Road vector data extraction method, device, equipment and medium based on three-dimensional scene

Abstract

The present application relates to a method, device, equipment and medium for extracting road vector data based on a three-dimensional scene, which extracts the road outer contour based on the road's feature identification in the three-dimensional scene to be subjected to road vector quantity extraction to obtain the corresponding road outer contour data, then extracts the road centerline intersection and the road intersection turning point based on the road outer contour data to obtain the road intersection center point data set and road intersection information, and obtains the road intersection data based on the road intersection center point data set and the road intersection information, and finally realizes the road vector data extraction in the three-dimensional scene based on the road outer contour data and the road intersection data. The present method can efficiently and quickly extract road vector data from the three-dimensional scene.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, equipment and medium for extracting road vector data based on a three-dimensional scene. Background Art

[0002] Extracting road vector data in a highly realistic virtual scene is an important task, which aims to obtain the precise geometry, location and attribute information of the road from a complex virtual environment. This has far-reaching significance for the application development of virtual scenes, such as autonomous driving simulation, urban planning simulation, and traffic flow analysis. By extracting road vector data, the road information in the virtual scene can be made more accurate, providing a reliable basis for subsequent scene analysis and decision-making.

[0003] However, the existing technology has many defects. On the one hand, when using multi-source trajectory data to extract the vector features of the trajectory points of road intersections and using clustering methods to segment them, the precision and accuracy of road intersection information extraction are low due to the limitations of data sampling rate and traditional clustering methods. On the other hand, the use of remote sensing image data to extract road intersections relies on machine learning algorithms to train feature detection networks. Although the detection accuracy is high, the cost is high and the timeliness is low. There are also problems such as ground object occlusion, difficulty in detecting small target intersections, and loss of semantic information. In addition, although the fusion of multi-source trajectory data and remote sensing images can complement each other, it poses new challenges to data fusion and matching, which restricts the effect and application of road vector data extraction. Summary of the invention

[0004] Based on this, it is necessary to provide a road vector data extraction method, device, equipment and medium based on a three-dimensional scene that can efficiently and accurately extract road vector data in response to the above technical problems.

[0005] A method for extracting road vector data based on a three-dimensional scene, the method comprising:

[0006] Obtaining a three-dimensional scene for road vector quantity extraction;

[0007] In the three-dimensional scene, the outer contour of the road is extracted based on the ground feature identification of the road to obtain corresponding outer contour data of the road;

[0008] Based on the road outer contour data, the road centerline intersection and the road intersection turning point are extracted to obtain a road intersection center point data set and road intersection information, and the road intersection data is obtained according to the road intersection center point data set and the road intersection information;

[0009] The road vector data in the three-dimensional scene is extracted according to the road outer contour data and the road intersection data.

[0010] In one embodiment, the extracting of the road outer contour based on the road feature identification includes:

[0011] Extracting mesh data and attribute information in a corresponding road static mesh component in the three-dimensional scene based on the road feature identification, wherein the attribute information includes geometric information;

[0012] Determine the outer boundary shape of the road based on the geometric information, generate outer contour data of the road, and obtain preliminary data of the road;

[0013] The preliminary data is converted from the coordinate system of the three-dimensional scene to the earth coordinate system to obtain the road outer contour data.

[0014] In one embodiment, the road contour data is in ShapeFile format and Geojson format.

[0015] In one embodiment, when extracting the road centerline intersection based on the road outer contour data:

[0016] Converting the road outline data in ShapeFile format into binary raster data, and refining the raster data;

[0017] Traverse the center point of each grid in the refined grid data to construct the initial road graph structure corresponding to the road centerline;

[0018] Based on the initial road graph structure, the Liang Youdong-Barsky algorithm is used to optimize the connection mode of the road centerline to obtain a simplified road graph structure.

[0019] Performing a centerline ring removal operation on the simplified road graph structure to obtain a road graph structure after the centerline ring is removed;

[0020] The minimum distance between adjacent intersections in the actual road intersection information is set as a threshold, all broken line segments smaller than the threshold are removed from the road graph structure after the centerline ring is removed, and the remaining broken line segments are reconnected to obtain a final road graph structure;

[0021] The centerline nodes that meet the preset requirements in the final road graph structure are extracted to construct the road intersection center point data set.

[0022] In one embodiment, the raster data is thinned using Zhang parallel fast thinning algorithm.

