present invention is applicable to both real - scene three - dimensional model data with texture information and without texture information, improving the applicability of the digital watermark embedding method.

[0087] Furthermore, the watermark information in the embodiments of the present invention is divided into multiple watermark information segments corresponding to the new vertices, with a relatively high embedding ratio. When the real - scene three - dimensional model data is cropped, it is easy to contain complete watermark information segments, having a certain cropping robustness. And by adopting the segmented embedding method, the watermark information can be completely embedded into the real - scene three - dimensional model data without setting up a database to store the complete watermark information, realizing the traceability of the source and copyright information of the real - scene three - dimensional model data without relying on the database.

[0088] Even further, the embodiments of the present invention realize the embedding of watermark information based on the proportional relationship. When extracting the watermark information, as long as the position offset of the new vertex containing the watermark information does not exceed 1 / 20 of the side length where it is located, it can ensure the correct extraction of the watermark information at this point, having good robustness to data compression with significantly reduced precision, and further improving the robustness of the digital watermark embedding method.

[0089] Since the total number of bits of the watermark information varies with the amount of information in the watermark information, the watermark information needs to be completely embedded in the real-scene three-dimensional model data. To achieve this purpose, in some embodiments of the present invention, such as Figure 4 shown, determining at least one carrier triangular face in the real-scene three-dimensional model data in step S102 includes:

[0090] S401. Obtain a plurality of target triangular faces in the real-scene three-dimensional model data;

[0091] S402. Determine the face storage bit number of each target triangular face, and determine the first ratio of the total bit number to the face storage bit number;

[0092] S403. Determine at least one carrier triangular face among the plurality of target triangular faces based on the first ratio; the number of at least one carrier triangular face is the same as the first ratio.

[0093] By determining the first ratio of the total bit number and the face storage bit number in the embodiments of the present invention, the number of carrier triangular faces can be determined based on the first ratio, and the three sides of the carrier triangular face are used as embedding units instead of adding vertices to one of the sides of the carrier triangular face, achieving the purpose of reducing the number of carrier triangular faces while embedding the complete watermark information, improving the embedding amount, that is, improving the information carrying capacity in the real-scene three-dimensional model data.

[0094] Based on Figure 2 the carrier triangular face, the face storage bit number is 9 bits.

[0095] Since the triangular faces in the real-scene three-dimensional model data are all adjacent to each other, that is, two adjacent triangular faces share one side. Based on this characteristic, in some embodiments of the present invention, such as Figure 5 shown, determining at least one carrier triangular face in the real-scene three-dimensional model data in step S102 includes:

[0096] S501. Obtain a plurality of target triangular faces in the real-scene three-dimensional model data;

[0097] S502. Obtain at least one target triangular face set based on the plurality of target triangular faces; the target triangular face set includes a characteristic triangular face and three neighbor triangular faces adjacent to the characteristic triangular face; the characteristic triangular face and the neighbor triangular faces are all carrier triangular faces;

[0098] S503. Determine the set storage bit number of each target triangular face set, and determine the second ratio of the total bit number to the set storage bit number;

[0099] S504. Determine at least one carrier triangular face set among at least one target triangular face set based on the second ratio; the number of at least one carrier triangular face set is the same as the second ratio.

[0100] In the embodiment of the present invention, by obtaining the geometry of the target triangular face, the setting of two new vertices on the common edge of adjacent triangular faces is avoided, resulting in the position conflict of the new vertices, and the reliability of watermark information embedding is improved.

[0101] Specifically, step S502 is as follows: Establish a neighbor triangular face relationship, which refers to the relationship based on which each triangular face adjacent to a large triangular face in the real-scene three-dimensional model data can be found. The input of the neighbor triangular face relationship is the sequence of triangular faces, and the output is the neighbor triangular faces of each triangular face.

[0102] It should be noted that: The set of triangular faces that must have and only have 3 neighbor triangular faces is the carrier triangular face set.

[0103] In a specific embodiment of the present invention, as Figure 6 shown, a carrier triangular face set includes 9 new vertices, which are respectively Figure 6 numbered 1-9 in. Then the set storage bit is 3bit×9 = 27bit.

[0104] It should be noted that: Different carrier triangular face sets do not have common edges.

[0105] As described above, the total number of new vertices set on each edge of the carrier triangular face in the embodiment of the present invention is the same, and the real-scene three-dimensional model data includes multiple initial triangular faces. There may be narrow triangular faces in the initial triangular faces. To avoid the low position resolution of the short side caused by the appearance of narrow triangular faces, and further cause the technical problem of low compression resistance, in some embodiments of the present invention, as Figure 7 shown, both step S401 and S501 include:

[0106] S701. Determine each interior angle of each initial triangular face;

[0107] S702. Judge whether the interior angle is within a preset interior angle range. If so, the initial triangular face is the target triangular face.

