Management system for safe storage and efficient display of digital cultural relic data

Abstract

The invention relates to the field of cultural relic digital protection, and discloses a management system for safe storage and efficient display of digital cultural relic data, and the system comprises a data preprocessing module which is used for collecting 3D point cloud data of a cultural relic and carrying out the denoising processing, and generating sparse matrix data; the model generation module is used for generating a preliminary 3D model of the cultural relic based on the sparse matrix; the intelligent repairing module is used for automatically repairing the preliminary 3D model to generate a complete 3D model; the storage encryption module is used for performing homomorphic encryption and compression storage on the complete 3D model; the index retrieval module is used for constructing an index based on the encrypted model data and realizing efficient retrieval; and the display management module is used for realizing multi-resolution display of the model through wavelet decomposition and controlling the access permission according to the user permission matrix. According to the method, high-precision modeling, safe storage and efficient display of cultural relic data can be realized, and the requirements of digital protection and display of cultural relics are met.

Description

Technical Field

[0001] The present invention relates to the technical field of digital protection of cultural relics, and in particular to a management system for secure storage and efficient display of digital cultural relic data. Background Art

[0002] In the field of digital cultural relic protection, with the continuous development of 3D scanning, digital storage and display technology, more and more cultural relics can be recorded and displayed with high precision through digital means. However, in the storage and management of cultural relic 3D data, there are a series of limiting factors in the existing technology, which affects the security, integrity and display effect of digital cultural relic data.

[0003] First, although the traditional 3D data acquisition method of cultural relics can record the surface details of the object, it often has noise interference, resulting in inaccurate model data and increasing the complexity of subsequent data processing. The existing technology has limited methods for processing noise data, which may cause the surface features of cultural relics to be blurred or lost during modeling. In addition, for the initially generated 3D model, the existing model repair method is difficult to effectively restore the defective area, which affects the overall accuracy and authenticity of the model and cannot fully meet the needs of cultural relic protection.

[0004] In terms of data storage and management, existing technologies have low storage efficiency when processing large-scale 3D cultural relic data, and lack effective data security measures. Most existing methods fail to effectively combine encryption and compression technology in the process of model data storage, and data is vulnerable to unauthorized access or tampering during transmission or storage, making data security difficult to guarantee. At the same time, the high-resolution data of cultural relic models takes up a lot of storage space, resulting in a decrease in system response speed, affecting the rapid access and retrieval of data.

[0005] In terms of display and permission management, the existing system is difficult to provide adaptive display effects according to the display capabilities of different devices, and the model loading speed is slow, which affects the user's interactive experience. In addition, user access permission control is not flexible enough, and there is a lack of security control measures for different users and access scenarios, resulting in insufficient security in data display and sharing. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a management system for the secure storage and efficient display of digital cultural relic data, which realizes high-precision modeling, secure storage, efficient retrieval and adaptive display of cultural relic 3D data, thereby meeting the needs of digital protection and display of cultural relics.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a management system for secure storage and efficient display of digital cultural relics data, comprising:

[0008] Data preprocessing module, used to collect and denoise the 3D point cloud data of cultural relics to form sparse matrix data;

[0009] A model generation module, used for generating a preliminary 3D model of the cultural relic based on the sparse matrix data;

[0010] An intelligent repair module, used for automatically repairing the preliminary 3D model to generate a complete 3D model;

[0011] A storage encryption module, used for homomorphically encrypting and compressing the complete 3D model for storage;

[0012] An index retrieval module, used to generate an index based on the data features of the complete 3D model to achieve efficient retrieval;

[0013] A display management module is used to adaptively display the complete 3D model based on the index retrieval result.

[0014] Preferably, the data preprocessing module includes:

[0015] The point cloud data acquisition unit is used to acquire the 3D point cloud data of the cultural relics and represent the acquired point cloud data as a point set P = {p i |p i =(x i ,y i ,z i )}, where p i Represents a three-dimensional coordinate point in space;

[0016] Sparse matrix construction unit, used to convert the point cloud data into a sparse matrix M∈ To represent the collected valid data points and missing points, the sparse matrix satisfies:

[0017]

[0018] Among them, p ij Indicates valid point cloud data value, and zero value indicates a vacant position where no data is collected;

[0019] The denoising processing unit is used to perform denoising on the sparse matrix data through wavelet transform to reduce the noise in the collected data and retain important features, thereby obtaining denoised sparse matrix data M′.

[0020] Preferably, the model generation module includes:

[0021] A point cloud feature extraction unit, used to extract spatial features of point cloud data from the denoised sparse matrix data output by the data preprocessing module, and determine the preliminary form of the model according to the extracted features;

[0022] A NURBS surface generating unit is used to generate a preliminary 3D surface model of the cultural relic by a non-uniform rational B-spline method based on the output of the feature extracting unit, wherein the NURBS surface satisfies the following parameterized representation:

[0023]

[0024] Among them, N i,p (u) and N j,q (v) is the B-spline basis function, P ij is the control point, w ij is the weight parameter, and u and v are parameters used to construct the surface model of the cultural relic;

[0025] The surface optimization unit is used to optimize the generated preliminary surface model by adjusting the position and weight parameters of the surface control points to make the surface model more consistent with the actual shape of the cultural relic surface.

