Lane line vector model construction method, device, electronic device and storage medium

Abstract

The present invention discloses a method, device, electronic device and storage medium for constructing a lane line vector model. The method comprises acquiring semantic point cloud data; extracting lane line point cloud data from the semantic point cloud data; projecting the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set; constructing a two-dimensional triangulation network according to the two-dimensional point set; extracting a two-dimensional lane line contour from the two-dimensional triangulation network; and constructing a lane line vector model according to the two-dimensional lane line contour. The solution provided by the present invention can construct a vector model based on semantic point cloud data at low cost and high precision, reduce vectorization cost, and improve vectorization efficiency and precision.

Description

Technical Field

[0001] The present invention relates to the field of crowdsourcing map technology, and in particular to a lane line vector model construction method, device, electronic device and storage medium. Background Art

[0002] Crowdsourced maps are high-precision maps that use road data collected by low-cost sensor hardware mounted on vehicles and transmitted to the cloud for data fusion. The raw data collected by low-cost sensor hardware is semantically recognized by the on-board computer program and output as a semantic point cloud with semantic attributes. The semantic point cloud needs to be vectorized and constructed into a vector model of road elements such as lane lines before data fusion can be performed. Here, in order to reduce data transmission costs, the vectorization process is usually performed in the on-board computer, so it is necessary to minimize resource consumption during the processing process.

[0003] In the prior art, point cloud clustering and line fitting methods are usually used to construct lane line vector models in high-precision maps from semantic point cloud data. This type of method has high computational cost and low production efficiency, and cannot meet the needs of low-cost, high-efficiency and high-precision construction of lane line vector models in crowdsourcing maps. Summary of the invention

[0004] In order to solve the technical problem that the existing technology cannot achieve both low cost and high precision when constructing a lane line vector model from semantic point cloud data, and it is difficult to meet the requirements of low-cost and high-efficiency construction of lane line vector models in crowdsourcing map production, the embodiments of the present invention provide a lane line vector model construction method, device, electronic device and storage medium.

[0005] The technical solution of the embodiment of the present invention is achieved as follows:

[0006] An embodiment of the present invention provides a method for constructing a lane line vector model, the method comprising: acquiring semantic point cloud data; extracting lane line point cloud data from the semantic point cloud data; projecting the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set; constructing a two-dimensional triangulated triangulated network based on the two-dimensional point set; extracting a two-dimensional lane line contour from the two-dimensional triangulated triangulated network; and constructing a lane line vector model based on the two-dimensional lane line contour.

[0007] In the above scheme, projecting the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set includes: projecting each three-dimensional point in the lane line point cloud data onto a horizontal two-dimensional plane to obtain a two-dimensional point corresponding to each three-dimensional point; assigning the elevation value of each three-dimensional point to the attribute of the corresponding two-dimensional point; and taking the set of two-dimensional points after attribute assignment as the two-dimensional point set.

[0008] In the above solution, constructing a two-dimensional triangulated triangulated network according to the two-dimensional point set includes: constructing a two-dimensional triangulated triangulated network according to the two-dimensional point set using a triangulation technique.

[0009] In the above scheme, extracting the two-dimensional lane line contour from the two-dimensional triangulation triangulation network includes: determining the external edges among all the edges in the two-dimensional triangulation triangulation network; determining the contour edges from all the external edges; and connecting the contour edges to obtain the two-dimensional lane line contour.

[0010] In the above scheme, constructing a lane line vector model based on the two-dimensional lane line contour includes: when the constructed vector model is a surface element vector model, using the attributes of the vertices of the two-dimensional lane line contour to restore the two-dimensional points to three-dimensional points as the contour points of the surface element to construct a lane line surface vector model.

[0011] In the above scheme, constructing a lane line vector model based on the two-dimensional lane line contour includes: when the constructed vector model is a line element vector model, extracting the lane line centerline according to the two-dimensional lane line contour, and constructing a lane line vector model based on the centerline.

[0012] In the above scheme, the lane line centerline is extracted according to the two-dimensional lane line contour, including: according to the two-dimensional lane line contour, vertices are extracted to construct triangulated triangulated networks respectively; the constructed triangulated triangulated networks are classified; according to the classification results, the triangulated triangulated networks are processed to obtain the constituent line segments of the skeleton line; the main skeleton line is obtained from the constituent line segments of the skeleton line; the attributes of the main skeleton line nodes are determined; and the two-dimensional points are restored to three-dimensional points by using the attributes of the main skeleton line nodes, and output as the lane line centerline.

