A method and system for constructing a cultural and tourism virtual scene based on a 3D model

Abstract

Embodiments of the present invention relate to the field of virtual culture and tourism technology, and specifically disclose a method and system for constructing a cultural and tourism virtual scene based on a three-dimensional model. Embodiments of the present invention download complete three-dimensional scene data of a target tourist attraction, obtain historical travel record data of a user; identify preferences from the historical travel record data; select a virtual culture and tourism guide; plan a travel route according to the travel preference information, generate a preferred travel route, and screen and process the complete three-dimensional scene data to construct a preferred virtual scene; optimize the guided tour of the preferred virtual scene for virtual culture and tourism to generate a cultural and tourism guided tour virtual scene. It can obtain the historical travel record data of the user, record the travel preference information of the user, construct a preferred virtual scene, select a virtual culture and tourism guide, and optimize and generate a cultural and tourism guided tour virtual scene, making the construction of the cultural and tourism virtual scene targeted and characteristic, so as to provide a cultural and tourism virtual scene that better meets the user's interests.

Description

Technical Field

[0001] The present invention belongs to the technical field of virtual culture and tourism, and particularly relates to a method and system for constructing a culture and tourism virtual scene based on a three-dimensional model. Background Art

[0002] Virtual culture and tourism is a new development model of the culture and tourism industry under the digital economy. It uses advanced technologies such as virtual reality (VR), augmented reality (AR), and mixed reality (MR) to deeply integrate the two major industries of culture and tourism, providing tourists with an immersive tourism experience.

[0003] Virtual culture and tourism is based on real tourist attractions. By simulating or constructing a virtual tourism environment beyond reality, it breaks the limitations of traditional tourism in space and time, enabling tourists to carry out virtual tourism activities as if they were on the spot.

[0004] In the prior art, the construction of virtual scenes in virtual culture and tourism only constructs a standard and general virtual scene of tourist attractions, and cannot analyze the tourism preferences of users for corresponding adjustments. The construction of cultural and tourism virtual scenes lacks pertinence and distinctiveness, and cannot provide users with cultural and tourism virtual scenes that better suit their interests. Summary of the Invention

[0005] The purpose of the embodiments of the present invention is to provide a method and system for constructing a culture and tourism virtual scene based on a three-dimensional model, aiming to solve the problems raised in the background art.

[0006] To achieve the above purpose, the embodiments of the present invention provide the following technical solutions:

[0007] A method for constructing a culture and tourism virtual scene based on a three-dimensional model, the method specifically includes the following steps:

[0008] Obtain the basic information of the target tourist attraction, download the complete three-dimensional scene data of the target tourist attraction, and interact with the user for permissions to obtain the user's historical travel record data;

[0009] Identify the preferences from the historical travel record data and record the user's tourism preference information;

[0010] Select a virtual culture and tourism guide according to the basic information of the scenic spot and the tourism preference information;

[0011] Plan a travel route according to the tourism preference information, generate a preferred travel route, and screen and process the complete three-dimensional scene data to construct a preferred virtual scene;

[0012] Optimize the virtual tour of the preferred virtual scene according to the preferred travel route and the virtual culture and tourism guide to generate a cultural and tourism guided virtual scene.

[0013] A cultural and tourism virtual scene construction system based on a 3D model, the system includes a complete data acquisition unit, a preference recognition and recording unit, a virtual tour guide selection unit, a virtual scene construction unit, and a scene tour optimization unit, where:

[0014] The complete data acquisition unit is used to obtain the basic information of the target tourist attraction, download the complete 3D scene data of the target tourist attraction, and interact with the user for permissions to obtain the user's historical travel record data;

[0015] The preference recognition and recording unit is used to perform preference recognition on the historical travel record data and record the user's travel preference information;

[0016] The virtual tour guide selection unit is used to select a virtual cultural and tourism tour guide according to the basic information of the scenic spot and the travel preference information;

[0017] The virtual scene construction unit is used to plan a travel route according to the travel preference information, generate a preferred travel route, and screen and process the complete 3D scene data to construct a preferred virtual scene;

[0018] The scene tour optimization unit is used to optimize the virtual cultural and tourism tour of the preferred virtual scene according to the preferred travel route and the virtual cultural and tourism tour guide to generate a cultural and tourism tour virtual scene.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] In the embodiment of the present invention, by downloading the complete 3D scene data of the target tourist attraction, obtaining the user's historical travel record data; performing preference recognition on the historical travel record data; selecting a virtual cultural and tourism tour guide; planning a travel route according to the travel preference information, generating a preferred travel route, and screening and processing the complete 3D scene data to construct a preferred virtual scene; optimizing the virtual cultural and tourism tour of the preferred virtual scene to generate a cultural and tourism tour virtual scene. It can obtain the user's historical travel record data, record the user's travel preference information, construct a preferred virtual scene, select a virtual cultural and tourism tour guide, and optimize and generate a cultural and tourism tour virtual scene, making the construction of the cultural and tourism virtual scene targeted and characteristic, so as to provide a cultural and tourism virtual scene that better meets the user's interests. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0022] Figure 1The flowchart of the method provided by the embodiment of the present invention is shown.

[0023] Figure 2 The flowchart of downloading complete 3D scene data in the method provided by the embodiment of the present invention is shown.

[0024] Figure 3 The flowchart of interacting with the user for permissions in the method provided by the embodiment of the present invention is shown.

[0025] Figure 4 The flowchart of recording travel preference information in the method provided by the embodiment of the present invention is shown.

[0026] Figure 5 The flowchart of selecting a virtual cultural and tourism guide in the method provided by the embodiment of the present invention is shown.

[0027] Figure 6 The flowchart of constructing a preference virtual scene in the method provided by the embodiment of the present invention is shown.

[0028] Figure 7 The flowchart of generating a cultural and tourism guide virtual scene in the method provided by the embodiment of the present invention is shown.

[0029] Figure 8 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0030] Figure 9 The structural block diagram of the complete data acquisition unit in the system provided by the embodiment of the present invention is shown.