[0026] Preferably, the intelligent repair module includes:

[0027] The sparse matrix completion unit is used to receive the sparse matrix data of the preliminary 3D model generated by the model generation module, and complete the missing area of ​​the sparse matrix by the nuclear norm minimization method. The optimization goal of the nuclear norm minimization is:

[0028] min||X|| * subject to X ij =M ij for observed entries

[0029] Among them, ||X|| * =∑ i σ i (X) is the nuclear norm of matrix X, σ i (X) is the singular value of the matrix, which is used to complete the missing model data by minimizing the rank of the matrix;

[0030] The alternating least squares solving unit is used to iteratively solve the nuclear norm minimization optimization problem of the sparse matrix completion unit by using the alternating least squares method. By gradually updating the matrix values ​​of rows and columns until the completion matrix X converges to a stable state, the completed high-precision sparse matrix X is obtained. * ;

[0031] Geometry repair unit for the sparse matrix X after completion * Restore the geometric structure of the preliminary 3D model so that the structure of the completed area is consistent with that of the original area of ​​the model;

[0032] The smoothing unit is used to smooth the 3D model output by the geometric structure repair unit to ensure the continuity and naturalness of the model surface. The smoothing is achieved by minimizing the following objective function:

[0033]

[0034] Among them, S(u,v) represents the surface of the repaired model, Represents the second-order gradient operator, which obtains a smooth 3D model by minimizing the quadratic rate of change of the surface model.

[0035] Preferably, the storage encryption module includes:

[0036] Homomorphic encryption unit, used to perform homomorphic encryption processing on the complete 3D model data generated by the intelligent repair module, and use the Paillier homomorphic encryption algorithm to encrypt the values ​​in the model data;

[0037] The singular value decomposition compression unit is used to compress the encrypted 3D model data after homomorphic encryption is completed, and decompose the encrypted data matrix using the singular value decomposition algorithm;

[0038] The encryption storage unit is used to store the 3D model data compressed by singular value decomposition in a storage medium.

[0039] Preferably, the index retrieval module includes:

[0040] A vector feature extraction unit is used to extract a feature vector F of the model from the encrypted 3D model data output by the storage encryption module. i}, the feature vector is used to represent the geometric shape, texture and other characteristics of the cultural relic model, providing a basis for subsequent similarity retrieval;

[0041] A local sensitive hash construction unit is used to construct a local sensitive hash index based on the feature vector set F extracted by the feature extraction unit, and a hash function family H = {h i} Map the feature vector into a hash table, where the hash function satisfies the following form:

[0042]

[0043] Among them, a is a random vector, b is a bias, and w is the bucket width. Similar feature vectors are mapped to the same hash bucket through hash mapping, thereby realizing similarity search;

[0044] An index storage unit, used to store the constructed local sensitive hash index structure into an index database;

[0045] A retrieval processing unit is used to receive the query conditions input by the user, call the corresponding local sensitive hash function from the index database to perform model matching based on the query conditions, implement fast similarity matching according to the feature vectors in the same bucket in the hash table, and output the 3D model data of the cultural relics that meet the conditions;

[0046] The result sorting unit is used to sort the multiple matching results output by the retrieval processing unit by similarity, and arrange the retrieval results from high to low according to the similarity scores between the feature vectors.

[0047] Preferably, the display management module includes:

[0048] The hierarchical loading unit is used to decompose the 3D model data into multiple resolution levels according to the data structure of the cultural relic 3D model, and realize hierarchical loading through wavelet decomposition. The decomposed model data meets the following requirements:

[0049]

[0050] Among them, X k is the low-resolution representation of the model at layer k, ψ k The corresponding wavelet basis function supports progressive loading through multi-resolution layered processing, ensuring that the 3D model of cultural relics can be displayed smoothly under different devices and network conditions;

[0051] Adaptive rendering unit, which automatically adjusts the loading resolution and detail level of 3D models according to the display resolution and performance parameters of the user's device to ensure high-resolution model display on high-performance devices;

[0052] The dynamic projection unit is used to adjust the projection angle of the 3D model in real time according to the change of the user's viewing angle to achieve multi-angle interactive display. The projection transformation is performed through projective geometry. The projection matrix T satisfies:

[0053]

[0054] Among them, f x and f y is the focal length, c x and c y It is the projection center to ensure the visual consistency of the model under different viewing angles;

[0055] The texture mapping unit is used to perform texture mapping and lighting processing on the surface of the 3D model. The real material and texture effects are superimposed on the model surface through texture mapping, and rendered in combination with the lighting model.

[0056] Preferably, the system further comprises an authority control module for controlling the access rights of users to ensure safe access to cultural relics data.

[0057] Preferably, the authority control module includes:

[0058] The permission matrix generation unit is used to represent the user set U and resource set R in the system as a permission matrix The permission matrix is ​​used to record the user's access rights to various resources and is defined as follows:

[0059]

[0060] Among them, A ij =1 means user i has access rights to resource j, A ij =0 means no permission;

[0061] The access request verification unit is used to receive the user's resource access request and verify whether the user has the access rights to the requested resource based on the permission matrix A. Specifically, it searches for the corresponding permission value A in the permission matrix. ij , if A ij =1, access is allowed. If A ij =0, access is denied and a prompt message indicating that the user does not have permission is returned;

[0062] A dynamic permission adjustment unit is used to dynamically adjust the permission values ​​in the permission matrix A according to the authorization request of the administrator or the permission requirement of the user, so as to update the user's access rights in real time;

[0063] A behavior recording unit is used to record the user's operation behavior and access log in the system, and store these behavior records in the behavior log database;

[0064] The anomaly detection unit is used to perform real-time analysis on the user access logs in the behavior recording unit, perform cluster analysis on user access behaviors through clustering algorithms, identify and mark abnormal access behaviors, and send alarm information to the administrator when abnormal access behaviors are detected to ensure the security of cultural relics data.