[0023] In one embodiment, performing a centerline ring removal operation on the simplified road graph structure includes:

[0024] The simplified road graph structure is enlarged, and a ring structure is located by traversing the edges of nodes whose degrees are greater than a preset value and limiting the step length, wherein the ring structure is composed of at least three nodes;

[0025] Removing nodes with a degree greater than 2 from the ring structure, if the removal fails, replacing the ring structure with the center point of the ring structure as a new node;

[0026] If the removal is successful, when the number of remaining nodes is equal to 2, the ring structure is removed successfully. When the number of remaining nodes is greater than 2, the center point of the ring structure is used as a new node to replace the ring structure.

[0027] In one embodiment, extracting the turning point of the road intersection based on the road outer contour data includes:

[0028] Based on the road contour data in Geojson format, extract the vertex association relationship set of the polygon and the connection relationship index between each vertex, and construct the outer ring structure point set and the hole structure point set according to the vertex association relationship set;

[0029] According to a preset distance threshold and the road intersection center point data set, determining the intersection point set of the road intersection from the outer ring structure point set and the hole structure point set;

[0030] The turning points of the road intersections are extracted according to the intersection point set and the road intersection center point data set.

[0031] The present application also provides a road vector data extraction device based on a three-dimensional scene, the device comprising:

[0032] A three-dimensional scene acquisition module is used to acquire a three-dimensional scene for road vector quantity extraction;

[0033] A road outer contour data extraction module is used to extract the road outer contour based on the road feature identification in the three-dimensional scene to obtain corresponding road outer contour data;

[0034] A road intersection and junction information extraction module, for extracting the road centerline intersection and the road intersection turning point based on the road outer contour data, obtaining a road intersection center point data set and road intersection information, and obtaining road intersection data according to the road intersection center point data set and the road intersection information;

[0035] The road vector data extraction module is used to extract the road vector data in the three-dimensional scene according to the road outer contour data and the road intersection data.

[0036] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0037] Obtaining a three-dimensional scene for road vector quantity extraction;

[0038] In the three-dimensional scene, the outer contour of the road is extracted based on the ground feature identification of the road to obtain corresponding outer contour data of the road;

[0039] Based on the road outer contour data, the road centerline intersection and the road intersection turning point are extracted to obtain a road intersection center point data set and road intersection information, and the road intersection data is obtained according to the road intersection center point data set and the road intersection information;

[0040] The road vector data in the three-dimensional scene is extracted according to the road outer contour data and the road intersection data.

[0041] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0042] Obtaining a three-dimensional scene for road vector quantity extraction;

[0043] In the three-dimensional scene, the outer contour of the road is extracted based on the ground feature identification of the road to obtain corresponding outer contour data of the road;

[0044] Based on the road outer contour data, the road centerline intersection and the road intersection turning point are extracted to obtain a road intersection center point data set and road intersection information, and the road intersection data is obtained according to the road intersection center point data set and the road intersection information;

[0045] The road vector data in the three-dimensional scene is extracted according to the road outer contour data and the road intersection data.

[0046] The above-mentioned road vector data extraction method, device, equipment and medium based on three-dimensional scene, in the three-dimensional scene to be subjected to road vector quantity extraction, extracts the road outer contour based on the road feature identification to obtain the corresponding road outer contour data, then extracts the road centerline intersection and the road intersection turning point based on the road outer contour data, obtains the road intersection center point data set and road intersection information, and obtains the road intersection data based on the road intersection center point data set and the road intersection information, and finally realizes the road vector data extraction in the three-dimensional scene based on the road outer contour data and the road intersection data. The method can be used to efficiently and quickly extract road vector data from the three-dimensional scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic diagram of a process of extracting road vector data based on a three-dimensional scene in one embodiment;

[0048] Figure 2 A schematic diagram of road outer contour data in an embodiment;

[0049] Figure 3 A schematic diagram of a set of points of an outer contour of a road in one embodiment;

[0050] Figure 4 A schematic diagram of intersection information extraction in one embodiment;

[0051] Figure 5 A schematic diagram of specific implementation steps of a method for extracting road vector data based on a three-dimensional scene in one embodiment;

[0052] Figure 6 is an example of an image in a process of extracting road data using the method in an embodiment, wherein: Figure 6 (a) is a schematic diagram of the road outer contour. Figure 6 (b) A schematic diagram showing the centerline of a road and its intersection points. Figure 6 (c) is a schematic diagram of road intersection extraction. Figure 6 (d) Schematic diagram showing road vector data display;

