A semi-supervised merging method and system for building cross-scale under scale factor constraints

Abstract

This invention discloses a semi-supervised method and system for cross-scale building merging under scale factor constraints. The method includes: Step 1: acquiring vector building data, a synthesized target scale, and relevant scale factors; Step 2: preprocessing the vector building data according to the relevant scale factors and constructing a building relationship proximity map; Step 3: obtaining various candidate line segments for building bridging surfaces based on the building relationship proximity map; Step 4: extracting features from the various candidate line segments for building bridging surfaces and inputting the features into a classification support vector machine for filtering; Step 5: merging vector buildings based on the filtered candidate line segments for building bridging surfaces. The semi-supervised cross-scale building merging method proposed in this invention, by introducing scale factors and semi-supervised learning techniques, achieves intelligent merging of buildings at different scales, improving the accuracy of geographic information representation and the practicality of maps.

Description

Technical Field

[0001] This invention relates to the field of cartographic generalization technology, and in particular to a method and system for semi-supervised merging of buildings across scales under scale factor constraints. Background Technology

[0002] Buildings, as a crucial component of geospatial vector databases, are fundamental to expressing spatial phenomena, supporting spatial analysis, and providing spatial services. They are core elements of large-scale urban maps and significantly impact map representation. When map scales are reduced, the resulting crowding and overlapping of map content necessitates operations such as combination, deletion, typification, or collapse of buildings. Among these methods, combination has consistently been the primary integration operator. Combining operations constitute the most common hierarchical structure among buildings on maps of different scales. It aims to combine separated buildings by filling space when the scale is reduced. By considering building shape, spatial distribution, and human perception, it replaces a group of visually inseparable adjacent buildings at a larger scale with individual buildings at a smaller scale, clearly representing the polygonal features of buildings on small-scale maps.

[0003] With the rapid development of computer cartography and geographic information systems, building merging has evolved from traditional manual mapping to automated cartographic generalization, achieving significant progress. For a long time, research on building polygon merging has been divided into two categories: merging methods based on raster data and merging methods based on vector data. Utilizing vector data for learning is more suitable for cartographic generalization processes, and this is expected to become an important research direction for future intelligent cartographic generalization. Building merging, as a crucial aspect of cartographic generalization, has always been one of the key focuses and challenges in cartographic generalization research both domestically and internationally.

[0004] Currently, users have higher requirements for the timeliness and consistency of digital spatial information. With the development of Geographic Information Systems (GIS) and remote sensing technology, map representation at different scales has become increasingly important. The scale factor, referring to the ratio of a distance on a map to the actual distance on the ground, is a core factor affecting map generalization. Currently, few studies on cartographic generalization consider the impact of the scale factor on the reasonableness of the results. Reasonable map generalization can effectively convey geographic information at different scales, while unreasonable generalization may lead to information distortion or loss. Therefore, studying the impact of the scale factor on map generalization is of great significance. The scale factor is a crucial parameter in the map generalization process, directly affecting the accuracy of geographic information representation and the user experience. Summary of the Invention

[0005] To at least partially address the problem of geographic information distortion or loss caused by inappropriate scale factor selection, this invention provides a semi-supervised method and system for cross-scale building merging under scale factor constraints. The method involves processing vector building data by acquiring scale factors, constructing candidate line segments for building bridging surfaces, filtering these candidate line segments using a single-class support vector machine, and then merging an appropriate number of buildings using these filtered candidate line segments. This invention addresses the problem of intelligent cross-scale building merging under scale factor constraints, enabling more effective transmission of geographic information and improving the practicality and readability of maps.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows:

[0007] The first aspect of this invention proposes a semi-supervised merging method for buildings across scales under scale factor constraints, comprising:

[0008] Step 1: Obtain vector building data, determine the target scale after integration and the relevant scale factors of the target scale, so as to facilitate the processing of vector building data and the construction of building relationship proximity map;

[0009] Step 2: Preprocess the vector building data according to the relevant scale factors, and construct a building relationship proximity map based on the preprocessed vector building data to facilitate subsequent feature extraction;

[0010] Step 3: Obtain candidate line segments for building bridging surfaces based on the building relationship proximity map;

[0011] Step 4: Extract features from candidate line segments of various building bridging surfaces, input the features into a classification support vector machine for filtering, and obtain the filtered candidate line segments of building bridging surfaces, which facilitates vector building merging;

[0012] Step 5: Merge vector buildings based on the selected candidate line segments of building bridging surfaces.