[0031] Figure 10 The structural block diagram of the virtual scene construction unit in the system provided by the embodiment of the present invention is shown. Detailed implementation manners

[0032] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0033] It can be understood that in the prior art, for the construction of virtual scenes in virtual cultural and tourism, only a standard and general virtual scene of tourist attractions is constructed, and it is not possible to analyze the travel preferences of users and make corresponding adjustments. The construction of cultural and tourism virtual scenes lacks pertinence and characteristics, and cannot provide users with cultural and tourism virtual scenes that more conform to their interests.

[0034] To solve the above problems, embodiments of the present invention obtain the basic information of the target tourist attraction, download the complete three-dimensional scene data of the target tourist attraction, interact with the user for permissions, and obtain the user's historical travel record data; identify preferences from the historical travel record data and record the user's travel preference information; select a virtual cultural and tourism guide according to the basic information of the attraction and the travel preference information; plan a travel route according to the travel preference information, generate a preferred travel route, and screen and process the complete three-dimensional scene data to construct a preferred virtual scene; optimize the guided tour of the preferred virtual scene according to the preferred travel route and the virtual cultural and tourism guide to generate a cultural and tourism guided virtual scene. It can obtain the user's historical travel record data, record the user's travel preference information, construct a preferred virtual scene, select a virtual cultural and tourism guide, and optimize the generation of a cultural and tourism guided virtual scene, making the construction of the cultural and tourism virtual scene targeted and characteristic, so as to provide the user with a cultural and tourism virtual scene that better suits their interests.

[0035] Figure 1 The flowchart of the method provided by an embodiment of the present invention is shown.

[0036] Specifically, a method for constructing a cultural and tourism virtual scene based on a three-dimensional model, the method specifically includes the following steps:

[0037] Step S101, obtain the basic information of the target tourist attraction, download the complete three-dimensional scene data of the target tourist attraction, interact with the user for permissions, and obtain the user's historical travel record data.

[0038] In an embodiment of the present invention, when the user needs to conduct virtual cultural and tourism through a mobile device, the virtual cultural and tourism requirements can be sent through the mobile device, the target of the virtual cultural and tourism requirements is identified to determine the target tourist attraction, then the basic information of the target tourist attraction is obtained, and the complete three-dimensional scene data of the target tourist attraction is matched and downloaded. At the same time, a historical permission application is generated and sent to the user's mobile device, and then the permission application feedback from the user through the mobile device is received. By judging whether to agree to the application for the permission application feedback, when it is determined that the user agrees to the historical permission application, the user's historical travel record data is obtained through the mobile device.

[0039] Specifically, Figure 2 The flowchart of downloading the complete three-dimensional scene data in the method provided by an embodiment of the present invention is shown.

[0040] Among them, in the preferred embodiment provided by the present invention, the steps of obtaining the basic information of the target tourist attraction, downloading the complete three-dimensional scene data of the target tourist attraction, interacting with the user for permissions, and obtaining the user's historical travel record data specifically include the following steps:

[0041] Step S1011: Receive the virtual cultural and tourism needs of the user and determine the target tourist attraction;

[0042] Step S1012: Obtain the basic information of the target tourist attraction;

[0043] Step S1013: Download the complete 3D scene data of the target tourist attraction;

[0044] Step S1014: Interact with the user regarding permissions and obtain the user's historical travel record data.

[0045] Specifically, Figure 3 FIG. shows the flow chart of interacting with the user regarding permissions in the method provided by the embodiment of the present invention.

[0046] Among them, in the preferred embodiment provided by the present invention, the interacting with the user regarding permissions and obtaining the user's historical travel record data specifically includes the following steps:

[0047] Step S10141: Generate and send a historical permission application to the user's mobile device;

[0048] Step S10142: Receive the permission application feedback from the mobile device;

[0049] Step S10143: According to the permission application feedback, when the historical permission application is approved, obtain the user's historical travel record data through the mobile device.

[0050] Furthermore, the method for constructing a cultural and tourism virtual scene based on a 3D model further includes the following steps:

[0051] Step S102: Identify the preferences from the historical travel record data and record the user's travel preference information.

[0052] In the embodiment of the present invention, the historical travel record data is classified and statistically analyzed according to a preset plurality of classification categories to obtain classification statistical data. Then, based on a preset standard threshold, the statistical proportions corresponding to the plurality of classification categories in the classification statistical data are compared to select a plurality of travel preference types, and then the plurality of travel preference types are sorted out to record the user's travel preference information.

[0053] It can be understood that the plurality of classification categories may include humanities, nature, entertainment, etc.

[0054] Specifically, Figure 4 FIG. shows the flow chart of recording travel preference information in the method provided by the embodiment of the present invention.

[0055] Among them, in the preferred embodiment provided by the present invention, the preference recognition of the historical travel record data and the recording of the user's travel preference information specifically include the following steps:

[0056] Step S1021: Classify and statistically analyze the historical travel record data to obtain classified statistical data;

[0057] Step S1022: Based on a preset standard threshold, compare the classified statistical data and select multiple travel preference types;

[0058] Step S1023: Organize the multiple travel preference types and record the user's travel preference information.

[0059] Among them, in the preferred embodiment provided by the present invention, the classification and statistical analysis of the historical travel record data to obtain classified statistical data specifically include the following steps:

[0060] Use the user ID as the index for the historical travel record data to establish a vertical data stack;

[0061] Slice the vertical data stack in time series according to three time granularities of week / month / year to establish a horizontal time axis and obtain a three-dimensional data structure;

[0062] Complement the data of the same user in the same period in the three-dimensional data structure by using the nearest neighbor interpolation method to obtain a three-dimensional data cube, where the first dimension is the user dimension, the second dimension is the time slice dimension, and the third dimension is the feature dimension;

[0063] Perform spatial projection along the user dimension to generate a feature query vector Q; perform time projection along the time slice dimension to generate a key vector K; perform linear projection along the feature dimension to generate a value vector V;

[0064] Perform attention calculation on the feature query vector Q, the key vector K, and the value vector V to obtain a weighted data cube;

[0065] Perform complex domain tensor decomposition and phase perturbation enhancement on the weighted data cube to construct spatio-temporal fusion features by combining time density distribution and nonlinear coupling. Calculate the confidence classification by fusing similarity and probability density for the spatio-temporal fusion features, and finally construct a three-dimensional histogram through frequency statistics and association rule mining to output the classification result.