[0053] Figure 7 is a structural block diagram of a road vector data extraction device based on a three-dimensional scene in one embodiment;

[0054] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] The existing road data extraction methods, including road intersection extraction based on multi-source trajectory data, road intersection information extraction based on high-resolution images, and road intersection extraction by fusing multi-source trajectory data and remote sensing image data, have been studied. For example, the vector features of intersection trajectory points extracted from multi-source trajectory data are used to segment intersection trajectory points by clustering and other methods. However, due to the limitations of data sampling rate and traditional clustering methods, the extraction precision of road intersection information is often not high and the accuracy is low. The previous method of extracting road intersections using remote sensing image data mainly relies on the popular machine learning algorithms at this stage to train the feature detection network of intersections and realize the automatic extraction of intersection positions. The detection results are highly accurate, but such methods have high costs, low timeliness, and problems such as ground object occlusion, difficulty in detecting intersections with small targets, and loss of intersection semantic information. When the current method of fusing multi-source trajectory data and remote sensing images is used to extract road intersection information, multi-channel and massive road intersection information can be integrated to complement the advantages and disadvantages of various types of data and make up for the shortcomings of the above two methods. However, new requirements are put forward for data fusion and matching.

[0057] In response to the above problems, Figure 1 As shown, a method for extracting road vector data based on a three-dimensional scene is provided, which specifically includes the following steps:

[0058] Step S100, obtaining a three-dimensional scene for road vector quantity extraction.

[0059] Step S110: In the three-dimensional scene, the outer contour of the road is extracted based on the ground feature identification of the road to obtain corresponding outer contour data of the road.

[0060] Step S120, based on the road outer contour data, the road centerline intersections and the road intersection turning points are extracted to obtain a road intersection center point data set and road intersection information, and the road intersection data is obtained according to the road intersection center point data set and the road intersection information.

[0061] Step S130, extracting road vector data in the three-dimensional scene according to the road outer contour data and the road intersection data.

[0062] In the application, the road vector data and road intersection information are extracted from the highly realistic virtual scene, and support is given to the downstream task data in the specific application scenario. Among them, the application scenario refers to the three-dimensional highly realistic virtual scene built by the Unreal Engine platform, and the existing Unreal Engine platform is mostly built by foreign countries. In order to avoid over-reliance on the Unreal Engine platform, improve the deduction operation rate, meet the needs of localization of the later system, and update the two-dimensional map by the extracted vector data, update the two-dimensional and three-dimensional road map in real time, and extract road vector data based on the three-dimensional scene built by the Unreal Engine, perform tasks and rules planning and decision-making functions, supplement and improve the simulation environment road data resources, and provide vector data resource support for simulation environment calculation analysis, two-dimensional and three-dimensional road map real-time synchronization update, etc., at the same time, considering the small volume of high-realistic simulation environment scene data sources, the difficulty of obtaining remote sensing image data, the low extraction accuracy of the current method, and the high time cost and slow operation rate of remote sensing image extraction. The problem of the road vector data extraction method based on the three-dimensional scene is proposed.

[0063] In step S100, the three-dimensional scene specifically refers to a high-fidelity scene, that is, a virtual environment with a high sense of reality created using an Unreal Engine (such as UnrealEngine, UE for short), which is very close to the real world in terms of visual effects, physical simulation, lighting and texture.

[0064] In step S110 , ground features refer to fixed objects distributed on the ground, such as buildings, roads, vegetation, etc.

[0065] In this embodiment, extracting the outer contour of the road based on the feature identification of the road includes: extracting the mesh data and attribute information of the corresponding static mesh component of the road in the three-dimensional scene based on the feature identification of the road, wherein the attribute information includes geometric information, determining the outer boundary shape of the road based on the geometric information, generating the outer contour data of the road, obtaining the preliminary data of the road, converting the preliminary data from the coordinate system of the three-dimensional scene to the earth coordinate system, and obtaining the outer contour data of the road, i.e., the road edge line, such as Figure 2 shown.

[0066] Specifically, according to the unique identification name of the scene entity, the mesh data and attribute information in the corresponding road static mesh component are obtained. The attribute information includes geometric information and attribute information, which provides the original data source for subsequent processing. Among them, the geometric information contains the basic representation of the shape of the road in three-dimensional space, such as the coordinates of the boundary points of the road, the length and width of the road section, etc., while the attribute information includes the type of road (such as expressway, urban branch road), the material of the road, etc.

