A method and device for processing ownership identification of a three-dimensional model of cultural relics

Abstract

The present application discloses a method and device for processing the identification of cultural relics three-dimensional models, the method comprising: dividing the cultural relics three-dimensional model into at least one area based on the geometric features of the cultural relics three-dimensional model to be processed; determining the sensitivity parameters corresponding to each area based on the area features corresponding to each area; determining the processing parameters corresponding to each area based on the sensitivity parameters corresponding to each area, and processing the original identification information corresponding to the area using the processing parameters to obtain the target identification; adding the target identification to the cultural relics three-dimensional model to obtain the target cultural relics three-dimensional model. In the present application, the corresponding processing parameters are determined according to the sensitive parameters of the cultural relics three-dimensional model area, ensuring that the embedded identification information is visually difficult to detect, improving the concealment and security of the identification information, and ensuring the copyright protection of the cultural relics three-dimensional model in public use.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and more specifically to a method and device for processing ownership identification of a three-dimensional model of cultural relics. Background Art

[0002] The digitization of 3D models of cultural relics has become an important means of cultural relic protection, display and research. The 3D models of cultural relics generated by 3D scanning and reconstruction technology can not only accurately record the shape and details of cultural relics, but also provide rich materials for virtual display, restoration planning and academic research. However, with the widespread application of digitized cultural relic models, copyright protection issues have also become prominent. Traditional copyright protection methods are mostly based on digital watermarking technology, that is, the copyright logo is embedded in the digital model in a hidden way so that the copyright ownership can be effectively confirmed in the event of a copyright dispute.

[0003] Three-dimensional models of cultural relics usually have complex geometric structures and delicate textures. Any obvious changes will affect the visual effect of the model and the accuracy of academic research. The digital watermarking methods in the existing technology often cannot be completely concealed when embedding information, and are easily detected and removed, thus losing their protective effect. Existing watermarking technology often shows weak robustness in the face of malicious attacks. The three-dimensional models of cultural relics will undergo multiple modifications, compression or format conversions due to various application requirements. After the operation of the existing watermarking technology, the embedded identification information is easily lost or destroyed, resulting in the inability to effectively protect the copyright information. Summary of the invention

[0004] In view of this, this application provides the following technical solutions:

[0005] A method for processing ownership identification of a three-dimensional model of a cultural relic, the method comprising:

[0006] Based on the geometric features of the three-dimensional model of the cultural relics to be processed, the three-dimensional model of the cultural relics is divided into at least one area;

[0007] Based on the regional features corresponding to each of the regions, determining a sensitivity parameter corresponding to each of the regions, wherein the sensitivity parameter is used to characterize the importance of the region in the three-dimensional model of the cultural relics;

[0008] Based on the sensitivity parameters corresponding to each of the regions, a processing parameter corresponding to each of the regions is determined, and the original title confirmation identification information corresponding to the region is processed using the processing parameter to obtain a target title confirmation identification;

[0009] The target title confirmation mark is added to the three-dimensional model of the cultural relic to obtain the three-dimensional model of the target cultural relic.

[0010] Optionally, dividing the three-dimensional model of cultural relics to be processed into at least one region based on geometric features of the three-dimensional model of cultural relics to be processed includes:

[0011] Preprocessing the initial data of the three-dimensional model of cultural relics to be processed to obtain the three-dimensional model target data of cultural relics corresponding to the three-dimensional model of cultural relics;

[0012] Generate a geometric feature matrix based on the geometric features of the target data of the three-dimensional model of cultural relics;

[0013] The geometric feature matrix is ​​processed based on a target clustering model to obtain at least one region corresponding to the three-dimensional model of the cultural relics; the target clustering model is a neural network model based on clustering regions by geometric features.

[0014] Optionally, the preprocessing of the initial data of the cultural relic three-dimensional model to be processed to obtain the cultural relic three-dimensional model target data corresponding to the cultural relic three-dimensional model includes:

[0015] Performing denoising processing on the initial data of the three-dimensional model of the cultural relic to be processed to obtain first data;

[0016] Performing smoothing on the first data to obtain second data;

[0017] The second data is format-converted based on the data format of the application scenario to obtain cultural relic three-dimensional model target data corresponding to the cultural relic three-dimensional model.

[0018] Optionally, determining a sensitivity parameter corresponding to each of the regions based on the region feature corresponding to each of the regions includes:

[0019] Based on the regional features corresponding to each of the regions, at least geometric dimension information, visual dimension information and structural dimension information corresponding to each of the regions are obtained;

[0020] The sensitivity parameters corresponding to each of the regions are determined based on the geometric dimension information, visual dimension information, structural dimension information corresponding to each of the regions and the weight parameters corresponding to each of the dimension information.

[0021] Optionally, determining the processing parameter corresponding to each of the regions based on the sensitivity parameter corresponding to each of the regions includes:

[0022] Determine a first level area and a second level area based on a sensitivity parameter corresponding to each of the areas, wherein the sensitivity parameter of the first level area is greater than a target parameter threshold, and the sensitivity parameter of the second level area is not greater than the target parameter threshold;

[0023] Determining a first processing parameter for the first level area based on the number of vertices in the first level area and a processing parameter threshold;

[0024] Based on the first processing parameter, a second processing parameter of the second level area is determined.

[0025] Optionally, the using the processing parameters to process the original title confirmation identification information corresponding to the area to obtain a target title confirmation identification includes:

[0026] Using the first processing parameter, performing a first processing on the first initial data of the original ownership confirmation identification information corresponding to the first level area;

[0027] Using the second processing parameters, performing second processing on the second initial data of the original title confirmation identification information corresponding to the second level area;

[0028] Based on the first processing and the second processing, a target ownership confirmation identifier is obtained.