[0065] The present invention also provides a management method for secure storage and efficient display of digital cultural relics data, comprising the following steps:

[0066] Obtain the 3D point cloud data of cultural relics, remove noise through wavelet transform, and generate a sparse matrix;

[0067] Generate a preliminary 3D model using NURBS based on sparse matrix;

[0068] The preliminary 3D model is intelligently repaired by minimizing the nuclear norm to generate a complete 3D model;

[0069] The complete 3D model is encrypted with Paillier homomorphic encryption and compressed with singular value decomposition;

[0070] The LSH algorithm is used to generate indexes for the encrypted 3D model features;

[0071] The 3D model is displayed in multiple resolutions through wavelet decomposition, and access rights are controlled based on the user permission matrix.

[0072] The present invention provides a management system for secure storage and efficient display of digital cultural relics data. It has the following beneficial effects:

[0073] 1. The present invention effectively filters out noise data in the acquisition process by denoising and sparse matrix modeling of 3D point cloud data, while retaining the key features of the surface of the cultural relics, providing an accurate data basis for subsequent 3D model generation and restoration, thereby improving the overall quality and detail accuracy of the model.

[0074] 2. The present invention uses NURBS surfaces to generate a preliminary 3D model, and then intelligently repairs the defective area by minimizing the nuclear norm, ensuring the integrity and smoothness of the cultural relic model. This process can truly reproduce the complex surface morphology of cultural relics, laying the foundation for the high-precision presentation of digital cultural relics.

[0075] 3. The present invention encrypts the complete 3D model data through the Paillier homomorphic encryption algorithm, ensuring the security of the cultural relic data during storage and transmission, and preventing unauthorized access and tampering. At the same time, the data storage space requirement is effectively reduced through singular value decomposition (SVD) compression, optimizing the storage efficiency of the system.

[0076] 4. The present invention uses the local sensitive hashing (LSH) algorithm to construct a feature index for the encrypted 3D model data, so that the system can quickly retrieve and match similar cultural relic models. This index structure improves the response speed and retrieval efficiency of the system and meets the rapid query requirements in large-scale management of cultural relic data.

[0077] 5. The present invention uses wavelet decomposition to perform multi-resolution layered display of 3D models, realizes adaptive loading and display on multiple terminal devices, and makes the display effect automatically adjusted according to the device performance. At the same time, the access rights are controlled through the user authority matrix, which ensures the security of cultural relics data display and avoids unauthorized access and operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a schematic diagram of the system architecture of the present invention;

[0079] Figure 2 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0080] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0081] Please refer to the attached Figure 1 The embodiment of the present invention provides a management system for secure storage and efficient display of digital cultural relics data. By collecting, denoising, repairing, encrypting, storing, indexing and adaptively displaying 3D point cloud data of cultural relics, the integrity, authenticity and efficient management of digital cultural relics data are achieved. The functions and principles of each module of the system are described in detail below.

[0082] Data preprocessing module

[0083] The data preprocessing module is used to collect, denoise and perform sparse matrix modeling on the 3D point cloud data of cultural relics, providing basic data support for subsequent model generation, restoration and display. The core of this module is to structure the original collected data into a sparse matrix form and remove the collection noise in the data through denoising technology to ensure the accuracy and clarity of the input data.

[0084] In this embodiment, the point cloud data acquisition unit is used to collect point cloud data from the cultural relic 3D scanning device to generate a set of three-dimensional coordinates of the surface of the cultural relic. Specifically, the three-dimensional structure information of the cultural relic is obtained by using a high-precision scanning device, and the collected point cloud data P = {p i |p i =(x i ,y i ,z i )}, where p i Represents a collected three-dimensional coordinate point, (x i ,y i ,z i ) is the position of the point in space. By acquiring point cloud data, a spatial description of the surface of the cultural relic is formed, and the point cloud data provides basic information for subsequent denoising and structural processing.

[0085] In this embodiment, the sparse matrix construction unit converts the point cloud data into a sparse matrix form for subsequent processing. The collected point cloud data set P is constructed into a sparse matrix The matrix elements are used to represent valid points and missing points in the point cloud data. Specifically, for the location where data exists, the corresponding element in the sparse matrix M assigns the value p of the point cloud data ij , and the vacant positions where no data is collected are represented by 0. The definition of the sparse matrix M is as follows:

[0086]

[0087] By constructing a sparse matrix, the structural characteristics of the collected data can be effectively represented, which facilitates the subsequent completion and repair of missing data.

[0088] In this embodiment, the denoising processing unit uses wavelet transform technology to denoise the point cloud data to reduce noise interference in data collection. Specifically, the valid data in the sparse matrix M is regarded as a signal, and multi-scale decomposition is performed on it through wavelet transform to convert the signal into frequency domain space. Assume that the collected signal is f(x), and its wavelet transform is expressed as:

[0089]

[0090] Among them, a represents the scale parameter, b represents the translation parameter, and ψ is the wavelet basis function. The wavelet basis function performs multi-resolution decomposition on the collected signal f(x) at different scales to separate the low-frequency component and the high-frequency noise component in the signal. By setting the threshold, the decomposed wavelet coefficients are threshold filtered to remove the high-frequency noise and retain only the low-frequency component, so that the denoised data matrix M′ can more accurately reflect the effective characteristics of the surface of the cultural relic.