[0067] Specifically, the outer contour of the road is extracted based on geometric information to obtain preliminary data of the road. Here, image-based methods such as edge detection algorithms and Hough transforms, or map data-based methods based on vector map data extraction and map matching algorithms can be used.

[0068] Furthermore, in a high-fidelity scene, the original coordinates may be based on a specific modeling coordinate system, but in order to facilitate integration and sharing with other geographic information data (such as map data), they need to be converted into the WGS84 coordinate system, that is, the geodetic coordinate system.

[0069] In this embodiment, the road contour data is in ShapeFile format and Geojson format. Both formats are commonly used vector data formats, which can well store and transmit road vector information (including road external contour). Through the previous three steps of processing, the data has been gradually converted from the original grid data to a format suitable for geographic information applications.

[0070] In step S120, when the road centerline intersection is extracted based on the road outer contour data: the road outer contour data in ShapeFile format is converted into binary raster data, and the raster data is refined, the center point of each grid in the refined raster data is traversed, and the initial road graph structure corresponding to the road centerline is constructed. Based on the initial road graph structure, the Liang Youdong-Barsky algorithm is used to optimize the connection mode of the road centerline to obtain a simplified road graph structure, and the centerline ring removal operation is performed on the simplified road graph structure to obtain a road graph structure after the centerline ring is removed. The minimum distance between adjacent intersections in the actual road intersection information is set as a threshold, all broken line segments less than the threshold are removed from the road graph structure after the centerline ring is removed, and the remaining broken line segments are reconnected to obtain the final road graph structure, extract the centerline nodes that meet the preset in the final road graph structure, and construct a road intersection center point data set.

[0071] In this embodiment, in raster data processing, thinning refers to the process of gradually reducing the width of a linear or block-shaped raster object with a certain width to a pixel width (for a linear object) or a single pixel representation (for a block-shaped object) through a specific algorithm. The purpose of this process is to extract the center line of the road. When the road vector data is converted into binary raster data, the road is represented as an area with a certain width in the raster data (for example, the value of the raster cell corresponding to the road is 1, and the value of the non-road area is 0). Through the thinning operation, the center line information of the road can be obtained.

[0072] Furthermore, Zhang parallel fast thinning algorithm is used to thin the raster data. Among them, "1" means valid, and "0" means invalid. Zhang parallel fast thinning algorithm can quickly process raster data by parallel computing while ensuring the quality of thinning. When processing large-scale road raster data, parallel computing can significantly improve processing efficiency.

[0073] In this embodiment, the center points of the grid points are gradually traversed based on the refined raster data to construct the initial graph structure of the raster data, that is, the refined raster data is vectorized, and the initial graph structure is the initial road graph structure. The initial road graph structure is composed of multiple centerline nodes and edges connecting two adjacent nodes. At the same time, the initial road graph structure records the basic topological information of the raster data, including the connection relationship between each pixel point.

[0074] Next, in order to obtain the accurate road centerline, the initial road graph structure is simplified by simplifying the road centerline, removing loops, and connecting adjacent points.

[0075] In this embodiment, the connection of the road centerline is optimized by, when simplifying the initial road graph structure, judging whether the center point connection line between the current pixel point B and the starting pixel point A passes through all the rectangular sequences between AB based on the Liang Youdong-Barsky algorithm according to the pixel points crossed by each segment of the vectorized broken line. If it can pass through, the original connection method is simplified and B is directly connected to A. If not, B is connected to the center point position of the traversable pixel point C, and C is used as the inflection point to search for the next traversable pixel point D.

[0076] In this embodiment, the centerline ring removal operation on the simplified road graph structure includes: enlarging the simplified road graph structure, traversing the edges of nodes with degrees greater than a preset value and locating the ring structure by step limit, wherein the ring structure is composed of at least three nodes, removing the nodes with degrees greater than 2 in the ring structure, if the removal fails, the center point of the ring structure is used as a new node to replace the ring structure, if the removal is successful, when the number of remaining nodes is equal to 2, the ring structure is removed successfully, and when the number of remaining nodes is greater than 2, the center point of the ring structure is used as a new node to replace the ring structure.

