A method for extracting road intersections based on the center line

Abstract

The present invention discloses a method for extracting road intersections based on the center line. The method specifically includes the following steps: S1, obtaining road surface data and its corresponding center line data, establishing a topological relationship of line-surface mixing, and extracting hanging nodes on the center line; S2, according to the extraction result of the hanging nodes, breaking the dimension-reduced road line elements at each hanging node; S3, obtaining intersection nodes, and extracting the intersection surface according to the broken road line elements and the intersection nodes. The method of the present invention can successfully extract intersection data and can be directly superimposed on the road surface data for display, achieving the purpose of highlighting intersections and quickly generating maps in road mapping, covering intersection types such as T-shaped intersections, cross-shaped intersections, Y-shaped intersections, X-shaped intersections, and ring-shaped intersections, and having strong generalization ability.

Description

Technical Field

[0001] The present invention belongs to the technical field of cartography, and particularly relates to a method for extracting road intersections based on centerlines. Background Art

[0002] Road intersections are important hubs commonly seen in traffic networks and are also important components of urban road systems. With the development of road construction, the crisscrossing road networks have become increasingly complex. In basic geographic information data, road intersections, as the necessary places for vehicle gathering, turning, and evacuation, are high-incidence areas of traffic accidents and have important application values in vehicle control, influencing factors of single-vehicle accident severity, crowd evacuation, particulate matter diffusion laws, etc.

[0003] Nowadays, many scholars have conducted a lot of research on the automatic extraction of road intersections. In terms of big data, Tang Luliang et al. identified turning point clusters and their center points using vehicle trajectory data, and used spatial clustering means to extract intersections and identify various intersection structures, but what was obtained were point clusters near intersections; in terms of image processing, Li Runsheng et al. proposed a method for extracting intersections from high-resolution images using feature semantic rules, but what was obtained was intersection information in a raster model; Chen Guang et al. proposed a method based on the quantitative mapping relationship between pixel structures and intersection structures, and endowed feature semantic information to locate the intersection center and intersection structure, but this method only obtained the intersection center point and intersection structure information in a raster model. The aforementioned studies have all extracted intersection information under different application backgrounds, but in the context of road mapping, the aforementioned methods cannot directly express the information of intersection planar elements during the mapping process, which affects the mapping efficiency of basic base maps. This is mainly because the data forms used by these methods and the limitations of the method application process result in the generated results being unable to be directly used as the vector intersection planar result data required by mapping requirements. Summary of the Invention

[0004] Aiming at the above deficiencies in the prior art, the purpose of the present invention is to provide a method for extracting road intersections based on centerlines. The method of the present invention successfully extracts the intersection planar data between road surface data by analyzing the topological features of road centerlines and the spatial distribution relationship with road surface data.

[0005] In order to achieve the above invention purpose, the technical solution adopted by the present invention is as follows:

[0006] Provide a method for extracting road intersections based on centerlines, the method comprising the following steps:

[0007] S1, obtain road surface data and its corresponding centerline data, establish a topological relationship of line-plane mixing, and extract the hanging nodes on the centerline;

[0008] S2. According to the hanging node extraction result, break the dimension-reduced road line features at each hanging node;

[0009] S3. Obtain intersection nodes, and extract the intersection surfaces according to the broken road line features and intersection nodes.

[0010] Furthermore, the specific method of step S1 includes the following sub-steps:

[0011] S1-1. Obtain road surface data and its corresponding center line data;

[0012] S1-2. Topologically preprocess the center line, segmentize the center line arcs, extract the topological nodes on the center line, and select the nodes with only one associated arc among all topological nodes as hanging nodes;

[0013] S1-3. Reduce the dimension of the road surface features and convert them into line features;

[0014] S1-4. Create an association relationship between the hanging nodes and the road line features; draw a rectangular box centered on each hanging node according to the association threshold (ε), search for the line features that spatially intersect with the rectangular box, associate this hanging node with the line feature and record it in dictionary form until all hanging nodes are traversed.

[0015] Furthermore, the specific method of step S3 includes the following sub-steps:

[0016] S3-1. Obtain intersection nodes: Using the topological relationship, obtain the nodes with the number of associated arcs greater than or equal to 3 among the center line topological nodes, and this node is the intersection node;

[0017] S3-2. Extract the boundary arc information associated with the intersection: Under the preset buffer threshold (θ), create a rectangular box with the intersection node, search for the boundary arcs that spatially intersect with the rectangular box, and when the number of boundary arcs is greater than or equal to 3, the extraction of the boundary arc information associated with the intersection is completed;

[0018] S3-3. Calculate the first nearest point: Calculate the first nearest point from the intersection node to the associated boundary arc;

[0019] S3-4. Calculate the second nearest point: Calculate the second nearest points from each first nearest point to the two adjacent associated boundary arcs;

[0020] S3-5. Construct the intersection surface: Connect the second nearest points in ascending order according to the azimuth angle size order between the intersection node and its related second nearest points to construct the surface, and obtain the intersection surface.

