Three-dimensional earth surface model data generation and rapid reading method, medium and system

Abstract

The invention provides a three-dimensional earth surface model data generation and fast reading method, medium and system, and belongs to the technical field of three-dimensional earth surface model data processing.The method comprises the steps that firstly, earth surface features are extracted through wavelet transform and singular value decomposition, then, self-adaptive partitioning is conducted based on spatial correlation analysis, and a three-dimensional earth surface model is obtained; and realizing accurate expression of the data by utilizing multi-scale grid division and morphological feature extraction. On this basis, data is organized by adopting hierarchical compression storage and a quadtree index structure, the importance of data blocks is determined through topological correlation analysis and viewpoint correlation evaluation, and finally rapid data reading based on a progressive loading strategy is realized. According to the method, the data processing and display efficiency is remarkably improved while the data integrity is ensured. The technical problems that in the prior art, rapid reading and efficient rendering of mass three-dimensional earth surface model data are difficult to achieve, and meanwhile the integrity and the display effect of the data are guaranteed are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of three-dimensional surface model data processing. Specifically, it relates to a method, medium and system for generating and quickly reading three-dimensional surface model data. Background Art

[0002] Three-dimensional surface model data has wide applications in fields such as geographic information systems, urban planning, and military simulations. Traditional methods for processing three-dimensional surface model data mainly include the regular grid method, the irregular triangular network method, and the hybrid grid method. The regular grid method divides the surface using a uniform grid size, which is simple to implement but difficult to adapt to complex terrains; the irregular triangular network method can perform adaptive meshing according to terrain features, but has a high computational complexity; the hybrid grid method combines the advantages of the two methods, but still has efficiency problems in data organization and management. With the development of remote sensing technology, the amount of three-dimensional surface model data has increased exponentially. Existing technologies face multiple challenges when processing massive data: First, data compression methods often use a uniform compression ratio, making it difficult to balance data volume and accuracy; second, traditional data organization structures (such as octrees, R-trees, etc.) have low retrieval efficiency when processing data at different scales; third, existing data loading strategies mostly use data blocks of a fixed size, making it difficult to adapt to dynamically changing display requirements. These problems often result in phenomena such as slow loading, stuttering display, or excessive memory occupancy in the actual application of three-dimensional surface models. Especially when processing high-resolution, large-scale surface data, how to achieve fast data reading and smooth display under limited computing and storage resources has become a core technical problem to be solved urgently.

[0003] In summary, there is a technical problem in the prior art of how to achieve fast reading and smooth display of massive three-dimensional surface model data under limited computing and storage resources. Summary of the Invention

[0004] In view of this, the present invention provides a method, medium and system for generating and quickly reading three-dimensional surface model data, which can solve the technical problem of how to achieve fast reading and smooth display of massive three-dimensional surface model data under limited computing and storage resources The present invention is implemented as follows: In the first aspect of the present invention, a method for generating and quickly reading three-dimensional surface model data includes the following steps: obtaining original surface elevation data and performing wavelet multi-scale decomposition to obtain multi-frequency surface feature data; performing singular value decomposition operation on the multi-frequency surface feature data to obtain a surface feature vector matrix; using the surface feature vector matrix to construct a surface spatial correlation matrix and performing regional division; generating multi-scale grid data based on the regional division result; constructing a quadtree structure to store the multi-scale grid data; calculating the topological correlation matrix between nodes in the quadtree structure; constructing a data reading evaluation function according to the topological correlation matrix; and implementing fast data reading by adopting an incremental loading strategy.

[0005] Among them, the steps of obtaining original surface elevation data and performing wavelet multi-scale decomposition specifically include: performing data integrity check on the original surface elevation data and removing outliers; performing 4-layer decomposition on the original surface elevation data by using Haar wavelet transform; performing normalization processing on the decomposed multi-frequency surface feature data; and evaluating the multi-frequency surface feature data by using a surface texture degree evaluation function.

[0006] Among them, the steps of performing singular value decomposition operation on multi-frequency surface feature data specifically include: constructing a covariance matrix; performing eigenvalue decomposition on the covariance matrix by using the Jacobi iterative method; selecting eigenvectors with a cumulative contribution rate reaching 90% to form a surface feature vector matrix; and evaluating the surface feature vector matrix by using a feature importance evaluation function.

[0007] Among them, the steps of constructing a surface spatial correlation matrix and performing regional division specifically include: constructing a surface spatial correlation matrix by using the Moran index method; calculating the spatial autocorrelation coefficient of a local area by using a sliding window method; optimizing the spatial autocorrelation coefficient by using a spatial association optimization function; performing regional division by using a spectral clustering algorithm; and constructing a surface partition weight matrix.

[0008] Among them, the steps of generating multi-scale grid data specifically include: performing grid meshing on each partition by using an adaptive quadtree division strategy; calculating the surface feature complexity of each grid cell; generating surface multi-scale grid data by means of recursive subdivision; and evaluating the surface multi-scale grid data by using a grid structure evaluation function.

[0009] Among them, the steps of constructing a quadtree structure to store multi-scale grid data specifically include: calculating the curvature matrix of grid nodes; extracting surface feature lines and feature points to form a surface morphology matrix; calculating the data compression factor; performing hierarchical compression storage on the surface multi-scale grid data; organizing the compressed surface multi-scale grid data using a quadtree structure; the steps of calculating the topological association matrix between nodes in the quadtree structure specifically include: calculating the spatial relationship between nodes; constructing a topological association matrix; calculating the connection strength using a topological evaluation function; optimizing the node connection method to ensure the rationality of the topological structure.

[0010] Among them, the steps of constructing a data reading evaluation function specifically include: calculating the importance value of each data block; constructing a view point association matrix; performing quantitative analysis on the view point association matrix using a view point importance function; organizing the data into blocks according to the view point association matrix.

[0011] Among them, the steps of implementing fast data reading using a progressive loading strategy specifically include: preferentially loading data blocks with high importance; monitoring the loading process using a loading performance function; preloading data blocks through a preloading mechanism; dynamically adjusting the loading batch size of data blocks.

[0012] The second aspect of the present invention provides a computer-readable storage medium, in which program instructions are stored. When the program instructions run on a computer, they are used to execute the above-mentioned method for generating and quickly reading three-dimensional surface model data.