[0066] Among them, in the preferred embodiment provided by the present invention, the weighted data cube is subjected to complex domain tensor decomposition and phase perturbation to enhance spatio-temporal feature representation, and spatio-temporal fusion features are constructed by combining time density distribution and non-linear coupling. The spatio-temporal fusion features are fused with similarity and probability density to calculate confidence classification, and finally a three-dimensional histogram is constructed through frequency statistics and association rule mining to output the classification result, which specifically includes the following steps:

[0067] Expand the real values in the weighted data cube to the complex domain to obtain the real part and the imaginary part, where the real part and the imaginary part form a three-dimensional tensor of complex numbers;

[0068] Perform orthogonal decomposition on the three-dimensional tensor of complex numbers along the corresponding dimensions to obtain an orthogonal basis system after low-rank decomposition. The orthogonal basis system after low-rank decomposition includes a user basis matrix, a core tensor, and a feature space basis;

[0069] In the orthogonal basis system after low-rank decomposition, add a rotation factor to the imaginary part of the user basis matrix and apply phase perturbation to the core tensor to obtain a user basis matrix with a rotation factor added and a core tensor with phase perturbation added;

[0070] Reconstruct the tensor by back projection using the user basis matrix with a rotation factor added, the core tensor with phase perturbation added, and the feature space basis to obtain a complex domain reconstructed tensor;

[0071] Use the sigmoid function to suppress extreme values for the real part in the complex domain reconstructed tensor to obtain a real part processing result; use the ReLU function to filter invalid phases for the imaginary part in the complex domain reconstructed tensor to obtain an imaginary part processing result; cross-couple the real part processing result and the imaginary part processing result to obtain a non-linearly enhanced feature matrix;

[0072] According to the timestamp of each data in the three-dimensional data cube and the current time, calculate the short-term decay effect of the time interval through the exponential function, calculate the long-term distribution form of the time interval through the error function, and give a seasonal weight according to the seasonal change. Linearly weight the short-term decay effect and the long-term distribution form with the seasonal weight to obtain a spatio-temporal correlation matrix;

[0073] Perform probability mapping on the spatio-temporal correlation matrix to generate a time density distribution matrix, where the row vector in the time density distribution matrix represents the time sensitivity distribution of a single user, and the column vector represents the comparison of the association strengths of different users within the same time slice;

[0074] Perform logarithmic transformation on each element in the time density distribution matrix and then perform standard processing by column to obtain a standardized time gate matrix;

[0075] The standardized time-gate matrix and the non-linearly enhanced feature matrix are respectively expanded in the frequency domain and then interwoven and spliced to obtain the spliced matrix. The spliced matrix is subjected to element-wise multiplication to enhance the coupling effect of time series-behavior, resulting in a spatio-temporal fusion feature tensor;

[0076] For a preset category, calculate the cosine similarity between the spatio-temporal fusion feature tensor and the centroid of the preset category to obtain a similarity result. Use the Gaussian kernel function to evaluate the compactness of the feature distribution in the spatio-temporal fusion feature tensor to obtain a probability density. Regularize and fuse the similarity result and the probability density to obtain a preliminary classification result with confidence;

[0077] For the preliminary classification result with confidence, count the occurrence frequency and duration distribution of each classification according to the user ID to obtain a structured statistical table. Mine the category association rules in the preliminary classification result with confidence across different confidence levels through the Apriori algorithm to obtain an association rule set;

[0078] Build a three-dimensional statistical histogram from the structured statistical table and the association rule set according to the time slice division parameter to obtain the final classification statistical data.

[0079] In the embodiments of the present invention, the present invention realizes a comprehensive expression of user behavior in space (user dimension), time (time granularity dimension) and content features (feature dimension) by constructing a three-dimensional data cube (user × time × feature) indexed by users, ensuring the diversity and integrity of data.

[0080] Vertically establishing a data stack ensures the systematicness of historical data. Horizontally slicing the time series enhances the ability to capture the time context, meeting the requirements of time series analysis. And the nearest neighbor interpolation method is used to fill in the missing data, effectively alleviating the problems of data sparsity and incompleteness and improving the reliability of subsequent analysis. At the same time, the exponential decay (capturing short-term memory) and error function distribution (characterizing long-term patterns) are adopted, which can avoid the limitations of traditional single-time decay models, making the statistical results reflect both the latest trends and retain historical patterns.

[0081] Through the extension from real numbers to complex numbers, the real and imaginary parts of complex numbers are used to capture more dimensions of implicit information and phase relationships. Orthogonal low-rank decomposition is used to reduce the dimension and extract features of high-dimensional complex data, which not only reduces noise but also retains important structural information, helping to reveal potential patterns.

[0082] Rotation factors and phase perturbations are implanted in the user base matrix and the core tensor to enhance the sensitivity and adaptability of the model to spatio-temporal dynamic changes, which is beneficial to capturing the non-linear and periodic characteristics of user behavior. Combining exponential decay, error function and seasonal weights, finely simulates the short-term decay effect, long-term distribution characteristics and seasonal variations of user behavior, improving the accuracy of time sensitivity.

[0083] The generated spatio-temporal correlation matrix and time density distribution matrix quantify the complex dependence relationships between users and time slices, supporting more accurate behavior modeling. The time series features of the data are captured by frequency domain transformation, and the element-wise product operation after interleaving and splicing strengthens the coupling of time series and behavior features, improving the model's recognition ability for time series dynamic behaviors.

[0084] Moreover, the feature query vector Q, key vector K, and value vector V are subjected to attention calculation. During subsequent tensor grading, weight generation and tensor decomposition form an iterative optimization relationship. Compared with traditional static decomposition methods, the recognition accuracy of key time slices can be effectively improved. And by replacing the traditional matrix chain operation with third-order tensor decomposition, the calculation efficiency is optimized. Also, the extension in the complex number domain reduces the number of model parameters, making it suitable for large-scale data environments such as culture and tourism.