[0077] Specifically, after the simplified road graph structure is enlarged 100 times, it is found that there is a ring structure between some pixels with a step size less than 4 (the ring structure with a step size less than 4 can be ignored and regarded as a pixel). First, locate the ring structure: traverse all vertices with a degree greater than 2. For one point A, traverse each edge of A. If you can return to point A within the step size of 4 by moving along the edge, you will find a ring. If you cannot find a point A, return directly and change to the next edge. Then remove the ring structure: remove the ring formed by three points, remove the vertex with a degree of 2. As long as one is deleted, the ring is released. If none is removed, take the center point of these three points and update the three center points to the new vertex; remove the ring formed by four points, remove the vertex with a degree of 2. If two can be deleted, the ring is released. If only one can be removed, or none are removed, take the center point of the four points and update the center point of the four points to the new vertex.

[0078] In this embodiment, the minimum distance between adjacent intersection points of the road centerline of the road intersection in the actual situation is set as a threshold, all the broken line segments in the road centerline with a ring structure less than the threshold are obtained, and the single broken line distance or continuous broken line segment distance less than the threshold distance is stored in the to-be-processed broken line file, all broken lines are traversed, the to-be-processed broken lines are removed, and the traversal is repeated again to connect adjacent points. Finally, all nodes with a degree greater than 2 are extracted to obtain a data set of the center points of the road intersection.

[0079] In this embodiment, the turning points of road intersections are extracted based on the road outer contour data, including: based on the road outer contour data in Geojson format, extracting the vertex association relationship set of polygons and the connection relationship index between each vertex, and constructing an outer ring structure point set and a hole structure point set according to the vertex association relationship set, according to a preset distance threshold and the road intersection center point data set, determining the intersection point set of the road intersection in the outer ring structure point set and the hole structure point set, and extracting the turning points of the road intersection according to the intersection point set and the road intersection center point data set.

[0080] Specifically, firstly, based on the road outer contour data in Geojson format, the outer contour points are traversed: according to the geojson data obtained from the road outer contour, the outer ring and hole of the polygon are obtained, and the vertex association relationship set D is created and the index is constructed: the outer ring point set A is obtained in counterclockwise order, and the hole point set B is obtained in clockwise order. Among them, the point set of the road outer contour, such as a ring, includes two circles, the inner one is called a hole, and the outer one is called a ring. The index refers to the connection relationship between the points, such as Figure 3 shown.

[0081] Furthermore, the road intersection point data is obtained, and when setting the distance threshold, the distance between the nearest point and the center point is set to 3.5, where the nearest point refers to the point with the smallest straight-line distance from the road center point in the above hole point and outer ring point set. According to the road intersection center point data set obtained above, the road centerline intersection point is located, and the points within the threshold range of the road centerline intersection point are scanned counterclockwise in the outer ring point set A and the hole point set B, and the points are sorted counterclockwise in sections to obtain the intersection point data of the road intersection.

[0082] Further, obtain intersection information, such as Figure 4 As shown, the adjacent segments of the intersection point data of the road intersection are extracted, including the first adjacent segment L1 and the second adjacent segment L2. The first point P1 is taken from the first adjacent segment L1, and the second point P2 is taken from the second adjacent segment L2. The triangle C x The circumference of P1P2 is the smallest, so the first point is P 1和 The second point P2 is regarded as the turning point of the same intersection and is marked as the information storage of the same intersection.

[0083] In step S130, the road vector data in the three-dimensional scene is finally extracted based on the road outer contour data and the road intersection data, namely the road intersection center point data set and the road intersection information.

[0084] like Figure 5 The figure shows a schematic diagram of specific implementation steps of a road vector data extraction method based on a three-dimensional scene.

[0085] like Figure 6 The figures are image examples of the process of road data extraction using the present method, wherein (a) is a schematic diagram of the road outer contour, (b) is a schematic diagram of the road centerline and its centerline intersection, (c) is a schematic diagram of road intersection extraction, and (d) is a schematic diagram of road vector data display.

[0086] In the above-mentioned road vector data extraction method based on three-dimensional scenes, the road outer contour is extracted based on the road feature identification in the three-dimensional scene where the road vector quantity is to be extracted to obtain the corresponding road outer contour data, and then based on the road outer contour data, the road centerline intersection and the road intersection turning point are extracted to obtain the road intersection center point data set and road intersection information, and the road intersection data is obtained based on the road intersection center point data set and the road intersection information, and finally the road vector data in the three-dimensional scene is extracted based on the road outer contour data and the road intersection data. This method can extract vector data from high-fidelity scenes, realize the extraction of geospatial data from high-fidelity scenes, provide more complete geospatial data resource support for downstream tasks, and realize the synchronous update of two-dimensional and three-dimensional road maps from the perspective of extracting road vector data from high-fidelity scenes, and ensure the consistency of two-dimensional and three-dimensional road maps.