[0029] Optionally, the adding the target ownership confirmation identifier to the three-dimensional model of the cultural relic to obtain the three-dimensional model of the target cultural relic includes:

[0030] Determining a target area of ​​the three-dimensional model of cultural relics;

[0031] Obtaining an embedding point set of the target area;

[0032] The target right confirmation identifier is allocated so that each identification information element corresponding to the target right confirmation identifier is allocated to each embedding point corresponding to the embedding point set, and the embedding point position after the identification information is embedded is obtained;

[0033] The embedded target title confirmation mark is optimized based on the embedding point position after the embedding mark information to obtain the target cultural relic three-dimensional model.

[0034] Optionally, the method further comprises:

[0035] Based on the target title confirmation identifier and the original title confirmation identifier information, verifying the target cultural relic three-dimensional model;

[0036] If the verification is passed, the target cultural relic three-dimensional model is encrypted to obtain an encrypted target cultural relic three-dimensional model.

[0037] Optionally, encrypting the target cultural relic three-dimensional model to obtain an encrypted target cultural relic three-dimensional model includes:

[0038] Based on the geometric features and topological structure of the target cultural relic three-dimensional model, the target cultural relic three-dimensional model is divided into at least one logical block;

[0039] Encrypting the logic blocks based on the encryption key corresponding to each of the logic blocks to obtain encrypted logic block data;

[0040] Based on each of the encrypted logical block data, an encrypted three-dimensional model of the target cultural relic is determined.

[0041] A device for processing ownership identification of a three-dimensional model of a cultural relic, the device comprising:

[0042] A division unit, used for dividing the three-dimensional model of cultural relics to be processed into at least one region based on the geometric features of the three-dimensional model of cultural relics to be processed;

[0043] A first determining unit is used to determine a sensitivity parameter corresponding to each of the regions based on a region feature corresponding to each of the regions, wherein the sensitivity parameter is used to characterize the importance of the region in the three-dimensional model of cultural relics;

[0044] A second determination unit is used to determine a processing parameter corresponding to each of the regions based on a sensitivity parameter corresponding to each of the regions, and use the processing parameter to process the original title confirmation identification information corresponding to the region to obtain a target title confirmation identification;

[0045] The adding unit is used to add the target title confirmation identifier to the three-dimensional model of the cultural relic to obtain the three-dimensional model of the target cultural relic.

[0046] Through the above technical solution, it can be known that the present application discloses a method and device for processing the title identification of a three-dimensional model of cultural relics, the method comprising: dividing the three-dimensional model of cultural relics into at least one area based on the geometric features of the three-dimensional model of cultural relics to be processed; determining the sensitivity parameter corresponding to each area based on the regional features corresponding to each area, the sensitivity parameter is used to characterize the importance of the data corresponding to the area; determining the processing parameters corresponding to each area based on the sensitivity parameters corresponding to each area, and using the processing parameters to process the title identification information corresponding to the area to obtain the target title identification; adding the target title identification to the three-dimensional model of cultural relics to obtain the target three-dimensional model of cultural relics. In the present application, the corresponding processing parameters are determined according to the sensitive parameters of the three-dimensional model area of ​​the cultural relics, ensuring that the embedded identification information is visually imperceptible, improving the concealment and security of the title identification information, and ensuring the copyright protection of the three-dimensional model of cultural relics in public use. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0048] Figure 1 A schematic diagram of a flow chart of a method for processing the identification of a three-dimensional model of cultural relics provided in an embodiment of the present application;

[0049] Figure 2 A schematic diagram of a process of area division provided in an embodiment of the present application;

[0050] Figure 3 A schematic diagram of the structure of a device for processing the identification of a three-dimensional model of cultural relics provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0052] The terms "first" and "second" in this application are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may include steps or units that are not listed.

[0053] An embodiment of the present application provides a method for processing the title identification of a three-dimensional model of cultural relics, which is applied to the scenario of copyright protection of the three-dimensional model of cultural relics. It mainly embeds the title identification of the three-dimensional model of cultural relics based on the differential privacy algorithm, thereby realizing effective protection of the copyright and ownership of the three-dimensional model of cultural relics.

[0054] See also Figure 1 , is a flow chart of a method for processing the identification of a cultural relic three-dimensional model provided in an embodiment of the present application. The method may include the following steps:

[0055] S101. Divide the three-dimensional model of cultural relics into at least one region based on geometric features of the three-dimensional model of cultural relics to be processed.

[0056] The three-dimensional model of cultural relics to be processed is a three-dimensional data model generated based on the actual three-dimensional image of cultural relics. In order to accurately extract the data features of the three-dimensional model of cultural relics, such as obtaining the corresponding geometric features, the data of the three-dimensional model of cultural relics can be pre-processed in advance to ensure that the data conforms to the current application scenario. Further, in order to accurately process the identification of the three-dimensional model of cultural relics, the three-dimensional model of cultural relics can be divided according to its geometric features to obtain at least one area. For example, cluster analysis can be performed based on geometric features, or a dedicated cluster analysis tool or algorithm can be used to automatically perform regional division to improve the efficiency of regional division.

[0057] In one implementation of an embodiment of the present application, the process of dividing the three-dimensional model of cultural relics into at least one region based on the geometric features of the three-dimensional model of cultural relics to be processed includes: preprocessing the initial data of the three-dimensional model of cultural relics to be processed to obtain the three-dimensional model target data of the cultural relics corresponding to the three-dimensional model of cultural relics; generating a geometric feature matrix based on the geometric features of the three-dimensional model target data of the cultural relics; processing the geometric feature matrix based on a target clustering model to obtain at least one region corresponding to the three-dimensional model of cultural relics; the target clustering model is a neural network model based on clustering regions by geometric features.