[0021] Furthermore, in step S3-2, according to the formula:

[0022] θ = σ * 2 α, α = {α|α ∈ [0, 10), α ∈ N}

[0023] Obtain a preset buffer threshold θ; where σ is the initial buffer threshold for intersection recognition, and α is the number of recognition times.

[0024] The beneficial effects of the present invention are as follows:

[0025] The method of the present invention can successfully extract intersection data and directly overlay it on the road surface data for display, achieving the purpose of highlighting intersections and quickly generating maps in road mapping, covering intersection types such as T-shaped intersections, cross-shaped intersections, Y-shaped intersections, X-shaped intersections, and roundabout intersections, with strong generalization ability. Description of the Drawings

[0026] Figure 1 It is a schematic diagram of the extraction of hanging nodes provided by an embodiment of the present invention;

[0027] Figure 2 It is a schematic diagram of the processing process of the special mode of hanging nodes provided by an embodiment of the present invention;

[0028] Figure 3 It is a schematic diagram of the interruption process of road line elements provided by an embodiment of the present invention;

[0029] Figure 4 It is a schematic diagram of the extraction of the associated boundary arc of an intersection provided by an embodiment of the present invention;

[0030] Figure 5 It is a schematic diagram of the calculation of the first nearest point provided by an embodiment of the present invention;

[0031] Figure 6 It is a schematic diagram of the process of obtaining the second nearest point provided by an embodiment of the present invention;

[0032] Figure 7 It is a schematic diagram of the intersection surface provided by an embodiment of the present invention;

[0033] Figure 8 It is a test result diagram of the intersection surface provided by an embodiment of the present invention. Detailed Embodiments

[0034] The following describes the detailed embodiments of the present invention to facilitate those skilled in the art to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the detailed embodiments. For those ordinary skilled in the art, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions made using the concept of the present invention are within the scope of protection.

[0035] Embodiment

[0036] A method for extracting road intersections based on the center line, the method comprising the following steps:

[0037] S1. Obtain road surface data and its corresponding center line data, establish a topological relationship of line-plane mixing, and extract hanging nodes on the center line;

[0038] As Figure 1 shown, first, preprocess the center line topologically, extract the topological nodes on the center line, and screen the nodes among all topological nodes that have only one associated arc segment as the initial candidate set of hanging nodes; then, reduce the dimension of the road surface elements to road line elements; finally, create an association relationship between the hanging nodes and the road line elements, that is, capture the hanging nodes around the road line elements, draw a rectangular box centered on each hanging node according to the association threshold (ε), and this association threshold (ε) is affected by the center line. In the method of the present invention, the center line is extracted from the road surface, and its data range size is consistent with the road surface. Therefore, the value of the association threshold (ε) is very small, and it can be set to 0.1 unit. Search for the line elements that spatially intersect with this rectangular box, associate this hanging node to this line element and record it in dictionary form until all hanging nodes are traversed. Only then is the task of extracting hanging nodes completed.

[0039] S2. According to the extraction result of the hanging nodes, break the reduced road line elements at each hanging node;

[0040] First, process the special cases often existing in the road surface as Figure 2 shown. Here, there is only one hanging node, and there is only one associated road line element (appearing as an "inner hole" connected end to end), which will affect the number of nearest points obtained in the next subsection and make it impossible to form a surface or the surface shape is not good. The specific judgment method is: whether the number of hanging nodes associated with the current road line element is equal to 1 and whether it is closed. If the conditions are met, calculate the nearest point from the intersection node to the road line element and the distance D temp , take twice of D temp as the buffer threshold to establish a rectangular box at the intersection, and use the intersection point that intersects with the road line element as the hanging node associated with the road line element; finally, based on the above updated hanging nodes, break the road line elements to obtain multiple boundary arcs that make up the road, as Figure 3 shown.

[0041] S3. Extract the intersection surface according to the broken road line elements and the intersection nodes.

[0042] Specifically, the extraction of the intersection surface is mainly divided into three parts: locating the intersection nodes, obtaining the nearest points, and forming the intersection surface.

[0043] Intersection Node Location: Using the topological relationship, obtain the nodes in the centerline topological nodes where the number of associated arc segments is greater than or equal to 3. This node is the intersection node; under the preset buffer threshold (θ), as Figure 4 shown, a rectangular frame is established at the intersection node, and the boundary arcs that spatially intersect with the rectangular frame are searched. When the number of boundary arcs is greater than or equal to 3, the extraction of the boundary arc information associated with the intersection is completed. Among them, the preset buffer threshold (θ) is the product of the initial intersection recognition buffer threshold (σ) and the number of recognition times (α). The boundary arcs around the intersection are captured within the number of recognition times (α), and once spatial intersection occurs, the recognition process can be terminated.