[0013] An embodiment of the present invention also provides a lane line vector model construction device, which includes: an acquisition module, used to acquire semantic point cloud data; a first extraction module, used to extract lane line point cloud data from the semantic point cloud data; a projection module, used to project the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set; a first construction module, used to construct a two-dimensional triangulation triangle network based on the two-dimensional point set; a second extraction module, used to extract a two-dimensional lane line contour from the two-dimensional triangulation triangle network; and a second construction module, used to construct a lane line vector model based on the two-dimensional lane line contour.

[0014] An embodiment of the present invention further provides an electronic device, comprising: a processor and a memory for storing a computer program that can be run on the processor; wherein the processor executes the steps of any of the above methods when running the computer program.

[0015] An embodiment of the present invention further provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0016] The lane line vector model construction method, device, electronic device and storage medium provided in the embodiment of the present invention obtain semantic point cloud data; extract lane line point cloud data from the semantic point cloud data; project the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set; construct a two-dimensional triangulation triangle network based on the two-dimensional point set; extract a two-dimensional lane line contour from the two-dimensional triangulation triangle network; and construct a lane line vector model based on the two-dimensional lane line contour. The solution provided by the present invention is based on semantic point cloud data, and utilizes the projection of point cloud data onto a two-dimensional plane to obtain a two-dimensional lane line contour, and then constructs a lane line vector model based on the two-dimensional lane line contour. The vector model can be constructed at low cost and high precision, reducing vectorization costs, improving vectorization efficiency and precision, so that the semantic point cloud vectorization process takes into account both high precision and low cost, and meets the needs of crowdsourcing map production. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a process of constructing a lane line vector model according to an embodiment of the present invention;

[0018] Figure 2 This is a schematic diagram of the structure of a lane line vector model building device according to an embodiment of the present invention;

[0019] Figure 3 The figure is a diagram showing the internal structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The present invention will be described in further detail below with reference to the accompanying drawings and embodiments.

[0021] The embodiment of the present invention provides a method for constructing a lane line vector model. Figure 1 As shown, the method includes:

[0022] Step 101: Acquire semantic point cloud data;

[0023] Step 102: extracting lane line point cloud data from the semantic point cloud data;

[0024] Step 103: Projecting the lane point cloud data onto a two-dimensional plane to obtain a two-dimensional point set;

[0025] Step 104: constructing a two-dimensional triangulation triangulation network according to the two-dimensional point set;

[0026] Step 105: extracting a two-dimensional lane line contour from the two-dimensional triangulated triangulated network;

[0027] Step 106: Construct a lane line vector model according to the two-dimensional lane line contour.

[0028] Specifically, this embodiment can obtain semantic point cloud data by processing sensor data collected by the sensor. Here, the sensor may include but is not limited to an image acquisition device and a laser radar, and the sensor data may include but is not limited to an image frame collected by the image acquisition device and a point cloud frame collected by the laser radar.

[0029] After obtaining semantic point cloud data, it can be extracted and denoised to obtain lane line point cloud data. Generally, lane line point cloud data can be obtained through specific algorithms, such as cluster analysis, connectivity analysis, etc., or lane line point cloud data can be directly selected manually. The lane line point cloud data obtained by the above method is generally of low accuracy. In order to further improve the accuracy of point cloud data, it is also possible to extract linear terrain road marking lines, remove noise, and obtain high-precision lane line point cloud data.

[0030] In one embodiment, projecting the lane point cloud data onto a two-dimensional plane to obtain a two-dimensional point set includes:

[0031] Projecting each three-dimensional point in the lane line point cloud data onto a horizontal two-dimensional plane to obtain a two-dimensional point corresponding to each three-dimensional point;

[0032] Assigning the elevation value of each three-dimensional point to the attribute of the corresponding two-dimensional point;

[0033] The set of two-dimensional points after attribute assignment is taken as the two-dimensional point set.

[0034] The two-dimensional plane in this embodiment is the horizontal XOY plane. Each three-dimensional point in the lane line point cloud data is projected onto the horizontal XOY plane to obtain the two-dimensional point corresponding to each three-dimensional point. The collection of all two-dimensional points is a two-dimensional point set. Here, the attribute H of the two-dimensional point in the two-dimensional point set can be the elevation value of the three-dimensional point. By assigning the elevation value of the three-dimensional point as an attribute to the attribute H of the projected two-dimensional point, it is convenient to determine the two-dimensional lane line contour from the two-dimensional point set later.

[0035] In one embodiment, constructing a two-dimensional triangulated triangulated network according to the two-dimensional point set includes:

[0036] A two-dimensional triangulation triangulated network is constructed according to the two-dimensional point set using triangulation technology.