[0058] Furthermore, the initial data of the three-dimensional model of cultural relics to be processed is preprocessed to obtain target data of the three-dimensional model of cultural relics corresponding to the three-dimensional model of cultural relics, including: denoising the initial data of the three-dimensional model of cultural relics to be processed to obtain first data; smoothing the first data to obtain second data; and formatting the second data based on the data format of the application scenario to obtain target data of the three-dimensional model of cultural relics corresponding to the three-dimensional model of cultural relics.

[0059] First, the initial data of the three-dimensional model of the cultural relic, that is, the original data set M0, is preprocessed, wherein the original data set may include point cloud data, mesh data or other three-dimensional model representation forms, and then the original data set is denoised, smoothed and format converted. Specifically, a denoising algorithm based on Gaussian filtering can be used to denoise the original data M0 of the three-dimensional model of the cultural relic, and the standard deviation of the Gaussian filter is set to σ to smooth the high-frequency noise in the original data, and the denoised three-dimensional model data set of the cultural relic is obtained as the first data M1. The denoised three-dimensional model data set of the cultural relic is smoothed, that is, the position vector v of each vertex is smoothed. i Adjust the vertex position so that the new vertex position satisfy:

[0060] ;

[0061] Among them, v irepresents the position vector of the vertex, λ1 is the smoothing coefficient, Represents the set of vertices adjacent to the current vertex, v j express The position vectors of the points in the set are obtained, and the three-dimensional model data set of the cultural relics after smoothing is represented by the second data M2.

[0062] The smoothed cultural relic 3D model dataset M2 is converted into a model format. According to the requirements of the target application scenario, the 3D model data format is converted from the initial format F0 to the target format F1, and the converted cultural relic 3D model dataset is represented by the target data M3.

[0063] Then, a geometric feature clustering analysis can be performed based on the target data corresponding to the preprocessed three-dimensional model of the cultural relics, and the three-dimensional model of the cultural relics can be divided into at least one area, and the geometric features of each area are similar.

[0064] S102: Determine a sensitivity parameter corresponding to each region based on a region feature corresponding to each region.

[0065] When performing regional division based on geometric features, regions with similar features are clustered together. For example, vertices or pixels in a three-dimensional model of cultural relics can be grouped according to their features, wherein the features include geometric attributes (such as vertex positions, normal vectors, etc.), colors, curvatures, and textures. Regional division is performed based on the geometric features of vertices or faces in the three-dimensional model of cultural relics to obtain at least one region. Then, according to the regional features corresponding to each region after the division, a sensitivity parameter is determined. Specifically, the sensitivity parameter is used to characterize the importance of the region in the three-dimensional model of cultural relics. Among them, the regional feature can be a feature that characterizes the geometric attribute characteristics of the region. For example, the complexity of the region can be determined by the regional feature, so as to determine the importance of the region in the three-dimensional model of cultural relics according to the complexity, and the sensitivity parameter is determined based on the importance, so that the obtained sensitivity parameter fully conforms to the characteristics of the region, and further makes the subsequent processing more accurate.

[0066] S103. Based on the sensitivity parameters corresponding to each area, determine the processing parameters corresponding to each area, and use the processing parameters to process the original title confirmation identification information corresponding to the area to obtain the target title confirmation identification.

[0067] The sensitivity parameter is used to characterize the importance of the area or the degree of influence of the confirmation mark on the area when the relevant confirmation mark is added to the area. For example, more sophisticated disturbance parameters can be used for high-sensitivity areas through sensitivity parameters, which effectively prevents the identification information from being detected, removed or tampered when it is publicly displayed or disseminated. At the same time, increasing the disturbance intensity in the low-sensitivity area improves the anti-interference ability of the identification information under malicious attacks. That is, the higher the score value corresponding to the sensitivity parameter of a certain area, the higher the demand for copyright protection in the area, and the area should be applied with a stricter copyright protection processing method. Specifically, after obtaining the sensitivity parameter, each area can be divided into a high-sensitivity area and a low-sensitivity area according to the relationship between the sensitivity parameter and the target parameter threshold, so as to determine different processing parameters for the corresponding area, and more accurately prevent attacks on the confirmation mark. In the embodiment of the present application, the processing parameters for each area can be differential privacy disturbance parameters as illustrated above, or other forms of processing parameters, and the present application does not limit this. Based on the processing parameters of each area, the corresponding right confirmation identification information of the area is processed to obtain the processed regional right confirmation identification information corresponding to the area, and then the right confirmation identification information of each area is combined to obtain the target right confirmation identification. Among them, the right confirmation identification generally refers to an identification used to determine the ownership of the rights of the work, product or data, etc. It may be a digital signature, copyright statement, special mark or other form of mark used to prove ownership or right holding. This application does not limit the specific form of the right confirmation identification.

[0068] S104: Add the target title confirmation identifier to the three-dimensional model of the cultural relic to obtain the target three-dimensional model of the cultural relic.

[0069] After obtaining the target title confirmation mark, it can be embedded into the corresponding selected area of ​​the three-dimensional model of the cultural relic. The embedding algorithm can be used to combine the title confirmation mark information with the geometric structure of different areas to obtain the target cultural relic three-dimensional model. That is, the target cultural relic three-dimensional model is a three-dimensional model of the cultural relic with the target title confirmation mark added, which can be used for subsequent display or dissemination to ensure its copyright.