[0091] In this embodiment, the denoised data of the sparse matrix M′ generated by the data preprocessing module will be used for model generation and repair processing in subsequent modules.

[0092] Model generation module

[0093] The model generation module generates a preliminary 3D model of the cultural relic through geometric feature extraction and surface modeling methods based on the denoised sparse matrix data output by the data preprocessing module. The core of this module is to convert sparse point cloud data into a continuous 3D surface model, generate complex surface structures using non-uniform rational B-splines (NURBS), and perform curvature optimization and morphological adjustment on the generated model, providing high-precision preliminary model input for the subsequent intelligent restoration module.

[0094] In this embodiment, the feature extraction unit is used to extract the geometric features of the point cloud data from the sparse matrix M' provided by the data preprocessing module to generate the preliminary form of the model. By scanning the distribution pattern of the data, the geometric structural characteristics of the surface area of ​​the model are determined, including the contour, boundary and local curvature information of the point cloud, providing a preliminary structural basis for the surface generation process.

[0095] In this embodiment, the NURBS surface generation unit generates a 3D model of the cultural relic using a non-uniform rational B-spline (NURBS) surface based on the geometric features output by the feature extraction unit. NURBS surface is a flexible surface representation method that can accurately describe complex surface morphology. The NURBS surface S (u, v) is defined by the following parameterized formula:

[0096]

[0097] in:

[0098] N i,p (u) and N j,q (v) are the B-spline basis functions in the u and v parameter directions, defining the geometric shape of the model in that direction;

[0099] w ij represents the control point P ij The weight of the control point P ij The coordinate points of the surface of the cultural relic are used to define the shape of the model surface.

[0100] The NURBS surface generation unit sets the above parameters according to the complexity of the surface of the cultural relic and the distribution of the control points to ensure that the generated surface accurately reflects the actual structure of the cultural relic.

[0101] In this embodiment, the preliminary 3D model generated by the model generation module will serve as the input of the intelligent repair module, providing an accurate basic form for subsequent model repair and completion.

[0102] Smart repair module

[0103] The intelligent repair module is used to repair and complete the preliminary 3D model output by the model generation module to generate complete 3D model data. The core of this module is to complete the defective area in the model with sparse matrix, and to optimize the structure and smooth the curvature of the completed surface to ensure the continuity and consistency between the repaired area and the original data.

[0104] In this embodiment, the sparse matrix completion unit is used to receive the sparse matrix data of the preliminary 3D model generated by the model generation module, and complete the missing parts of the sparse matrix. In order to achieve reasonable filling of data, the sparse matrix completion process adopts the nuclear norm minimization method to infer the missing data by minimizing the rank of the matrix to restore the surface structure of the cultural relic. Specifically, the sparse matrix completion problem is expressed as the following optimization goal:

[0105] min||X|| * subject to X ij =M ij for observed entries

[0106] Among them, X is the matrix to be completed, M ij Indicates the existing data location, ||X|| * =∑ i σ i (X) is the nuclear norm of the matrix, σi (X) is the singular value of the matrix. By minimizing the nuclear norm, a low-rank matrix completion solution X can be obtained. * , thereby completing the missing data in the preliminary 3D model.

[0107] In this embodiment, the alternating least squares solution unit uses the alternating least squares method (ALS) to iteratively solve the above-mentioned nuclear norm minimization problem. The ALS method alternately optimizes and updates the rows and columns of the matrix to gradually approach the optimal solution. Specifically, in each iteration, one dimension is fixed and the element value of the other dimension is updated until the solution result X of the matrix X is * Converge to a stable state, thus achieving efficient completion of sparse matrices.

[0108] In this embodiment, after completing the sparse matrix completion, the geometric structure repair unit performs geometric structure repair on the completed data. Specifically, by analyzing the surface morphology of the model, the completed area is structurally consistent with the original area of ​​the preliminary 3D model. The geometric structure repair unit uses the control points and weight information in the model generation module to adjust the geometric morphology of the completed area so that the repaired model can accurately reflect the original morphology of the cultural relic.

[0109] In this embodiment, the smoothing processing unit is used to smooth the model surface after the geometric structure is repaired to eliminate local mutations and ensure that the model surface is smooth and natural. To achieve smooth optimization of the surface, the smoothing processing unit optimizes the curvature by minimizing the quadratic change rate so that the local changes of the model surface remain continuous. The objective function of smooth optimization is as follows:

[0110]

[0111] Among them, S(u,v) represents the surface of the repaired model, It represents the second-order gradient operator. By adjusting the positions and weights of the control points, the smoothing unit can reduce the local change rate of the surface, so that the surface of the repaired 3D model remains globally consistent and smooth.

[0112] In this embodiment, the complete 3D model data generated by the intelligent restoration module will be used as the input of the storage encryption module, providing an accurate 3D model basis for the safe storage and display of cultural relics data.

[0113] Storage encryption module

[0114] The storage encryption module is used to encrypt and compress the complete 3D model data generated by the intelligent restoration module. Through homomorphic encryption and singular value decomposition (SVD) compression technology, this module can ensure the security of cultural relic model data during storage and transmission, and effectively reduce storage space requirements.

