City-level geographic scene incremental data governance method and system

Abstract

The invention discloses a city-level geographic scene incremental data governance method, which belongs to the technical field of geographic scene updating and comprises the following steps of S1, arranging existing data; s2, dividing existing data into a plurality of basic geographic grid units which are continuous in space and mutually overlapped according to urban local coordinates; s3, the quadtree grid subdivision structure is divided into geographic grid models in a refined mode; s4, during incremental updating, collecting incremental to-be-updated data, performing block processing and channel processing, quickly positioning a to-be-updated position based on a spatial index structure, extracting boundary information, analyzing a topological relation between the boundary information and existing data, and performing splicing, dodging processing, color correction, feathering processing and deformation processing with the existing data based on the topological relation to complete updating; the invention further discloses a city-level geographic scene incremental data governance system. According to the method, the problems that in the prior art, indexes are missed in an updating area, color deviation exists in updating data, and updating patches are staggered can be well solved.

Description

Technical Field

[0001] The present invention belongs to the field of geographic scene updating technology, and specifically relates to a city-level geographic scene incremental data management method and system. Background Art

[0002] Urban-level realistic geographic scenes refer to the high-precision and comprehensive digital reproduction of urban spaces using technical means such as high-resolution imaging and three-dimensional modeling.

[0003] With the continuous advancement of urban construction, real-life geographic scenes need to be updated regularly to maintain their timeliness and accuracy.

[0004] However, the current updates of city-level realistic geographic scenes involve problems such as missing indexes in the update area, color deviation in the update data, and misaligned patches in the update.

[0005] In view of this, a city-level geographic scenario incremental data governance method and system are designed to solve the above problems. Summary of the Invention

[0006] In order to solve the problems raised in the above background technology, the present invention provides a method and system for managing incremental data of urban-level geographic scenes, which can effectively solve the problems existing in the prior art, such as missing indexes in updated areas, color deviations in updated data, and misaligned patches in updated patches, thereby improving the updating efficiency and effect of urban-level realistic geographic scenes.

[0007] Another object of the present invention is to provide a city-level geographic scene incremental data governance system.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method for managing incremental data of city-level geographic scenarios, comprising the following steps:

[0009] S1: Organize existing city-level real-world geographic scene data;

[0010] S2: Divide the existing multi-scale city-level real-life geographic scene data into multiple spatially continuous and overlapping basic geographic grid units according to the city's local coordinates;

[0011] S3: multiple basic geographic grid cells are used as initial grid cells and divided into geographic grid models using a quadtree grid subdivision structure;

[0012] S3: Geocode the geographic grid model and build an incremental city-level real-world geographic scene data spatial index structure;

[0013] S4: During incremental updates, collect incremental city-level realistic geographic scene data to be updated, perform block denoising and enhancement processing, and adjust brightness by channel. Based on the constructed spatial index structure of incremental city-level realistic geographic scene data, quickly locate its position to be updated, extract boundary information, analyze the topological relationship between the boundary information and the existing city-level realistic geographic scene data, and splice it with the existing city-level realistic geographic scene data based on the topological relationship. Perform light balancing on the spliced ​​image based on the multi-scale retinex algorithm, perform color correction on the spliced ​​image after light balancing, and feather the spliced ​​edges of the color-corrected spliced ​​image. Perform grid deformation based on the Laplace coordinate grid deformation algorithm to solve the problem of spliced ​​image edge mismatch. Perform deformation processing on the raised model facets to make the spliced ​​image edges match, and complete the update.

[0014] Furthermore, in step S3, the geographic coding includes grid row and column numbers, grid subdivision levels and grid subdivision units. The grid row and column numbers are expressed as the coordinates of the southwest corner of the grid based on the city local coordinate system. The grid subdivision level is expressed as the level of the grid unit in the basic geographic grid unit. The grid subdivision unit coding rule is expressed by encoding the four grid cells in the order of upper left, lower left, lower right and upper right as 0, 1, 2 and 3 respectively.

