A spatial data governance method based on data twins

Abstract

The present invention discloses a spatial data governance method based on data twins, which relates to the field of spatial data processing. The present invention divides the target area into several sub-areas, obtains and pre-processes the spatial data on the sub-areas, and obtains standard spatial data; then the standard remote sensing image and standard point cloud data are segmented respectively, the remote sensing feature set and point cloud feature set of the object are extracted, and a three-dimensional stereogram of the target area is generated by feature matching; by analyzing the standard remote sensing image and point cloud data, the remote sensing update level and the point cloud update level are obtained respectively, and according to the specific levels of the two update levels, it is determined whether to update the three-dimensional stereogram of the sub-area. The present invention can reduce resource waste while ensuring the modeling accuracy of the target area through multi-source data fusion and differentiated update strategy.

Description

Technical Field

[0001] The present invention relates to the field of spatial data processing, and specifically to a spatial data governance method based on data twins. Background Art

[0002] Data twin refers to the creation of a virtual model corresponding to a physical entity in the virtual world by collecting multi-source data and comprehensively processing the multi-source data;

[0003] However, when updating the virtual model of the created physical entity in the virtual world, since not all parts of the physical entity will change, adopting a fixed strategy (processing multi-source data once each time it is collected) will result in a large waste of resources. In addition, since the quality of the collected multi-source data may not be as good as the previous collection, the accuracy of the virtual model will also decrease.

[0004] How to analyze the collected multi-source data and adopt differentiated update strategies to update the virtual model based on the analysis results is the problem we need to solve. To this end, we now provide a spatial data governance method based on data twins. Summary of the Invention

[0005] The purpose of this invention is to provide a spatial data governance method based on data twins.

[0006] The purpose of the present invention can be achieved through the following technical solutions: A spatial data governance method based on data twins, comprising:

[0007] Divide the target area into several sub-areas, obtain the spatial data of each sub-area and pre-process it to obtain the standard spatial data of each sub-area. The standard spatial data includes standard remote sensing images and standard point cloud data.

[0008] Perform image segmentation on the standard remote sensing image to obtain the remote sensing feature set corresponding to each object in the sub-area;

[0009] Perform point cloud segmentation on the standard point cloud data to obtain the point cloud feature set of each object in the sub-area;

[0010] Perform feature matching on the remote sensing feature set corresponding to the objects in the standard remote sensing image of the sub-region and the point cloud feature set corresponding to the objects in the standard point cloud data to obtain a three-dimensional stereogram of the target area;

[0011] Analyze the standard remote sensing images corresponding to each sub-region to obtain the remote sensing update level corresponding to each sub-region;

[0012] Analyze the standard point cloud data corresponding to each sub-area to obtain the point cloud update level corresponding to each sub-area;

[0013] Whether to update the three-dimensional stereogram of the sub-area is determined based on the remote sensing update level and the point cloud update level of the sub-area.

[0014] Preferably, the spatial data includes remote sensing images and point cloud data;

[0015] Perform radiation correction, geometric correction and image enhancement on remote sensing images to obtain standard remote sensing images;

[0016] The point cloud data is denoised and filtered to obtain standard point cloud data.

[0017] Preferably, the process of performing image segmentation on the standard remote sensing image to obtain the remote sensing feature set corresponding to each object in the sub-region is:

[0018] Use the Canny algorithm to perform convolution operations on standard remote sensing images, identify the pixel points at the edge of the object, and obtain the pixel point images corresponding to each object;

[0019] Perform expansion and erosion operations on the pixel image to obtain the contour image of each object;

[0020] Feature extraction is performed on the contour image to obtain the remote sensing feature set corresponding to each object. The remote sensing feature set includes the area, perimeter and center point coordinates of the contour image.

[0021] Preferably, the process of performing point cloud segmentation on the standard point cloud data to obtain the point cloud feature set of each object in the sub-region is as follows:

[0022] Extract the geometric features of each point in the point cloud data as the classification features of the corresponding points;

[0023] Based on the classification features, the k-means clustering algorithm is used to perform cluster analysis on the points in the point cloud data to obtain the point sets corresponding to each object in the target area;

[0024] The three-dimensional shape of the corresponding object is restored based on the point set, and the point cloud feature set corresponding to the object is obtained based on the three-dimensional shape of the object.

