Automatic road vector map generation method and system based on laser point cloud data

Abstract

The invention discloses an automatic road vector map generation method and system based on laser point cloud data, and the method comprises the steps: obtaining all points in the laser point cloud data, setting a separation model of ground points and non-ground points, calculating the separation value of each point in the laser point cloud data, separating the ground points from the non-ground points, and generating a road vector map. Retaining ground points; extracting feature vectors of all points, setting a road marking extraction model, calculating a clustering set of the road marking, and identifying the road marking in the laser point cloud data; setting a vectorization model of the road marking, vectorizing the identified road marking, and generating a vector value of each road marking; and obtaining real road network data, setting a road network fusion optimization model, and optimizing the vector values of the road markings to enable the vector values of the road markings to be closest to the real road network data, thereby generating a road vector map.

Description

Technical Field

[0001] The present invention belongs to the technical field of map generation, and more specifically, relates to an automatic road vector map generation method and system based on laser point cloud data. Background Art

[0002] Vector map generation includes data collection, processing, storage and visualization. The following are some major technological advances and applications:

[0003] Data collection:

[0004] Remote sensing technology: Using remote sensing equipment such as satellites and drones to obtain geographic information, combined with technologies such as spectroscopy and laser radar (LiDAR), the accuracy and resolution of the data are improved.

[0005] Data processing:

[0006] Automated extraction: Use image processing and machine learning algorithms to automatically extract object information from images and convert raster data into vector format.

[0007] Big data technology: Use big data processing frameworks such as Hadoop and Spark to accelerate the processing and analysis of large-scale geographic data.

[0008] Storage and Management:

[0009] Cloud computing: Cloud platforms (such as Google Cloud, AWS, etc.) provide efficient storage solutions and support large-scale vector data storage and sharing.

[0010] Databases: Geographic Information System (GIS) databases such as PostGIS allow for the storage and management of complex spatial data, supporting efficient query and analysis.

[0011] Visualization:

[0012] Web map services: Open source libraries such as Mapbox and Leaflet allow users to easily display and interact with vector maps on web pages.

[0013] 3D visualization: Combining WebGL and other technologies, 3D visualization of vector maps is achieved, improving user experience.

[0014] Application areas:

[0015] Navigation and Positioning: Applications used for real-time navigation and traffic management, such as Google Maps and Amap.

[0016] Urban Planning: Assist in urban planning and management, conduct environmental monitoring and disaster management.

[0017] Geographic analysis: used for geographic data analysis, spatial decision support, etc., and widely used in ecology, resource management and other fields.

[0018] However, the vector map generated by the technical solutions in the existing technology is not accurate enough, so there is an urgent need for a high-precision vector map generation solution. Summary of the invention

[0019] In order to solve the above technical problems, the present invention proposes an automatic road vector map generation method based on laser point cloud data, comprising:

[0020] Get all the points in the laser point cloud data, set the separation model of ground points and non-ground points, calculate the separation value of each point in the laser point cloud data, so as to separate the ground points from the non-ground points and retain the ground points;

[0021] Extract the feature vectors of all points, set up a road marking extraction model, calculate the clustering set of road markings, and identify the road markings in the laser point cloud data;

[0022] Setting a vectorization model for road markings, vectorizing the identified road markings, and generating a vector value for each road marking;

[0023] The real road network data is obtained, and a road network fusion optimization model is set to optimize the vector value of the road marking so that the vector value of the road marking is closest to the real road network data, thereby generating a road vector map.

[0024] Furthermore, the separation model of ground points and non-ground points includes:

[0025]

[0026] Where D(p) is the separation value of point p in the laser point cloud data. When the separation value D of point p in the laser point cloud data exceeds the preset separation threshold, point p in the laser point cloud data is a ground point. N is the number of points in the laser point cloud data. a1 is the first adjustment factor of the separation model. b1 is the second adjustment factor of the separation model. z i is the i-th point p in the laser point cloud data i The height eigenvector of , c′ is the third adjustment factor of the separation model, d1 is the fourth adjustment factor of the separation model, N neighbors (p i ) is the i-th point p in the laser point cloud data i The number of neighboring points, α is the fifth adjustment factor of the separation model, p i is the i-th point in the laser point cloud data, p j is the jth point in the neighborhood point set, N′(i) is the i-th point p in the laser point cloud data iis the set of neighborhood points, and σ is the adjustment factor used to control the neighborhood influence range.