[0015] Furthermore, in step S4, the specific steps of performing light uniforming processing on the stitched image based on the multi-scale retinex algorithm include:

[0016] S401: Set the incident light image to L(x,y), the reflected light image to R(x,y), and the original image to I'(x,y). The expression of the single-scale retinex algorithm is:

[0017] I'(x,y)=L(x,y)·R(x,y)

[0018] To further obtain an incident light image with uniform brightness, the above formula can be converted to the logarithmic domain, and the expression is:

[0019] logR(x,y)=logI'(x,y)-logL(x,y)

[0020] make:

[0021]

[0022] Assume that the incident light image L(x,y) is obtained by convolving the Gaussian function F(x,y) with the original image I'(x,y). Substituting the above formula into the solution, we can get:

[0023]

[0024] Where: F(x,y) represents the Gaussian function model, Expressed as a scale parameter, λ must satisfy:

[0025] ∫∫F(x,y)dxdy=1;

[0026] S402: When uniform lighting, first select the three scale parameters of large, medium and small. Then perform single-scale operations separately, and then perform weighted average calculation on the results to obtain the stitched image after uniform light.

[0027] Furthermore, in step S4, the specific steps of performing color correction on the stitched image after the light uniforming process include:

[0028] S403: Divide the uniformly lighted stitched image into nine evenly spaced areas according to a preset rule;

[0029] S404: Identify features with significant characteristics in each area;

[0030] S405: Calculate the statistics of each feature object in different color channels;

[0031] S406: Select the feature object with the largest statistical value in each area as a calibration feature point;

[0032] S407: Recording the grayscale value of the selected calibration feature point;

[0033] S408: Using the least squares method, the grayscale correction value of the image is fitted according to the grayscale values ​​of the calibrated feature points;

[0034] S409: performing grayscale correction on each pixel of the image according to the grayscale correction value of the fitted image.

[0035] Furthermore, in step S4, the grid deformation algorithm based on Laplace coordinates is used to perform grid deformation on the spliced ​​image to solve the edge mismatch problem. The specific steps of deforming the raised model surface include:

[0036] S410: Given a mesh vertex v, its Laplace coordinate is defined as the difference between the weighted average of the positions of the surrounding neighborhood vertices and the position of v itself, expressed as:

[0037]

[0038] Where: N(v) represents the set of vertices v, ω(v,ω) represents the weight function used to express the relationship function between vertices v and w, p w With p v Represented as the position vectors of vertices v and w respectively;

[0039] S411: Find a new set of positions by constraining the vertex v and minimize the energy function solution:

[0040]

[0041] Where: L v 'Represented as the Laplace coordinate of vertex v after transformation;

[0042] S412: Selecting significant feature points to establish deformation constraint conditions;

[0043] S413: Solving the linear equations using the least squares method to obtain the vertices of the deformed model;

[0044] S414: updating the coordinates of the deformed model vertices obtained by the solution to achieve deformation processing of the convex model surface.

[0045] The city-level geographic scene incremental data governance system includes:

[0046] Existing data sorting module, sorting existing city-level real-life geographic scene data;

[0047] The segmentation module divides the existing multi-scale city-level real-life geographic scene data into multiple spatially continuous and overlapping basic geographic grid units according to the city's local coordinates;

[0048] The subdivision module uses multiple basic geographic grid units as initial grid units and divides them into geographic grid models using a quadtree grid subdivision structure;

[0049] The geocoding module geocodes the geographic grid model and constructs an incremental city-level real-world geographic scene data spatial index structure;

[0050] Incremental update data collection module, collects incremental city-level real-life geographic scene data to be updated;

[0051] The incremental update data processing module performs block denoising and enhancement processing on the incremental city-level real-life geographic scene data to be updated, as well as channel-by-channel brightness adjustment.

[0052] The incremental update data splicing module quickly locates the location to be updated based on the constructed spatial index structure of incremental city-level realistic geographic scene data, extracts boundary information, analyzes the topological relationship between boundary information and existing city-level realistic geographic scene data, and splices it with the existing city-level realistic geographic scene data based on the topological relationship;

[0053] Incremental update data dodging processing module, which performs dodging processing on the stitched image based on the multi-scale retinex algorithm;

[0054] Incrementally update the data color correction module to perform color correction on the stitched image after light uniformity processing;

[0055] Incrementally update the data feathering processing module to feather the stitching edges of the color-corrected stitching image;

[0056] The incremental update data edge processing module uses the Laplace coordinate-based grid deformation algorithm to perform grid deformation on the stitched image edge mismatch problem and deform the raised model surface to make the stitched image edge match.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] The present invention integrates existing data, performs grid division based on the existing data, establishes an incremental real-scene three-dimensional data spatial index structure, then performs color correction on the updated image based on the Retinex algorithm, and finally performs surface correction based on the grid transformation of the Laplace operator. This method can effectively solve the problems existing in the prior art, such as missing indexes in the updated area, color deviations in the updated data, and misaligned updated patches, thereby improving the updating efficiency and effectiveness of urban-level real-scene geographic scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flow chart of the city-level geographic scenario incremental data management method of the present invention;