[0037] Triangulation technology (Delaunay triangulation technology for short) is the process of generating a set of triangles for a given set of plane points. That is, for any given set of plane points, there is only one triangulation method that satisfies the so-called "maximum-minimum angle" optimization criterion, that is, the sum of all minimum interior angles is the largest, which is Delaunay triangulation. This triangulation method follows the "maximum minimum angle" and "empty circumscribed circle" criteria. The "maximum minimum angle" criterion is that in the absence of singularity, the sum of the minimum angles of the Delaunay triangulation is greater than the sum of the minimum angles of any triangle formed by non-Delaunay triangulation, and the sum of the minimum interior angles of the triangle is the largest, so that the divided triangles will not have a situation where the interior angle is too small, which is more conducive to the subsequent calculation of the finite element.

[0038] This embodiment uses triangulation technology to construct and obtain a two-dimensional triangulated triangulated network. Because triangulation technology requires relatively low computing resources, it only takes up relatively low computing costs.

[0039] In one embodiment, extracting the two-dimensional lane line contour from the two-dimensional triangulated triangulated network includes:

[0040] Determine external edges among all edges in the two-dimensional triangulation triangulation network;

[0041] determining a silhouette edge from all of said exterior edges;

[0042] The contour edges are connected to obtain a two-dimensional lane line contour.

[0043] In this embodiment, all edges in all two-dimensional triangulation triangulation networks are first traversed, and the edges whose lengths exceed the first threshold and the other two edges of the triangles in which they are located are determined as external edges and added to the external edge set L1. Here, the first threshold can be set according to the situation. Then, each edge in the external edge set L1 is traversed to determine whether the reverse edge of each edge is marked. If the reverse edge of the edge is not marked, the edge is determined to be a contour edge and added to the contour edge set L2. Finally, all edges in the contour edge set L2 are connected in a counterclockwise order to obtain several closed two-dimensional polygon contours, and each two-dimensional polygon contour is output as a lane line contour. In this way, a two-dimensional lane line contour is obtained.

[0044] After obtaining the two-dimensional lane line contour, different methods may be used to construct the lane line vector model according to the type of lane line vector model. Specifically, the types of lane line vector models are divided into two categories: surface element vector model and line element vector model. The following will describe the construction process of the lane line vector model in detail for the above two types of lane line vector models.

[0045] The first one:

[0046] In one embodiment, constructing a lane line vector model according to the two-dimensional lane line profile includes:

[0047] When the constructed vector model is a surface element vector model, the attributes of the two-dimensional lane line contour vertices are used to restore the two-dimensional points to three-dimensional points as the contour points of the surface element to construct the lane line surface vector model.

[0048] The lane line vector model in this embodiment is a surface element vector model. For the surface element vector model, the two-dimensional points can be restored to three-dimensional points according to the attribute H of the two-dimensional lane line contour vertices, and used as the contour points of the surface element to construct the lane line surface vector model.

[0049] In one embodiment, constructing a lane line vector model according to the two-dimensional lane line profile includes:

[0050] When the constructed vector model is a line element vector model, the lane line center line is extracted according to the two-dimensional lane line contour, and the lane line vector model is constructed based on the center line.

[0051] The lane line vector model in this embodiment is a line element vector model. For the line element vector model, the lane line center line can be extracted according to the contour of each lane line vector object, and the line vector model can be constructed based on the center line.

[0052] Specifically, in one embodiment, extracting the lane centerline according to the two-dimensional lane contour includes:

[0053] According to the two-dimensional lane line contour, extract vertices to construct triangulated triangulated networks respectively;

[0054] Classifying the constructed triangulated triangulated network;

[0055] According to the classification result, the triangulated triangulated network is processed to obtain the constituent line segments of the skeleton line;

[0056] Obtaining a main skeleton line from the constituent line segments of the skeleton line;

[0057] Determining the attributes of the main skeleton line nodes;

[0058] The attributes of the main skeleton line nodes are used to restore the two-dimensional points to three-dimensional points, which are output as lane center lines.

[0059] That is, in this embodiment, the lane center line is extracted according to the contour of each lane vector object, specifically including:

[0060] Step 1: Construct a Delaunay triangulation based on contour points;

[0061] In this step, a Delaunay triangulation is constructed based on the contour points, and vertices are extracted for each two-dimensional lane line contour to construct a Delaunay triangulation.

[0062] Step 2: Divide the triangles of the Delaunay triangulation into three categories: A, B, and C. A triangle has one adjacent triangle, B triangle has two adjacent triangles, and C triangle has three adjacent triangles.