[0070] The present application discloses a method for processing the identification of a cultural relic three-dimensional model, the method comprising: dividing the cultural relic three-dimensional model into at least one region based on the geometric features of the cultural relic three-dimensional model to be processed; determining the sensitivity parameter corresponding to each region based on the regional features corresponding to each region, the sensitivity parameter being used to characterize the importance of the region in the cultural relic three-dimensional model; determining the processing parameters corresponding to each region based on the sensitivity parameters corresponding to each region, and processing the original identification information corresponding to the region using the processing parameters to obtain a target identification; adding the target identification to the cultural relic three-dimensional model to obtain a target cultural relic three-dimensional model. In the present application, the corresponding processing parameters are determined according to the sensitive parameters of the cultural relic three-dimensional model region, ensuring that the embedded identification information is visually difficult to detect, improving the concealment and security of the identification information, and ensuring the copyright protection of the cultural relic three-dimensional model in public use.

[0071] The specific implementation of the embodiments of the present application is described below.

[0072] In the embodiment of the present application, the geometric features of the three-dimensional model of the cultural relics can be obtained according to the target data of the three-dimensional model of the cultural relics, and the geometric information of the vertices, edges and faces in the three-dimensional model of the cultural relics can be extracted to generate a geometric feature matrix G:

[0073] ;

[0074] Where i represents the i-th row in the geometric feature matrix G, j represents the j-th column in the geometric feature matrix; g ij Represents the relationship measure between the i-th vertex and the j-th geometric feature in the three-dimensional model of the cultural relic, where the value range of i is [1, n] and the value range of j is [1, m]; where n is the number of vertices in the three-dimensional model of the cultural relic, and m is the number of geometric features. In the three-dimensional model of the cultural relic, the vertex refers to the basic component unit of the geometric body on the model, which is the structure of the model. The vertex is the intersection of multiple edges on the surface of the model. The vertices are connected by edges to form polygonal faces (usually triangles or quadrilaterals), thus forming the shape of the three-dimensional object.

[0075] Then, the geometric feature matrix can be processed by the target clustering model, which is a neural network model based on clustering regions by geometric features. For example, the target clustering model can be an improved k-means clustering algorithm, thereby dividing the three-dimensional model of cultural relics into multiple regions. k represents the kth region divided by the three-dimensional model of the cultural relic, where k=1,2,…,n, and n is the number of vertices in the three-dimensional model of the cultural relic. k It can be expressed by the following formula:

[0076] ;

[0077] Among them, g i represents the eigenvector of the i-th vertex, c k For region R k The center of the cluster, λ2 is the balance parameter of the cluster, Represents the geometric eigenvalues ​​of the contrast.

[0078] In one implementation of the embodiment of the present application, based on the regional features corresponding to each region, the sensitivity parameter corresponding to each region is determined, including: based on the regional features corresponding to each region, at least the geometric dimension information, visual dimension information and structural dimension information corresponding to each region are obtained; according to the geometric dimension information, visual dimension information, structural dimension information corresponding to each region and the weight parameter corresponding to each dimension information, the sensitivity parameter corresponding to each region is determined. That is, in the embodiment of the present application, a weighted comprehensive evaluation can be performed based on the geometric complexity, visual saliency and structural importance in the region to obtain the sensitivity parameter, and S k For example, S k It can be calculated by the following calculation method:

[0079] ;

[0080] Among them, |R k | is region R k The number of vertices, is the mean of the set features in the three-dimensional model of cultural relics. In the above formula, v i and n i are the position vector and normal vector of the fixed point, T i is the geometric transformation matrix associated with the vertex, det(T i ) represents the matrix T i The determinant of ; α, β and γ are the corresponding weight coefficients, which balance the effects of geometric complexity, visual saliency and structural importance on sensitivity.

[0081] See also Figure 2, is a schematic diagram of the regional division process provided by the embodiment of the present application. Through the above-mentioned analysis of the geometric features of the cultural relics three-dimensional model data set, and then clustering analysis of the geometric feature matrix, a plurality of regions are obtained, and further, they can be divided into first-level regions and second-level regions according to the sensitivity parameters corresponding to each region, wherein the sensitivity parameter of the first-level region is greater than the target parameter threshold, and the sensitivity parameter of the second-level region is not greater than the target parameter threshold; based on the number of vertices in the first-level region and the processing parameter threshold, the first processing parameter of the first-level region is determined; based on the first processing parameter, the second processing parameter of the second-level region is determined. Specifically, the first-level region may be a high-sensitivity region, and the second-level region may be a low-sensitivity region. Further, the processing of the title confirmation identification information corresponding to the region using the processing parameters to obtain the target title confirmation identification includes: using the first processing parameter to perform a first processing on the first initial data of the original title confirmation identification information corresponding to the first-level region; using the second processing parameter to perform a second processing on the second initial data of the original title confirmation identification information corresponding to the second-level region; based on the first processing and the second processing, the target title confirmation identification is obtained.

[0082] In addition to dividing the area levels based on the target parameter threshold, in one implementation of the present application, the sensitivity score S k As input, a random forest model is used to classify features of different geometric regions:

[0083] ;

[0084] Where T is the number of decision trees in the random forest, σ t is the standard deviation of each tree, w t and θ t is the parameter of the random forest model; the sensitivity level of each area is determined by maximizing the probability density function, and the area of ​​the three-dimensional model of cultural relics is divided into high-sensitivity area and low-sensitivity area.

[0085] Specifically, for the geometric feature area classified as a high sensitivity area, R high Represents that the differential privacy perturbation parameter is selected as , the fine differential privacy perturbation parameter represents the privacy budget when performing differential privacy protection in a highly sensitive area. The fine differential privacy perturbation parameter satisfies the following conditions:

[0086] .