[0087] Furthermore, the vector road data extracted from the highly realistic scenes can be analyzed and decided on the localized platform, and the information memory is smaller than that of the same level three-dimensional solid model. Therefore, the subsequent analysis can get rid of the dependence on the Unreal Engine, promote the localization process of the system, and improve the computing speed of the deduction system. At the same time, the extraction of vector road data for highly realistic scenes can ensure the real-time synchronization update of the later two-dimensional road vector map and the three-dimensional highly realistic scene map, and ensure the consistency of the two-dimensional and three-dimensional road maps. The extraction of road vector data for highly realistic scenes can supplement and improve the data resources of the roads and intersections in the simulation environment, and provide vector data resource support for the simulation environment calculation analysis, the real-time synchronization update of the two-dimensional and three-dimensional road maps, etc.

[0088] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0089] In one embodiment, Figure 7 As shown, a road vector data extraction device based on a three-dimensional scene is provided, comprising: a three-dimensional scene acquisition module 200, a road outer contour data extraction module 210, a road intersection and junction information extraction module 220 and a road vector data extraction module 230, wherein:

[0090] A three-dimensional scene acquisition module 200 is used to acquire a three-dimensional scene for road vector quantity extraction;

[0091] A road outer contour data extraction module 210 is used to extract the outer contour of the road based on the feature identification of the road in the three-dimensional scene to obtain corresponding road outer contour data;

[0092] A road intersection and junction information extraction module 220 is used to extract the road centerline intersection and the road intersection turning point based on the road outer contour data, obtain the road intersection center point data set and the road intersection information, and obtain the road intersection data according to the road intersection center point data set and the road intersection information;

[0093] The road vector data extraction module 230 is used to extract the road vector data in the three-dimensional scene according to the road outer contour data and the road intersection data.

[0094] The specific definition of the road vector data extraction device based on the three-dimensional scene can be found in the definition of the road vector data extraction method based on the three-dimensional scene above, which will not be repeated here. Each module in the above-mentioned road vector data extraction device based on the three-dimensional scene can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0095] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a road vector data extraction method based on a three-dimensional scene is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse, etc.

[0096] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0097] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0098] Obtaining a three-dimensional scene for road vector quantity extraction;

[0099] In the three-dimensional scene, the outer contour of the road is extracted based on the ground feature identification of the road to obtain corresponding outer contour data of the road;

[0100] Based on the road outer contour data, the road centerline intersection and the road intersection turning point are extracted to obtain a road intersection center point data set and road intersection information, and the road intersection data is obtained according to the road intersection center point data set and the road intersection information;

[0101] The road vector data in the three-dimensional scene is extracted according to the road outer contour data and the road intersection data.

[0102] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0103] Obtaining a three-dimensional scene for road vector quantity extraction;

[0104] In the three-dimensional scene, the outer contour of the road is extracted based on the ground feature identification of the road to obtain corresponding outer contour data of the road;

[0105] Based on the road outer contour data, the road centerline intersection and the road intersection turning point are extracted to obtain a road intersection center point data set and road intersection information, and the road intersection data is obtained according to the road intersection center point data set and the road intersection information;

[0106] The road vector data in the three-dimensional scene is extracted according to the road outer contour data and the road intersection data.