[0025] Preferably, the process of obtaining a three-dimensional image of the target area is:

[0026] For each object in the standard remote sensing image, obtain the matching value with each object in the standard point cloud data one by one;

[0027] A matching value threshold is preset, and when the matching value is less than the matching value threshold, an association relationship is established between the object in the standard remote sensing image corresponding to the matching value and the object in the standard point cloud data corresponding to the matching value;

[0028] Generate a 3D model of the object using the standard remote sensing image and standard point cloud data corresponding to the object and map it into the virtual space to obtain a 3D stereogram of each sub-area of ​​the target area;

[0029] The three-dimensional stereograms of the sub-areas are combined to obtain a three-dimensional stereogram of the target area.

[0030] Preferably, the process of analyzing the standard remote sensing images corresponding to each sub-region to obtain the remote sensing update level corresponding to each sub-region is as follows:

[0031] S1: Acquire each object and its corresponding coordinates in the standard remote sensing image; and simultaneously acquire each object and its corresponding coordinates in the sub-region storage image;

[0032] S2: Obtain the spatial distances between the objects in the standard remote sensing image and the objects in the stored image one by one through the coordinates of the objects in the standard remote sensing image and the coordinates of the objects in the stored image of the sub-region;

[0033] S3: Preset spatial distance threshold. When the spatial distance is less than the spatial distance threshold, it is considered that the object in the standard remote sensing image corresponding to the spatial distance is successfully matched with the object in the stored image.

[0034] S4: Obtain the number of successful matches, the number of objects in the standard remote sensing image, and the number of objects in the stored image. If the three numbers are not completely equal, determine that the remote sensing update level is A. If the three numbers are completely equal, determine whether the objects in the successfully matched standard remote sensing image are the same as the objects in the stored image. If they are not the same, determine that the remote sensing update level is A. If they are the same, proceed to step S5.

[0035] S5: Analyze the standard remote sensing image to obtain a corresponding quality index, and determine the remote sensing update level based on the obtained quality index.

[0036] Preferably, the standard remote sensing image is analyzed to obtain a corresponding quality index, and the process of determining the remote sensing update level based on the obtained quality index is as follows:

[0037] Obtain the color space corresponding to the standard remote sensing image, obtain several color intervals through the color space, and obtain the pixel frequency value in each color interval;

[0038] The information entropy formula is used to process the pixel frequency values ​​in all color intervals to obtain the color index of the standard remote sensing image;

[0039] Convert the standard remote sensing image into a grayscale image, use the Laplace algorithm to get the value of each pixel in the grayscale image as the grayscale change value of the corresponding pixel, and obtain the variance of the grayscale change values ​​of all pixels as the pixel wave value;

[0040] Obtain the intensity value of each pixel in the grayscale image, and use the standard deviation formula to process the intensity values ​​of all pixels on the grayscale image to obtain the global contrast of the grayscale image;

[0041] Preset several global contrast intervals, different intervals correspond to different contrast scores, match the global contrast interval where the global contrast is located, and thus obtain the contrast score corresponding to the standard remote sensing image;

[0042] The color index, pixel wave value and global contrast are fused to obtain the quality index of the standard remote sensing image;

[0043] Obtain the quality index corresponding to the stored image, preset the image update condition, and determine whether the image update condition is met based on the quality index of the standard remote sensing image. When the image update condition is met, the remote sensing update level is determined to be B, otherwise the remote sensing update level is determined to be C.

[0044] Preferably, the process of analyzing the standard point cloud data corresponding to each sub-region to obtain the point cloud update level corresponding to each sub-region is as follows:

[0045] Obtain the total number of points in the sub-region and the volume of the sub-region of the standard point cloud data, and divide the total number of points by the volume to obtain the average point density of the sub-region;

[0046] Obtain the normal vector angles of all neighboring points in the standard point cloud data to obtain an angle set, preset a normal vector angle threshold, and the number of elements in the angle set that are greater than the normal vector angle threshold is used as the anomaly index;

[0047] According to the average point density and anomaly index, the quality index of the standard point cloud data is obtained;

[0048] Obtain the quality index of the stored point cloud, preset the point cloud update conditions, and determine whether the point cloud update conditions are met based on the quality index of the standard point cloud data. When the point cloud update conditions are met, the point cloud update level is determined to be B, otherwise, the point cloud update level is determined to be C.