[0060] Figure 2 This is a schematic diagram of the scale division of real-scene three-dimensional data of the present invention;

[0061] Figure 3 This is a schematic diagram of the urban grid space subdivision of the present invention;

[0062] Figure 4 Schematic diagram of the coding rules for grid cell division according to the present invention;

[0063] Figure 5 This is a schematic diagram of a terrain-level real-scene three-dimensional model of the present invention;

[0064] Figure 6 Schematic diagram of grid division and coding of the present invention;

[0065] Figure 7 Update the plot color correction comparison diagram for the present invention;

[0066] Figure 8 This is a schematic diagram of the updated plot after Laplace grid correction of the present invention;

[0067] Figure 9 This is a framework diagram of the city-level geographic scenario incremental data governance system of the present invention;

[0068] In the figure: 1. Existing data sorting module; 2. Segmentation module; 3. Subdivision and segmentation module; 4. Geocoding module; 5. Incremental update data acquisition module; 6. Incremental update data processing module; 7. Incremental update data splicing module; 8. Incremental update data uniformity processing module; 9. Incremental update data color correction module; 10. Incremental update data feathering processing module; 11. Incremental update data edge processing module. DETAILED DESCRIPTION

[0069] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0070] See attached Figure 1 ,A city-level geographic scenario incremental data governance method, including the following steps:

[0071] S1: Organize existing city-level real-world geographic scene data;

[0072] S2: Divide the existing multi-scale city-level real-life geographic scene data into multiple spatially continuous and overlapping basic geographic grid units according to the city's local coordinates;

[0073] See attached Figure 2 The existing multi-scale city-level real-life geographic scene data includes large scale, medium scale, small scale and small scale. Among them, the large scale includes the terrain-level 3D terrain scene based on DEM and DOM collection, the medium scale includes the city-level 3D geographic scene based on oblique photography collection, the small scale includes the component-level 3D geographic scene built based on software 3D model, and the small scale includes the component-level 3D city scene integrated with point cloud data;

[0074] S3: Using multiple basic geographic grid cells as initial grid cells, splitting them using a quadtree grid splitting structure, recursively checking each grid cell. If the preset splitting conditions are met, the grid cell is split into four sub-grid cells. Repeating the above checking and splitting process for each newly generated sub-grid cell until the preset splitting level is reached, thus forming a geographic grid model.

[0075] See attached Figure 3 ,The basic geographic grid cell includes metadata, i.e., spatial extent, grid code, and tile address;

[0076] The quadtree grid subdivision structure divides the basic geographic grid unit into preset levels. The first-level subdivision grid includes large-scale data, namely heterogeneous 3D data A, including 3D terrain data, namely terrain tiles and image data. The second-level subdivision grid includes medium-scale data, namely heterogeneous 3D data B, including manual models and inclined models, namely manual model tiles and inclined model tiles. The n-level subdivision grid includes small-scale data, namely heterogeneous 3D data C, including point cloud models, namely point cloud model tiles.

[0077] S3: Geocode the geographic grid model and build an incremental city-level real-world geographic scene data spatial index structure;

[0078] See attached Figure 4 ,geographic coding includes grid row and column numbers, grid subdivision levels and grid subdivision units. Grid row and column numbers are expressed as the coordinates of the southwest corner of the grid based on the city local coordinate system. The grid subdivision level is expressed as the level of the grid unit in the basic geographic grid unit. The grid subdivision unit coding rule is expressed by coding the four grid units in the order of upper left, lower left, lower right and upper right as 0, 1, 2 and 3 respectively.

[0079] S4: During incremental updates, the incremental city-level realistic geographic scene data to be updated is collected, and denoising and enhancement processing is performed on the blocks, and brightness adjustment processing is performed on the channels. The location to be updated is quickly located based on the constructed spatial index structure of the incremental city-level realistic geographic scene data, boundary information is extracted, and the topological relationship between the boundary information and the existing city-level realistic geographic scene data is analyzed. Based on the topological relationship, the data is spliced ​​with the existing city-level realistic geographic scene data. The spliced ​​image is evened based on the multi-scale retinex algorithm, and color correction is performed on the evened spliced ​​image. The spliced ​​edges of the color-corrected spliced ​​image are feathered. The grid deformation algorithm based on the Laplace coordinate is used to grid-deform the spliced ​​image to solve the problem of spliced ​​image edge mismatch. The convex model patch is deformed to make the spliced ​​image edges match, and the update is completed.