[0013] Furthermore, the relevant scale factors include a detail threshold that can be synthesized, a spacing threshold for dividing and merging groups, an area threshold for the synthesized open space, and an area threshold for the synthesized independent buildings, which facilitates vector building merging.

[0014] Furthermore, the preprocessing of the vector building data according to relevant scale factors specifically includes:

[0015] Remove buildings whose area on the map is too small to be combined after scaling changes, to facilitate the merging of independent buildings;

[0016] Remove shared arcs between buildings to merge topologically adjacent buildings, facilitating the merging of topologically adjacent buildings.

[0017] Furthermore, the step of obtaining multiple candidate line segments for building bridging surfaces based on the building relationship proximity map specifically includes:

[0018] In the building relationship proximity diagram, adjacent buildings are connected by connecting edges. The two buildings connected by the connecting edge are merged into a merge group for easier subsequent processing.

[0019] Construct a minimum bounding rectangle outside the merged group, and generate adjacent edges between the two merged groups under the constraints of the minimum bounding rectangle;

[0020] Candidate line segments for building bridging surfaces are constructed based on adjacent edges. These candidate line segments include candidate line segments perpendicular to the endpoints of adjacent edges, candidate line segments extending from adjacent edges, and candidate line segments perpendicular to the angle bisector of the angle between the extensions of adjacent edges.

[0021] Furthermore, the step of constructing candidate line segments for building bridging surfaces based on adjacent edges specifically includes:

[0022] In the merged group, extract the endpoints of the adjacent edges of the buildings and draw perpendicular lines to the adjacent edges of other buildings in the merged group. The perpendicular lines are candidate line segments for perpendicular lines to the endpoints of the adjacent edges.

[0023] The adjacent edge of the extended building polygon intersects the adjacent edge of another building in the merged group, and the extended line segment is a candidate line segment for the extension line of the adjacent edge;

[0024] In the merged group, extend the adjacent sides of the building polygon and obtain the intersection point. Remove the intersection points located on the adjacent line or outside the smallest bounding rectangle of the merged group, and retain the remaining intersection points as valid intersection points. Draw the angle bisector of the valid intersection point, and draw a perpendicular line through the angle bisector to intersect the adjacent side. The perpendicular line segment of the angle bisector of the adjacent side is a candidate line segment of the perpendicular line segment of the angle bisector of the adjacent side extension line.

[0025] Furthermore, the features extracted from candidate line segments of various building bridging surfaces include the spatial location features of the candidate line segments, the geometric features of the candidate line segments, the connection features between the candidate line segments and adjacent edges, and the intersection features between the candidate line segments and the minimum bounding rectangle of the building.

[0026] Furthermore, the spatial location features of the candidate line segment include the coordinates of its two endpoints and the coordinates of its midpoint;

[0027] The geometric features of the candidate line segments include the length, direction, and degree of variation of the candidate line segments; the degree of variation includes the absolute value of the difference between the direction of a candidate line segment and the average direction of all candidate line segments in the building merging group;

[0028] The connection characteristics between the candidate line segment and the adjacent edge include the absolute value of the difference between the direction of the candidate line segment and the direction of the line connecting the centroid of the minimum bounding rectangle of the building, and the minimum distance between the two endpoints of the candidate line segment and the endpoints of the adjacent edge.

[0029] The intersection characteristics of the candidate line segment and the minimum bounding rectangle of the building include the two intersection points of the candidate line segment and the minimum bounding rectangle and the minimum distance between the endpoints of the minimum bounding rectangle.

[0030] A second aspect of this invention proposes a semi-supervised merging system for buildings across scales under scale factor constraints, comprising:

[0031] The collection module is used to acquire vector building data, determine the integrated target scale and the relevant scale factors of the target scale, so as to facilitate the processing of vector building data and the construction of building relationship proximity maps;

[0032] The processing module is used to preprocess the vector building data according to relevant scale factors, and to construct a building relationship proximity map based on the preprocessed vector building data to facilitate subsequent feature extraction.

[0033] The candidate line segment module is used to obtain various candidate line segments for building bridging surfaces based on the building relationship proximity map.