[0044] The buffer threshold (θ) is calculated by the following formula:

[0045] θ = σ * 2 α , α = {α|α ∈ [0, 10), α ∈ N}

[0046] In the formula, the preset maximum value of the number of recognition times (α) is 10 times, but the 10th operation is not included. This value can be changed accordingly according to actual needs.

[0047] Calculation of the Nearest Point: For the first calculation of the nearest point, as Figure 5 shown, calculate the first nearest point from the intersection node to the associated boundary arc; for the second calculation of the nearest point, this step is to make the shape of the intersection surface more ideal. As Figure 6 shown, calculate the second nearest points on the two associated boundary arcs before and after each first nearest point. It should be noted here that the calculated second nearest points may coincide with the first nearest points, but it does not affect the construction of the intersection surface.

[0048] Construction of the Intersection Surface: Connect the second nearest points. According to the ascending order of the azimuth angles between the intersection node and its relevant second nearest points, connect the second nearest points in ascending order to construct the surface, and obtain the intersection surface as Figure 7 shown.

[0049] As Figure 8 shown, the second column is the global intersection extraction result map of the experimental area, and the left and right columns are the local intersection extraction result maps of the second column respectively. From top to bottom, each row corresponds to the extraction results of common intersection surfaces of three different sets of experimental data. It can be seen from the Figure 8 experimental results of using the method of the present invention for road intersection extraction that this method successfully extracts intersection data and can directly overlay it on the road surface data for display, achieving the purpose of highlighting intersections and quickly generating maps in road mapping, covering T-shaped intersections, cross-shaped intersections, Y-shaped intersections, X-shaped intersections, ring-shaped intersections, and the more complex intersection types in the upper left corner of the figure. Such complex intersections are usually at the viaduct ramp exits.

[0050] It is obvious to those skilled in the art that the present invention is not limited to the details of the above-described exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, in all respects, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Accordingly, all changes that fall within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention.

[0051] In addition, it should be understood that although this specification is described in terms of embodiments, not every embodiment contains only one independent technical solution. This narrative style of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for extracting road intersections based on the center line, characterized in that It includes the following steps: S1. Obtain the road surface data and its corresponding centerline data, establish a topological relationship of line-plane mixing, and extract the hanging nodes on the centerline; S2. According to the extraction result of the hanging nodes, break the dimension-reduced road line features at each hanging node; S3. Obtain the intersection nodes, and extract the intersection surfaces according to the broken road line features and the intersection nodes; The specific method of step S1 includes the following sub-steps: S1-1. Obtain the road surface data and its corresponding centerline data; S1-2. Perform topological preprocessing on the centerline, segmentize the centerline arcs, extract the topological nodes on the centerline, and select the nodes with only one associated arc among all topological nodes as the hanging nodes; S1-3. Reduce the dimension of the road surface features and convert them into line features; S1-4. Create an association relationship between the hanging nodes and the road line features; draw a rectangular box centered on each hanging node according to the association threshold ε, search for the line features that spatially intersect with the rectangular box, associate this hanging node with the line feature and record it in dictionary form until all hanging nodes are traversed.

2. The method for extracting road intersections based on the center line according to claim 1, wherein The specific method of step S3 includes the following sub-steps: S3-1. Obtain the intersection nodes: Using the topological relationship, obtain the nodes with the number of associated arcs greater than or equal to 3 among the centerline topological nodes, and this node is the intersection node; S3-2. Extract the boundary arc information associated with the intersection: Under the preset buffer threshold θ, establish a rectangular box with the intersection node, search for the boundary arcs that spatially intersect with the rectangular box, and when the number of boundary arcs is greater than or equal to 3, the extraction of the boundary arc information associated with the intersection is completed; S3-3. Calculate the first nearest point: Calculate the first nearest point from the intersection node to the associated boundary arc; S3-4. Calculate the second nearest point: Calculate the second nearest points from each first nearest point to the two adjacent associated boundary arcs; S3-5. Construct the intersection surface: According to the azimuth angle size order between the intersection node and its related second nearest points, connect the second nearest points in ascending order to construct the surface and obtain the intersection surface.

3. The method for extracting a road intersection based on a center line according to claim 2, wherein In step S3-2, according to the formula: θ = σ * 2 α , α = {α | α ∈ [0, 10), α ∈ N} Obtain the preset buffer threshold θ; where σ is the initial buffer threshold for intersection recognition, and α is the number of recognition times.

Images