[0108] It should be understood that: The target triangular face should be as equilateral as possible.

[0109] In the embodiment of the present invention, by determining the target triangular face in the initial triangular face based on the interior angle, the target triangular face is approximated as an equilateral triangle, thereby improving the resolution of the new vertices. When the real-scene three-dimensional model data is compressed, the extraction accuracy of the new vertex position information is still ensured, that is: improving the compression resistance performance of the digital watermark embedding method.

[0110] Specifically, the preset interior angle range is 61.5°±10°.

[0111] To ensure the integrity and effectiveness of the watermark information, in some embodiments of the present invention, such as Figure 8 shown, determining the watermark information in binary data format in step S101 includes:

[0112] S801. Obtain the initial watermark information, perform a cyclic redundancy check on the initial watermark information, and generate a cyclic redundancy check code;

[0113] S802. Merge the cyclic redundancy check code to the tail of the initial watermark information to obtain the initial verified watermark information;

[0114] S803. Convert the initial verified watermark information into binary data format to obtain the watermark information.

[0115] In the embodiments of the present invention, by performing a cyclic redundancy check on the initial watermark information, generating a cyclic redundancy check code, and merging the cyclic redundancy check code to the tail of the initial watermark information, the generated watermark information carries the cyclic redundancy check code. Furthermore, when extracting the watermark information, the integrity and effectiveness of the watermark information can be ensured based on the cyclic redundancy check code.

[0116] Since in the embodiments of the present invention, the watermark information is segmented and each watermark information segment is embedded into the real scene three-dimensional model data respectively. To further ensure the effectiveness and integrity of the watermark information, in some embodiments of the present invention, such as Figure 9 shown, before storing the watermark information segment to the newly added vertex, it further includes:

[0117] S901. Convert the number of segments of multiple watermark information segments into binary data format to obtain the binary number of segments;

[0118] S902. Merge the binary number of segments to the head of each watermark information segment to obtain the merged watermark information segment;

[0119] S903. Perform a check on the merged watermark information segment to obtain a check code, and merge the check code to the tail of the merged watermark information segment to obtain the verified watermark information segment.

[0120] In the embodiments of the present invention, by performing a check on each watermark information segment, two-level checks on the watermark information are realized, that is, the segmentation of the watermark information segment and the check of the overall watermark information, further ensuring the integrity and effectiveness of the watermark information. Moreover, in the embodiments of the present invention, by merging the number of segments of the watermark information segment to the head of the watermark information segment, when extracting the watermark information, the watermark information can be quickly spliced according to the number of segments, improving the extraction efficiency of the watermark information.

[0121] In some embodiments of the present invention, such as Figure 10 shown, embedding the watermark information segment into the real scene three-dimensional model data based on the corresponding relationship in step S105 includes:

[0122] S1001, determining the vertex position of the newly added vertex based on the corresponding relationship and the watermark information segment;

[0123] S1002, dividing the carrier triangular face into four triangular faces based on the vertex positions, and using the triangular face composed of the newly added vertices among the four triangular faces as a replacement triangular face;

[0124] S1003: Replace the carrier triangular face with the substitute triangular face to obtain real-scene three-dimensional model data embedded with watermark information.

[0125] When embedding watermark information in the embodiment of the present invention, the shapes of the four triangular faces after segmentation do not change compared with the original triangular faces, that is, the geometric accuracy of the real-scene 3D model data is not changed, thereby ensuring the high fidelity effect of the real-scene 3D model data embedded with the watermark information.

[0126] Since the watermark information segments do not carry sequence information when they are embedded into each newly added vertex, in order to ensure the orderly embedding and accurate reading of the watermark information, in some embodiments of the present invention, such as Figure 11 As shown, embedding the watermark information segment into the real scene 3D model data based on the corresponding relationship in step S105 includes:

[0127] S1101, obtaining the inner angles of each triangular face of the carrier;

[0128] S1102, determining an embedding order of multiple newly added vertices based on the inner angles of the triangle surface;

[0129] S1103: embed the watermark information segment into the real-scene 3D model data based on the embedding order and the corresponding relationship.