[0049] Preferably, the process of jointly determining whether to update the three-dimensional stereogram of the sub-region according to the remote sensing update level and the point cloud update level of the sub-region is:

[0050] If the remote sensing update level is A or both the remote sensing update level and the point cloud update level are B, the 3D stereogram of the sub-area is updated using both the standard remote sensing image and the standard point cloud data.

[0051] If the remote sensing update level is B and the point cloud update level is C, the 3D stereogram of the sub-area is updated using the standard remote sensing image and the stored point cloud;

[0052] If the remote sensing update level is C and the point cloud update level is B, the 3D stereogram of the sub-area is updated using standard point cloud data and stored images;

[0053] If the remote sensing update level and the point cloud update level are both C, the 3D stereogram of the sub-area will not be updated.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] The present invention divides the target area into several sub-areas, and obtains the remote sensing update level by comparing and analyzing the remote sensing images of the sub-areas collected and the stored images stored in the database; obtains the point cloud update level by comparing and analyzing the point cloud data of the sub-areas collected and the stored point cloud stored in the database. The remote sensing update level and the point cloud update level are combined to jointly determine the update method of each sub-area of ​​the target area, thereby ensuring the modeling accuracy of the target area while reducing resource waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0057] Figure 1 This is a schematic diagram of the present invention.

[0058] Figure 2 A flow chart of the present invention for analyzing standard remote sensing images;

[0059] Figure 3 This is a flow chart of analyzing standard point cloud data according to the present invention; DETAILED DESCRIPTION

[0060] like Figure 1 As shown in the figure, a spatial data governance method based on data twins includes the following steps:

[0061] Divide the target area into several sub-areas, generate a unique identifier for each sub-area, obtain the spatial data of each sub-area and pre-process it to obtain the standard spatial data of each sub-area;

[0062] The spatial data includes: remote sensing images and point cloud data;

[0063] The standard spatial data includes: standard remote sensing images and standard point cloud data;

[0064] Perform radiation correction, geometric correction and image enhancement on remote sensing images to obtain standard remote sensing images;

[0065] Denoise and filter the point cloud data to obtain standard point cloud data;

[0066] Specifically, remote sensing images of each sub-area of ​​the target area are obtained using sensors carried by satellite remote sensing or aerial remote sensing platforms; point cloud data of each sub-area of ​​the target area are obtained using ground-based laser scanners;

[0067] Perform image segmentation on the standard remote sensing image to obtain the remote sensing feature set corresponding to each object in the sub-area;

[0068] Use the Canny algorithm to perform convolution operations on standard remote sensing images, identify the pixel points at the edge of the object, and obtain the pixel point images corresponding to each object;

[0069] Perform expansion and erosion operations on the pixel image to obtain the contour image of each object;

[0070] Perform feature extraction on the contour image to obtain the remote sensing feature set corresponding to each object. The remote sensing feature set includes the area, perimeter and center point coordinates of the contour image;

[0071] Perform point cloud segmentation on the standard point cloud data to obtain the point cloud feature set of each object in the sub-area;

[0072] Extract the geometric features of each point in the point cloud data as the classification features of the corresponding points;

[0073] In detail, the geometric features include: normal vector, curvature, number of neighborhood points, eigenvalues, and eigenvectors;

[0074] Based on the classification features, the k-means clustering algorithm is used to perform cluster analysis on the points in the point cloud data to obtain the point sets corresponding to each object in the target area;

[0075] The three-dimensional shape of the corresponding object is restored based on the point set, and the point cloud feature set corresponding to the object is obtained based on the three-dimensional shape of the object;