[0027] Furthermore, the road marking extraction model includes:

[0028]

[0029] Among them, C is the cluster set of road markings, which represents the set of extracted road marking center points, c j is the center point of the jth road marking, F(p i ) is the i-th point p in the laser point cloud data i , λ is the first adjustment factor of the road marking extraction model, M is the number of center points, p′ is the second adjustment factor of the road marking extraction model, μ is the third adjustment factor of the road marking extraction model, d′(c j , c k ) is the center point c of the jth road marking j and the center point c of the kth road marking k The distance between them, η is the adjustment factor used to control the influence of distance.

[0030] Furthermore, the vectorized model of the road marking includes:

[0031]

[0032] Where V(C) is the vector value of each road marking in the cluster set C of road markings, x i′ is the horizontal coordinate of the i′th node on the road marking, y i′ is the ordinate of the i′th node on the road marking, β is the first adjustment factor of the vector model of the road marking, L i′+1 is the i′+1th node on the road marking, L i′ is the i′th node on the road marking, p″ is the second adjustment factor of the vector model of the road marking, θ is the direction angle of the road marking, and K is the number of nodes on the road marking.

[0033] Furthermore, the road network fusion optimization model includes:

[0034]

[0035] Among them, M is the number of road markings in the real road network data, R j′ is the vector value of the j′th road marking in the real road network data, Length(V j′ (C)) is the length of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, Curvature (V j′(C)) is the curvature of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, δ is the adjustment factor of the road network fusion optimization model, V j′ (C) is the vector value of the j′th road marking in the cluster set C of road markings, is the mean of the vector values ​​of all road markings in the cluster set C of road markings.

[0036] The present invention also proposes an automatic road vector map generation system based on laser point cloud data, comprising:

[0037] The separation module is used to obtain all points in the laser point cloud data, set the separation model of ground points and non-ground points, calculate the separation value of each point in the laser point cloud data, so as to separate the ground points from the non-ground points and retain the ground points;

[0038] The road marking recognition module is used to extract the feature vectors of all points, set the road marking extraction model, calculate the clustering set of road markings, and identify the road markings in the laser point cloud data;

[0039] A vectorization module, used for setting a vectorization model of road markings, vectorizing the identified road markings, and generating a vector value for each road marking;

[0040] The optimization module is used to obtain real road network data, set up a road network fusion optimization model, optimize the vector value of the road marking, and make the vector value of the road marking closest to the real road network data, thereby generating a road vector map.

[0041] Furthermore, the separation model of ground points and non-ground points includes:

[0042]

[0043] Where D(p) is the separation value of point p in the laser point cloud data. When the separation value D of point p in the laser point cloud data exceeds the preset separation threshold, point p in the laser point cloud data is a ground point. N is the number of points in the laser point cloud data. a1 is the first adjustment factor of the separation model. b1 is the second adjustment factor of the separation model. z i is the i-th point p in the laser point cloud data i The height eigenvector of , c′ is the third adjustment factor of the separation model, d1 is the fourth adjustment factor of the separation model, N neighbors (p i ) is the i-th point p in the laser point cloud data i The number of neighboring points, α is the fifth adjustment factor of the separation model, p i is the i-th point in the laser point cloud data, p jis the jth point in the neighborhood point set, N′(i) is the i-th point p in the laser point cloud data i is the set of neighborhood points, and σ is the adjustment factor used to control the neighborhood influence range.

[0044] Furthermore, the road marking extraction model includes:

[0045]

[0046] Among them, C is the cluster set of road markings, which represents the set of extracted road marking center points, c j is the center point of the jth road marking, F(p i ) is the i-th point p in the laser point cloud data i , λ is the first adjustment factor of the road marking extraction model, M is the number of center points, p′ is the second adjustment factor of the road marking extraction model, μ is the third adjustment factor of the road marking extraction model, d′(c j , c k ) is the center point c of the jth road marking j and the center point c of the kth road marking k The distance between them, η is the adjustment factor used to control the influence of distance.