[0130] Specifically, the inner angle of the triangle face that is closest to the preset angle is used as the starting point, and all newly added vertices are traversed in a counterclockwise direction as the embedding order of all newly added vertices. Figure 6 As shown in the figure, the newly added vertex with serial number 1 is the starting point. For the central triangle face, the newly added vertices with serial numbers 2 and 3 are determined in counterclockwise order. For the three neighbor triangle faces, the triangle face in the upper left corner is used as the starting point, and the remaining two neighbor triangle faces are traversed counterclockwise. Each neighbor triangle face also finds the starting point separately, and then determines the serial number of the newly added vertex based on the counterclockwise order, and the following is obtained: Figure 6 The newly added vertex numbers are in the order of 1-9.

[0131] Specifically, the preset angle is 61.5°.

[0132] It should be understood that the preset angle can also be adaptively adjusted according to other requirements, which will not be elaborated here.

[0133] When the real scene three-dimensional model data includes texture information, to avoid texture distortion caused by the existence of newly added vertices, in some embodiments of the present invention, after step S105, it further includes:

[0134] Determine the new texture coordinates of the newly added vertices based on the texture coordinates of the vertices in the carrier triangular face, and assign the new texture coordinates to the newly added vertices.

[0135] Specifically, the texture coordinates of the newly added vertices are calculated by interpolation. Given the texture coordinates of the original edge endpoints and the ratio of the distances from the newly added vertex to the two endpoints, the texture coordinates of the newly added point are interpolated using the inverse distance weighting method.

[0136] Similarly, other point attributes are maintained similar to the texture attributes.

[0137] In the embodiments of the present invention, the texture coordinates of the newly added vertices are interpolated according to the principle of texture mapping during triangulation, ensuring the fidelity of the real scene three-dimensional model data embedded with watermark information.

[0138] In a specific embodiment of the present invention, the watermark information includes long watermarks and short watermarks.

[0139] The long watermark includes 32 segments of watermark information segments, each segment of watermark information segment is 27 bits, where 13 bits store the segment number and check information, and the remaining 14 bits are used to store the real user information; 32 segments of watermark information segments can store 14 bit × 32 / 8 = 54 byte of complete information, where 2 bytes are used for complete watermark verification, and finally the information length allowed for the user to store is 52 byte.

[0140] The short watermark includes 16 segments of watermark information segments, each segment of watermark information segment is 27 bits, where 12 bits are used to store the segment number and check information, and the remaining 15 bits are used to store user information; 16 segments of watermark information segments can store 15 bit × 16 / 8 = 30 byte of complete information, where 2 bytes are taken out to store the check code for complete watermark verification, so finally the information length allowed for the user to store is 28 byte.

[0141] One Chinese character requires two bytes, so the information capacity of the long watermark is 26 Chinese characters, and the information capacity of the short watermark is 14 Chinese characters.

[0142] In a specific embodiment of the present invention, taking a short watermark as an example, the digital watermark embedding method of real-life 3D model data is as follows: first, the cyclic redundancy check code of the input user information is calculated, and the check code is merged into the end of the user information to generate the user information with check; then the user information with check is converted into a binary number and divided into 16 segments, each segment is 30×8 / 16 = 15 bits long, the segment number (0-15) is converted into binary and added to the front of the binary user information, the segment number + the check code of the binary user information is calculated, and the check code is placed after the binary user information, and finally 27 bits of segmented user information with segment number and check code are spliced; finally, the 27 bits are divided into 9 parts, each with 3 bits, which are used to calculate the positions of nine newly added vertices in the triangle set. Secondly, the carrier triangle face set is searched in the real-life 3D model data, wherein the triangle face with an inner angle within the preset inner angle range is regarded as the target triangle face, and the triangle face in the carrier triangle face set must have and only have 3 neighbor triangle faces before the watermark information can be embedded. According to the proportion of the watermark information and the original vertices of the triangle, the new vertices are calculated, and the original triangle is divided into four small triangles using the original vertices and the three new vertices. The original divided triangle is replaced with the triangle composed of the three new vertices, and the remaining three new triangles are inserted into the triangle list of the original model; when all the segmentation information is written into the original model by replacing the original triangle with the divided triangle, the embedding of the watermark information is completed. When adding a new vertex to the original edge, the texture coordinates of the new vertex need to be calculated by interpolation; the texture coordinates of the new point are interpolated using the inverse distance weight method when the texture coordinates of the original edge endpoints and the ratio of the distance from the new vertex to the two endpoints are known.

[0143] In summary, the digital watermark embedding method for real-scene 3D model data proposed in the embodiment of the present invention has the following advantages:

[0144] 1. The geometric accuracy and texture of the original model are not changed, and the high-fidelity effect of the 3D model can be achieved;

[0145] 2. The watermark can be effectively maintained during basic processing such as model rotation, translation, scaling and cropping.