[0115] In this embodiment, the homomorphic encryption unit is used to homomorphically encrypt the complete 3D model data output by the intelligent repair module to ensure that limited mathematical operations can be performed on the data in an encrypted state. Specifically, the Paillier homomorphic encryption algorithm is used to encrypt the values ​​in the model data matrix. The characteristics of the Paillier encryption algorithm enable the encrypted data to support homomorphic addition operations without decryption, which are defined as follows:

[0116] E(a+b)=E(a)·E(b)

[0117] Among them, E represents the encryption operation, and a and b are the values ​​of the 3D model data. Through this encryption method, the confidentiality of the model data during transmission and storage can be ensured to prevent unauthorized access and tampering.

[0118] In this embodiment, the singular value decomposition compression unit compresses the encrypted 3D model data after homomorphic encryption to reduce the data storage space requirement. Specifically, the encrypted data matrix E(X) is decomposed using a singular value decomposition (SVD) algorithm, and SVD decomposes the matrix E(X) into the following form:

[0119] E(X)=UΣV T

[0120] Among them, U and V are orthogonal matrices, and Σ is a diagonal matrix that contains the singular values ​​of the matrix E(X). By retaining the singular values ​​of the principal components that have a greater impact on the data and discarding the smaller singular values, the amount of data can be greatly reduced while maintaining the main features of the data. The compressed data finally generated is used to reduce storage occupancy and improve storage efficiency.

[0121] In this embodiment, the encrypted storage unit is used to store the 3D model data after homomorphic encryption and SVD compression processing in the database. The stored data exists in an encrypted form to ensure that data leakage and unauthorized access can be prevented even at the storage medium level. The encrypted storage unit can flexibly select the storage medium according to the system requirements and storage conditions, and provide a data reading and transmission interface with other modules.

[0122] In this embodiment, the data integrity verification unit is used to verify the integrity of the encrypted stored 3D model data to ensure that the data has not been tampered with during storage and transmission. The integrity verification is achieved by generating hash values ​​for the stored and read data and comparing them. If the two hash values ​​are consistent, it proves that the data has not changed; if they are inconsistent, there may be data damage or tampering.

[0123] In this embodiment, the encrypted and compressed 3D model data processed by the storage encryption module will be used as the input of the index retrieval module to ensure that the data can be efficiently retrieved and used on the basis of secure storage.

[0124] Index retrieval module

[0125] The index retrieval module is used to efficiently build and retrieve the encrypted and compressed 3D model data generated by the storage encryption module, so as to support users to quickly find cultural relic models based on feature similarity. Through feature extraction, local sensitive hashing (LSH) construction and efficient retrieval processing, this module can quickly find similar models in a large amount of cultural relic data to meet the needs of efficient retrieval.

[0126] In this embodiment, the feature extraction unit is used to extract the model's feature vector F={f i}, each eigenvector f i It contains information about the geometric shape, texture characteristics, and spatial structure of the 3D model. By analyzing the main features of the model, the feature extraction unit extracts the high-dimensional feature vector of the model for subsequent similarity matching and retrieval.

[0127] In this embodiment, the locality sensitive hash (LSH) construction unit is used to construct an efficient hash index structure based on the feature vector set FFF extracted by the feature extraction unit. The LSH algorithm preserves the similarity information of the features by mapping high-dimensional features to low-dimensional space. Specifically, the LSH construction unit uses a hash function family H = {h i}Map the feature vector into a hash table. Each hash function satisfies the following definition:

[0128]

[0129] Among them, a is a random vector, b is the bias, and w is the bucket width. This mapping process maps similar feature vectors to the same or similar hash buckets to form feature similarity clusters, thereby speeding up the retrieval of similar models.

[0130] In this embodiment, the index storage unit is used to store the constructed local sensitive hash index structure into the index database. The stored index structure can support the rapid retrieval operation of large-scale cultural relics model data. During the index storage process, the index storage unit optimizes the LSH structure to reduce space usage and improve retrieval efficiency.

[0131] In this embodiment, the retrieval processing unit is used to perform similarity retrieval in the index database based on the query conditions input by the user. Specifically, the retrieval processing unit realizes fast matching of the 3D models of cultural relics that meet the conditions by searching for the same or similar hash buckets in the LSH hash table according to the query feature vector of the user. The matching results include high-similarity model data in the same hash bucket and the second-highest-similarity model data in adjacent buckets.

[0132] In this embodiment, the result sorting unit is used to score the similarity of the multiple matching results output by the search processing unit, and sort them from high to low according to the similarity. The result sorting unit calculates similarity metrics such as cosine similarity or Euclidean distance between feature vectors, and sorts the search results according to the scores to ensure that the cultural relic model that best meets the user's needs is displayed first.

[0133] In this embodiment, the search results processed by the index search module will be used as the input of the display management module to provide accurate data support for the adaptive display and interaction of cultural relics data.

[0134] Display management module

[0135] The display management module is used to adaptively display the 3D model data of cultural relics output by the index retrieval module. The display management module enables users to conveniently view the details of the cultural relic model from different perspectives and devices through layered loading, adaptive rendering, dynamic projection and user interaction functions.