[0080] The specific steps for performing block denoising and enhancement processing on the incremental city-level scene data to be updated include:

[0081] Divide the collected incremental city-level real-life geographic scene data to be updated according to the set block size to obtain multiple small blocks of images;

[0082] Process each small image block, including denoising and enhancement;

[0083] Reassemble the processed block images into a complete image;

[0084] The specific steps of performing channel-by-channel brightness adjustment processing on the incremental city-level real-life geographic scene data to be updated include:

[0085] Separate the complete image into multiple monochrome channel images;

[0086] Process each monochrome channel image, including adjusting brightness;

[0087] Reassemble the processed monochrome channel images into a complete image;

[0088] Analyze the topological relationship between boundary information and existing city-level real-world geographic scene data using existing software, including ArcGIS and QGIS;

[0089] Based on existing software, it can be combined with existing city-level real-life geographic scene data based on topological relationships, including GIS;

[0090] The specific steps of uniforming the stitched image based on the multi-scale retinex algorithm include:

[0091] S401: Set the incident light image to L(x,y), the reflected light image to R(x,y), and the original image to I'(x,y). The expression of the single-scale retinex algorithm is:

[0092] I'(x,y)=L(x,y)·R(x,y)

[0093] To further obtain an incident light image with uniform brightness, the above formula can be converted to the logarithmic domain, and the expression is:

[0094] logR(x,y)=logI'(x,y)-logL(x,y)

[0095] make:

[0096]

[0097] Assume that the incident light image L(x,y) is obtained by convolving the Gaussian function F(x,y) with the original image I'(x,y). Substituting the above formula into the solution, we can get:

[0098]

[0099] Where: F(x,y) represents the Gaussian function model, Expressed as a scale parameter, λ must satisfy:

[0100] ∫∫F(x,y)dxdy=1;

[0101] S402: When uniform lighting, first select the three scale parameters of large, medium and small. Then perform single-scale operations separately, and then perform weighted average calculation on the results to obtain the stitched image after uniform light;

[0102] The specific steps for color correction of the stitched image after light uniformity processing include:

[0103] S403: Divide the uniformly lighted stitched image into nine evenly spaced areas according to a preset rule;

[0104] S404: Identify features with significant characteristics in each area;

[0105] S405: Calculate the statistics of each feature object in different color channels;

[0106] S406: Select the feature object with the largest statistical value in each area as a calibration feature point;

[0107] S407: Recording the grayscale value of the selected calibration feature point;

[0108] S408: Using the least squares method, the grayscale correction value of the image is fitted according to the grayscale values ​​of the calibrated feature points;

[0109] S409: performing grayscale correction on each pixel of the image according to the grayscale correction value of the fitted image;

[0110] Feathering the stitching edges of the color-corrected stitched image using existing software, including Photoshop;

[0111] The Laplace coordinate-based grid deformation algorithm is used to deform the grid to solve the edge mismatch problem of the stitched image. The specific steps for deforming the convex model patch include:

[0112] S410: Given a mesh vertex v, its Laplace coordinate is defined as the difference between the weighted average of the positions of the surrounding neighborhood vertices and the position of v itself, expressed as:

[0113]

[0114] Where: N(v) represents the set of vertices v, ω(v,ω) represents the weight function used to express the relationship function between vertices v and w, p w With p v Represented as the position vectors of vertices v and w respectively;

[0115] S411: Find a new set of positions by constraining the vertex v and minimize the energy function solution:

[0116]

[0117] Where: L v 'Represented as the Laplace coordinate of vertex v after transformation;

[0118] S412: Selecting significant feature points to establish deformation constraint conditions;

[0119] S413: Solving the linear equations using the least squares method to obtain the vertices of the deformed model;

[0120] S414: updating the coordinates of the deformed model vertices obtained by the solution to achieve deformation processing of the convex model surface.