[0013] The third aspect of the present invention provides a system for generating and quickly reading three-dimensional surface model data, including the above-mentioned computer-readable storage medium. The system can be any one of a computer, a server, and a single-chip microcomputer. The computer-readable storage medium is set inside the system, and a microprocessor for executing the program instructions stored in the computer-readable storage medium is set inside the system.

[0014] Compared with the prior art, a method, medium and system for generating and quickly reading three-dimensional surface model data provided by the present invention. The method for generating and quickly reading three-dimensional surface model data proposed by the present invention establishes a complete data processing and management framework through the organic combination of technologies such as multi-scale decomposition, feature extraction, adaptive partitioning, and progressive loading. This method can automatically adjust the data compression ratio and grid density according to the complexity of surface features, realizing the optimal utilization of data storage space. In terms of data organization, the present invention adopts a multi-level index structure based on a quadtree and introduces a spatial correlation analysis and view-point correlation degree evaluation mechanism, enabling the data retrieval and loading process to be dynamically adjusted according to actual display requirements. By constructing the topological relationship between data blocks and optimizing the data block size, this method significantly improves the data reading efficiency and rendering performance. At the same time, the introduction of the progressive loading strategy ensures the continuity and smoothness of the display process. The present invention successfully solves the core problems in the processing of massive three-dimensional surface model data. Through the accurate extraction and multi-level organization of data features, it realizes the optimal balance of data storage and reading efficiency. That is to say, the present invention solves the technical problem of how to quickly read and smoothly display massive three-dimensional surface model data under limited computing and storage resources, providing reliable technical support for the real-time display of large-scale surface data. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flowchart of the method of the present invention; Figure 2 is a distribution characteristic diagram of surface texture degrees of different landform types in Embodiment 2; Figure 3 is a cumulative contribution rate change curve diagram of eigenvalue decomposition in Embodiment 2. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0017] As Figure 1 shown, it is a flowchart of a method for generating and quickly reading three-dimensional surface model data provided by the first aspect of the present invention. This method includes the following steps: S01. Obtain the original surface elevation data, perform wavelet multi-scale decomposition on the original surface elevation data to obtain multi-frequency surface feature data, and evaluate the multi-frequency surface feature data using a surface texture degree evaluation function; S02. Perform singular value decomposition operation on the multi-frequency surface feature data to obtain a surface feature vector matrix, and evaluate the surface feature vector matrix using a feature importance evaluation function; S03. Construct a surface spatial correlation matrix using the surface feature vector matrix, calculate the surface spatial correlation coefficient, and normalize the surface spatial correlation coefficient using a spatial correlation optimization function; S04. Perform adaptive zoning on the surface area according to the surface spatial correlation coefficient to obtain a surface zoning weight matrix, and evaluate the surface zoning weight matrix using a regional distribution evaluation function; S05. Calculate the grid density of each zone using the surface zoning weight matrix to generate surface multi-scale grid data, and optimize the surface multi-scale grid data using a grid structure evaluation function; S06. Calculate the grid node curvature matrix according to the surface multi-scale grid data, construct a surface morphology matrix, and perform quantitative analysis on the surface morphology matrix using a morphology evaluation function; S07. Calculate the data compression factor using the surface morphology matrix, construct a compression optimization function to determine the compression threshold, and perform hierarchical compression storage on the surface multi-scale grid data; S08. Construct a quadtree structure, store the compressed surface multi-scale grid data in the quadtree structure, and evaluate the retrieval efficiency of the quadtree structure using an index performance function; S09. Calculate the topological correlation matrix between nodes in the quadtree structure, and calculate the connection strength of the topological correlation matrix using a topological evaluation function; S10. Construct a data reading evaluation function according to the topological correlation matrix, and the data reading evaluation function is used to calculate the importance value of data blocks; S11. Calculate a view point correlation matrix using the data reading evaluation function, and perform quantitative analysis on the view point correlation matrix using a view point importance function; S12. Organize data blocks according to the view point correlation matrix, and optimize the data block size using a storage evaluation function; S13. Read the data blocks using an incremental loading strategy, and dynamically adjust the incremental loading strategy using a loading performance function.

[0018] The specific implementation of the above steps is described in detail below. The specific implementation of step S01 is: first, the acquired original surface elevation data is checked for data integrity, and outliers are removed. The outlier judgment standard is that the deviation from the surrounding data exceeds 3 times the standard deviation. Then, the original surface elevation data is decomposed into multiple scales using Haar wavelet transform, and the number of decomposition layers is set to 4 layers. Each layer decomposes to obtain high-frequency components and low-frequency components of the corresponding scale. In the decomposition process, the completeness and orthogonality of the decomposition results are ensured by introducing orthogonal basis functions. Then, the decomposed multi-frequency surface feature data are normalized so that the data is distributed in the range of 0 to 1. Finally, the surface texture evaluation function is used to evaluate the processing results. The evaluation function measures the complexity of the surface texture by calculating the variance and entropy value of the pixel values ​​in the local area. The larger the evaluation value, the more complex the texture. The purpose of this step is to extract the multi-scale features of the surface and lay the foundation for subsequent processing.

[0019] The specific implementation method of step S02 is: construct a covariance matrix for multi-frequency surface feature data, and the size of the matrix is ​​determined by the data dimension. Then, the Jacobi iteration method is used to perform eigenvalue decomposition on the covariance matrix, and the iterative convergence threshold is set to 0.0001. According to the eigenvalues ​​obtained by the decomposition, they are arranged in order from large to small, and the eigenvectors with a cumulative contribution rate of 90% are selected to form a surface eigenvector matrix. The eigenvector matrix is ​​evaluated using a feature importance evaluation function, which comprehensively considers the eigenvalue size and the spatial distribution characteristics of the eigenvector, and the eigenvector with an evaluation value greater than 0.8 is retained. The role of this step is to reduce the data dimension and extract the main feature information.

[0020] The specific implementation method of step S03 is: based on the surface feature vector matrix, the Moran index method is used to construct the surface spatial correlation matrix, and the matrix elements represent the degree of correlation between different positions. The spatial autocorrelation coefficient of the local area is calculated by the sliding window method, and the window size is set to one tenth of the data range. The calculated surface spatial correlation coefficient is processed using the minimum and maximum normalization method to ensure that the value distribution is between 0 and 1. Then the normalized coefficient is optimized using the spatial correlation optimization function, which uses an exponential decay model to describe the distance attenuation characteristics of the spatial correlation intensity, and the attenuation coefficient is set to 0.3. The purpose of this step is to quantify the spatial distribution law of surface features.