[0136] In this embodiment, the hierarchical loading unit is used to perform multi-resolution hierarchical processing on the 3D model data of the cultural relics to support progressive loading and adaptive display. The model data matrix is ​​hierarchically represented by wavelet decomposition so that the model data can be displayed at different resolutions. Specifically, the hierarchical representation after wavelet decomposition is as follows:

[0137]

[0138] Among them, X k represents the low-resolution representation of the kth layer, ψ k is the corresponding wavelet basis function. The hierarchical loading unit first loads low-resolution data during the initial display, and gradually loads higher-resolution data when the device performance allows, ensuring smooth display under different device and network conditions.

[0139] In this embodiment, the adaptive rendering unit is used to automatically adjust the display resolution and detail level of the 3D model according to the display resolution and performance parameters of the user device. Specifically, the adaptive rendering unit selects a suitable resolution level for rendering according to the computing power, memory and display resolution parameters of the device. Loading higher resolution model data on high-performance devices provides a more refined display effect; loading lower resolution models on low-performance devices ensures the smoothness of the display process.

[0140] In this embodiment, the dynamic projection unit is used to adjust the projection angle of the 3D model in real time according to the change of the user's perspective, so that the user can obtain a multi-angle interactive experience in the model display. The dynamic projection unit realizes the perspective adjustment through projective geometry, and its projection matrix T is expressed as:

[0141]

[0142] Among them, f x and f y is the focal length, c x and c y By adjusting the parameters of the projection matrix in real time, the dynamic projection unit ensures the visual consistency of the cultural relic model at different viewing angles, allowing users to observe the structural details of the cultural relic model from multiple angles.

[0143] In this embodiment, the texture mapping unit is used to perform texture mapping and lighting processing on the surface of the 3D model to make the displayed cultural relic model more realistic. The texture mapping unit applies high-precision texture data to the surface of the 3D model and renders it in combination with the lighting model to simulate the lighting effect in the real environment. The combination of texture mapping and lighting processing can enhance the layering and visual effect of the cultural relic model, so that users can clearly identify the texture and shape of the cultural relic during the display process.

[0144] In this embodiment, the user interaction unit is used to receive and respond to the user's operation on the cultural relic model, including interactive operations such as rotation, scaling and translation. The user interaction unit updates the display state of the 3D model in real time according to the user's input signal, allowing the user to explore and observe the details of the cultural relic in a variety of ways. For example, when the user rotates, the model will adjust the display angle accordingly; when the user scales, the model will display more details or the overall appearance in a zoomed-in or zoomed-out state.

[0145] In this embodiment, the display management module supports the display of data output by the index retrieval module on multiple terminal devices with different resolutions and viewing angles through the collaborative work of various units, and provides users with a rich interactive experience.

[0146] Permission control module

[0147] The permission control module is used to manage the user's access rights to the cultural relic 3D model data. Through the permission matrix, access verification, dynamic adjustment, behavior recording and anomaly detection functions, the permission control module ensures that only authorized users can access and operate the cultural relic data in the system, ensuring the security of the system and the confidentiality of the data.

[0148] In this embodiment, the permission matrix generation unit is used to construct a permission matrix of users and resources to record the access rights of each user to the cultural relics data. Assume that the user set is U and the resource set is R, and define the permission matrix The matrix elements represent the user's access rights to resources, as follows:

[0149]

[0150] The construction of the permission matrix A enables the system to easily find and update the access rights of each user.

[0151] In this embodiment, the access request verification unit is used to verify the user's authority based on the authority matrix A when the user accesses the resource. Specifically, after receiving the user's access request, the access request verification unit queries the corresponding authority value A in the authority matrix A. ij If A ij = 1, the user is allowed to access the resource; if A ij = 0, access is denied and a prompt message indicating that there is no permission is returned. This permission verification mechanism ensures that only users who meet the permission can access the corresponding resources.

[0152] In this embodiment, the dynamic permission adjustment unit is used to dynamically adjust the permission matrix A according to the authorization of the system administrator or the change requirements of the user's permissions. During the use of the system, the user's permissions may change, such as adding new access permissions or canceling existing permissions. The dynamic permission adjustment unit modifies the elements in the permission matrix according to the administrator's instructions, so that the permission management is real-time and flexible, and can adapt to the permission requirements of different user roles.

[0153] In this embodiment, the behavior recording unit is used to record the user's operation behavior and access log for subsequent security monitoring and analysis. Every time a user accesses cultural relic data or performs an operation, the behavior recording unit will write the behavior into the log, including information such as the accessed resources, operation time, and operation type, and store the log in the behavior record database. The behavior record provides detailed data support for subsequent anomaly detection and security analysis.

[0154] In this embodiment, the anomaly detection unit is used to perform real-time analysis on the access log of the behavior recording unit to identify abnormal access behaviors. The anomaly detection unit uses a cluster analysis algorithm to analyze the user's access pattern and marks the anomaly by identifying access behaviors that deviate greatly from the normal behavior pattern. When abnormal access behavior is detected, the anomaly detection unit will send an alarm message to the system administrator to indicate possible security risks.

[0155] In this embodiment, the permission control module ensures that the system can safely and effectively control the user's access rights through the collaborative work of the above-mentioned units, prevent unauthorized access, and improve the security and confidentiality of cultural relics data.

[0156] In general, the present invention constructs a complete cultural relic 3D data management process through the collaborative work of data preprocessing, model generation, intelligent repair, storage encryption, index retrieval, display management and authority control modules. The system realizes the acquisition and denoising of 3D point cloud data, the generation and repair of models, encrypted and compressed storage, efficient feature-based retrieval and multi-terminal adaptive display, and ensures secure access to data through authority control. Through this system, the digital data of cultural relics can be presented in a high-precision, safe and smooth manner, providing effective technical support for the protection, research and display of cultural relics.