[0107] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0108] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0109] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A road vector data extraction method based on a three-dimensional scene, characterized in that: The method comprises: Obtaining a three-dimensional scene for road vector quantity extraction; In the three-dimensional scene, an outer contour of a road is extracted based on a land feature identifier of the road to obtain corresponding outer contour data of the road. Specifically, mesh data and attribute information in a corresponding static mesh component of the road are extracted in the three-dimensional scene based on the land feature identifier of the road, wherein the attribute information includes geometric information, an outer boundary shape of the road is determined based on the geometric information, outer contour data of the road is generated, preliminary data of the road is obtained, and the preliminary data is converted from a coordinate system of the three-dimensional scene to a geodetic coordinate system to obtain the outer contour data of the road; Based on the road outer contour data, the road centerline intersections and the road intersection turning points are extracted to obtain a road intersection center point data set and road intersection information, and the road intersection data is obtained according to the road intersection center point data set and the road intersection information, wherein, when the road centerline intersections are extracted based on the road outer contour data: the road outer contour data in ShapeFile format is converted into binary raster data, and the raster data is refined, the center point of each grid in the refined raster data is traversed, and an initial road graph structure corresponding to the road centerline is constructed, based on Initial road graph structure, using Liang Youdong-Barsky algorithm to optimize the connection mode of road centerlines to obtain a simplified road graph structure, performing centerline ring removal operation on the simplified road graph structure to obtain a road graph structure after centerline ring removal, setting the minimum distance between adjacent intersections in the actual road intersection information as a threshold, removing all broken line segments less than the threshold in the road graph structure after centerline ring removal, and reconnecting the remaining broken line segments to obtain a final road graph structure, extracting the preset centerline nodes in the final road graph structure, and constructing the road intersection center point data set; The road vector data in the three-dimensional scene is extracted according to the road outer contour data and the road intersection data.

2. The road vector data extraction method according to claim 1, characterized in that: The road outline data is in ShapeFile format and Geojson format.

3. The road vector data extraction method according to claim 2, characterized in that: The raster data is thinned using Zhang parallel fast thinning algorithm.

4. The road vector data extraction method according to claim 3, characterized in that: The centerline ring removal operation on the simplified road graph structure includes: The simplified road graph structure is enlarged, and a ring structure is located by traversing the edges of nodes whose degrees are greater than a preset value and limiting the step length, wherein the ring structure is composed of at least three nodes; Removing nodes with a degree greater than 2 from the ring structure, if the removal fails, replacing the ring structure with the center point of the ring structure as a new node; If the removal is successful, when the number of remaining nodes is equal to 2, the ring structure is removed successfully. When the number of remaining nodes is greater than 2, the center point of the ring structure is used as a new node to replace the ring structure.

5. The road vector data extraction method according to claim 4, characterized in that: Extracting the turning point of the road intersection based on the road outer contour data includes: Based on the road contour data in Geojson format, extract the vertex association relationship set of the polygon and the connection relationship index between each vertex, and construct the outer ring structure point set and the hole structure point set according to the vertex association relationship set; According to a preset distance threshold and the road intersection center point data set, determining the intersection point set of the road intersection from the outer ring structure point set and the hole structure point set; The turning points of the road intersections are extracted according to the intersection point set and the road intersection center point data set.

6. A road vector data extraction device based on a three-dimensional scene, characterized in that: The device comprises: A three-dimensional scene acquisition module is used to acquire a three-dimensional scene for road vector quantity extraction; A road outer contour data extraction module is used to extract the outer contour of the road in the three-dimensional scene based on the object identification of the road to obtain the corresponding road outer contour data. Specifically, the mesh data and attribute information in the corresponding road static mesh component extracted in the three-dimensional scene based on the object identification of the road, wherein the attribute information includes geometric information, and the outer boundary shape of the road is determined based on the geometric information, and the outer contour data of the road is generated to obtain preliminary data of the road, and the preliminary data is converted from the coordinate system of the three-dimensional scene to the earth coordinate system to obtain the road outer contour data; The road intersection and junction information extraction module is used to extract the road centerline intersection points and road intersection turning points based on the road outer contour data, obtain the road intersection center point data set and road intersection information, and obtain the road intersection data according to the road intersection center point data set and road intersection information, wherein, when the road centerline intersection points are extracted based on the road outer contour data: the road outer contour data in ShapeFile format is converted into binary raster data, and the raster data is refined, the center point of each grid in the refined raster data is traversed, and the initial Road graph structure, based on the initial road graph structure, using the Liang Youdong-Barsky algorithm to optimize the connection mode of the road centerline to obtain a simplified road graph structure, performing a centerline ring removal operation on the simplified road graph structure to obtain a road graph structure after the centerline ring is removed, setting the minimum distance between adjacent intersections in the actual road intersection information as a threshold, removing all broken line segments less than the threshold in the road graph structure after the centerline ring is removed, and reconnecting the remaining broken line segments to obtain a final road graph structure, extracting the preset centerline nodes in the final road graph structure, and constructing the road intersection center point data set; The road vector data extraction module is used to extract the road vector data in the three-dimensional scene according to the road outer contour data and the road intersection data.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images