[0034] The filtering module is used to extract features of candidate line segments for various building bridging surfaces. The features are then input into a classification support vector machine for filtering to obtain the filtered candidate line segments for building bridging surfaces, which facilitates vector building merging.

[0035] The merging module is used to merge vector buildings based on the filtered candidate line segments of building bridging surfaces.

[0036] A third aspect of the present invention provides an electronic device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements a semi-supervised merging method for building cross-scale under scale factor constraints as described in the first aspect above.

[0037] A fourth aspect of the present invention provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device on which the storage medium is located to execute a semi-supervised merging method for buildings across scales under scale factor constraints as described in the first aspect above.

[0038] The beneficial effects of this invention are:

[0039] (1) Improving the accuracy of geographic information representation: By introducing a scale factor as a key constraint in the merging process, this invention can ensure that the merging results of buildings at different scales are more accurate. This helps to reduce information distortion or loss, more realistically reflect the spatial distribution and morphological characteristics of geographic entities, and improve the accuracy of geographic information representation.

[0040] (2) Enhanced map usability and readability: This invention intelligently merges buildings, making the representation of buildings on the map more concise and clear. The merged buildings not only retain key spatial structural information but also reduce redundant details, improving the map's readability and usability. Users can obtain the information they need more quickly, improving the efficiency of map usage.

[0041] (3) Improved Data Processing Efficiency: Compared to traditional manual merging methods, this invention employs semi-supervised learning technology to automate the merging of buildings. This significantly reduces the need for manual intervention and improves the speed and efficiency of data processing. Furthermore, through optimized algorithms and parallel processing techniques, this invention can handle large-scale datasets, meeting the needs of real-time applications.

[0042] (4) Enhanced system flexibility and adaptability: The cross-scale merging method proposed in this invention has strong flexibility and adaptability. It can handle building data at different scales and intelligently merge data according to actual needs. In addition, the system architecture and algorithm design of this invention are easy to extend and modify to adapt to new needs and technical challenges that may arise in the future.

[0043] (5) Promoting Geographic Information Data Sharing and Interoperability: By achieving cross-scale merging of building data, this invention helps promote the sharing and interoperability of geographic information data. The merged building data is more standardized and unified, facilitating data exchange and integration between different systems. This helps break down information silos and promotes the integration and utilization of geographic information resources. Attached Figure Description

[0044] Figure 1 The flowchart illustrates a semi-supervised merging method for buildings across scales under scale factor constraints, as provided in an embodiment of the present invention.

[0045] Figure 2 This is a schematic diagram of the vector buildings after merging, provided as an embodiment of the present invention.

[0046] Figure 3 The following is a flowchart illustrating a semi-supervised merging method for buildings across scales under scale factor constraints, provided as an embodiment of the present invention.

[0047] Figure 4 This is a schematic diagram of the adjacent edges of the merged group provided in an embodiment of the present invention.

[0048] Figure 5 This is a schematic diagram illustrating the construction of three types of candidate line segments provided in an embodiment of the present invention.

[0049] Figure 6 This is a schematic diagram illustrating the features of candidate line segments for building bridging surfaces provided in an embodiment of the present invention.

[0050] Figure 7 This is a schematic diagram illustrating feature extraction of candidate line segments for building bridging surfaces provided in an embodiment of the present invention.

[0051] Figure 8 This is an architecture diagram of a semi-supervised merging system for buildings across scales under scale factor constraints, provided as an embodiment of the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0053] Example 1

[0054] like Figure 1 As shown, a semi-supervised merging method for buildings across scales under scale factor constraints includes:

[0055] S101: Obtain vector building data and determine the integrated target scale and related scale factors of the target scale.

[0056] Specifically, this involves searching for relevant scale factors for the target scale in online resources, literature, and library books, including:

[0057] (1) Relevant scaling factors include detail thresholds that can be synthesized.

[0058] (2) The spacing threshold for dividing the merged groups.

[0059] (3) Area threshold of the overall empty space expression.

[0060] (4) Threshold for the area of ​​independent buildings that can be integrated.

[0061] Taking the requirements of 1:1000 or 1:5000 topographic maps as an example: Convex and concave areas less than 1mm on the map can generally be represented collectively; building spacing less than 1.5mm on the map can be represented collectively; minor streets and alleys can be appropriately omitted; open spaces can be represented by omissions or omissions based on the characteristics of residential areas in the north and south, with the omission index generally being 4mm on the map.2 -9mm 2 For detached houses, those with a length less than 1.0 mm and a width less than 0.7 mm on the map are represented by symbols not based on scale.