[0121] As attached Figure 9 As shown, a city-level geographic scenario incremental data governance system includes:

[0122] Existing data sorting module 1, sorting existing city-level real-life geographic scene data;

[0123] The segmentation module 2 segments the existing multi-scale city-level real-life geographic scene data into multiple spatially continuous and overlapping basic geographic grid units according to the city's local coordinates;

[0124] The subdivision module 3 uses a plurality of basic geographic grid units as initial grid units and divides them into a geographic grid model using a quadtree grid subdivision structure;

[0125] Geocoding module 4 geocodes the geographic grid model and constructs an incremental city-level real-world geographic scene data spatial index structure;

[0126] Incremental update data collection module 5, collects incremental city-level real-life geographic scene data to be updated;

[0127] Incremental update data processing module 6, which performs block denoising and enhancement processing and channel brightness adjustment processing on the collected incremental city-level real-life geographic scene data to be updated;

[0128] The incremental update data splicing module 7 quickly locates the location to be updated based on the constructed spatial index structure of the incremental city-level realistic geographic scene data, extracts boundary information, analyzes the topological relationship between the boundary information and the existing city-level realistic geographic scene data, and splices the data with the existing city-level realistic geographic scene data based on the topological relationship;

[0129] Incremental update data dodging processing module 8, which performs dodging processing on the stitched image based on the multi-scale retinex algorithm;

[0130] The incremental update data color correction module 9 performs color correction on the spliced ​​image after the uniform light processing;

[0131] The incremental update data feathering processing module 10 performs feathering processing on the splicing edges of the color-corrected splicing image;

[0132] The incremental update data edge processing module 11 performs grid deformation on the spliced ​​image edge mismatch problem based on the grid deformation algorithm of Laplace coordinates, and deforms the convex model surface to make the spliced ​​image edge match.

[0133] As attached Figure 5 As shown, the real scene 3D model is generated by combining the real scene 3D construction to verify the above scheme;

[0134] As attached Figure 6 As shown, the above spatial model is subjected to spatial multi-scale data grid segmentation and encoding;

[0135] As attached Figure 7 As shown in Figure 2, the color correction results of the real-scene 3D data are performed using the multi-scale retinex algorithm after incremental update.

[0136] As attached Figure 8 As shown in the figure, the result of grid correction using Laplace grid coordinate transformation is used to correct the mismatch of the updated data patches.

[0137] By selecting relevant feature points to analyze the texture, using color difference and histogram similarity to analyze local texture information, and using standard deviation and average gradient to evaluate global texture information, taking a certain updated plot as an example, the changes in its color features are shown in the following table:

[0138] Color Correction Analysis Sheet

[0139]

[0140] Select characteristic points to analyze the grid, use the total station to obtain points to analyze the grid, and evaluate the conversion results with conversion error and point error as shown in the table. Taking a certain updated plot as an example, the changes in its grid point positions are shown in the following table:

[0141] Grid conversion analysis table

[0142]

[0143] The results show that the updated plots of the above scheme have clear spatial indexes, and the texture of the updated plots further reduces the color difference with the original plots while retaining the original color characteristics. The deformed grid retains the topological relationship with the original grid while better fitting the updated plots, showing certain applicability.

[0144] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A city-level geographic scene incremental data governance method, characterized in that: The following steps are involved: S1: Organize the existing city-level real-life geographic scene data; S2: Divide the existing multi-scale city-level real-life geographic scene data into multiple spatially continuous and overlapping basic geographic grid units according to the city’s local coordinates; S3: multiple basic geographic grid units are used as initial grid units and divided into geographic grid models using a quadtree grid division structure; S3: Geocode the geographic grid model and construct an incremental city-level real-life geographic scene data spatial index structure; S4: During incremental update, collect incremental city-level realistic geographic scene data to be updated, perform block denoising and enhancement processing, and adjust brightness by channel, quickly locate the location to be updated based on the constructed spatial index structure of incremental city-level realistic geographic scene data, extract boundary information, analyze the topological relationship between boundary information and existing city-level realistic geographic scene data, splice with existing city-level realistic geographic scene data based on the topological relationship, perform uniform light processing on the spliced ​​image based on the multi-scale retinex algorithm, perform color correction on the spliced ​​image after uniform light processing, perform feathering processing on the spliced ​​edge of the spliced ​​image after color correction, perform grid deformation on the spliced ​​image edge mismatch problem based on the grid deformation algorithm of Laplace coordinates, and deform the convex model facets to make the spliced ​​image edge match, and complete the update.

2. According to claim 1, a city-level geographic scene incremental data management method is characterized by: In step S3, the geographic coding includes grid row and column numbers, grid subdivision levels and grid subdivision units. The grid row and column numbers are expressed by the coordinates of the southwest corner of the grid based on the city local coordinate system. The grid subdivision level is expressed by the level of the grid unit in the basic geographic grid unit. The grid subdivision unit coding rule is expressed by encoding the four grid cells in the order of upper left, lower left, lower right and upper right as 0, 1, 2 and 3 respectively.