[0063] In this step, the triangles in the triangulation network are classified into three categories: A, B, and C. Category A triangles have one adjacent triangle, category B triangles have two adjacent triangles, and category C triangles have three adjacent triangles.

[0064] Step 3: Connect the centroid of the C-type triangle and the midpoints of its three sides, the midpoints of the two shared sides of the B-type triangle and the adjacent triangles, and the vertex of the A-type triangle and the midpoint of the shared sides of the adjacent triangles as the line segments of the skeleton line;

[0065] In this step, a skeleton line segment is constructed by connecting the centroid of the C-type triangle and the midpoints of the three sides, the midpoints of the two shared sides of the B-type triangle and the adjacent triangles, and the vertex of the A-type triangle and the midpoint of the shared side of the adjacent triangles as the skeleton line segment.

[0066] Step 4: Extract the main skeleton line and calculate the attribute H of the nodes on the main skeleton line. The attribute H of the centroid of the C-type triangle is the average of the attributes H of the three vertices, and the attribute H of the midpoint of the shared edge is the average of the attributes H of the two vertices.

[0067] In this step, the main skeleton line is extracted, and the Prim algorithm is used to generate the minimum spanning tree for each component line of the skeleton line, and the longest path is extracted as the main skeleton line. In addition, the main skeleton line node attribute H is calculated, where the attribute H of the center of gravity of the C-type triangle is the average of the three vertex attributes H, the attribute H of the midpoint of the shared edge is the average of the two vertex attributes H, and the attribute H of the vertex of the A-type triangle remains unchanged.

[0068] Step 5: Restore the two-dimensional points to three-dimensional points according to the attribute H of the nodes in the main skeleton line, and output them as the center line of the lane line.

[0069] In this step, the lane centerline is constructed based on the main skeleton line, and the two-dimensional points are restored to three-dimensional points according to the attribute H of the nodes in the main skeleton line, and the output is the lane centerline.

[0070] The lane line vector model construction method, device, electronic device and storage medium provided in the embodiment of the present invention obtain semantic point cloud data; extract lane line point cloud data from the semantic point cloud data; project the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set; construct a two-dimensional triangulation triangle network based on the two-dimensional point set; extract a two-dimensional lane line contour from the two-dimensional triangulation triangle network; and construct a lane line vector model based on the two-dimensional lane line contour. The solution provided by the present invention is based on semantic point cloud data, and utilizes the projection of point cloud data onto a two-dimensional plane to obtain a two-dimensional lane line contour, and then constructs a lane line vector model based on the two-dimensional lane line contour. The vector model can be constructed at low cost and high precision, reducing vectorization costs, improving vectorization efficiency and precision, so that the semantic point cloud vectorization process takes into account both high precision and low cost, and meets the needs of crowdsourcing map production.

[0071] In order to implement the method of the embodiment of the present invention, the embodiment of the present invention also provides a lane line vector model construction device, such as Figure 2 As shown, the lane line vector model construction device 200 includes: an acquisition module 201, a first extraction module 202, a projection module 203, a first construction module 204, a second extraction module 205 and a second construction module 206; wherein,

[0072] An acquisition module 201 is used to acquire semantic point cloud data;

[0073] A first extraction module 202, used to extract lane line point cloud data from the semantic point cloud data;

[0074] A projection module 203 is used to project the lane point cloud data onto a two-dimensional plane to obtain a two-dimensional point set;

[0075] A first construction module 204 is used to construct a two-dimensional triangulation triangulation network according to the two-dimensional point set;

[0076] A second extraction module 205 is used to extract a two-dimensional lane line contour from the two-dimensional triangulation triangulation network;

[0077] The second construction module 206 is used to construct a lane line vector model according to the two-dimensional lane line contour.

[0078] In actual application, the acquisition module 201, the first extraction module 202, the projection module 203, the first construction module 204, the second extraction module 205 and the second construction module 206 can be implemented by a processor in the lane line vector model construction device.

[0079] It should be noted that: when the above-mentioned device provided in the above-mentioned embodiment is executed, only the division of the above-mentioned program modules is used as an example. In actual application, the above-mentioned processing can be assigned to different program modules as needed, that is, the internal structure of the terminal is divided into different program modules to complete all or part of the above-mentioned processing. In addition, the above-mentioned device provided in the above-mentioned embodiment and the above-mentioned method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0080] In order to implement the method of the embodiment of the present invention, the embodiment of the present invention also provides a computer program product, which includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of the above method.