[0062] S102: Preprocess the vector building data according to the relevant scale factors, and construct a building relationship proximity map based on the preprocessed vector building data.

[0063] S103: Obtain multiple candidate line segments for building bridging surfaces based on the building relationship proximity diagram.

[0064] S104: Extract features of candidate line segments for various building bridging surfaces, input the features into a classification support vector machine for filtering, and obtain the filtered candidate line segments for building bridging surfaces.

[0065] Specifically, the candidate line segments for various building bridge surfaces were divided into three datasets: 60% training set, 20% validation set, and 20% test set. The 15% training set was manually labeled. The features of the extracted candidate line segments were standardized and then input into an Ordinary Classification Support Vector Machine (OCSVM). A kernel function was used to construct the OCSVM, and it was trained using a training set consisting of partially labeled and unlabeled data. After training, the trained OCSVM model was used to classify the test set and predict the classes of candidate line segments to be retained. Cross-validation was used to select the optimal parameters of the OCSVM classifier.

[0066] To evaluate the classification performance of the algorithm, it is necessary to compare it with the results of human classification to confirm the classification effect. The accuracy (P), recall (R), and F-value are calculated using the following formulas:

[0067]

[0068] Where P is the classification accuracy, R is the recall, F is the F-value, TP is the number of correctly classified cases, FP is the number of misclassified cases, and FN is the number of missed cases.

[0069] S105: Merge vector buildings based on the selected candidate line segments of building bridging surfaces.

[0070] Specifically, the building bridging surface is formed by the candidate line segments of the selected building bridging surface and the adjacent edges of the building, and then extracted together with the edges of the original merged group to form the outer contour of the building, thus realizing the merging of vector buildings.

[0071] During the merging process, the extracted outer contour is processed based on a scale factor to integrate convex and concave details, and the extracted outer contour is represented using open space to make the merged result closer to the true value. Unreasonable shapes in the extracted outer contour, such as bridging surface intersections, are reprocessed after bridging surface reconstruction. Specifically, after obtaining the features of candidate line segments for building bridging surfaces, such as... Figure 2 As shown in (a), the minimum convexity and concavity details of the extracted outer contour are synthesized, as follows: Figure 2 As shown in (b), the extracted outer contour is represented using empty space to make the merged result closer to the true value, as shown in [the diagram]. Figure 2 As shown in (c), unreasonable shapes in the extracted outer contour, such as the intersection of bridging surfaces, are reconstructed and processed.

[0072] When generating candidate line segments for bridging surfaces, this invention not only considers the right-angled features, geometric features, and spatial structural relationships of the building polygons, but also extracts multiple features of the candidate line segments for comprehensive evaluation. These features include spatial location features, geometric features, connection features with adjacent edges, and intersection features with the building's minimum bounding rectangle. Through multi-feature fusion and comprehensive evaluation, this invention can more accurately select the optimal candidate line segments for merging, improving the quality and reliability of the merging results.

[0073] like Figure 3 As shown, this invention preprocesses vector building data by acquiring the target scale and its relevant scale factors. Then, it constructs a proximity map of building relationships within the vector buildings, identifies candidate line segments for building bridging surfaces, and uses a single-class support vector machine to filter these candidate line segments. Finally, it merges the vector buildings based on the filtered candidate line segments. This approach addresses the challenge of intelligent cross-scale building merging under scale factor constraints, enabling more effective communication of geographic information and improving the map's usability and readability.

[0074] Example 2

[0075] Based on the above embodiments, this invention provides a specific process for constructing candidate line segments and extracting features for building bridging surfaces, specifically including:

[0076] A Delaunay triangulation is constructed using preprocessed vector building data, skeleton lines are generated, and finally, a building proximity map is produced. In the proximity map, each pair of buildings is connected by a connecting edge. The length of the connecting edge is extracted, and based on scale factor constraints (the spacing threshold for dividing merge groups), the connecting edges between buildings that need to be merged are retained, and the two buildings connected by the connecting edge are grouped into one merge group. For example... Figure 4 As shown, neighbor edges are constructed based on the merged groups.