[0146] 3. Ability to withstand data compression with significant reduction in accuracy.

[0147] 4. It can carry large amounts of information in watermarks, achieving traceability without relying on databases.

[0148] On the other hand, the embodiment of the present invention also provides a method for extracting digital watermarks from real-scene 3D model data. Figure 12 As shown, the digital watermark extraction method of real scene 3D model data includes:

[0149] S1201. Determine new vertices from the real-scene three-dimensional model data;

[0150] S1202. Determine the first part of the edge where the new vertex is located and the ratio of the edge;

[0151] S1203. Determine the storage bit positions based on the ratio and the correspondence between the ratio and the storage bit positions. The storage bit positions store the watermark information segments;

[0152] S1204. Concatenate multiple watermark information segments to obtain the watermark information.

[0153] In a specific embodiment of the present invention, taking the short watermark as an example, steps S1201 to S1204 are specifically as follows:

[0154] First, extract a carrier triangular surface set from the real-scene three-dimensional model data. Determine whether the carrier triangular surface set is formed by triangulating a large triangular surface through the collinearity condition. This triangulated triangular surface contains part of the information of the segmented watermark; after judging the triangulation result, restore the original large triangular surface according to the reverse process of the triangulation.

[0155] Specifically, as Figure 13 shown, when points A, B, and F are collinear, the original large triangular surface can be restored. As Figure 14 shown, when points A, B, and F are not collinear, the original large triangular surface cannot be restored.

[0156] Secondly, establish neighbor relationships for all restored original large triangular surfaces. If the three neighbors of an original triangular surface are also original triangular surfaces, then their set is very likely to contain a segment of watermark information; find all such sets for the extraction of segmented information.

[0157] As Figure 15 shown, Figure 15 in the left side, there are a total of four original triangular surfaces labeled A, B, C, and D respectively. Then the set composed of these four original triangular surfaces as shown on the Figure 15 right side is the carrier triangular surface set.

[0158] Next, judge whether the interior angle of the central triangular surface in Figure 15 is within the preset interior angle range, determine the interior angle closest to the preset angle among the interior angles, and use this interior angle as the starting point to determine the positions of 9 new vertices in the counterclockwise order. Calculate 9 ratio values based on the positions of the new vertices, determine 9 3-bit binary numbers based on the 9 ratio values, and arrange the 9 binary numbers into a 27-bit watermark information segment in the embedding order. The last 8 bits of the watermark information segment are the check codes. Calculate an 8-bit check code using the first 19 bits and compare it with the extracted check code. If the two check codes are the same, it indicates that the segmented watermark is successfully extracted.

[0159] Finally, multiple segments of watermark information are merged. The first 4 bits of each watermark information segment record the number of segments of the segmented watermark information segment, and the middle 15 bits are the binary watermark information. The 15 bits of each watermark information segment are filled into the complete watermark according to the number of segments determined by the first 4 bits, totaling 15 bit × 16 / 8 = 30 bytes. The complete watermark information still needs to be verified. The first 28 bytes are taken and the cyclic redundancy check method is used to calculate a 2-byte check code, which is compared with the 2-byte check code extracted at the end. If they are the same, the verification passes. After passing the verification, the digital watermark embedded in the real scene three-dimensional model data in advance is extracted.

[0160] Those skilled in the art can understand that all or part of the processes of implementing the methods of the above embodiments can be completed by instructing relevant hardware (such as processors, controllers, etc.) through computer programs, and the computer programs can be stored in a computer-readable storage medium. Among them, the computer-readable storage medium is a magnetic disk, an optical disc, a read-only memory, or a random access memory, etc.

[0161] The above has introduced in detail a method for digital watermark embedding and extraction of real scene three-dimensional model data. Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A digital watermark embedding method for real-scene three-dimensional model data, characterized in that: include: Determining watermark information in a binary data format and determining the total number of bits of the watermark information; Determine at least one carrier triangle face in the real-scene three-dimensional model data, wherein the carrier triangle face includes three edges; Adding vertices to each of the edges to obtain newly added vertices, wherein the newly added vertices divide the edge into a first portion of edges and a second portion of edges; Constructing a correspondence between the ratio of the first part of edges and the edges and the storage bits; the correspondence indicates that different positions of the newly added vertices on the edges represent different binary information; The watermark information is segmented based on the storage bits and the total bits to generate a plurality of watermark information segments corresponding to the newly added vertices, and the watermark information segments are embedded into the real-scene three-dimensional model data based on the corresponding relationship.