[0157] Please see attached Figure 2 The present invention also provides a management method for secure storage and efficient display of digital cultural relics data, comprising the following steps:

[0158] Step S1: Acquire the 3D point cloud data of the cultural relic and perform wavelet transform denoising on it, converting the collected point cloud data into a sparse matrix form so as to retain the effective features of the surface of the cultural relic after denoising.

[0159] Step S2: Based on the sparse matrix data generated in step S1, a preliminary 3D model is generated using non-uniform rational B-splines (NURBS) so that the model has the basic form of the surface of the cultural relic.

[0160] Step S3: Perform intelligent repair on the preliminary 3D model generated in step S2, fill in the missing data by minimizing the nuclear norm, generate complete 3D model data, and ensure the continuity and consistency of the model surface.

[0161] Step S4: Encrypt and compress the complete 3D model data generated in step S3. The data is encrypted using the Paillier homomorphic encryption algorithm and compressed using singular value decomposition (SVD) to ensure the security and storage efficiency of the model data.

[0162] Step S5: Use the Local Sensitive Hash (LSH) algorithm to construct a feature index for the encrypted 3D model data, and map the model features into a hash table to facilitate subsequent rapid retrieval of similar cultural relic models.

[0163] Step S6: The 3D model is displayed in multiple resolutions and layers through wavelet decomposition, and access rights are controlled according to the user authority matrix to ensure that the model data is adaptively loaded on different devices and to restrict unauthorized access.

[0164] This method realizes the complete process of cultural relic data from collection, denoising, modeling, restoration to storage, retrieval and display, effectively improving the security, management efficiency and display effect of cultural relic digital data.

[0165] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A management system for secure storage and efficient display of digital cultural relics data, characterized in that: include: Data preprocessing module, used to collect and denoise the 3D point cloud data of cultural relics to form sparse matrix data; A model generation module, used for generating a preliminary 3D model of the cultural relic based on the sparse matrix data; An intelligent repair module, used for automatically repairing the preliminary 3D model to generate a complete 3D model; A storage encryption module, used for homomorphically encrypting and compressing the complete 3D model for storage; An index retrieval module, used to generate an index based on the data features of the complete 3D model to achieve efficient retrieval; A display management module is used to adaptively display the complete 3D model based on the index retrieval result.

2. A management system for secure storage and efficient display of digital cultural relics data according to claim 1, characterized in that: The data preprocessing module comprises: The point cloud data acquisition unit is used to acquire the 3D point cloud data of the cultural relics and represent the acquired point cloud data as a point set P = {p i |p i =(x i ,y i ,z i )}, where p i Represents a three-dimensional coordinate point in space; A sparse matrix construction unit, used to convert the point cloud data into a sparse matrix To represent the collected valid data points and missing points, the sparse matrix satisfies: Among them, p ij Indicates valid point cloud data value, and zero value indicates a vacant position where no data is collected; The denoising processing unit is used to perform denoising on the sparse matrix data through wavelet transform to reduce the noise in the collected data and retain important features, thereby obtaining denoised sparse matrix data M′.

3. A management system for secure storage and efficient display of digital cultural relics data according to claim 1, characterized in that: The model generation module includes: A point cloud feature extraction unit, used to extract spatial features of point cloud data from the denoised sparse matrix data output by the data preprocessing module, and determine the preliminary form of the model according to the extracted features; A NURBS surface generating unit is used to generate a preliminary 3D surface model of the cultural relic by a non-uniform rational B-spline method based on the output of the feature extracting unit, wherein the NURBS surface satisfies the following parameterized representation: Among them, N i,p (u) and N j,q (v) is the B-spline basis function, P ij is the control point, w ij is the weight parameter, and u and v are parameters used to construct the surface model of the cultural relic; The surface optimization unit is used to optimize the generated preliminary surface model by adjusting the position and weight parameters of the surface control points to make the surface model more consistent with the actual shape of the cultural relic surface.

4. A management system for secure storage and efficient display of digital cultural relics data according to claim 1, characterized in that: The intelligent repair module comprises: The sparse matrix completion unit is used to receive the sparse matrix data of the preliminary 3D model generated by the model generation module, and complete the missing area of ​​the sparse matrix by the nuclear norm minimization method. The optimization goal of the nuclear norm minimization is: min||X|| * subjectto X ij =M ij for observed entries Among them, ||X|| * =Σ i σ i (X) is the nuclear norm of matrix X, σ i (X) is the singular value of the matrix, which is used to complete the missing model data by minimizing the rank of the matrix; The alternating least squares solving unit is used to iteratively solve the nuclear norm minimization optimization problem of the sparse matrix completion unit by using the alternating least squares method. By gradually updating the matrix values ​​of rows and columns until the completion matrix X converges to a stable state, the completed high-precision sparse matrix X is obtained. * ; Geometry repair unit for the sparse matrix X after completion * Restore the geometric structure of the preliminary 3D model so that the structure of the completed area is consistent with that of the original area of ​​the model; The smoothing unit is used to smooth the 3D model output by the geometric structure repair unit to ensure the continuity and naturalness of the model surface. The smoothing is achieved by minimizing the following objective function: Among them, S(u,v) represents the surface of the repaired model, Represents the second-order gradient operator, which obtains a smooth 3D model by minimizing the quadratic rate of change of the surface model.