[0087] Among them, δ is the processing parameter threshold, which represents the overall privacy leakage risk threshold. is the number of vertices in the high-sensitivity area. Based on the differential privacy perturbation parameter, the Laplace noise perturbation processing is performed on the identification information of the high-sensitivity area, that is, the first processing is performed. The identification information after noise perturbation is It is expressed as:

[0088] .

[0089] Among them, w high It is the original confirmation identification information. Indicates The original identification information can be unique information embedded in the three-dimensional model of the cultural relic (such as copyright, watermark, ownership and other identification information), which is used to identify the ownership or copyright of the three-dimensional model of the cultural relic. In the three-dimensional model of the cultural relic, the original identification information can be embedded in the three-dimensional model in combination with the vertex, geometric structure and texture data of the three-dimensional model.

[0090] For the geometric feature region R classified as a low sensitivity region low , choose a differential privacy perturbation parameter that increases the perturbation strength , the differential privacy perturbation parameter with increased perturbation intensity represents the privacy budget for differential privacy protection in low-sensitivity areas. The differential privacy perturbation parameter with increased perturbation intensity satisfies .

[0091] For the identification information w in the low-sensitivity area low Perform enhanced noise disturbance processing, and the processed identification information It is expressed as:

[0092] .

[0093] For example, the original title confirmation identification information is generated based on the characteristics of the three-dimensional model of the cultural relics and the regional division results: , where W represents the set of identity information for confirmation, including the identity information after noise disturbance and processed identification information , l is the length of the ownership confirmation information.

[0094] For each highly sensitive area R k , according to the sensitivity level S of the sensitive area k And the optimized differential privacy perturbation parameters , for the confirmation identification information w k Perform disturbance processing to obtain the post-disturbance confirmation identification information w k '; For each low sensitivity area R j , according to the low sensitivity level of the non-sensitive area and the corresponding differential privacy perturbation parameter , corresponding to the confirmation information w j Perform relatively strong disturbance processing to obtain the post-disturbance confirmation identification information ; Post-disturbance confirmation identification information for all areas and Integrate to form the final disturbance confirmation identification information set:

[0095] .

[0096] Among them, W' is the final disturbance right confirmation identification information set, n high and n low are the number of high sensitivity areas and low sensitivity areas, respectively, A k and B j are the weight matrices for high sensitivity areas and low sensitivity areas, respectively. and are the identification information after disturbance in high sensitivity area and low sensitivity area respectively. and are the differential privacy parameters for high sensitivity areas and low sensitivity areas, respectively. Indicates region R k The inverse characteristic matrix of Indicates region R j The positive characteristic matrix represents the diffusion ability of the geometric structure in the low-sensitivity area to the disturbance information. After adding the identification information to the relevant area of ​​the cultural relic 3D model, the geometric position of the embedded point will change, mainly because the embedded identification information will cause a slight disturbance to the geometric structure, and the change in position is determined by the identification information. Together with the scaling factor η, the marker information is implicitly embedded in the 3D model, so that the position is changed in order to deeply embed the marker information into the geometric structure without significantly affecting the appearance of the model.

[0097] In one implementation of the embodiment of the present application, adding the target ownership confirmation mark to the three-dimensional model of the cultural relic to obtain the three-dimensional model of the target cultural relic includes:

[0098] Determine the target area of ​​the three-dimensional model of the cultural relics; obtain the embedding point set of the target area; allocate the target title confirmation identifier so that each identification information element corresponding to the target title confirmation identifier is allocated to each embedding point corresponding to the embedding point set, and obtain the embedding point position after the identification information is embedded; optimize the embedded target title confirmation identifier based on the embedding point position after the identification information is embedded, and obtain the target cultural relics three-dimensional model.

[0099] For example, select the selected area R of the three-dimensional model of the cultural relic s , and determine the set of embedded points P in the selected area.

[0100] The target identification information set after disturbance processing Assign each identification information element to the corresponding embedding point p i , so that the embedded identity information is combined with the geometry within the selected region:

[0101] ;

[0102] in, is the embedding point position after embedding the identification information, η is the scaling factor used to control the embedding depth, n i is the embedding point p i The corresponding normal vector is obtained by using the geometric structure combination algorithm to combine the assigned identification information with the geometric features in the selected area, and the objective function is minimized. Optimize embedded information to make it stable and imperceptible:

[0103] .

[0104] Among them, µ is the weight parameter used to balance the distance between embedded points, is the embedding point p i The neighborhood point set of the cultural relic embedded with the identification information is adjusted to obtain the final target cultural relic three-dimensional model M after the identification information is embedded f Specifically, the purpose of minimizing the objective function is to make the value of the objective function as small as possible, and to minimize the impact of embedded identification information on the geometric structure of the three-dimensional model of the cultural relics through optimization, thereby ensuring the visual consistency of the model and the concealment of the embedded information.

[0105] After obtaining the target cultural relic three-dimensional model, in order to further improve the security of the target cultural relic three-dimensional model, the cultural relic three-dimensional model data after the embedded identification information can also be encrypted and stored in segments through an adaptive encryption algorithm to ensure that the identification information embedded in the model data can still be accurately identified after being compressed, format converted or partially reconstructed. The mechanism of segmented storage and segment-by-segment decryption ensures the security of the model data. Even if the model is modified or processed, the validity of the identification information is still maintained, preventing the loss or damage of copyright information during long-term use or in a dynamic environment, and providing strong support for the long-term protection of the cultural relic three-dimensional model. That is, in the embodiment of the present application, it also includes: based on the target confirmation identification and the original confirmation identification information, the target cultural relic three-dimensional model is verified; if the verification is passed, the target cultural relic three-dimensional model is encrypted to obtain the encrypted target cultural relic three-dimensional model.