[0077] Based on the adjacent edges, three types of candidate line segments for building bridging surfaces are constructed: candidate line segments of perpendicular lines from the endpoints of adjacent edges, candidate line segments of the extensions of adjacent edges, and candidate line segments of perpendicular lines from the angle bisectors of the angles between the extensions of adjacent edges.

[0078] like Figure 5 As shown in (a), the endpoints of the adjacent edges of buildings in the merged group are extracted and perpendicular to the adjacent edges of another building in the merged group. This perpendicular line is a candidate line segment for the perpendicular line from the endpoints of the adjacent edges.

[0079] like Figure 5 As shown in (b), the adjacent edge of the extended building polygon intersects the adjacent edge of another building in the merged group, and the extended line segment is a candidate line segment for the extension line of the adjacent edge.

[0080] like Figure 5 As shown in (c), extend the adjacent sides of the building polygon in the merged group to obtain the intersection point. Remove the intersection points located on the adjacent line or outside the smallest bounding rectangle of the merged group, and retain the remaining intersection points as valid intersection points. Draw the angle bisector with the valid intersection point, and draw a perpendicular line through the angle bisector to intersect the adjacent side. The perpendicular line segment of the angle bisector of the adjacent side is the candidate line segment of the perpendicular line segment of the angle bisector of the adjacent side extension line.

[0081] like Figure 6 As shown, the features extracted from multiple candidate line segments of building bridging surfaces include the spatial location features of the candidate line segments, the geometric features of the candidate line segments, the connection features between the candidate line segments and adjacent edges, and the intersection features between the candidate line segments and the minimum bounding rectangle of the buildings.

[0082] The spatial location features of candidate line segments include the coordinates of their two endpoints and the coordinates of their midpoint. Specifically, the spatial location coordinates of the two endpoints and the coordinates of the midpoint of each candidate line segment are extracted. The midpoint coordinates of each candidate line segment are calculated by averaging the coordinates of the two endpoints. If the endpoint coordinates are (x1, y1) and (x2, y2), then the coordinates of the midpoint are:

[0083]

[0084] Where, x m Let x be the x-coordinate of the midpoint, and y be the y-coordinate of the midpoint. m Let (x1, y1) be the y-axis coordinate of the midpoint, and (x2, y2) be the coordinates of the two endpoints of the candidate line segment.

[0085] The geometric features of candidate line segments include their length, direction, and degree of variation.

[0086] The degree of variation includes the absolute value of the difference between the direction of a candidate line segment and the average direction of all candidate line segments in the building's merged group.

[0087] Specifically, the direction of the candidate line segment is represented by the following formula:

[0088] θ = atan2(y2-y1, x2-x1)

[0089] Where θ is the direction of the candidate line segment, atan2 is the arctangent function, which takes into account the signs of the numerator and denominator, and the return value range is [-π, π], that is, [-180°, 180°], and (x1, y1) and (x2, y2) are the coordinates of the two endpoints of the candidate line segment, respectively.

[0090] The connection characteristics between the candidate line segment and the adjacent edge include the absolute value of the difference between the direction of the candidate line segment and the direction of the line connecting the centroid of the minimum bounding rectangle of the building, and the minimum distance between the two endpoints of the candidate line segment and the endpoints of the adjacent edge.

[0091] The intersection characteristics of the candidate line segment and the minimum bounding rectangle of the building include the two intersection points of the candidate line segment and the minimum bounding rectangle and the minimum distance between the endpoints of the minimum bounding rectangle.

[0092] The connection characteristics between candidate line segments and adjacent edges include the consistency between the direction of the candidate line segment and the main direction of the building, and the minimum distance between the two endpoints of the line segment and the endpoints of the adjacent edges.

[0093] The consistency between the candidate line segment direction and the main direction of the building is determined by calculating the absolute value of the difference between the line segment direction and the direction of the centroid of the minimum bounding rectangle of the building, based on the generated minimum bounding rectangle.

[0094] like Figure 7 As shown in (a), the minimum distance between the endpoints of the adjacent edge is the closest distance between the two endpoints of the candidate line segment and the endpoints of the adjacent edge, respectively, based on the generated adjacent edges.