[0076] Specifically, a top view corresponding to the object is obtained through the three-dimensional shape of the object, and the area, perimeter and center point coordinates of the top view are obtained as a point cloud feature set corresponding to the object;

[0077] Perform feature matching on the remote sensing feature set corresponding to the objects in the standard remote sensing image of the sub-region and the point cloud feature set corresponding to the objects in the standard point cloud data to obtain a three-dimensional stereogram of the target area;

[0078] For each object in the standard remote sensing image, obtain the matching value with each object in the standard point cloud data one by one;

[0079] Specifically, after normalizing the values ​​in the remote sensing feature set and the point cloud feature set, the formula is used:

[0080] ;

[0081] Calculate the matching value between objects in standard remote sensing images and objects in standard point cloud data ,in Representative area, represents the circumference, represents the horizontal axis, represents the vertical axis, 、 and Represent the weight influence factors corresponding to area, perimeter and coordinates respectively;

[0082] A matching value threshold is preset, and when the matching value is less than the matching value threshold, an association relationship is established between the object in the standard remote sensing image corresponding to the matching value and the object in the standard point cloud data corresponding to the matching value;

[0083] Generate a 3D model of the object using the standard remote sensing image and standard point cloud data corresponding to the object and map it into the virtual space to obtain a 3D stereogram of each sub-area of ​​the target area;

[0084] Combining the three-dimensional stereograms of each sub-region to obtain a three-dimensional stereogram of the target region;

[0085] Analyze the standard remote sensing images corresponding to each sub-region to obtain the remote sensing update level corresponding to each sub-region;

[0086] S1: Acquire each object and its corresponding coordinates in the standard remote sensing image; and simultaneously acquire each object and its corresponding coordinates in the sub-region storage image;

[0087] In detail, the stored image refers to the standard remote sensing image of the sub-area stored in the database;

[0088] S2: Obtain the spatial distances between the objects in the standard remote sensing image and the objects in the stored image one by one through the coordinates of the objects in the standard remote sensing image and the coordinates of the objects in the stored image of the sub-region;

[0089] S3: Preset spatial distance threshold. When the spatial distance is less than the spatial distance threshold, it is considered that the object in the standard remote sensing image corresponding to the spatial distance is successfully matched with the object in the stored image.

[0090] S4: Obtain the number of successful matches, the number of objects in the standard remote sensing image, and the number of objects in the stored image. If the three numbers are not completely equal, determine that the remote sensing update level is A. If the three numbers are completely equal, determine whether the objects in the successfully matched standard remote sensing image are the same as the objects in the stored image. If they are not the same, determine that the remote sensing update level is A. If they are the same, proceed to step S5.

[0091] S5: Analyze the standard remote sensing image to obtain a corresponding quality index, and determine the remote sensing update level based on the obtained quality index;

[0092] S5-1: Obtain the color space corresponding to the standard remote sensing image, obtain several color intervals through the color space, and obtain the pixel frequency value in each color interval;

[0093] In detail, taking the RGB color space as an example, a division value k is preset. Since the RGB color space contains three channels, the value range of each channel is 0-255, a total of 256 values. Dividing 256 by k gives the number of intervals t. The number of intervals must be an integer, so a total of t*t*t color intervals can be obtained.

[0094] Specifically, if the division value is 16, the value of t is also 16, then each channel in the RGB color space will be divided into 16 intervals, [0,15], [16,31], [32,47] and so on. Since there are three channels in total, 16*16*16 color intervals can be obtained;

[0095] Get the RGB value of each pixel and match the RGB values ​​of all pixels with the color interval. Count the number of successful matches in each color interval as the number of pixels in the corresponding color interval. Divide the number of pixels in the color interval by the total number of pixels in the RGB color space to get the pixel frequency value of the color interval.