[0021] The specific implementation of step S04 is as follows: According to the normalized surface spatial correlation coefficient, the spectral clustering algorithm is used for regional division, and the number of clusters is adaptively determined by the silhouette coefficient. Calculate the centroid position and boundary features for each cluster region, and construct a surface partition weight matrix, where the matrix elements represent the relative importance of each region. Use the regional distribution evaluation function to evaluate the partition result. This function considers two indicators: the balance of regional area and the smoothness of the boundary. A evaluation value greater than 0.7 indicates that the partition result is reasonable. Adjust the partition boundary through an iterative optimization method until the evaluation value meets the requirements. The role of this step is to achieve a reasonable division of the surface area.

[0022] The specific implementation of step S05 is as follows: Based on the surface partition weight matrix, an adaptive quadtree partitioning strategy is used to perform grid meshing on each partition. The initial grid size is set to 4 times the data resolution. Calculate the surface feature complexity for each grid cell, and subdivide the regions with a complexity higher than the threshold of 0.6. Generate surface multi-scale grid data through a recursive subdivision method, and the maximum number of subdivision layers does not exceed 6 layers. Use the grid structure evaluation function to evaluate the grid quality. This function comprehensively considers the regularity of the grid shape and the gradual change of the size. The grids with an evaluation value greater than 0.8 remain unchanged, otherwise, they are optimized and adjusted. The purpose of this step is to generate a multi-scale grid structure that adapts to the surface features.

[0023] The specific implementation of step S06 is as follows: For each node in the surface multi-scale grid data, use the discrete differential geometry method to calculate the principal curvature and Gaussian curvature, and construct a grid node curvature matrix. Then, based on the curvature information, extract the surface feature lines and feature points to form a surface morphology matrix. Use the morphology evaluation function to perform quantitative analysis on the feature extraction result. This function evaluates the significance of the surface morphology features by calculating the continuity of the feature lines and the distribution density of the feature points. A evaluation value greater than 0.75 indicates that the feature extraction is effective. The role of this step is to achieve an accurate expression of the surface morphology features.

[0024] The specific implementation of step S07 is as follows: Using the feature information in the surface morphology matrix, an adaptive compression algorithm is used to calculate the data compression factor, and the value range of the compression factor is from 0.1 to 0.9. Construct a compression optimization function to determine the optimal compression threshold. This function balances two indicators: the compression ratio and the reconstruction error. When the reconstruction error is less than 0.05 and the compression ratio is greater than 80%, the compression threshold is considered appropriate. Perform hierarchical compression storage on the surface multi-scale grid data according to the determined compression threshold. The important feature regions use a low compression ratio, and the flat regions use a high compression ratio. The purpose of this step is to achieve efficient compression storage of the data.

[0025] The specific implementation of step S08 is as follows: The compressed multi-scale grid data of the ground surface is organized using a quadtree structure, and the depth of the tree is determined by the grid level. During the construction process, first, the root node range is determined, and then the child nodes are recursively divided until the finest level is reached. For each node, the storage location index and adjacency relationship information are additionally stored. The retrieval efficiency of the quadtree structure is evaluated using an index performance function, which calculates the average retrieval time and memory occupancy. When the retrieval time is less than 10 milliseconds and the memory occupancy is less than the preset value, it indicates that the index structure is reasonable. The function of this step is to establish an efficient data management structure.

[0026] The specific implementation of step S09 is as follows: Based on the quadtree structure, the spatial relationship between nodes is calculated to construct a topological association matrix. The matrix elements represent the connection relationship and hierarchical relationship between nodes, and graph theory methods are used to analyze the connectivity and reachability of nodes. A topological evaluation function is used to calculate the connection strength, which considers the distance and hierarchical difference between nodes. Node pairs with a connection strength greater than 0.6 maintain the topological relationship. By optimizing the node connection method, the rationality of the topological structure is ensured. The purpose of this step is to establish the association relationship between data blocks.

[0027] The specific implementation of step S10 is as follows: A data reading evaluation function is constructed according to the topological association matrix, which comprehensively considers the size, level, and access frequency of data blocks. The importance value is calculated for each data block, and the weighted summation method is used for importance calculation, and the weight coefficients are dynamically adjusted through historical access records. When the importance value of a data block is greater than 0.7, it is marked as a priority loading object. The function of this step is to achieve a reasonable allocation of the data block loading priority.

[0028] The specific implementation of step S11 is as follows: The association degree between the viewpoint position and each data block is calculated through the data reading evaluation function to construct a viewpoint association matrix. The calculation of the association degree considers three factors: the viewpoint distance, the viewing direction, and the occlusion relationship. A viewpoint importance function is used to perform a quantitative analysis on the association matrix, which is based on the frustum culling principle and calculates the contribution degree of the data block to the visual effect. When the contribution degree is greater than 0.65, the corresponding data block is included in the rendering sequence. The purpose of this step is to optimize the loading order of data blocks.

[0029] The specific implementation of step S12 is as follows: The data is organized in blocks according to the viewpoint association matrix, and the size of each data block is dynamically determined according to the storage evaluation function. The storage evaluation function comprehensively considers the data access efficiency and storage space utilization rate. When the data block size is between 32 kilobytes and 256 kilobytes, the storage efficiency is optimal. An index table is established for the organized data blocks to record the position information and attribute information of the data blocks. The function of this step is to achieve a reasonable block storage of data.

[0030] The specific implementation of step S13 is as follows: The progressive loading strategy is adopted to achieve fast data reading. First, the data blocks with high importance are loaded, and then the secondary data blocks are gradually loaded. The loading performance function is used to monitor and adjust the loading process. This function records the loading time and display effect of the data blocks. When the loading time exceeds 50 milliseconds, the batch size is automatically adjusted. Through the preloading mechanism, the data blocks that may be needed are preloaded in advance, and the preloading quantity does not exceed 30% of the currently displayed data volume. The purpose of this step is to ensure the real-time and continuous data loading.

[0031] The following is a detailed description of the functions or calculation processes involved in the present invention.