[0047] Furthermore, the vectorized model of the road marking includes:

[0048]

[0049] Where V(C) is the vector value of each road marking in the cluster set C of road markings, x i′ is the horizontal coordinate of the i′th node on the road marking, y i′ is the ordinate of the i′th node on the road marking, β is the first adjustment factor of the vector model of the road marking, L i′+1 is the i′+1th node on the road marking, L i′ is the i′th node on the road marking, p″ is the second adjustment factor of the vector model of the road marking, θ is the direction angle of the road marking, and K is the number of nodes on the road marking.

[0050] Furthermore, the road network fusion optimization model includes:

[0051]

[0052] Among them, M is the number of road markings in the real road network data, R j′ is the vector value of the j′th road marking in the real road network data, Length(V j′(C)) is the length of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, Curvature (V j′ (C)) is the curvature of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, δ is the adjustment factor of the road network fusion optimization model, V j′ (C) is the vector value of the j′th road marking in the cluster set C of road markings, is the mean of the vector values ​​of all road markings in the cluster set C of road markings.

[0053] In general, the above technical solution conceived by the present invention has the following beneficial effects compared with the prior art:

[0054] Through the technical solution of the present invention, the accuracy of traffic network in the project can be greatly improved, making the generated vector map closer to the real road. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a flow chart of the method of embodiment 1 of the present invention;

[0056] Figure 2 It is a system structure diagram of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0057] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0058] The method provided by the present invention can be implemented in the following terminal environment, and the terminal may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method described in the following embodiment.

[0059] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts in the entire terminal, and executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and calling data stored in the storage medium.

[0060] The storage medium may include a random access memory (RAM) or a read-only memory (ROM). The storage medium may be used to store instructions, programs, codes, code sets or instructions.

[0061] The display screen is used to display the user interface of each application.

[0062] In addition, those skilled in the art can understand that the structure of the above terminal does not constitute a limitation on the terminal, and the terminal may include more or fewer components, or combine certain components, or arrange the components differently. For example, the terminal also includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, and a power supply, which will not be described in detail here.

[0063] Example 1

[0064] like Figure 1 As shown, the embodiment of the present invention proposes an automatic road vector map generation method based on laser point cloud data, comprising:

[0065] Step 101, obtaining all points in the laser point cloud data, setting a separation model for ground points and non-ground points, calculating a separation value for each point in the laser point cloud data, thereby separating the ground points from the non-ground points and retaining the ground points;

[0066] Specifically, the separation model of ground points and non-ground points includes:

[0067]

[0068] Where D(p) is the separation value of point p in the laser point cloud data. When the separation value D of point p in the laser point cloud data exceeds the preset separation threshold, point p in the laser point cloud data is a ground point. N is the number of points in the laser point cloud data. a1 is the first adjustment factor of the separation model. b1 is the second adjustment factor of the separation model. z i is the i-th point p in the laser point cloud data i The height eigenvector of , c′ is the third adjustment factor of the separation model, d1 is the fourth adjustment factor of the separation model, N neighbors (p i ) is the i-th point p in the laser point cloud data i The number of neighboring points, α is the fifth adjustment factor of the separation model, p i is the i-th point in the laser point cloud data, p j is the jth point in the neighborhood point set, N′(i) is the i-th point p in the laser point cloud data i is the set of neighborhood points, and σ is the adjustment factor used to control the neighborhood influence range.

[0069] Step 102, extracting feature vectors of all points, setting a road marking extraction model, calculating a cluster set of road markings, and identifying road markings in the laser point cloud data;

[0070] Specifically, the road marking extraction model includes:

[0071]

[0072] Among them, C is the cluster set of road markings, which represents the set of extracted road marking center points, c j is the center point of the jth road marking, F(p i ) is the i-th point p in the laser point cloud data i , λ is the first adjustment factor of the road marking extraction model, M is the number of center points, p′ is the second adjustment factor of the road marking extraction model, μ is the third adjustment factor of the road marking extraction model, d′(c j ,c k ) is the center point c of the jth road marking j and the center point c of the kth road marking k The distance between them, η is the adjustment factor used to control the influence of distance.