[0085] Furthermore, the method for constructing a culture and tourism virtual scene based on a three-dimensional model further includes the following steps:

[0086] Step S103, select a virtual culture and tourism guide according to the basic information of the scenic spot and the tourism preference information.

[0087] In the embodiments of the present invention, according to the basic information of the scenic spot, multiple relevant historical figures related to the target tourist scenic spot are determined. Then, based on big data technology, the three-dimensional images of multiple relevant historical figures are obtained from the Internet. And according to the tourism preference information, preference selection is performed on the three-dimensional images of multiple figures, and a virtual culture and tourism guide is selected from the three-dimensional images of multiple figures.

[0088] Specifically, Figure 5 shows the flow chart of selecting a virtual culture and tourism guide in the method provided by the embodiments of the present invention.

[0089] Among them, in the preferred implementation manner provided by the present invention, the step of selecting a virtual culture and tourism guide according to the basic information of the scenic spot and the tourism preference information specifically includes the following steps:

[0090] Step S1031, determine multiple relevant historical figures according to the basic information of the scenic spot;

[0091] Step S1032, obtain the three-dimensional images of multiple relevant historical figures;

[0092] Step S1033, select a virtual culture and tourism guide from the three-dimensional images of multiple figures according to the tourism preference information.

[0093] Furthermore, the method for constructing a culture and tourism virtual scene based on a three-dimensional model further includes the following steps:

[0094] Step S104: According to the travel preference information, plan a travel route to generate a preferred travel route, and screen and process the complete three-dimensional scene data to construct a preferred virtual scene.

[0095] In an embodiment of the present invention, based on the basic information of scenic spots, determine the locations of multiple tourist attractions in the target tourist attraction. Then, according to the travel preference information, perform location preference matching on the locations of multiple tourist attractions to determine the preferred scenic spot locations that multiple users are interested in. Import the electronic map of the scenic spots of the target tourist attraction, mark the locations of multiple preferred scenic spots in the electronic map of the scenic spots, and then plan a travel route to generate a preferred travel route. Furthermore, based on the multiple preferred scenic spot locations in the preferred travel route, screen the relevant three-dimensional scenes in the complete three-dimensional scene data, and then perform cross-dimensional video processing of the three-dimensional real scene according to the preferred travel route to construct a preferred virtual scene.

[0096] Specifically, Figure 6 The flowchart of constructing a preferred virtual scene in the method provided by the embodiment of the present invention is shown.

[0097] Among them, in the preferred embodiment provided by the present invention, the step of according to the travel preference information, planning a travel route to generate a preferred travel route, and screening and processing the complete three-dimensional scene data to construct a preferred virtual scene specifically includes the following steps:

[0098] Step S1041: According to the travel preference information, perform location preference matching on the basic information of the scenic spots to determine multiple preferred scenic spot locations;

[0099] Step S1042: Import the electronic map of the scenic spots of the target tourist attraction;

[0100] Step S1043: In the electronic map of the scenic spots, plan a travel route for multiple preferred scenic spot locations to generate a preferred travel route;

[0101] Step S1044: Based on the preferred travel route, perform partial screening and route sceneization on the complete three-dimensional scene data to construct a preferred virtual scene.

[0102] Among them, in the preferred embodiment provided by the present invention, the step of according to the travel preference information, performing location preference matching on the basic information of the scenic spots to determine multiple preferred scenic spot locations specifically includes the following steps:

[0103] Decompose the travel preference vector into time-sensitive features and static preference features to obtain a structured preference feature group, perform hierarchical parsing on the scenic spot attribute tensor, and separate the basic attributes and dynamic popularity attributes to obtain a hierarchical scenic spot attribute set;

[0104] Input the structured preference feature group into the LSTM network to extract temporal pattern features, calculate the dynamic weight by combining the temporal pattern features with the time decay factor of the scenic spot visit popularity, and obtain the time decay weight matrix;

[0105] Perform cross-modal convolution on the structured preference feature group and the hierarchical scenic spot attribute set to obtain the cross-modal convolution result, and extract the real part feature component from the cross-modal convolution result;

[0106] Extract the cultural semantic features in the hierarchical scenic spot attribute set, perform conjugate transpose transformation on the cultural semantic features, and perform non-linear interaction calculation on the transpose transformation and the user preference features to obtain the semantic association strength vector;

[0107] Bitwise superimpose the real part feature component and the semantic association strength vector, and perform amplitude normalization processing on the superimposed result to obtain the fusion feature vector;

[0108] Perform non-linear activation and feature space mapping processing on the fusion feature vector in sequence to obtain the activated matching degree score;

[0109] Take the median of the activated matching degree scores as the dynamic threshold, and sample the scenic spots with activated matching degree scores greater than the dynamic threshold according to the score gradient to obtain the preliminary screening scenic spot set;

[0110] Calculate the geodesic distance between the scenic spots in the preliminary screening scenic spot set, generate the adjacency matrix, and perform path-aware optimization on the adjacency matrix to obtain multiple preferred scenic spot locations.

[0111] In the embodiment of the present invention, through the parallel calculation of the real part channel (cross-modal convolution) and the imaginary part channel (semantic association), the limitation of traditional single-channel matching is broken through. Compared with the traditional cosine similarity algorithm, the parsing error of complex preferences can be effectively reduced. And the synergistic effect of the LSTM network and the time decay factor is adopted to realize the real-time evolution of the weight matrix, which can more accurately reflect the current preferences of users. And through the dual constraint mechanism of the topological relationship graph and the matching degree score, while ensuring the preference matching degree, the standard deviation of the distance between adjacent scenic spots is reduced.