[0106] Furthermore, the target cultural relic three-dimensional model is encrypted to obtain an encrypted target cultural relic three-dimensional model, including: dividing the target cultural relic three-dimensional model into at least one logical block based on the geometric features and topological structure of the target cultural relic three-dimensional model; encrypting the logical blocks based on an encryption key corresponding to each of the logical blocks to obtain encrypted logical block data; and determining the encrypted target cultural relic three-dimensional model based on each of the encrypted logical block data.

[0107] For example, a 3D model of a cultural relic with ownership information embedded in it M f Encryption processing is performed, and an adaptive encryption algorithm based on the structural characteristics of the model is used to divide the three-dimensional model of the cultural relics into multiple logical blocks B according to the geometric characteristics and topological structure of the three-dimensional model of the cultural relics. i (i=1,2,…,m), where each logical block contains several model vertices and face data:

[0108] ;

[0109] in, represents the jth vertex in the i-th logical block, and the value range of j is [1, k i ],k i is the number of vertices in the i-th logical block; represents the lth patch in the ith logical block, and the value range of l is [1, l i ],l i is the number of patches in the i-th logical block.

[0110] For each logical block B i Generate an independent symmetric encryption key K i and the initial vector IV i , and use the adaptive encryption algorithm to encrypt the logical block to generate the encrypted logical block data C i :

[0111] ;

[0112] Among them, C i is the encrypted data of the ith logical block, Indicates the use of key K i and the AES encryption operation of the initial vector IVi.

[0113] The encrypted logical block C i The data segments of each logical block are stored in a distributed storage system in an independent manner in different physical locations and the key K of each logical block is stored in a distributed storage system in a distributed manner. iThe initial vector IVi is securely stored in the key management system; at the display end, according to the display requirements of the cultural relic 3D model, the corresponding logical block data C is downloaded and decrypted block by block through the access control mechanism of the logical block. i , and reorganize the encrypted data set of the three-dimensional model of the cultural relic in the memory: C f ={C1,C2,…,C m}.

[0114] Use the key K stored in the key management system i and initial vector IVi, for each logical block data C i Decrypt and restore the original 3D model block data B i :

[0115] .

[0116] The decrypted logical block data B i Reassemble and generate a complete 3D model of the cultural relic M f .

[0117] Furthermore, in the embodiment of the present application, the three-dimensional model of the cultural relic after the embedded identification is verified, that is, the three-dimensional model of the target cultural relic needs to be verified to determine whether the three-dimensional model of the target cultural relic meets the actual application requirements. Specifically, the verification process of the target three-dimensional model can be realized by extracting the title confirmation identification information embedded in the three-dimensional model of the target cultural relic and comparing it with the original title confirmation identification information.

[0118] Specifically, the identification information in the three-dimensional model of the target cultural relic after the embedded identification can be extracted first, and the perturbed identification information of the title confirmation can be extracted from the embedded area of ​​the three-dimensional model of the cultural relic using the reverse perturbation algorithm. .

[0119] The extracted identification information W embed Perform anti-disturbance processing and use the same differential privacy parameters as in the embedding process to perform reverse calculations to restore the original identification information W rec . i Indicates the embedded identification information inserted in the stage of embedding the ownership confirmation identification into the original cultural relic 3D model, that is, the identification information that has not been processed in any way; represents the embedded identification information extracted from the target cultural relic model, that is, the identification information after noise processing. Therefore, it can be expressed as follows: , calculate and obtain the embedded identification information w in the original three-dimensional model of the cultural relic i :

[0120] ;

[0121] in, is the differential privacy perturbation parameter used for high-sensitivity areas, is the Laplace noise added in the embedding process. Then, according to w i Determine the restored ownership identification information W rec That is, the information extracted from the embedded identification information after the anti-disturbance processing is used to restore the original identification information as much as possible.

[0122] The restored ownership identification information W rec Compare it with the original ownership identification information W, and use the hash verification algorithm to compare whether the hash values ​​of the two are consistent:

[0123] Hash(W rec ) = Hash(W);

[0124] Based on the comparison results, verify whether the embedded ownership identification information has been successfully extracted and restored. If the comparison results are consistent, the identification information embedding verification is passed; if not, record the abnormal situation in the model and output an error report.

[0125] In the embodiment of the present application, the embedding of identification information in different areas is differentiated according to the geometric features and regional sensitivity of the three-dimensional model of cultural relics. By adopting a sophisticated embedding algorithm in the high-sensitivity area, the embedded identification information is closely integrated with the geometric structure of the area, thereby enhancing the stability of the identification information in key areas. In the low-sensitivity area, strong disturbance and encryption measures are adopted to further improve the concealment and anti-attack ability of the identification information. The adaptive embedding algorithm design not only optimizes the embedding effect of the title confirmation mark, but also ensures the copyright protection strength of the three-dimensional model of cultural relics in different application scenarios.

[0126] See also Figure 3 In an embodiment of the present application, a device for processing the identification of a cultural relic three-dimensional model is also provided, and the device comprises:

[0127] A division unit 201 is used to divide the three-dimensional model of cultural relics to be processed into at least one area based on the geometric features of the three-dimensional model of cultural relics to be processed;

[0128] A first determining unit 202, configured to determine a sensitivity parameter corresponding to each of the regions based on a region feature corresponding to each of the regions, wherein the sensitivity parameter is used to characterize the importance of the region in the three-dimensional model of cultural relics;

[0129] The second determination unit 203 is used to determine the processing parameters corresponding to each of the regions based on the sensitivity parameters corresponding to each of the regions, and use the processing parameters to process the original title confirmation identification information corresponding to the region to obtain a target title confirmation identification;

[0130] The adding unit 204 is used to add the target ownership confirmation identifier to the cultural relic three-dimensional model to obtain the target cultural relic three-dimensional model.