2. The digital watermark embedding method for real-scene 3D model data according to claim 1, characterized in that: The determining of at least one carrier triangular face in the real-scene three-dimensional model data comprises: Acquire multiple target triangular faces in the real-scene three-dimensional model data; Determine the face storage bit position of each of the target triangular faces, and determine a first ratio of the total bit position to the face storage bit position; At least one carrier triangular facet among the plurality of target triangular facets is determined based on the first ratio; the number of the at least one carrier triangular facet is the same as the first ratio.

3. The digital watermark embedding method for real-scene 3D model data according to claim 1, characterized in that: The determining of at least one carrier triangular face in the real-scene three-dimensional model data comprises: Acquire multiple target triangular faces in the real-scene three-dimensional model data; At least one target triangular face set is obtained based on the multiple target triangular faces; the target triangular face set includes a characteristic triangular face and three neighboring triangular faces adjacent to the characteristic triangular face; the characteristic triangular face and the neighboring triangular faces are both the carrier triangular faces; Determine the set storage bit position of each target triangle face set, and determine a second ratio of the total bit position to the set storage bit position; At least one carrier triangle face set in the at least one target triangle face set is determined based on the second ratio; the number of the at least one carrier triangle face set is the same as the second ratio.

4. The digital watermark embedding method for real-scene 3D model data according to claim 2 or 3, characterized in that: The real-scene 3D model data includes a plurality of initial triangular faces; and the step of obtaining a plurality of target triangular faces in the real-scene 3D model data includes: Determine each interior angle of each of the initial triangular faces; It is determined whether the inner angle is within a preset inner angle range, and if so, the initial triangular face is the target triangular face.

5. The digital watermark embedding method for real-scene 3D model data according to claim 1, characterized in that: The watermark information in binary data format is determined, including: Acquire initial watermark information, and perform a cyclic redundancy check on the initial watermark information to generate a cyclic redundancy check code; Merging the cyclic redundancy check code into the tail of the initial watermark information to obtain initial check watermark information; The initial verification watermark information is converted into a binary data format to obtain the watermark information.

6. The digital watermark embedding method for real-scene 3D model data according to claim 1, characterized in that: Before storing the watermark information segment to the newly added vertex, the method further includes: Convert the segment numbers of the plurality of watermark information segments into a binary data format to obtain a binary segment number; Merging the binary segment number into the header of each watermark information segment to obtain a merged watermark information segment; The merged watermark information segment is verified to obtain a verification code, and the verification code is merged into the end of the merged watermark information segment to obtain a verification watermark information segment.

7. The digital watermark embedding method for real-scene 3D model data according to claim 1, characterized in that: The embedding of the watermark information segment into the real-scene three-dimensional model data based on the corresponding relationship includes: Determining the vertex position of the newly added vertex based on the corresponding relationship and the watermark information segment; Divide the carrier triangular face into four triangular faces based on the vertex positions, and use the triangular face composed of the newly added vertices among the four triangular faces as a replacement triangular face; The carrier triangular face is replaced by the replacement triangular face to obtain real-scene three-dimensional model data embedded with the watermark information.

8. The digital watermark embedding method for real-scene 3D model data according to claim 1, characterized in that: The embedding of the watermark information segment into the real-scene three-dimensional model data based on the corresponding relationship includes: Obtaining the inner angles of each triangular face of the carrier triangular face; Determining an embedding order of a plurality of newly added vertices based on the interior angles of the triangular surface; The watermark information segment is embedded into the real-scene three-dimensional model data based on the embedding order and the corresponding relationship.

9. The digital watermark embedding method for real-scene 3D model data according to claim 1, characterized in that: The method further comprises: The newly added texture coordinates of the newly added vertex are determined based on the texture coordinates of each vertex in the carrier triangle surface, and the newly added texture coordinates are assigned to the newly added vertex.

10. A method for extracting digital watermarks from real-scene three-dimensional model data, characterized in that: The method is used to extract watermark information embedded by a digital watermark embedding method based on real-scene 3D model data, wherein the digital watermark embedding method for real-scene 3D model data is the digital watermark embedding method for real-scene 3D model data according to any one of claims 1 to 9, and the digital watermark extraction method for real-scene 3D model data comprises: Determine the newly added vertices from the real-life 3D model data; Determine the ratio of the first part of the edge where the newly added vertex is located to the edge; Determining a storage bit position based on the ratio and a corresponding relationship between the ratio and the storage bit position, wherein the watermark information segment is stored in the storage bit position; Multiple watermark information segments are concatenated to obtain watermark information.

Images