[0096] S5-2: Use the information entropy formula to process the pixel frequency values ​​in all color intervals to obtain the color index of the standard remote sensing image ;

[0097] In detail, the information entropy formula is:

[0098] ;

[0099] in, The number representing the color range, Representative No. The pixel frequency value of the color interval;

[0100] In detail, the larger the color index is, the better the richness and uniformity of the colors in the image are, that is, the higher the color quality of the image is;

[0101] S5-3: Convert the standard remote sensing image into a grayscale image, use the Laplace algorithm to obtain the value of each pixel in the grayscale image as the grayscale change value of the corresponding pixel, and obtain the variance of the grayscale change values ​​of all pixels as the pixel wave value;

[0102] In detail, the value of the pixel is the value obtained by calculating the Laplacian operator;

[0103] Specifically, a larger pixel variance indicates a more dispersed distribution of pixel values ​​in the image, which means that there are more areas with drastic changes in pixel values ​​in the image, that is, the image contains more edge and detail information, which indicates that the image clarity is higher; conversely, a smaller variance indicates a lower image clarity;

[0104] S5-4: Obtain the intensity value of each pixel in the grayscale image, and process the intensity values ​​of all pixels in the grayscale image using the standard deviation formula to obtain the global contrast of the grayscale image;

[0105] S5-5: Preset several global contrast intervals, different intervals correspond to different contrast scores, match the global contrast interval where the global contrast is located, and thus obtain the contrast score corresponding to the standard remote sensing image;

[0106] Specifically, it is not good to have a global contrast that is too high or too low, because when setting the contrast score corresponding to the contrast interval, it should be high in the middle and low on both sides;

[0107] S5-6: Fusion processing is performed on the color index, pixel wave value, and global contrast to obtain the quality index of the standard remote sensing image;

[0108] Specifically, weighted influence factors corresponding to the color index, pixel wave value, and global contrast are preset, and the color index, pixel wave value, and global contrast are weighted based on the preset weighted influence factors to obtain a quality index of the standard remote sensing image;

[0109] In detail, the larger the quality index, the higher the quality of the standard remote sensing image;

[0110] S5-7: Based on the process of steps S5-1 to S5-6, a quality index corresponding to the stored image is obtained, an image update condition is preset, and whether the image update condition is met is determined based on the quality index of the standard remote sensing image. If the image update condition is met, the remote sensing update level is determined to be B, otherwise the remote sensing update level is determined to be C;

[0111] Specifically, the quality index of the standard remote sensing image is divided by the quality index of the stored image to obtain a remote sensing index ratio, and a remote sensing index ratio threshold is preset. If the remote sensing index ratio is greater than the remote sensing index ratio threshold, it means that the preset image condition is met;

[0112] Analyze the standard point cloud data corresponding to each sub-area to obtain the point cloud update level corresponding to each sub-area;

[0113] Obtain the total number of points in the sub-region and the volume of the sub-region of the standard point cloud data, and divide the total number of points by the volume to obtain the average point density of the sub-region;

[0114] Obtain the normal vector angles of all neighboring points in the standard point cloud data to obtain an angle set, preset a normal vector angle threshold, and the number of elements in the angle set that are greater than the normal vector angle threshold is used as the anomaly index;

[0115] In detail, a large angle between the normal vectors of neighboring points will affect the modeling of the object, so the larger the anomaly index is, the worse the quality of the standard point cloud data is;

[0116] According to the average point density and anomaly index, the quality index of the standard point cloud data is obtained;

[0117] In detail, the average point density and abnormal index After normalization, enter the formula:

[0118] ;

[0119] in, and Represents the average point density and abnormal index The corresponding weight impact factor;

[0120] Obtain the quality index of the stored point cloud, preset the point cloud update conditions, and determine whether the point cloud update conditions are met based on the quality index of the standard point cloud data. If the point cloud update conditions are met, the point cloud update level is determined to be B, otherwise, the point cloud update level is determined to be C;

[0121] Specifically, the stored point cloud refers to the standard point cloud data of the sub-region stored in the database; the quality index of the standard point cloud data is divided by the quality index of the stored point cloud to obtain the point cloud index ratio, and a point cloud index ratio threshold is preset. If the point cloud index ratio is greater than the point cloud index ratio threshold, it means that the point cloud update condition is met;

[0122] Determine whether to update the three-dimensional image of the sub-area based on the remote sensing update level and the point cloud update level of the sub-area;

[0123] If the remote sensing update level is A or both the remote sensing update level and the point cloud update level are B, the 3D stereogram of the sub-area is updated using both the standard remote sensing image and the standard point cloud data.