[0112] Among them, in the preferred implementation manner provided by the present invention, the step of generating a preferred travel route by planning a travel route for multiple preferred scenic spot locations in the scenic spot electronic map specifically includes the following steps:

[0113] Perform Gaussian kernel density estimation on multiple preferred scenic spot locations to generate a preferred location probability distribution map;

[0114] Separate the scenic spot electronic map, extract the terrain layer, facility layer, and pedestrian flow layer features to obtain a hierarchical map feature set;

[0115] Divide the preference location probability distribution map along the time axis into equally spaced sampling points, and bind the preference location point coordinates to the time points to generate a spatio-temporal location coordinate sequence;

[0116] Evaluate the position value of each preference location point in the spatio-temporal location coordinate sequence to obtain the value of each preference location point, calculate the value change gradient of adjacent preference location points, and obtain the path value change rate;

[0117] Perform cross-channel interactive convolution operations on each layer in the hierarchical map feature set using deformable convolutional kernels to obtain the dynamic convolution results of the fused map features;

[0118] Perform a tensor product operation on the dynamic convolution results of the fused map features and the path value change rate to generate a spatio-temporal correlation field, and then perform progressive integration on the spatio-temporal correlation field along the time axis to generate a spatio-temporal integration field;

[0119] Extract the user's historical behavior vector from the tourism preference information, use the LSTM network to extract the temporal behavior features in the user's historical behavior vector, calculate the similarity between the temporal behavior features and the user's historical behavior vector, and perform normalization to obtain the dynamic attention weight;

[0120] Calculate the three-dimensional geodesic distance value of each preference location point in the preference location probability distribution map, and introduce the terrain slope to the three-dimensional geodesic distance value for constraint correction to obtain the scene adaptation distance;

[0121] Convert the spatio-temporal integration field into a path benefit term, and combine the dynamic attention weight and the scene distance into a cost term to establish a benefit-cost game balance equation;

[0122] Abstract the scenic spot electronic map into a weighted graph structure, where the nodes in the weighted graph structure represent the scenic spot locations, and the edge weights are the spatio-temporal integration field, the dynamic attention weight, and the scene distance results;

[0123] Taking maximizing the total spatio-temporal integration value of the path coverage area, maximizing the weighted average of the dynamic attention weights, and minimizing the total path scene distance as the optimization objectives, perform heuristic search on the scenic spot electronic map based on the A* algorithm, and use the benefit-cost game balance equation to detect the benefit-cost conflict points in the path in real time during the search process;

[0124] When the spatio-temporal integration value of a certain path drops by more than the threshold and the attention weight is higher than the average value, perform local backtracking to the nearest valid branch point, perform cost re-estimation, and re-expand the search along the optimal benefit direction. After the search is completed, generate an optimized path, and perform smoothing processing and scene binding processing on the optimized path to obtain the preferred tourism route.

[0125] In the embodiments of the present invention, the time dimension and the spatial coordinates are jointly encoded to establish a five-dimensional computing space (X, Y, Z, time, preference intensity), so that the generated path can simultaneously meet the spatial accessibility, time rationality, and preference matching degree, effectively improving the spatio-temporal matching accuracy. And a dual-channel optimization mechanism is designed to form a game balance through maximizing spatio-temporal value and minimizing behavioral cost to solve the problem of preference deviation caused by single-objective optimization.

[0126] Further, the method for constructing a cultural and tourism virtual scene based on a three-dimensional model further includes the following steps:

[0127] Step S105, according to the preferred tourist route and the virtual cultural and tourism guide, optimize the virtual tour of the preferred virtual scene to generate a cultural and tourism guided virtual scene.

[0128] In the embodiments of the present invention, according to the preferred tourist route, route guide information is generated, where the route guide information is a guide introduction along the positions of each preferred scenic spot in the preferred tourist route. Then, according to the virtual cultural and tourism guide, the route guide information is voice-processed to generate a route guide voice, and in the preferred virtual scene, the virtual tour of the virtual cultural and tourism guide and the route guide voice is optimized to generate a cultural and tourism guided virtual scene, so that the cultural and tourism guided virtual scene has a guide introduction along the positions of each preferred scenic spot in the preferred tourist route.

[0129] Specifically, Figure 7 shows a flowchart of generating a cultural and tourism guided virtual scene in the method provided by the embodiments of the present invention.

[0130] Among them, in the preferred embodiment provided by the present invention, the optimizing the virtual tour of the preferred virtual scene according to the preferred tourist route and the virtual cultural and tourism guide to generate a cultural and tourism guided virtual scene specifically includes the following steps:

[0131] Step S1051, generate route guide information according to the preferred tourist route;

[0132] Step S1052, voice-process the route guide information according to the virtual cultural and tourism guide to generate a route guide voice;

[0133] Step S1053, in the preferred virtual scene, optimize the virtual tour of the virtual cultural and tourism guide and the route guide voice to generate a cultural and tourism guided virtual scene.

[0134] Further, Figure 8 shows an application architecture diagram of the system provided by the embodiments of the present invention.

[0135] Among them, in another preferred embodiment provided by the present invention, a cultural and tourism virtual scene construction system based on a three-dimensional model includes:

[0136] A complete data acquisition unit 101, configured to acquire the basic information of the target tourist attraction, download the complete three-dimensional scene data of the target tourist attraction, and interact with the user for permissions to obtain the user's historical travel record data.

[0137] In the embodiment of the present invention, when the user needs to conduct virtual cultural and tourism through a mobile device, the mobile device can send a virtual cultural and tourism requirement. The complete data acquisition unit 101 performs target recognition on the virtual cultural and tourism requirement to determine the target tourist attraction, then acquires the basic information of the target tourist attraction, matches the complete three-dimensional scene data of the target tourist attraction, downloads the complete three-dimensional scene data, and at the same time, generates a historical permission application and sends the historical permission application to the user's mobile device, and then receives the permission application feedback from the user through the mobile device. By judging whether to agree to the application for the permission application feedback, when it is determined that the user agrees to the historical permission application, the user's historical travel record data is acquired through the mobile device.

[0138] Specifically, Figure 9 FIG. shows the structural block diagram of the complete data acquisition unit 101 in the system provided by the embodiment of the present invention.