[0095] The intersection characteristics of the candidate line segment building's minimum bounding rectangle include the two intersection points of the line segment with the minimum bounding rectangle and the minimum distance between the endpoints of the minimum bounding rectangle. Based on the constructed merged group of minimum bounding rectangles, the minimum distance between the two intersection points of the candidate line segment with the minimum bounding rectangle and the endpoints of the minimum bounding rectangle is calculated. For example... Figure 7 (b) The lengths of dis1 and dis2 are the minimum distances between the two intersections of the red candidate line and the minimum bounding rectangle and the endpoints of the minimum bounding rectangle.

[0096] Example 3

[0097] Corresponding to the methods mentioned above, such as Figure 8 As shown, a semi-supervised merging system for buildings across scales under scale factor constraints includes:

[0098] The data collection module is used to acquire vector building data and determine the integrated target scale and related scale factors of the target scale.

[0099] The processing module is used to preprocess the vector building data according to relevant scale factors and construct a building relationship proximity map based on the preprocessed vector building data.

[0100] The candidate line segment module is used to obtain various candidate line segments for building bridging surfaces based on the proximity of buildings.

[0101] The filtering module is used to extract features of candidate line segments for various building bridging surfaces, input the features into a classification support vector machine for filtering, and obtain the filtered candidate line segments for building bridging surfaces.

[0102] The merging module is used to merge vector buildings based on the filtered candidate line segments of building bridging surfaces.

[0103] It should be noted that the building cross-scale semi-supervised merging system under scale factor constraints provided in this embodiment of the invention is to implement the building cross-scale semi-supervised merging method under scale factor constraints in the above embodiments. Its specific functions can be referred to the above method embodiments, and will not be repeated here.

[0104] Example 4

[0105] Corresponding to the above method, this embodiment of the invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a semi-supervised merging method for buildings across scales under scale factor constraints as described in the above embodiment.

[0106] This invention provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the storage medium is located to execute a semi-supervised merging method for buildings across scales under scale factor constraints as described in the above embodiments.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A semi-supervised merging method for buildings across scales under scale factor constraints, characterized in that, include: Step 1: Obtain vector building data, determine the integrated target scale and the relevant scale factors of the target scale; Step 2: Preprocess the vector building data according to relevant scale factors, and construct a building relationship proximity map based on the preprocessed vector building data; Step 3: Obtain candidate line segments for building bridging surfaces based on the building relationship proximity map; Step 4: Extract features from various candidate line segments of building bridging surfaces, input the features into a classification support vector machine for filtering, and obtain the filtered candidate line segments of building bridging surfaces; Step 5: Merge vector buildings based on the selected candidate line segments of building bridging surfaces; The process of obtaining multiple candidate line segments for building bridging surfaces based on the building relationship proximity map specifically includes: In the building relationship proximity diagram, adjacent buildings are connected by connecting edges, and the two buildings connected by the connecting edge are merged into a merge group; Construct a minimum bounding rectangle outside the merged group, and generate adjacent edges between the two merged groups under the constraints of the minimum bounding rectangle; Candidate line segments for building bridging surfaces are constructed based on adjacent edges. These candidate line segments include candidate line segments perpendicular to the endpoints of adjacent edges, candidate line segments extending from adjacent edges, and candidate line segments perpendicular to the angle bisector of the included angle of the extended lines of adjacent edges. The process of constructing candidate line segments for building bridging surfaces based on adjacent edges specifically includes: In the merged group, extract the endpoints of the adjacent edges of the buildings and draw perpendicular lines to the adjacent edges of another building in the merged group. The perpendicular lines are candidate line segments for perpendicular lines to the endpoints of the adjacent edges. Extend the adjacent edge of the building polygon to intersect the adjacent edge of another building in the merge group. The line segment obtained by the extension operation is the candidate line segment of the adjacent edge extension line. In the merged group, extend the adjacent sides of the building polygon and obtain the intersection point. Remove the intersection points located on the adjacent line or outside the smallest bounding rectangle of the merged group, and retain the remaining intersection points as valid intersection points. Draw the angle bisector of the valid intersection point, and draw a perpendicular line through the angle bisector to intersect the adjacent side. The perpendicular line segment of the angle bisector of the adjacent side is a candidate line segment of the perpendicular line segment of the angle bisector of the adjacent side extension line.