[0124] If the remote sensing update level is B and the point cloud update level is C, the 3D stereogram of the sub-area is updated using the standard remote sensing image and the stored point cloud;

[0125] If the remote sensing update level is C and the point cloud update level is B, the 3D stereogram of the sub-area is updated using standard point cloud data and stored images;

[0126] If the remote sensing update level and point cloud update level are both C, the 3D stereogram of the sub-area will not be updated;

[0127] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any modification or equivalent replacement of the above embodiments made according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of the technical solution of the present invention.

Claims

1. A spatial data governance method based on data twins, characterized by: The following steps are involved: Divide the target area into several sub-areas, obtain the spatial data of each sub-area and pre-process it to obtain the standard spatial data of each sub-area. The standard spatial data includes standard remote sensing images and standard point cloud data. Perform image segmentation on the standard remote sensing image to obtain the remote sensing feature set corresponding to each object in the sub-area; Perform point cloud segmentation on the standard point cloud data to obtain the point cloud feature set of each object in the sub-area; Perform feature matching on the remote sensing feature set corresponding to the objects in the standard remote sensing image of the sub-region and the point cloud feature set corresponding to the objects in the standard point cloud data to obtain a three-dimensional stereogram of the target area; Analyze the standard remote sensing images corresponding to each sub-region to obtain the remote sensing update level corresponding to each sub-region; Analyze the standard point cloud data corresponding to each sub-area to obtain the point cloud update level corresponding to each sub-area; Determine whether to update the three-dimensional image of the sub-region according to the remote sensing update level and the point cloud update level of the sub-region; The process of analyzing the standard remote sensing images corresponding to each sub-region and obtaining the remote sensing update level corresponding to each sub-region is as follows: S1: Acquire each object and its corresponding coordinates in the standard remote sensing image; and simultaneously acquire each object and its corresponding coordinates in the sub-region storage image; S2: Obtain the spatial distances between the objects in the standard remote sensing image and the objects in the stored image one by one through the coordinates of the objects in the standard remote sensing image and the coordinates of the objects in the stored image of the sub-region; S3: Preset spatial distance threshold. When the spatial distance is less than the spatial distance threshold, it is considered that the object in the standard remote sensing image corresponding to the spatial distance is successfully matched with the object in the stored image. S4: Obtain the number of successful matches, the number of objects in the standard remote sensing image, and the number of objects in the stored image. If the three numbers are not completely equal, determine that the remote sensing update level is A. If the three numbers are completely equal, determine whether the objects in the successfully matched standard remote sensing image are the same as the objects in the stored image. If they are not the same, determine that the remote sensing update level is A. If they are the same, proceed to step S5. S5: Analyze the standard remote sensing image to obtain a corresponding quality index, and determine the remote sensing update level based on the obtained quality index.

2. A spatial data governance method based on data twins according to claim 1, characterized in that: The spatial data includes remote sensing images and point cloud data; Perform radiation correction, geometric correction and image enhancement on remote sensing images to obtain standard remote sensing images; The point cloud data is denoised and filtered to obtain standard point cloud data.

3. A spatial data governance method based on data twins according to claim 2, characterized in that: The process of segmenting a standard remote sensing image and obtaining the remote sensing feature set corresponding to each object in the sub-area is as follows: Use the Canny algorithm to perform convolution operations on standard remote sensing images, identify the pixel points at the edge of the object, and obtain the pixel point images corresponding to each object; Perform expansion and erosion operations on the pixel image to obtain the contour image of each object; Feature extraction is performed on the contour image to obtain the remote sensing feature set corresponding to each object. The remote sensing feature set includes the area, perimeter and center point coordinates of the contour image.