[0139] Among them, in the preferred embodiment provided by the present invention, the complete data acquisition unit 101 specifically includes:

[0140] A requirement receiving module 1011, configured to receive the virtual cultural and tourism requirement of the user and determine the target tourist attraction;

[0141] An information acquisition module 1012, configured to acquire the basic information of the target tourist attraction;

[0142] A data download module 1013, configured to download the complete three-dimensional scene data of the target tourist attraction;

[0143] A permission interaction module 1014, configured to interact with the user for permissions to obtain the user's historical travel record data.

[0144] Furthermore, the cultural and tourism virtual scene construction system based on a three-dimensional model further includes:

[0145] A preference recognition and recording unit 102, configured to perform preference recognition on the historical travel record data and record the user's travel preference information.

[0146] In the embodiment of the present invention, the preference recognition and recording unit 102 classifies and statistically analyzes the historical tourism record data according to a plurality of preset classification categories to obtain classification statistical data, and then compares the statistical ratios corresponding to the plurality of classification categories in the classification statistical data based on a preset standard threshold to select a plurality of tourism preference types, and further sorts out the plurality of tourism preference types to record the tourism preference information of the user.

[0147] The virtual tour guide selection unit 103 is configured to select a virtual cultural and tourism tour guide according to the basic information of the scenic spot and the tourism preference information.

[0148] In the embodiment of the present invention, the virtual tour guide selection unit 103 determines a plurality of relevant historical figures related to the target tourist scenic spot according to the basic information of the scenic spot, and then obtains the three-dimensional images of the plurality of relevant historical figures from the Internet based on big data technology, and performs preference selection on the three-dimensional images of the plurality of figures according to the tourism preference information, and selects a virtual cultural and tourism tour guide from the three-dimensional images of the plurality of figures.

[0149] The virtual scene construction unit 104 is configured to plan a tourism route according to the tourism preference information to generate a preferred tourism route, and screen and process the complete three-dimensional scene data to construct a preferred virtual scene.

[0150] In the embodiment of the present invention, the virtual scene construction unit 104 determines a plurality of tourist scenic spot locations in the target tourist scenic spot according to the basic information of the scenic spot, and then performs position preference matching on the plurality of tourist scenic spot locations according to the tourism preference information to determine a plurality of preferred scenic spot locations that the user is interested in, and imports the electronic map of the scenic spot of the target tourist scenic spot, marks the positions of the plurality of preferred scenic spot locations in the electronic map of the scenic spot, and then plans a tourism route to generate a preferred tourism route, and further screens the relevant three-dimensional scenes in the complete three-dimensional scene data based on the plurality of preferred scenic spot locations in the preferred tourism route, and performs cross-dimensional video processing on the three-dimensional real scene according to the preferred tourism route to construct a preferred virtual scene.

[0151] Specifically, Figure 10 The block diagram of the virtual scene construction unit 104 in the system provided by the embodiment of the present invention is shown.

[0152] Among them, in the preferred embodiment provided by the present invention, the virtual scene construction unit 104 specifically includes:

[0153] The preference matching module 1041 is configured to perform position preference matching on the basic information of the scenic spot according to the tourism preference information to determine a plurality of preferred scenic spot locations;

[0154] The map import module 1042 is configured to import the electronic map of the scenic spot of the target tourist scenic spot;

[0155] A route planning module 1043, configured to plan a travel route for multiple preferred scenic spot locations in the scenic spot electronic map to generate a preferred travel route;

[0156] A scene construction module 1044, configured to perform partial screening and route sceneization on the complete three-dimensional scene data based on the preferred travel route to construct a preferred virtual scene.

[0157] Further, the cultural and tourism virtual scene construction system based on the three-dimensional model further includes:

[0158] A scene tour optimization unit 105, configured to optimize the virtual cultural and tourism tour of the preferred virtual scene according to the preferred travel route and the virtual cultural and tourism guide to generate a cultural and tourism tour virtual scene.

[0159] In an embodiment of the present invention, the scene tour optimization unit 105 generates route tour information according to the preferred travel route, where the route tour information is a tour introduction along each preferred scenic spot location in the preferred travel route, and then performs voice processing on the route tour information according to the virtual cultural and tourism guide to generate a route tour voice, and in the preferred virtual scene, performs virtual cultural and tourism tour optimization of the virtual cultural and tourism guide and the route tour voice to generate a cultural and tourism tour virtual scene, so that the cultural and tourism tour virtual scene has a tour introduction along each preferred scenic spot location of the preferred travel route.

[0160] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown in sequence according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages do not necessarily have to be executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0161] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in this application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0162] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0163] The above embodiments merely represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the invention patent of the present invention should be subject to the appended claims.

[0164] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a cultural and tourism virtual scene based on a 3D model, characterized in that, The method specifically includes the following steps: Obtain the basic information of the target tourist attraction, download the complete 3D scene data of the target tourist attraction, and interact with the user for permissions to obtain the user's historical travel record data; Identify preferences from the historical travel record data and record the user's travel preference information; Select a virtual cultural and tourism guide according to the basic information of the attraction and the travel preference information; Plan a travel route according to the travel preference information, generate a preferred travel route, and screen and process the complete 3D scene data to construct a preferred virtual scene; Optimize the virtual tour of the preferred virtual scene according to the preferred travel route and the virtual cultural and tourism guide to generate a cultural and tourism guided virtual scene; Among them, the steps of obtaining the basic information of the target tourist attraction, downloading the complete 3D scene data of the target tourist attraction, and interacting with the user for permissions to obtain the user's historical travel record data specifically include the following steps: Receive the user's virtual cultural and tourism needs and determine the target tourist attraction; Obtain the basic information of the target tourist attraction; Download the complete 3D scene data of the target tourist attraction; Interact with the user for permissions to obtain the user's historical travel record data; The steps of identifying preferences from the historical travel record data and recording the user's travel preference information specifically include the following steps: Classify and count the historical travel record data to obtain classified statistical data; Compare the classified statistical data based on a preset standard threshold and select multiple travel preference types; Organize the multiple travel preference types and record the user's travel preference information; The steps of classifying and counting the historical travel record data to obtain classified statistical data specifically include the following steps: Establish a vertical data stack with the historical travel record data indexed by user ID; Slice the vertical data stack in three time granularities of week / month / year to establish a horizontal time axis and obtain a three-dimensional data structure; Complement the data of the same user in the same period in the three-dimensional data structure using the nearest neighbor interpolation method to obtain a three-dimensional data cube, where the first dimension is the user dimension, the second dimension is the time slice dimension, and the third dimension is the feature dimension; Perform spatial projection along the user dimension to generate a feature query vector Q; perform time projection along the time slice dimension to generate a key vector K; perform linear projection along the feature dimension to generate a value vector V; Perform attention calculation on the feature query vector Q, the key vector K, and the value vector V to obtain a weighted data cube; Perform complex domain tensor decomposition and phase perturbation on the weighted data cube to enhance spatio-temporal feature representation, combine time density distribution and non-linear coupling to construct spatio-temporal fusion features, use similarity and probability density fusion for the spatio-temporal fusion features to calculate confidence classification, and finally construct a three-dimensional histogram through frequency statistics and association rule mining to output the classification result.