2. The method for semi-supervised merging of buildings across scales under scale factor constraints according to claim 1, characterized in that, The relevant scale factors include the detail threshold that can be synthesized, the spacing threshold for dividing and merging groups, the area threshold for the overall open space, and the area threshold for individual buildings that can be synthesized.

3. The method for semi-supervised merging of buildings across scales under scale factor constraints according to claim 1, characterized in that, The preprocessing of vector building data based on relevant scale factors specifically includes: Remove buildings whose area on the map is too small to be composited after scaling; Remove shared arcs between buildings to merge adjacent buildings in the topology.

4. The method for semi-supervised merging of buildings across scales under scale factor constraints according to claim 1, characterized in that, The extraction of features for candidate line segments of various building bridging surfaces specifically includes the spatial location features of the candidate line segments, the geometric features of the candidate line segments, the connection features between the candidate line segments and adjacent edges, and the intersection features between the candidate line segments and the minimum bounding rectangle of the building.

5. The method for semi-supervised merging of buildings across scales under scale factor constraints according to claim 4, characterized in that, The spatial location features of the candidate line segment include the coordinates of the two endpoints and the midpoint of the candidate line segment; The geometric features of the candidate line segments include the length, direction, and degree of variation of the candidate line segments; the degree of variation includes the absolute value of the difference between the direction of a candidate line segment and the average direction of all candidate line segments in the building merging group; The connection characteristics between the candidate line segment and the adjacent edge include the absolute value of the difference between the direction of the candidate line segment and the direction of the line connecting the centroid of the minimum bounding rectangle of the building, and the minimum distance between the two endpoints of the candidate line segment and the endpoints of the adjacent edge. The intersection characteristics of the candidate line segment and the minimum bounding rectangle of the building include the two intersection points of the candidate line segment and the minimum bounding rectangle and the minimum distance between the endpoints of the minimum bounding rectangle.

6. A semi-supervised merging system for buildings across scales under scale factor constraints, characterized in that, include: The collection module is used to acquire vector building data and determine the integrated target scale and related scale factors of the target scale. The processing module is used to preprocess the vector building data according to relevant scale factors, and to construct a building relationship proximity map based on the preprocessed vector building data; The candidate line segment module is used to obtain various candidate line segments for building bridging surfaces based on the building relationship proximity map. The filtering module is used to extract features of candidate line segments for various building bridging surfaces, input the features into a classification support vector machine for filtering, and obtain the filtered candidate line segments for building bridging surfaces. The merging module is used to merge vector buildings based on the filtered candidate line segments of building bridging surfaces; The process of obtaining multiple candidate line segments for building bridging surfaces based on the building relationship proximity map specifically includes: In the building relationship proximity diagram, adjacent buildings are connected by connecting edges, and the two buildings connected by the connecting edge are merged into a merge group; Construct a minimum bounding rectangle outside the merged group, and generate adjacent edges between the two merged groups under the constraints of the minimum bounding rectangle; Candidate line segments for building bridging surfaces are constructed based on adjacent edges. These candidate line segments include candidate line segments perpendicular to the endpoints of adjacent edges, candidate line segments extending from adjacent edges, and candidate line segments perpendicular to the angle bisector of the included angle of the extended lines of adjacent edges. The process of constructing candidate line segments for building bridging surfaces based on adjacent edges specifically includes: In the merged group, extract the endpoints of the adjacent edges of the buildings and draw perpendicular lines to the adjacent edges of another building in the merged group. The perpendicular lines are candidate line segments for perpendicular lines to the endpoints of the adjacent edges. Extend the adjacent edge of the building polygon to intersect the adjacent edge of another building in the merge group. The line segment obtained by the extension operation is the candidate line segment of the adjacent edge extension line. In the merged group, extend the adjacent sides of the building polygon and obtain the intersection point. Remove the intersection points located on the adjacent line or outside the smallest bounding rectangle of the merged group, and retain the remaining intersection points as valid intersection points. Draw the angle bisector of the valid intersection point, and draw a perpendicular line through the angle bisector to intersect the adjacent side. The perpendicular line segment of the angle bisector of the adjacent side is a candidate line segment of the perpendicular line segment of the angle bisector of the adjacent side extension line.

7. An electronic device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements a semi-supervised merging method for building cross-scale under scale factor constraints as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the storage medium is located to perform a semi-supervised merging method for buildings across scales under scale factor constraints as described in any one of claims 1 to 5.

Images