[0073] Step 103, setting a vectorization model for road markings, vectorizing the identified road markings, and generating a vector value for each road marking;

[0074] Specifically, the vectorized model of the road marking includes:

[0075]

[0076] Where V(C) is the vector value of each road marking in the cluster set C of road markings, x i′ is the horizontal coordinate of the i′th node on the road marking, y i′ is the ordinate of the i′th node on the road marking, β is the first adjustment factor of the vector model of the road marking, L i′+1 is the i′+1th node on the road marking, L i′ is the i′th node on the road marking, p″ is the second adjustment factor of the vector model of the road marking, θ is the direction angle of the road marking, and K is the number of nodes on the road marking.

[0077] Step 104, obtaining real road network data, and setting a road network fusion optimization model to optimize the vector value of the road markings so that the vector value of the road markings is closest to the real road network data, thereby generating a road vector map.

[0078] Specifically, the road network fusion optimization model includes:

[0079]

[0080] Among them, M is the number of road markings in the real road network data, R j′ is the vector value of the j′th road marking in the real road network data, Length(V j′(C)) is the length of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, Curvature (V j′ (C)) is the curvature of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, δ is the adjustment factor of the road network fusion optimization model, V j′ (C) is the vector value of the j′th road marking in the cluster set C of road markings, is the mean of the vector values ​​of all road markings in the cluster set C of road markings.

[0081] Specifically, this embodiment also includes calculating the longitudinal section slope and the cross section slope of the road within the set range. Specifically, firstly, the ground points are obtained according to the separation model of the ground points and the non-ground points, and then the laser point cloud data is sliced ​​to obtain the point set of the ground points in each slice interval, and the variables are set:

[0082] p i =(x i ,y i , z i ): the i-th point p in the laser point cloud data i , contains spatial coordinates and elevation information, where z i is the i-th point p i 's elevation;

[0083] H long (x): longitudinal slope function, which represents the slope along the x-direction (usually the direction of travel of the road);

[0084] H cross (y): cross-sectional slope function, representing the slope along the y-direction (usually the lateral width direction of the road);

[0085] Δz i，i+1 ..: two points p i and p i+1 The height difference between i，i+1 =z i+1 -z i .

[0086] Calculation of longitudinal slope

[0087] The longitudinal slope is usually calculated along the travel direction of the road (usually the x-axis direction). The longitudinal slope can be calculated by the following formula:

[0088]

[0089] Among them, ||x i+1 -x i ||2 represents point p i+1 With point pi The distance in the x-direction (Euclidean distance) is used to normalize the height difference, α″′ is a weight coefficient that represents the nonlinear sensitivity to slope changes, and N is the number of longitudinal points selected at a given interval d.

[0090] Calculation of cross-sectional slope

[0091] The cross-sectional slope is usually calculated along the transverse width direction of the road (usually the y-axis direction). The cross-sectional slope can be calculated using the following formula:

[0092]

[0093] Among them, ‖y i+1 -y i ‖2 represents point p i+1 With point p i The distance in the y-direction (Euclidean distance) is used to normalize the height difference, β″′ is a weight coefficient that represents the nonlinear sensitivity to the change of the cross-sectional slope, and M is the number of cross-sectional points selected at a given interval d.

[0094] Example 2

[0095] like Figure 2 As shown, an embodiment of the present invention further provides an automatic road vector map generation system based on laser point cloud data, comprising:

[0096] The separation module is used to obtain all points in the laser point cloud data, set the separation model of ground points and non-ground points, calculate the separation value of each point in the laser point cloud data, so as to separate the ground points from the non-ground points and retain the ground points;

[0097] Specifically, the separation model of ground points and non-ground points includes:

[0098]

[0099] Where D(p) is the separation value of point p in the laser point cloud data. When the separation value D of point p in the laser point cloud data exceeds the preset separation threshold, point p in the laser point cloud data is a ground point. N is the number of points in the laser point cloud data. a1 is the first adjustment factor of the separation model. b1 is the second adjustment factor of the separation model. z i is the i-th point p in the laser point cloud data i The height eigenvector of , c′ is the third adjustment factor of the separation model, d1 is the fourth adjustment factor of the separation model, N neighbors (p i ) is the i-th point p in the laser point cloud data i The number of neighboring points, α is the fifth adjustment factor of the separation model, p iis the i-th point in the laser point cloud data, p j is the jth point in the neighborhood point set, N′(i) is the i-th point p in the laser point cloud data i is the set of neighborhood points, and σ is the adjustment factor used to control the neighborhood influence range.