4. A spatial data governance method based on data twins according to claim 3, characterized in that: The process of performing point cloud segmentation on standard point cloud data and obtaining the point cloud feature set of each object in the sub-area is as follows: Extract the geometric features of each point in the point cloud data as the classification features of the corresponding points; Based on the classification features, the k-means clustering algorithm is used to perform cluster analysis on the points in the point cloud data to obtain the point sets corresponding to each object in the target area; The three-dimensional shape of the corresponding object is restored based on the point set, and the point cloud feature set corresponding to the object is obtained based on the three-dimensional shape of the object.

5. A spatial data governance method based on data twins according to claim 4, characterized in that: The process of obtaining a three-dimensional image of the target area is as follows: For each object in the standard remote sensing image, obtain the matching value with each object in the standard point cloud data one by one; A matching value threshold is preset, and when the matching value is less than the matching value threshold, an association relationship is established between the object in the standard remote sensing image corresponding to the matching value and the object in the standard point cloud data corresponding to the matching value; Generate a 3D model of the object using the standard remote sensing image and standard point cloud data corresponding to the object and map it into the virtual space to obtain a 3D stereogram of each sub-area of ​​the target area; The three-dimensional stereograms of the sub-areas are combined to obtain a three-dimensional stereogram of the target area.

6. A spatial data governance method based on data twins according to claim 5, characterized in that: The standard remote sensing image is analyzed to obtain the corresponding quality index. The process of determining the remote sensing update level based on the obtained quality index is as follows: Obtain the color space corresponding to the standard remote sensing image, obtain several color intervals through the color space, and obtain the pixel frequency value in each color interval; The information entropy formula is used to process the pixel frequency values ​​in all color intervals to obtain the color index of the standard remote sensing image; Convert the standard remote sensing image into a grayscale image, use the Laplace algorithm to get the value of each pixel in the grayscale image as the grayscale change value of the corresponding pixel, and obtain the variance of the grayscale change values ​​of all pixels as the pixel wave value; Obtain the intensity value of each pixel in the grayscale image, and use the standard deviation formula to process the intensity values ​​of all pixels on the grayscale image to obtain the global contrast of the grayscale image; Preset several global contrast intervals, different intervals correspond to different contrast scores, match the global contrast interval where the global contrast is located, and thus obtain the contrast score corresponding to the standard remote sensing image; The color index, pixel wave value and global contrast are fused to obtain the quality index of the standard remote sensing image; Obtain the quality index corresponding to the stored image, preset the image update condition, and determine whether the image update condition is met based on the quality index of the standard remote sensing image. When the image update condition is met, the remote sensing update level is determined to be B, otherwise the remote sensing update level is determined to be C.

7. A spatial data governance method based on data twins according to claim 6, characterized in that: The process of analyzing the standard point cloud data corresponding to each sub-area and obtaining the point cloud update level corresponding to each sub-area is as follows: Obtain the total number of points in the sub-region and the volume of the sub-region of the standard point cloud data, and divide the total number of points by the volume to obtain the average point density of the sub-region; Obtain the normal vector angles of all neighboring points in the standard point cloud data to obtain an angle set, preset a normal vector angle threshold, and the number of elements in the angle set that are greater than the normal vector angle threshold is used as the anomaly index; According to the average point density and anomaly index, the quality index of the standard point cloud data is obtained; Obtain the quality index of the stored point cloud, preset the point cloud update conditions, and determine whether the point cloud update conditions are met based on the quality index of the standard point cloud data. When the point cloud update conditions are met, the point cloud update level is determined to be B, otherwise, the point cloud update level is determined to be C.

8. The spatial data governance method based on data twins according to claim 7 is characterized in that: The process of jointly determining whether to update the 3D image of a sub-region based on the remote sensing update level and the point cloud update level of the sub-region is as follows: If the remote sensing update level is A or both the remote sensing update level and the point cloud update level are B, the 3D stereogram of the sub-area is updated using both the standard remote sensing image and the standard point cloud data. If the remote sensing update level is B and the point cloud update level is C, the 3D stereogram of the sub-area is updated using the standard remote sensing image and the stored point cloud; If the remote sensing update level is C and the point cloud update level is B, the 3D stereogram of the sub-area is updated using standard point cloud data and stored images; If the remote sensing update level and the point cloud update level are both C, the 3D stereogram of the sub-area will not be updated.

Images