[0032] 1. The evaluation function of the surface texture degree is specifically expressed as follows: ; In the formula, is the surface texture degree; is the element of the spatial weight matrix; is the elevation value of adjacent grid points; is the average elevation of the local area; is the number of grid points in the local area; is the standard deviation of the elevation in the local area; is the elevation entropy value of the local area; is the weight coefficient.

[0033] 2. The evaluation function of the feature importance is specifically expressed as follows: ; In the formula, is the feature importance; is the th eigenvalue; is the spatial distance of the feature vector; is the Gaussian kernel parameter; is the feature space correlation; is the feature aggregation degree; is the weight coefficient.

[0034] 3. The spatial association optimization function is specifically expressed as follows: ; In the formula, is the spatial correlation coefficient; is the spatial weight; is the spatial position at which the attribute value; is the average attribute; is the number of sampling points; is the spatial distance attenuation term; is the weight coefficient.

[0035] 4. The area distribution evaluation function is specifically expressed as follows: ; In the formula, is the area distribution uniformity; is the area of the th area; is the average area; is the total number of areas; is the smoothness of the area boundary; is the regularity of the area shape; is the weight coefficient.

[0036] 5. The grid structure evaluation function is specifically expressed as follows: ; In the formula, is the grid quality; is the area of the grid cell; is the length of the three sides of the grid cell; is the grid gradient; is the uniformity of the grid vertex distribution; is the weight coefficient.

[0037] 6. The morphology evaluation function is specifically expressed as follows: ; In the formula, is the morphological eigenvalue; is the principal curvature; is the feature point density; is the continuity of the feature line; is the weight coefficient.

[0038] 7. The compression optimization function is specifically expressed as follows: ; In the formula, is the compression efficiency; is the size of the original data; is the size of the compressed data; is the reconstruction error; is the data redundancy; is the weight coefficient.

[0039] 8. The view point importance function is specifically expressed as follows: ; In the formula, is the view point importance; is the viewing angle; is the view point distance; is the occlusion degree; is the visual attention value; is the weight coefficient.

[0040] 9. The index performance function is specifically expressed as follows: ; In the formula, is the index performance value; is the average retrieval time; is the total memory occupancy; is the used memory; is the tree height difference degree; is the weight coefficient.

[0041] 10. The topology evaluation function is specifically expressed as follows: ; In the formula, is the topology connection strength; is the node connection degree; is the distance between nodes; is the hierarchical connectivity; is the node degree distribution; is the weight coefficient.

[0042] 11. The data reading evaluation function is specifically expressed as follows: ; In the formula, is the data block reading priority; is the data block size; is the data block level; is the access frequency; is the weight coefficient; is the random perturbation term.

[0043] 12. The storage evaluation function is specifically expressed as follows: ; In the formula, is the storage efficiency; is the data block size; is the average access time; is the space utilization rate; is the data dispersion degree; is the weight coefficient.

[0044] 13. The loading performance function is specifically expressed as follows: ; In the formula, is the loading performance value; is the number of loaded data blocks; is the loading time; is the display quality; is the memory occupancy rate; is the weight coefficient.

[0045] The surface feature vector matrix is specifically represented as follows: ; In the formula, is the th eigenvalue of the th sample point; is the number of sample points; is the feature dimension.

[0046] The surface spatial correlation matrix is specifically represented as follows: ; In the formula, is the spatial correlation coefficient between position and position ; is the number of spatial positions.

[0047] The topological association matrix is specifically represented as follows: ; In the formula, is the topological association strength between node and node ; is the total number of nodes.

[0048] The parameter acquisition method for each function is as follows: The elements of the spatial weight matrix are calculated by the inverse distance weighting method. The specific steps are: calculating the distance between spatial point pairs, constructing a distance matrix, and normalizing; the elevation value is directly obtained from the original data; the eigenvalue is obtained by singular value decomposition; the Gaussian kernel parameter is determined by cross-validation, and the value range is from 0.1 to 1.0; the weight coefficients are all optimized and determined by the grid search method, and the value range is from 0 to 1; the curvature value is calculated by the discrete differential geometry method; the viewpoint parameters are calculated from the viewpoint position and direction, and the distance is normalized to the range of 0 to 1; the retrieval time is obtained by taking the average value through multiple tests; the memory occupancy is obtained by real-time monitoring through system calls; the node connectivity is calculated by graph theory algorithms; the data block parameters are obtained by real-time statistics; the weight coefficients are optimized and determined by machine learning methods.

[0049] The construction principles of these functions are as follows: The surface texture evaluation function combines spatial autocorrelation and information entropy, taking into account local elevation changes and complexity; the feature importance evaluation function combines eigenvalue magnitudes and spatial distribution characteristics, using an exponential decay model to describe spatial correlation; the spatial association optimization function is improved based on the Moran index, adding a distance decay term to better describe spatial relationships; the regional distribution evaluation function considers area uniformity and shape regularity, using the form of the square difference to measure the degree of non-uniformity; the grid structure evaluation function is based on the measurement of grid shape quality, combining the area and side length ratio relationship; the morphology evaluation function integrates local curvature features and global distribution features to comprehensively describe the surface morphology; the compression optimization function balances the compression ratio and reconstruction quality, introducing redundancy to evaluate data characteristics; the viewpoint importance function uses a distance decay model, comprehensively considering the line-of-sight direction and occlusion relationship; the index performance function comprehensively considers retrieval speed and memory efficiency, introducing tree structure balance evaluation; the topology evaluation function is based on graph theory methods, considering node connection density and hierarchical relationships; the data reading evaluation function uses a weighted summation model, introducing random perturbations to increase adaptability; the storage evaluation function balances storage space and access efficiency, considering data distribution characteristics; the loading performance function combines loading speed and display effect to optimize the user experience.

[0050] The derivation and establishment process of each function will be described in detail below. 1. The derivation and establishment process of the surface texture evaluation function: First, using the principle of spatial autocorrelation, a basic model is constructed, which only considers the spatial distribution characteristics of elevation values; then, a standard deviation term is introduced to describe the data dispersion degree, and an entropy term is introduced to describe the data complexity, resulting in an improved model ; the weight coefficient is optimized through the least squares method. This function can simultaneously reflect the local change characteristics and overall complexity of the surface.