2. The method for constructing a cultural and tourism virtual scene based on a three-dimensional model according to claim 1, wherein The steps of interacting with the user for permissions to obtain the user's historical travel record data specifically include the following steps: Generate and send a historical permission application to the user's mobile terminal; Receive the permission application feedback from the mobile terminal; According to the feedback of the permission application, when agreeing to the historical permission application, the historical travel record data of the user is obtained through the mobile terminal.

3. The method for constructing a cultural and tourism virtual scene based on a three-dimensional model according to claim 1, characterized in that, Performing complex domain tensor decomposition and phase perturbation enhanced spatio-temporal feature characterization on the weighted data cube, constructing fused features by combining time density distribution and non-linear coupling, calculating confidence classification by fusing similarity and probability density for the spatio-temporal fused feature tensor, and finally constructing a three-dimensional histogram output classification result through frequency statistics and association rule mining specifically includes the following steps: Expanding the real values in the weighted data cube to the complex domain to obtain a real part and an imaginary part, where the real part and the imaginary part form a three-dimensional tensor of complex numbers; Performing orthogonal decomposition on the three-dimensional tensor of complex numbers along the corresponding dimensions to obtain an orthogonal basis system after low-rank decomposition, and the orthogonal basis system after low-rank decomposition includes a user basis matrix, a core tensor, and a feature space basis; In the orthogonal basis system after low-rank decomposition, adding a rotation factor to the imaginary part of the user basis matrix and applying phase perturbation to the core tensor to obtain a user basis matrix with a rotation factor added and a core tensor with phase perturbation added; Reconstructing the tensor by back projection using the user basis matrix with a rotation factor added, the core tensor with phase perturbation added, and the feature space basis to obtain a reconstructed tensor in the complex domain; Using the sigmoid function to suppress extreme values for the real part of the reconstructed tensor in the complex domain to obtain a processed result for the real part; using the ReLU function to filter invalid phases for the imaginary part of the reconstructed tensor in the complex domain to obtain a processed result for the imaginary part; cross-coupling the processed result for the real part and the processed result for the imaginary part to obtain a non-linearly enhanced feature matrix; Calculating the short-term decay effect of the time interval through an exponential function according to the timestamp of each data in the three-dimensional data cube and the current time, calculating the long-term distribution pattern of the time interval through an error function, and giving a seasonal weight according to seasonal changes, and linearly weighting the short-term decay effect and the long-term distribution pattern using the seasonal weight to obtain a spatio-temporal correlation matrix; Performing probability mapping on the spatio-temporal correlation matrix to generate a time density distribution matrix, where the row vector in the time density distribution matrix represents the time sensitivity distribution of a single user, and the column vector represents the comparison of the correlation strengths of different users within the same time slice; Performing logarithmic transformation on each element in the time density distribution matrix and then performing standard processing by column to obtain a standardized time gate matrix; Performing frequency domain expansion on the standardized time gate matrix and the non-linearly enhanced feature matrix respectively and then performing interleaved splicing to obtain a spliced matrix, and performing element-wise multiplication on the spliced matrix to enhance the coupling effect of time series - behavior to obtain a spatio-temporal fused feature tensor; Presetting categories, calculating the cosine similarity between the spatio-temporal fused feature tensor and the centroid of the preset categories to obtain a similarity result, evaluating the compactness of the feature distribution in the spatio-temporal fused feature tensor using the Gaussian kernel function to obtain a probability density, and fusing the similarity result and the probability density using regularization to obtain a preliminary classification result with confidence; Statistically analyze the frequency and duration distribution of each classification according to the user ID for the preliminary classification results with confidence, obtain a structured statistical table, and mine the category association rules in the preliminary classification results across confidence levels through the Apriori algorithm to obtain an association rule set; Establish a three-dimensional statistical histogram based on the structured statistical table and the association rule set according to the time slice division parameters to obtain the final classification statistical data.

4. The method for constructing a cultural and tourism virtual scene based on a three-dimensional model according to claim 3, wherein, The selection of the virtual cultural and tourism guide according to the basic scenic spot information and the tourism preference information specifically includes the following steps: Determine multiple relevant historical figures according to the basic scenic spot information; Obtain the three-dimensional human figures of multiple relevant historical figures; Select a virtual cultural and tourism guide from the three-dimensional human figures according to the tourism preference information; Among them, the tourism route planning according to the tourism preference information, generating a preferred tourism route, and screening and processing the complete three-dimensional scene data to construct a preferred virtual scene specifically includes the following steps: Perform location preference matching on the basic scenic spot information according to the tourism preference information to determine multiple preferred scenic spot locations; Import the scenic spot electronic map of the target tourist scenic spot; Plan a tourism route for multiple preferred scenic spot locations in the scenic spot electronic map to generate a preferred tourism route; Based on the preferred tourism route, perform partial screening and route sceneization on the complete three-dimensional scene data to construct a preferred virtual scene.