[0081] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiment of the present invention, the embodiment of the present invention also provides an electronic device (computer device). Specifically, in one embodiment, the computer device can be a terminal, and its internal structure diagram can be as follows: Figure 3 As shown. The computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05 and a memory (not shown in the figure) connected through a system bus. Among them, the processor A01 of the computer device is used to provide computing and control capabilities. The memory of the computer device includes an internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A06. The network interface A02 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 A01, the method of any one of the above embodiments is implemented. The display screen A04 of the computer device can be a liquid crystal display or an electronic ink display, and the input device A05 of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0082] Those skilled in the art will understand that Figure 3 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.

[0083] The device provided by the embodiment of the present invention includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the method of any one of the above embodiments is implemented.

[0084] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0085] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0086] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0088] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0089] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0090] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0091] It can be understood that the memory of the embodiment of the present invention can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAM bus random access memory (DRRAM, Direct Rambus Random Access Memory).The memories described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memories.

[0092] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0093] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for constructing a lane line vector model, It is characterized in that The method comprises: Get semantic point cloud data; Extracting lane line point cloud data from the semantic point cloud data; Projecting the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set; Constructing a two-dimensional triangulation triangulation network according to the two-dimensional point set; Extracting a two-dimensional lane line contour from the two-dimensional triangulated triangulated network; Constructing a lane line vector model according to the two-dimensional lane line contour; Among them, the constructing of a lane line vector model based on the two-dimensional lane line contour includes: when the constructed vector model is a surface element vector model, using the attributes of the vertices of the two-dimensional lane line contour to restore the two-dimensional points to three-dimensional points as the contour points of the surface elements to construct a lane line surface vector model; and / or, when the constructed vector model is a line element vector model, extracting the lane line centerline according to the two-dimensional lane line contour, and constructing a lane line vector model based on the centerline.

2. The method according to claim 1, It is characterized in that The step of projecting the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set includes: Projecting each three-dimensional point in the lane line point cloud data onto a horizontal two-dimensional plane to obtain a two-dimensional point corresponding to each three-dimensional point; Assigning the elevation value of each three-dimensional point to the attribute of the corresponding two-dimensional point; The set of two-dimensional points after attribute assignment is taken as the two-dimensional point set.

3. The method according to claim 1, It is characterized in that The step of constructing a two-dimensional triangulated triangulated network according to the two-dimensional point set comprises: A two-dimensional triangulation triangulated network is constructed according to the two-dimensional point set using triangulation technology.

4. The method according to claim 1, It is characterized in that The step of extracting a two-dimensional lane line contour from the two-dimensional triangulated triangulated network comprises: Determine external edges among all edges in the two-dimensional triangulation triangulation network; determining a silhouette edge from all of said exterior edges; The contour edges are connected to obtain a two-dimensional lane line contour.

5. The method according to claim 1, It is characterized in that The extracting the lane line centerline according to the two-dimensional lane line contour comprises: According to the two-dimensional lane line contour, extract vertices to construct triangulated triangulated networks respectively; Classifying the constructed triangulated triangulated network; According to the classification result, the triangulated triangulated network is processed to obtain the constituent line segments of the skeleton line; Obtaining a main skeleton line from the constituent line segments of the skeleton line; Determining the attributes of the main skeleton line nodes; The attributes of the main skeleton line nodes are used to restore the two-dimensional points to three-dimensional points, which are output as lane center lines.

6. A lane line vector model construction device, It is characterized in that The device comprises: Acquisition module, used to acquire semantic point cloud data; A first extraction module, used to extract lane line point cloud data from the semantic point cloud data; A projection module, used for projecting the lane line point cloud data onto a two-dimensional plane to obtain a two-dimensional point set; A first construction module is used to construct a two-dimensional triangulation triangulation network according to the two-dimensional point set; A second extraction module, used to extract a two-dimensional lane line contour from the two-dimensional triangulated triangulation network; A second construction module is used to construct a lane line vector model according to the two-dimensional lane line contour; Among them, the second construction module is used for: when the constructed vector model is a surface element vector model, using the attributes of the two-dimensional lane line contour vertices to restore the two-dimensional points to three-dimensional points as the contour points of the surface element to construct the lane line surface vector model; and / or, when the constructed vector model is a line element vector model, extracting the lane line centerline according to the two-dimensional lane line contour, and constructing the lane line line vector model based on the centerline.

7. An electronic device, It is characterized in that include: A processor and a memory for storing a computer program capable of running on the processor; wherein, When the processor is used to run the computer program, the steps of the method according to any one of claims 1 to 5 are performed.

8. A storage medium having a computer program stored therein, It is 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