[0051] 2. The derivation and establishment process of the feature importance evaluation function: Based on the eigenvalue decomposition theory, the initial model is , which only considers the eigenvalue magnitudes; considering the spatial distribution characteristics of features, a Gaussian kernel function is introduced, resulting in ; further, feature correlation and aggregation degree are introduced, resulting in the final model . The parameter is determined through cross-validation, and is optimized through grid search.

[0052] 3. Process of Deriving and Establishing the Spatial Association Optimization Function: Starting from the Moran's Index, a basic model is constructed ; A distance decay term is introduced Considering the influence of spatial position, an optimized model is obtained ; Among them It is calculated using an exponential decay function Determined by maximum likelihood estimation

[0053] 4. Process of Deriving and Establishing the Regional Distribution Evaluation Function: Initially, area difference measurement is adopted ; Boundary smoothness is introduced And shape regularity to obtain a complete model ; Among them Calculated through boundary curvature Calculated through circularity index, and the weight coefficient is optimized by simulated annealing algorithm

[0054] 5. Process of Deriving and Establishing the Grid Structure Evaluation Function: Based on the grid quality evaluation theory, the initial model is ; Grid gradient is introduced And vertex distribution uniformity to obtain an optimized model ; Among them Calculated through the grid side length ratio Calculated through the variance of Thiessen polygon area

[0055] 6. Process of Deriving and Establishing the Morphology Evaluation Function: Starting from differential geometry theory, the initial model is ; Feature point density is introduced And feature line continuity to obtain a complete model ; Among them, the principal curvature is calculated by least squares fitting Obtained through kernel density estimation Evaluated through curve fitting error

[0056] 7. Process of Deriving and Establishing the Compression Optimization Function: The basic model is expressed by the compression ratio ; Reconstruction error is introduced And data redundancy to obtain an optimized model ; Among them Calculated through root mean square error Evaluated through information entropy, and the weight coefficient is optimized by genetic algorithm

[0057] 8. Process of deriving and establishing the viewpoint importance function: Initially adopt the line-of-sight attenuation model ; Introduce the occlusion degree and the visual attention value , to obtain the complete model ; Among them is calculated through ray tracing, is obtained through saliency detection.

[0058] 9. Process of deriving and establishing the index performance function: The basic model considers the retrieval time ; Introduce the memory utilization rate and the balance of the tree structure to obtain the optimized model ; The parameters are obtained through performance testing and adjusted dynamically.

[0059] 10. Process of deriving and establishing the topology evaluation function: Starting from graph theory, the initial model is ; Introduce the hierarchical connectivity and the node degree distribution to obtain the complete model ; The parameters are determined through graph structure analysis.

[0060] The second aspect of the present invention provides a computer-readable storage medium, in which program instructions are stored. When the program instructions run on a computer, they are used to execute the above method for generating and quickly reading three-dimensional surface model data.

[0061] The third aspect of the present invention provides a system for generating and quickly reading three-dimensional surface model data, including the above computer-readable storage medium. The system is any one of a computer, a server, and a single-chip microcomputer. The computer-readable storage medium is arranged inside the system, and a microprocessor for executing the program instructions stored in the computer-readable storage medium is arranged inside the system.

[0062] Specifically, the principle of the present invention is as follows: The technical solution of the present invention is based on the principles of data feature analysis and multi-level organization. First, multi-frequency features of the ground surface are extracted through wavelet multi-scale decomposition. This method can effectively separate the ground surface detail information at different scales, providing a theoretical basis for subsequent feature extraction and data compression. The introduction of singular value decomposition further reduces the data dimension, retaining the main feature information while reducing redundant data. Spatial correlation analysis is another important theoretical basis of the present invention. By constructing a spatial correlation matrix and calculating the correlation coefficient, the distribution law of ground surface features can be accurately described, providing a basis for regional division. The implementation of adaptive partitioning processing depends on the spectral clustering algorithm, which can automatically determine the optimal number of partitions and boundary positions according to data features. In terms of data organization, the present invention adopts the quadtree structure because of its good spatial partitioning characteristics and hierarchical management capabilities. By establishing a topological correlation matrix between nodes, the spatial relationship and hierarchical relationship between data blocks can be effectively described, providing support for rapid data retrieval. The design of the progressive loading strategy is based on the view point correlation analysis. By calculating the contribution degree of data blocks to the display effect, a reasonable allocation of loading priorities is achieved.

[0063] The following provides a specific Embodiment 1 of the present invention. The specific implementation of each step in this Embodiment 1 is described in detail as follows.

[0064] The specific implementation of step S01 is as follows: First, perform a data integrity check on the acquired original ground surface elevation data. Identify outliers by constructing a local elevation statistical model, calculate the elevation difference value between each data point and its neighboring points. When the difference value exceeds 3 times the local standard deviation, it is determined as an outlier. Then, perform multi-scale decomposition on the original ground surface elevation data using the Haar wavelet transform. The decomposition process is recursive. Each decomposition can obtain a low-frequency component and three high-frequency components, corresponding to the detail features in the horizontal, vertical, and diagonal directions respectively. For the decomposed multi-frequency ground surface feature data, use the ground surface texture evaluation function for evaluation. The expression of this function is , where the spatial weight is calculated by inverse distance weighting , is the distance between spatial point pairs, is the distance attenuation exponent, and its value range is from 2 to 3; the local elevation average is calculated by a sliding window, and the window size is set to one-tenth of the data range; the standard deviation term adopts an improved local variance calculation method, introducing a directional weight to enhance the ability to extract edge features; the entropy term is calculated by the normalized elevation histogram, reflecting the degree of uncertainty of the data. The weight coefficient is obtained by fitting a large amount of experimental data using the least squares method. Generally The value range is from 0.3 to 0.5, and the value range is from 0.2 to 0.4. The main purpose of this step is to extract the multi-scale features of the surface, providing a basis for subsequent feature analysis and regional division.