3. According to claim 1, a city-level geographic scene incremental data management method is characterized by: In step S4, the specific steps of performing light homogenization processing on the stitched image based on the multi-scale retinex algorithm include: S401: Set the incident light image to L(x, y), the reflected light image to R(x, y), and the original image to I'(x, y). Then the expression of the single-scale retinex algorithm is: I'(x,y)=L(x,y)·R(x,y) In order to further obtain an incident light image with uniform brightness, the above formula can be transformed into the logarithmic domain, and the expression is: logR(x,y)=logI'(x,y)-logL(x,y) make: Assume that the incident light image L(x,y) is obtained by convolving the Gaussian function F(x,y) with the original image I'(x,y), and substitute the above formula into the solution: Where: F(x,y) represents the Gaussian function model, Expressed as a scale parameter, λ needs to satisfy: ∫∫F(x,y)dxdy=1; S402: When uniform light, first select the three scale parameters of large, medium and small. Then perform single-scale operations separately, and then perform weighted average calculation on the results to obtain the stitched image after uniform light.

4. According to claim 1, a city-level geographic scene incremental data management method is characterized by: In step S4, the specific steps of color correction of the stitched image after the uniform light processing include: S403: Divide the uniformly lighted spliced ​​image into nine uniform areas according to a preset rule; S404: Identify the features with significant characteristics in each area; S405: Calculate the statistics of each feature object in different color channels; S406: Select the characteristic feature with the largest statistical value in each area as a calibration feature point; S407: Recording the grayscale value of the selected calibration feature point; S408: Using the least square method, the grayscale correction value of the image is fitted according to the grayscale value of the calibrated feature point; S409: Perform grayscale correction on each pixel of the image according to the grayscale correction value of the fitted image.

5. According to claim 1, a city-level geographic scene incremental data management method is characterized by: In step S4, the grid deformation algorithm based on Laplace coordinates is used to perform grid deformation on the spliced ​​image edge mismatch problem, and the specific steps of deforming the convex model surface patch include: S410: Given a mesh vertex v, its Laplace coordinate is defined as the difference between the weighted average of the positions of the surrounding neighborhood vertices and the position of v itself, expressed as: Where: N(v) represents the set of vertices v, ω(v,ω) represents the weight function used to express the relationship function between vertices v and w, and p w With p v Represented as the position vectors of vertices v and w respectively; S411: Find a set of new positions by constraining the vertex v and minimize the energy function solution: Where: L v 'Represents the Laplace coordinates of vertex v after transformation; S412: Selecting significant feature points to establish deformation constraint conditions; S413: solving the linear equations using the least square method to obtain the vertices of the deformed model; S414: updating the vertex coordinates of the deformed model according to the solution, so as to realize the deformation processing of the convex model surface.

6. A city-level geographic scene incremental data management system according to claims 1-5, characterized in that: include: An existing data sorting module (1) sorts the existing city-level real-life geographic scene data; A segmentation module (2) is used to segment the existing multi-scale city-level real-life geographic scene data into a plurality of spatially continuous and overlapping basic geographic grid units according to the local coordinates of the city; A subdivision module (3) uses a plurality of basic geographic grid units as initial grid units and divides them into a geographic grid model using a quadtree grid division structure; A geocoding module (4) performs geocoding on the geographic grid model and constructs an incremental city-level real-life geographic scene data spatial index structure; An incremental update data collection module (5) collects incremental city-level real-life geographic scene data to be updated; The incremental update data processing module (6) performs block denoising and enhancement processing and channel brightness adjustment processing on the collected incremental city-level real-life geographic scene data to be updated; An incremental update data splicing module (7) is used to quickly locate the location to be updated based on the constructed incremental city-level real-life geographic scene data spatial index structure, extract boundary information, analyze the topological relationship between the boundary information and the existing city-level real-life geographic scene data, and splice the data with the existing city-level real-life geographic scene data based on the topological relationship; An incremental update data homogenization processing module (8) performs homogenization processing on the spliced ​​image based on a multi-scale retinex algorithm; An incremental update data color correction module (9) performs color correction on the spliced ​​image after the uniform light processing; An incremental update data feathering processing module (10) performs feathering processing on the splicing edge of the spliced ​​image after color correction; The incremental update data edge processing module (11) performs grid deformation on the spliced ​​image edge mismatch problem based on the grid deformation algorithm of Laplace coordinates, and performs deformation processing on the raised model surface to make the spliced ​​image edge match.

Images