[0131] Optionally, the dividing unit includes:

[0132] A preprocessing subunit is used to preprocess the initial data of the cultural relic three-dimensional model to be processed to obtain the cultural relic three-dimensional model target data corresponding to the cultural relic three-dimensional model;

[0133] A first generating subunit is used to generate a geometric feature matrix based on the geometric features of the target data of the three-dimensional model of cultural relics;

[0134] The first processing subunit is used to process the geometric feature matrix based on a target clustering model to obtain at least one region corresponding to the three-dimensional model of the cultural relics; the target clustering model is a neural network model based on clustering regions by geometric features.

[0135] Optionally, the preprocessing subunit is configured to:

[0136] Performing denoising processing on the initial data of the three-dimensional model of the cultural relic to be processed to obtain first data;

[0137] Performing smoothing on the first data to obtain second data;

[0138] The second data is format-converted based on the data format of the application scenario to obtain cultural relic three-dimensional model target data corresponding to the cultural relic three-dimensional model.

[0139] Optionally, the first determining unit includes:

[0140] A first acquisition subunit is used to obtain at least geometric dimension information, visual dimension information and structural dimension information corresponding to each of the regions based on the region features corresponding to each of the regions;

[0141] The first determination subunit is used to determine the sensitivity parameter corresponding to each of the regions according to the geometric dimension information, visual dimension information, structural dimension information corresponding to each of the regions and the weight parameter corresponding to each dimension information.

[0142] Optionally, the second determining unit includes:

[0143] A second determining subunit is used to determine a first-level area and a second-level area based on a sensitivity parameter corresponding to each of the areas, wherein the sensitivity parameter of the first-level area is greater than a target parameter threshold, and the sensitivity parameter of the second-level area is not greater than the target parameter threshold;

[0144] A third determining subunit, configured to determine a first processing parameter of the first level area based on the number of vertices in the first level area and a processing parameter threshold;

[0145] The fourth determining subunit is configured to determine a second processing parameter of the second level area based on the first processing parameter.

[0146] Optionally, the second determining unit further includes:

[0147] A second processing subunit is used to perform a first processing on the first initial data of the original title confirmation identification information corresponding to the first level area by using the first processing parameter;

[0148] A third processing subunit is used to perform a second processing on the second initial data of the original title confirmation identification information corresponding to the second level area by using the second processing parameter;

[0149] The second acquisition subunit is used to obtain a target ownership confirmation identifier based on the first processing and the second processing.

[0150] Optionally, the adding unit is configured as:

[0151] Determining a target area of ​​the three-dimensional model of cultural relics;

[0152] Obtaining an embedding point set of the target area;

[0153] The target right confirmation identifier is allocated so that each identification information element corresponding to the target right confirmation identifier is allocated to each embedding point corresponding to the embedding point set, and the embedding point position after the identification information is embedded is obtained;

[0154] The embedded target title confirmation mark is optimized based on the embedding point position after the embedding mark information to obtain the target cultural relic three-dimensional model.

[0155] Optionally, the device further comprises:

[0156] A verification unit, used to verify the target cultural relic three-dimensional model based on the target title confirmation identifier and the original title confirmation identifier information;

[0157] The encryption unit is used to encrypt the target cultural relic three-dimensional model if the verification is passed to obtain an encrypted target cultural relic three-dimensional model.

[0158] Optionally, the encryption unit is configured to:

[0159] Based on the geometric features and topological structure of the target cultural relic three-dimensional model, the target cultural relic three-dimensional model is divided into at least one logical block;

[0160] Encrypting the logic blocks based on the encryption key corresponding to each of the logic blocks to obtain encrypted logic block data;

[0161] Based on each of the encrypted logical block data, an encrypted three-dimensional model of the target cultural relic is determined.

[0162] It should be noted that the specific implementation of each unit and sub-unit in this embodiment can refer to the corresponding content in the previous text and will not be described in detail here.

[0163] In another embodiment of the present application, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for processing the ownership identification of the three-dimensional model of cultural relics as described above is implemented.

[0164] In another embodiment of the present application, an electronic device is further provided, and the electronic device may include:

[0165] A memory, used to store applications and data generated by the operation of the applications;

[0166] The processor is used to execute the application program to implement the above-mentioned method for processing the identification of the three-dimensional model of cultural relics.

[0167] It should be noted that the specific implementation of the processor in this embodiment can refer to the corresponding content in the previous text and will not be described in detail here.

[0168] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0169] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0170] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0171] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for processing the identification of a three-dimensional model of a cultural relic, characterized in that: The method comprises: Based on the geometric features of the three-dimensional model of the cultural relics to be processed, the three-dimensional model of the cultural relics is divided into at least one area; Based on the regional features corresponding to each of the regions, determining a sensitivity parameter corresponding to each of the regions, wherein the sensitivity parameter is used to characterize the importance of the region in the three-dimensional model of the cultural relics; Based on the sensitivity parameters corresponding to each of the areas, the processing parameters corresponding to each of the areas are determined, and the original ownership identification information corresponding to the areas is processed using the processing parameters to obtain the target ownership identification; wherein, based on the sensitivity parameters corresponding to each of the areas, the processing parameters corresponding to each of the areas are determined, including: based on the sensitivity parameters corresponding to each of the areas, a first-level area and a second-level area are determined, the sensitivity parameter of the first-level area is greater than a target parameter threshold, and the sensitivity parameter of the second-level area is not greater than the target parameter threshold; based on the number of vertices in the first-level area and the processing parameter threshold, a first processing parameter of the first-level area is determined; based on the first processing parameter, a second processing parameter of the second-level area is determined; wherein the processing parameter threshold represents an overall privacy leakage risk threshold; the processing parameters include a differential privacy perturbation parameter; The target title confirmation mark is added to the three-dimensional model of the cultural relic to obtain the three-dimensional model of the target cultural relic.