[0065] The specific implementation of step S02 is: constructing a covariance matrix for the multi-frequency surface feature data, and the calculation formula for the matrix elements is , where represents the th eigenvalue of the th sample, represents the mean of the th feature, is the number of samples. Using the Jacobi iterative method to perform eigenvalue decomposition on the covariance matrix, the non-diagonal elements are eliminated through the rotation matrix during the iteration process, and the iteration termination condition is that the sum of the squares of the non-diagonal elements is less than the set threshold of 0.0001. The eigenvectors obtained from the eigenvalue decomposition are sorted according to the corresponding eigenvalue magnitudes, and the eigenvector matrix of the surface features is formed by selecting the eigenvectors with a cumulative contribution rate reaching 90%. Introducing the feature importance evaluation function to evaluate the importance of the eigenvectors, where is the th eigenvalue, the spatial distance is calculated through the Euclidean distance of the eigenvectors in the feature space, the Gaussian kernel parameter is determined through cross-validation, with a value range of 0.1 to 1.0, the feature space correlation is calculated through the Pearson correlation coefficient, and the feature aggregation degree is obtained through local density estimation. The weight coefficient is optimized and determined through the grid search method, with a value range of 0 to 1. The main purpose of this step is to reduce the data dimension and extract the main feature information.

[0066] The specific implementation of step S03 is: based on the surface feature vector matrix, using the improved Moran's index method to construct the surface spatial correlation matrix, and the expression of the spatial association optimization function is , where the spatial weight is calculated using an adaptive kernel function, and the kernel bandwidth is determined through cross-validation. The distance decay term adopts the exponential decay model , where is the spatial distance, is the feature correlation length, determined through variogram analysis. The weight coefficient Calculated by maximum likelihood estimation, the general value range is from 0.2 to 0.4. The spatial autocorrelation coefficient of the local area is calculated by the sliding window method. The window size is set to one-tenth of the data range, and the window overlap rate is set to 50%. The calculated spatial autocorrelation coefficient is processed by the min-max normalization method to ensure that the values are distributed between 0 and 1. The main function of this step is to quantify the spatial distribution law of surface features.

[0067] The specific implementation of step S04 is as follows: According to the normalized spatial autocorrelation coefficient, the spectral clustering algorithm is used for region division. First, a similarity matrix is constructed , and the calculation formula of the matrix elements is , where is the Gaussian kernel parameter. Then, the Laplacian matrix is calculated, where is the degree matrix, and the diagonal element . Solve the generalized eigenvalue problem , select the first eigenvectors to construct a new feature space. Through the k-means clustering algorithm is used for clustering in the new feature space, and the number of clusters is adaptively determined by the silhouette coefficient. The expression of the regional distribution evaluation function is , where the regional area is obtained by the polygon area calculation formula, the boundary smoothness is calculated by the boundary curvature, and the shape regularity is calculated by the circularity index. The weight coefficient is optimized and determined by the simulated annealing algorithm, and the value range is from 0 to 1. The main purpose of this step is to achieve a reasonable division of the surface area.

[0068] The specific implementation of step S05 is as follows: Based on the surface partition weight matrix, an adaptive quadtree partitioning strategy is used to perform grid meshing on each partition. The initial grid size is set to 4 times the data resolution. The surface feature complexity is calculated for each grid cell, and the complexity calculation uses the grid structure evaluation function , where is the grid cell area, are the three side lengths of the grid cell, and the grid gradient is calculated by the side length ratio , are the maximum and minimum side lengths respectively, and the vertex distribution uniformity is calculated by the variance of the Thiessen polygon area. When the complexity is higher than the threshold of 0.6, the grid cell is subdivided, and the subdivision process is recursive, with a maximum subdivision level of 6 layers. The weight coefficient Determined by genetic algorithm optimization, with a value range from 0 to 1. The main purpose of this step is to generate a multi-scale grid structure adapted to surface features.

[0069] The specific implementation of step S06 is as follows: for each node in the surface multi-scale grid data, the principal curvature and Gaussian curvature are calculated using discrete differential geometry methods. First, a local quadratic surface fitting model is constructed , and the coefficients are solved using the least squares method . The Gaussian curvature is calculated based on the fitting coefficients and the mean curvature , and then the principal curvature is obtained. The expression of the morphological evaluation function is , where the feature point density is calculated by kernel density estimation , is the bandwidth parameter, is the kernel function; the feature line continuity is evaluated by the curve fitting error, and the continuity parameter of the feature line is calculated using the spline interpolation method. The weight coefficient is determined by cross-validation, with a value range from 0.2 to 0.5. The main role of this step is to achieve an accurate expression of surface morphological features.

[0070] The specific implementation of step S07 is as follows: using the feature information in the surface morphology matrix, an adaptive compression algorithm is used to calculate the data compression factor. The expression of the compression optimization function is , where are the storage sizes of the original data and the compressed data respectively, and the reconstruction error is calculated by the root mean square error , are the original value and the reconstructed value respectively, and the data redundancy is calculated by the information entropy , is the probability of the data value occurring. The value range of the compression factor is from 0.1 to 0.9, and the weight coefficient is determined by genetic algorithm optimization. The main purpose of this step is to achieve efficient compressed storage of data.

[0071] The specific implementation of step S08 is as follows: the compressed surface multi-scale grid data is organized using a quadtree structure. The expression of the index performance function is , where the retrieval time is averaged by multiple tests, the memory occupancy rate is monitored in real time through system calls, and the tree height difference is calculated by the hierarchical distribution variance , is the node hierarchical depth. The weight coefficient Optimized by the gradient descent method, the value range is from 0.1 to 0.3. The function of this step is to establish an efficient data management structure.

[0072] The specific implementation of step S09 is: calculating the spatial relationship between nodes based on the quadtree structure and constructing a topological association matrix. The expression of the topological evaluation function is , where the node connectivity is calculated by the graph theory algorithm, and the distance between nodes adopts the Euclidean distance, and the hierarchical connectivity is evaluated by the depth-first search algorithm, and the node degree distribution is calculated by the degree distribution entropy , is the proportion of nodes with degree . The weight coefficient is determined by the grid search method. The main purpose of this step is to establish the association relationship between data blocks.

[0073] The specific implementation of step S10 is: constructing a data reading evaluation function according to the topological association matrix, and the function expression is , where the data block size is obtained by byte counting, and the data block level is the depth of the node in the tree, and the access frequency is counted by the historical access records. The random perturbation term obeys the normal distribution with a mean of 0 and a standard deviation of 0.1. The weight coefficient is optimized and determined by the machine learning method. The function of this step is to realize the reasonable allocation of the data block loading priority.