5. A management system for secure storage and efficient display of digital cultural relics data according to claim 1, characterized in that: The storage encryption module comprises: Homomorphic encryption unit, used to perform homomorphic encryption processing on the complete 3D model data generated by the intelligent repair module, and use the Paillier homomorphic encryption algorithm to encrypt the values ​​in the model data; The singular value decomposition compression unit is used to compress the encrypted 3D model data after homomorphic encryption is completed, and decompose the encrypted data matrix using the singular value decomposition algorithm; The encryption storage unit is used to store the 3D model data compressed by singular value decomposition in a storage medium.

6. A management system for secure storage and efficient display of digital cultural relics data according to claim 1, characterized in that: The index retrieval module includes: A vector feature extraction unit is used to extract a feature vector F of the model from the encrypted 3D model data output by the storage encryption module. i }, the feature vector is used to represent the geometric shape, texture and other characteristics of the cultural relic model, providing a basis for subsequent similarity retrieval; A local sensitive hash construction unit is used to construct a local sensitive hash index based on the feature vector set F extracted by the feature extraction unit, and a hash function family H = {h i } Map the feature vector into a hash table, where the hash function satisfies the following form: Among them, a is a random vector, b is a bias, and w is the bucket width. Similar feature vectors are mapped to the same hash bucket through hash mapping, thereby realizing similarity search; An index storage unit, used to store the constructed local sensitive hash index structure into an index database; A retrieval processing unit is used to receive the query conditions input by the user, call the corresponding local sensitive hash function from the index database to perform model matching based on the query conditions, implement fast similarity matching according to the feature vectors in the same bucket in the hash table, and output the 3D model data of the cultural relics that meet the conditions; The result sorting unit is used to sort the multiple matching results output by the retrieval processing unit by similarity, and arrange the retrieval results from high to low according to the similarity scores between the feature vectors.

7. A management system for secure storage and efficient display of digital cultural relics data according to claim 1, characterized in that: The display management module includes: The hierarchical loading unit is used to decompose the 3D model data into multiple resolution levels according to the data structure of the cultural relic 3D model, and realize hierarchical loading through wavelet decomposition. The decomposed model data meets the following requirements: Among them, X k is the low-resolution representation of the model at layer k, ψ k The corresponding wavelet basis function supports progressive loading through multi-resolution layered processing, ensuring that the 3D model of cultural relics can be displayed smoothly under different devices and network conditions; Adaptive rendering unit, which automatically adjusts the loading resolution and detail level of 3D models according to the display resolution and performance parameters of the user's device to ensure high-resolution model display on high-performance devices; The dynamic projection unit is used to adjust the projection angle of the 3D model in real time according to the change of the user's viewing angle to achieve multi-angle interactive display. The projection transformation is performed through projective geometry. The projection matrix T satisfies: Among them, f x and f y is the focal length, c x and c y It is the projection center to ensure the visual consistency of the model under different viewing angles; The texture mapping unit is used to perform texture mapping and lighting processing on the surface of the 3D model. The real material and texture effects are superimposed on the model surface through texture mapping, and rendered in combination with the lighting model.

8. A management system for secure storage and efficient display of digital cultural relics data according to claim 1, characterized in that: The system also includes a permission control module for controlling the user's access rights to ensure safe access to cultural relic data.

9. A management system for secure storage and efficient display of digital cultural relics data according to claim 8, characterized in that: The authority control module includes: The permission matrix generation unit is used to represent the user set U and resource set R in the system as a permission matrix The permission matrix is ​​used to record the user's access rights to various resources and is defined as follows: Among them, A ij =1 means user i has access rights to resource j, A ij =0 means no permission; The access request verification unit is used to receive the user's resource access request and verify whether the user has the access rights to the requested resource based on the permission matrix A. Specifically, it searches for the corresponding permission value A in the permission matrix. ij , if A ij =1, access is allowed. If A ij =0, access is denied and a prompt message indicating that the user does not have permission is returned; A dynamic permission adjustment unit is used to dynamically adjust the permission values ​​in the permission matrix A according to the authorization request of the administrator or the permission requirement of the user, so as to update the user's access rights in real time; A behavior recording unit is used to record the user's operation behavior and access log in the system, and store these behavior records in the behavior log database; The anomaly detection unit is used to perform real-time analysis on the user access logs in the behavior recording unit, perform cluster analysis on user access behaviors through clustering algorithms, identify and mark abnormal access behaviors, and send alarm information to the administrator when abnormal access behaviors are detected to ensure the security of cultural relics data.

10. A method for managing the secure storage and efficient display of digital cultural relic data, applied to the management system for secure storage and efficient display of digital cultural relic data as claimed in any one of claims 1 to 9, characterized in that: The following steps are involved: Obtain the 3D point cloud data of cultural relics, remove noise through wavelet transform, and generate a sparse matrix; Generate a preliminary 3D model using NURBS based on sparse matrix; The preliminary 3D model is intelligently repaired by minimizing the nuclear norm to generate a complete 3D model; The complete 3D model is encrypted with Paillier homomorphic encryption and compressed with singular value decomposition; The LSH algorithm is used to generate indexes for the encrypted 3D model features; The 3D model is displayed in multiple resolutions through wavelet decomposition, and access rights are controlled based on the user permission matrix.

Images