2. The method according to claim 1, characterized in that The step of dividing the three-dimensional model of cultural relics to be processed into at least one region based on the geometric features of the three-dimensional model of cultural relics to be processed comprises: Preprocessing the initial data of the three-dimensional model of cultural relics to be processed to obtain the three-dimensional model target data of cultural relics corresponding to the three-dimensional model of cultural relics; Generate a geometric feature matrix based on the geometric features of the target data of the three-dimensional model of cultural relics; The geometric feature matrix is ​​processed based on a target clustering model to obtain at least one region corresponding to the three-dimensional model of the cultural relics; the target clustering model is a neural network model based on clustering regions by geometric features.

3. The method according to claim 2, characterized in that The preprocessing of the initial data of the three-dimensional model of cultural relics to be processed to obtain the three-dimensional model target data of cultural relics corresponding to the three-dimensional model of cultural relics includes: Performing denoising processing on the initial data of the three-dimensional model of the cultural relic to be processed to obtain first data; Performing smoothing on the first data to obtain second data; The second data is format-converted based on the data format of the application scenario to obtain cultural relic three-dimensional model target data corresponding to the cultural relic three-dimensional model.

4. The method according to claim 1, characterized in that: The determining of the sensitivity parameter corresponding to each of the regions based on the region feature corresponding to each of the regions includes: Based on the regional features corresponding to each of the regions, at least geometric dimension information, visual dimension information and structural dimension information corresponding to each of the regions are obtained; The sensitivity parameters corresponding to each of the regions are determined based on the geometric dimension information, visual dimension information, structural dimension information corresponding to each of the regions and the weight parameters corresponding to each of the dimension information.

5. The method according to claim 1, characterized in that The using the processing parameters to process the original title confirmation identification information corresponding to the area to obtain the target title confirmation identification includes: Using the first processing parameter, performing a first processing on the first initial data of the original ownership confirmation identification information corresponding to the first level area; Using the second processing parameters, performing second processing on the second initial data of the original title confirmation identification information corresponding to the second level area; Based on the first processing and the second processing, a target ownership confirmation identifier is obtained.

6. The method according to claim 1, characterized in that The step of adding the target title confirmation mark to the three-dimensional model of the cultural relic to obtain the three-dimensional model of the target cultural relic includes: Determining a target area of ​​the three-dimensional model of cultural relics; Obtaining an embedding point set of the target area; The target right confirmation identifier is allocated so that each identification information element corresponding to the target right confirmation identifier is allocated to each embedding point corresponding to the embedding point set, and the embedding point position after the identification information is embedded is obtained; The embedded target title confirmation mark is optimized based on the embedding point position after the embedding mark information to obtain the target cultural relic three-dimensional model.

7. The method according to claim 1, characterized in that The method further comprises: Based on the target title confirmation identifier and the original title confirmation identifier information, verifying the target cultural relic three-dimensional model; If the verification is passed, the target cultural relic three-dimensional model is encrypted to obtain an encrypted target cultural relic three-dimensional model.

8. The method according to claim 7, characterized in that The step of encrypting the target cultural relic three-dimensional model to obtain the encrypted target cultural relic three-dimensional model comprises: Based on the geometric features and topological structure of the target cultural relic three-dimensional model, the target cultural relic three-dimensional model is divided into at least one logical block; Encrypting the logic blocks based on the encryption key corresponding to each of the logic blocks to obtain encrypted logic block data; Based on each of the encrypted logical block data, an encrypted three-dimensional model of the target cultural relic is determined.

9. A device for processing the identification of a cultural relic three-dimensional model, characterized in that: The device comprises: A division unit, used for dividing the three-dimensional model of cultural relics to be processed into at least one region based on the geometric features of the three-dimensional model of cultural relics to be processed; A first determining unit is used to determine a sensitivity parameter corresponding to each of the regions based on a region feature corresponding to each of the regions, wherein the sensitivity parameter is used to characterize the importance of the region in the three-dimensional model of cultural relics; The second determination unit is used to determine the processing parameters corresponding to each of the areas based on the sensitivity parameters corresponding to each of the areas, and use the processing parameters to process the original title confirmation identification information corresponding to the areas to obtain the target title confirmation identification; wherein, the processing parameters corresponding to each of the areas based on the sensitivity parameters corresponding to each of the areas are determined, including: determining the first-level area and the second-level area based on the sensitivity parameters corresponding to each of the areas, the sensitivity parameter of the first-level area is greater than the target parameter threshold, and the sensitivity parameter of the second-level area is not greater than the target parameter threshold; determining the first processing parameter of the first-level area based on the number of vertices in the first-level area and the processing parameter threshold; determining the second processing parameter of the second-level area based on the first processing parameter; wherein the processing parameter threshold represents the overall privacy leakage risk threshold; the processing parameters include differential privacy perturbation parameters; The adding unit is used to add the target title confirmation identifier to the three-dimensional model of the cultural relic to obtain the three-dimensional model of the target cultural relic.

Images