[0074] The specific implementation of step S11 is: calculating the association degree between the viewpoint position and each data block by using the data reading evaluation function and constructing a viewpoint association matrix. First, determine the viewpoint position coordinates and the line-of-sight direction vector , and calculate the distance and angle between the viewpoint and the center point of the data block. The expression of the viewpoint importance function is , where the line-of-sight angle is calculated by the vector dot product , is the center coordinate of the data block; the viewing distance is the Euclidean distance from the viewpoint to the center of the data block; the occlusion degree is calculated by the ray tracing algorithm, considering the occlusion relationship between data blocks; the visual attention value is calculated by the saliency detection algorithm, adopting the multi-scale contrast feature extraction method. The weight coefficient Optimized by reinforcement learning method, with a value range of 0.2 to 0.4. The main purpose of this step is to optimize the loading order of data blocks.

[0075] The specific implementation of step S12 is: organizing data into blocks according to the viewpoint association matrix. The storage evaluation function expression is , where the data block size has a value range of 32 to 256 kilobytes, the average access time is obtained through performance testing, and the space utilization rate is calculated by the ratio of the actual storage space to the theoretical storage space. The data dispersion is calculated by the distribution entropy of data blocks on the storage medium , is the probability of the data block in the th storage unit. The weight coefficient is determined by the Bayesian optimization method, with a value range of 0.1 to 0.3. An index table is established for the organized data blocks, recording information such as data block identifiers, spatial ranges, compression ratios, storage locations, etc. The main function of this step is to achieve reasonable block storage of data.

[0076] The specific implementation of step S13 is: adopting an incremental loading strategy to achieve fast data reading. The loading performance function expression is , where the number of loaded data blocks is counted in real time by a counter, the loading time is measured by a high-precision timer, and the display quality is calculated by an image quality evaluation algorithm, including indicators such as resolution integrity and texture clarity. The memory occupancy rate is obtained through the system monitoring interface. When the loading time exceeds 50 milliseconds, it is optimized by adjusting the batch size. The adjustment of the batch size adopts an adaptive control algorithm and changes dynamically according to the system load and network status. The preloading mechanism is designed using a prediction model, which predicts the data blocks that may be needed at the next moment based on user interaction behavior and viewpoint change trends. The preloading quantity does not exceed 30% of the current displayed data volume. The weight coefficient is optimized by an online learning method and dynamically adjusted according to the system operation state. The main purpose of this step is to ensure the real-time and continuous data loading.

[0077] The optimization algorithms involved in the above steps, including genetic algorithms, simulated annealing algorithms, gradient descent methods, etc., have their parameter settings as follows: the population size of the genetic algorithm is set to 100, the crossover probability is 0.8, the mutation probability is 0.1, and the maximum number of iterations is 1000; the initial temperature of the simulated annealing algorithm is set to 100, the cooling coefficient is 0.95, and the termination temperature is 0.01; the learning rate of the gradient descent method is set to 0.01, and the convergence threshold is 0.0001. These parameter values are the optimal configurations obtained through a large number of experimental verifications and can be appropriately adjusted according to specific application scenarios.

[0078] In the optimization process of the weight coefficients in all evaluation functions, the cross-validation method is adopted. The dataset is divided into a training set and a validation set with a ratio of 8:2, and the optimal weight value is determined by minimizing the error on the validation set. The design of the evaluation function takes into account the balance between the diversity of data features and computational efficiency. By introducing multiple evaluation indicators and weight coefficients, a comprehensive evaluation of different features is achieved.

[0079] To better understand and implement the present invention, the following provides Example 2 of a specific application scenario of the present invention: A certain research institute uses the method for generating and quickly reading three-dimensional surface model data of the present invention to process high-resolution terrain data. The total area of the research area is approximately 28,600 square kilometers, and the original data comes from multi-temporal airborne lidar scanning, containing approximately 28 billion point cloud data points. The research team conducts data processing work based on the method of the present invention, and the specific implementation process is as follows.

[0080] The first stage: Data preprocessing and feature extraction. First, the quality of the original point cloud data is checked and outliers are identified. The local statistical feature analysis method is used to calculate the local elevation standard deviation of each point. The detection window size is set to 200 meters × 200 meters, and the outlier determination threshold is 3.5 times the standard deviation. The detection results are shown in Table 1 below: Table 1 Detection Results Table

[0081] The improved Haar wavelet transform is used for 4-layer multi-scale decomposition. The feature data and its statistical features obtained from each layer of decomposition are shown in Table 2 below: Table 2 Feature Data Table

[0082] The surface texture degree is calculated using an improved spatial autocorrelation model, and the parameters are set as follows: the spatial weight decay exponent , the texture degree weight coefficient , the entropy value weight coefficient . The calculation results show that the surface texture characteristics of the research area have obvious spatial variability, and the statistical results are shown in Table 3 below: Table 3 Statistical Data Table

[0083] Figure 2 Shows the distribution characteristics of surface texture degrees of different geomorphic types, including the mean value, standard deviation, and value range.

[0084] The second stage: Feature decomposition and data dimensionality reduction. Construct a covariance matrix for multi-frequency surface feature data and perform eigenvalue decomposition. Set the Gaussian kernel parameter , the correlation weight of the feature space , the aggregation weight of the feature . The evaluation results of feature importance are shown in the following table: Table 4 Feature evaluation table

[0085] Figure 3 Shows the cumulative contribution rate change curve of eigenvalue decomposition, reflecting the basis for feature selection.

[0086] The third stage: Spatial correlation analysis and regional division. Calculate the spatial correlation features based on the eigenvector matrix, and use the improved Moran's index method to construct a spatial correlation optimization function. Set the parameters as follows: distance decay coefficient , spatial correlation threshold 0.65. The spatial correlation analysis results calculated are shown in Table 5 below: Table 5 Spatial correlation analysis table

[0087] Based on the spatial correlation analysis results, use the spectral clustering algorithm for regional division. Set the parameters as follows: Gaussian kernel bandwidth 0.5, clustering number adaptive threshold 0.85. The evaluation of the regional division results is shown in Table 6 below: Table 6 Regional division result table

[0088] The fourth stage: Grid generation and optimization. Adopt the adaptive quadtree dissection strategy, set the initial grid size to 25 meters, and perform multi-level subdivision according to the surface complexity. The grid quality evaluation results are shown in Table 7 below: Table 7 Network quality evaluation table

[0089] The fifth stage: Data compression and index construction. Adopt the morphological feature adaptive compression algorithm, and set the parameters of the compression optimization function: reconstruction error weight , redundancy weight . The evaluation of the compression effect is shown in Table 8 below: Table 8 Compression effect evaluation table