5. The method for constructing a cultural and tourism virtual scene based on a 3D model according to claim 4, wherein The performing location preference matching on the basic scenic spot information according to the tourism preference information to determine multiple preferred scenic spot locations specifically includes the following steps: Decompose the tourism preference vector into time-sensitive features and static preference features to obtain a structured preference feature group, perform hierarchical parsing on the scenic spot attribute tensor, and separate the basic attributes and dynamic popularity attributes to obtain a hierarchical scenic spot attribute set; Input the structured preference feature group into the LSTM network to extract time series pattern features, calculate the dynamic weight by combining the time series pattern features with the time decay factor of the scenic spot access popularity to obtain a time decay weight matrix; Perform cross-modal convolution on the structured preference feature group and the hierarchical scenic spot attribute set to obtain a cross-modal convolution result, and extract the real part feature component from the cross-modal convolution result; Extract the cultural semantic features in the hierarchical scenic spot attribute set, perform conjugate transpose transformation on the cultural semantic features, and perform non-linear interaction calculation on the transpose transformation and the user preference features to obtain a semantic association strength vector; Overlay the real part feature component and the semantic association strength vector bit by bit, and perform amplitude normalization processing on the overlay result to obtain a fusion feature vector; Perform non-linear activation and feature space mapping processing on the fusion feature vector in sequence to obtain an activated matching degree score; Take the median of the activated matching degree scores as the dynamic threshold, and sample the scenic spots with activated matching degree scores greater than the dynamic threshold according to the score gradient to obtain a preliminary screening scenic spot set; Calculate the geodesic distance between the scenic spots in the preliminary screening scenic spot set, generate an adjacency matrix, and perform path perception optimization on the adjacency matrix to obtain multiple preferred scenic spot locations.

6. The method for constructing a cultural and tourism virtual scene based on a three-dimensional model according to claim 5, wherein In the scenic spot electronic map, planning a travel route for multiple preferred scenic spot locations to generate a preferred travel route specifically includes the following steps: Performing Gaussian kernel density estimation on multiple preferred scenic spot locations to generate a probability distribution map of preferred locations; Separating the scenic spot electronic map, extracting the terrain layer, facility layer, and crowd flow layer features to obtain a hierarchical map feature set; Dividing the probability distribution map of preferred locations along the time axis into equally spaced sampling points, and binding the coordinates of the preferred location points to the time points to generate a spatio-temporal position coordinate sequence; Evaluating the position value of each preferred location point in the spatio-temporal position coordinate sequence to obtain the value of each preferred location point, calculating the value change gradient of adjacent preferred location points to obtain the path value change rate; Performing cross-channel interactive convolution operations on each layer in the hierarchical map feature set using deformable convolution kernels to obtain the dynamic convolution results of the fused map features; Performing a tensor product operation on the dynamic convolution results of the fused map features and the path value change rate to generate a spatio-temporal correlation field, and then performing progressive integration on the spatio-temporal correlation field along the time axis to generate a spatio-temporal integration field; Extracting the user's historical behavior vector from the travel preference information, using an LSTM network to extract the temporal behavior features in the user's historical behavior vector, calculating the similarity between the temporal behavior features and the user's historical behavior vector, and normalizing it to obtain the dynamic attention weight; Calculating the three-dimensional geodesic distance value of each preferred location point in the probability distribution map of preferred locations, introducing the terrain slope to constrain and correct the three-dimensional geodesic distance value to obtain the scene adaptation distance; Converting the spatio-temporal integration field into a path benefit term, and combining the dynamic attention weight and the scene distance as a cost term to establish a benefit-cost game balance equation; Abstracting the scenic spot electronic map into a weighted graph structure, where the nodes in the weighted graph structure represent the scenic spot locations, and the edge weights are the spatio-temporal integration field, dynamic attention weight, and scene distance results; Taking maximizing the total spatio-temporal integration value of the path coverage area, maximizing the weighted average value of the dynamic attention weight, and minimizing the total scene distance of the path as the optimization objectives, performing heuristic search on the scenic spot electronic map, and using the benefit-cost game balance equation to detect the benefit-cost conflict points in the path in real time during the search process; When the spatio-temporal integration value of a certain path drops by more than the threshold and the attention weight is higher than the average value, perform local backtracking to the nearest valid branch point, perform cost re-estimation, and re-expand the search along the optimal benefit direction. After the search is completed, generate an optimized path, perform smoothing processing and scene binding processing on the optimized path to obtain a preferred travel route.

7. The method for constructing a cultural and tourism virtual scene based on a three-dimensional model according to claim 6, wherein According to the preferred travel route and the virtual cultural and tourism guide, optimizing the virtual cultural and tourism tour of the preferred virtual scene to generate a cultural and tourism tour virtual scene specifically includes the following steps: Generating route tour information according to the preferred travel route; Performing voice processing on the route tour information according to the virtual cultural and tourism guide to generate route tour voice; In the preferred virtual scene, performing virtual cultural and tourism tour optimization of the virtual cultural and tourism guide and the route tour voice to generate a cultural and tourism tour virtual scene.

8. A cultural and tourism virtual scene construction system based on a three-dimensional model, the system is applied to the cultural and tourism virtual scene construction method based on a three-dimensional model according to any one of claims 1 to 7, characterized in that, The system includes a complete data acquisition unit, a preference recognition and recording unit, a virtual tour guide selection unit, a virtual scene construction unit, and a scene tour optimization unit, where: The complete data acquisition unit is used to obtain the basic information of the target tourist attraction, download the complete three-dimensional scene data of the target tourist attraction, and interact with the user for permissions to obtain the user's historical travel record data; The preference recognition and recording unit is used to perform preference recognition on the historical travel record data and record the user's travel preference information; The virtual tour guide selection unit is used to select a virtual cultural and tourism tour guide according to the basic information of the attraction and the travel preference information; The virtual scene construction unit is used to plan a travel route according to the travel preference information, generate a preferred travel route, and screen and process the complete three-dimensional scene data to construct a preferred virtual scene; The scene tour optimization unit is used to optimize the virtual cultural and tourism tour of the preferred virtual scene according to the preferred travel route and the virtual cultural and tourism tour guide, and generate a cultural and tourism tour virtual scene.

Images