[0090] Sixth stage: Performance testing and effect comparison. Traditional methods generally include: Method 1: The method based on regular grids divides the ground surface by grids of a fixed size, stores data of different resolutions using a pyramid hierarchical structure, compresses the data using a general compression algorithm, and gradually loads the data of the display area in a block manner according to the viewpoint position and display ratio. Method 2: The method based on irregular triangular meshes first extracts terrain feature points to construct a Delaunay triangular mesh, generates a multi-level model through an edge-collapse mesh simplification algorithm, organizes the data using a tree structure, and dynamically loads triangular mesh data of different precisions according to the viewpoint parameters. Method 3: The method based on hybrid grids processes the ground surface by dividing it into regions and then using regular grids and irregular triangular meshes respectively, manages the data using a partition compression strategy and a multi-way tree structure, and combines a preloading mechanism to achieve dynamic reading and display of the data. Next, performance testing is carried out in a standard test environment (CPU: Intel Xeon 2.8 GHz, memory: 256 GB, video memory: 32 GB), comparing Traditional Method 1 and the method of the present invention. The results are shown in Table 9 as follows: Table 9 Performance testing data table

[0091] The main improvements of the method of the present invention compared with traditional technologies are as follows: 1. Through multi-scale decomposition and feature extraction, the data processing efficiency is increased by 3.6 times, and a high ground surface feature fidelity is maintained.

[0092] 2. By adopting an adaptive compression algorithm, the storage space is saved by 82.9% while ensuring that the reconstruction accuracy is greater than 0.98.

[0093] 3. Based on the improved quadtree index structure, the retrieval response time is reduced from 85 milliseconds to 12 milliseconds.

[0094] 4. Through the view-dependent progressive loading strategy, the display frame rate is increased to 42.6 frames per second, achieving a smooth interaction experience.

[0095] 5. The overall performance of the system is increased by more than 80% on average compared with the existing technology, meeting the real-time processing requirements of large-scale three-dimensional ground surface data.

[0096] It should be noted that the detailed explanations of the variables involved in the present invention are shown in Table 10 below.

[0097] Table 10 Variable explanation table

[0098] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for generating and quickly reading three-dimensional surface model data, characterized in that: The following steps are involved: The original surface elevation data is obtained and subjected to wavelet multi-scale decomposition to obtain multi-frequency surface feature data; a singular value decomposition operation is performed on the multi-frequency surface feature data to obtain a surface feature vector matrix; a surface spatial correlation matrix is ​​constructed using the surface feature vector matrix and regional division is performed; multi-scale grid data is generated based on the regional division result; a quadtree structure is constructed to store the multi-scale grid data; a topological association matrix between nodes in the quadtree structure is calculated; a data reading evaluation function is constructed according to the topological association matrix; and a progressive loading strategy is adopted to realize fast data reading.

2. The method for generating and quickly reading three-dimensional surface model data according to claim 1, characterized in that: The steps of obtaining original surface elevation data and performing wavelet multi-scale decomposition specifically include: performing data integrity check on the original surface elevation data and removing outliers; performing 4-layer decomposition on the original surface elevation data using Haar wavelet transform; normalizing the decomposed multi-frequency surface feature data; and evaluating the multi-frequency surface feature data using a surface texture evaluation function.

3. The method for generating and quickly reading three-dimensional surface model data according to claim 1, characterized in that: The steps of performing singular value decomposition operation on multi-frequency surface feature data specifically include: constructing a covariance matrix; performing eigenvalue decomposition on the covariance matrix using the Jacobi iteration method; selecting eigenvectors with a cumulative contribution rate of 90 percent to form a surface feature vector matrix; and evaluating the surface feature vector matrix using a feature importance evaluation function.

4. The method for generating and quickly reading three-dimensional surface model data according to claim 1, characterized in that: The steps of constructing a surface spatial correlation matrix and performing regional division specifically include: constructing a surface spatial correlation matrix using the Moran index method; calculating the spatial autocorrelation coefficient of a local area using a sliding window method; optimizing the spatial autocorrelation coefficient using a spatial association optimization function; performing regional division using a spectral clustering algorithm; and constructing a surface partition weight matrix.

5. The method for generating and quickly reading three-dimensional surface model data according to claim 1, characterized in that: The steps of generating multi-scale grid data specifically include: using an adaptive quadtree partitioning strategy to grid each partition; calculating the surface feature complexity of each grid unit; generating surface multi-scale grid data by recursive subdivision; and evaluating the surface multi-scale grid data using a grid structure evaluation function.

6. The method for generating and quickly reading three-dimensional surface model data according to claim 1, characterized in that: The steps of constructing a quadtree structure to store multi-scale grid data specifically include: calculating a grid node curvature matrix; extracting surface feature lines and feature points to form a surface morphology matrix; calculating a data compression factor; compressing and storing the surface multi-scale grid data in layers; organizing the compressed surface multi-scale grid data using a quadtree structure; the steps of calculating a topological association matrix between nodes in the quadtree structure specifically include: calculating the spatial relationship between nodes; constructing a topological association matrix; calculating the connection strength using a topological evaluation function; and optimizing the node connection method to ensure the rationality of the topological structure.

7. The method for generating and quickly reading three-dimensional surface model data according to claim 1, characterized in that: The steps of constructing a data reading evaluation function specifically include: calculating the importance value of each data block; constructing a viewpoint association matrix; performing quantitative analysis on the viewpoint association matrix using a viewpoint importance function; and organizing data into blocks according to the viewpoint association matrix.

8. The method for generating and quickly reading three-dimensional surface model data according to claim 1, characterized in that: The steps of using progressive loading strategy to realize fast data reading include: loading high-importance data blocks first; monitoring the loading process by using loading performance function; loading data blocks in advance by preloading mechanism; and dynamically adjusting the loading batch size of data blocks.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, and when the program instructions are run in a computer, they are used to execute the three-dimensional surface model data generation and rapid reading method according to any one of claims 1 to 8.

10. A three-dimensional surface model data generation and rapid reading system, characterized in that: The system comprises the computer-readable storage medium as claimed in claim 9, wherein the system is any one of a computer, a server, and a single-chip microcomputer, the computer-readable storage medium is arranged in the system, and the system is provided with a microprocessor for executing program instructions stored in the computer-readable storage medium.

Images