[0100] The road marking recognition module is used to extract the feature vectors of all points, set the road marking extraction model, calculate the clustering set of road markings, and identify the road markings in the laser point cloud data;

[0101] Specifically, the road marking extraction model includes:

[0102]

[0103] Among them, C is the cluster set of road markings, which represents the set of extracted road marking center points, c j is the center point of the jth road marking, F(p i ) is the i-th point p in the laser point cloud data i , λ is the first adjustment factor of the road marking extraction model, M is the number of center points, p′ is the second adjustment factor of the road marking extraction model, μ is the third adjustment factor of the road marking extraction model, d′(c j ,c k ) is the center point c of the jth road marking j and the center point c of the kth road marking k The distance between them, η is the adjustment factor used to control the influence of distance.

[0104] A vectorization module, used for setting a vectorization model of road markings, vectorizing the identified road markings, and generating a vector value for each road marking;

[0105] Specifically, the vectorized model of the road marking includes:

[0106]

[0107] Where V(C) is the vector value of each road marking in the cluster set C of road markings, x i′ is the horizontal coordinate of the i′th node on the road marking, y i′ is the ordinate of the i′th node on the road marking, β is the first adjustment factor of the vector model of the road marking, L i′+1 is the i′+1th node on the road marking, L i′ is the i′th node on the road marking, p″ is the second adjustment factor of the vector model of the road marking, θ is the direction angle of the road marking, and K is the number of nodes on the road marking.

[0108] The optimization module is used to obtain real road network data, set up a road network fusion optimization model, optimize the vector value of the road marking, and make the vector value of the road marking closest to the real road network data, thereby generating a road vector map.

[0109] Specifically, the road network fusion optimization model includes:

[0110]

[0111] Among them, M is the number of road markings in the real road network data, R j′ is the vector value of the j′th road marking in the real road network data, Length(V j′ (C)) is the length of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, Curvature (V j′ (C)) is the curvature of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, δ is the adjustment factor of the road network fusion optimization model, V j′ (C) is the vector value of the j′th road marking in the cluster set C of road markings, is the mean of the vector values ​​of all road markings in the cluster set C of road markings.

[0112] Example 3

[0113] The embodiment of the present invention further provides a storage medium storing a plurality of instructions, wherein the instructions are used to implement the method for automatically generating a road vector map based on laser point cloud data.

[0114] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0115] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: Step 101, obtaining all points in the laser point cloud data, setting a separation model of ground points and non-ground points, calculating a separation value of each point in the laser point cloud data, thereby separating the ground points from the non-ground points and retaining the ground points;

[0116] Specifically, the separation model of ground points and non-ground points includes:

[0117]

[0118] Where D(p) is the separation value of point p in the laser point cloud data. When the separation value D of point p in the laser point cloud data exceeds the preset separation threshold, point p in the laser point cloud data is a ground point. N is the number of points in the laser point cloud data. a1 is the first adjustment factor of the separation model. b1 is the second adjustment factor of the separation model. z i is the i-th point p in the laser point cloud data i The height eigenvector of , c′ is the third adjustment factor of the separation model, d1 is the fourth adjustment factor of the separation model, N neighbors (p i ) is the i-th point p in the laser point cloud data i The number of neighboring points, α is the fifth adjustment factor of the separation model, p i is the i-th point in the laser point cloud data, p j is the jth point in the neighborhood point set, N′(i) is the i-th point p in the laser point cloud data i is the set of neighborhood points, and σ is the adjustment factor used to control the neighborhood influence range.

[0119] Step 102, extracting feature vectors of all points, setting a road marking extraction model, calculating a cluster set of road markings, and identifying road markings in the laser point cloud data;

[0120] Specifically, the road marking extraction model includes:

[0121]

[0122] Among them, C is the cluster set of road markings, which represents the set of extracted road marking center points, c j is the center point of the jth road marking, F(p i ) is the i-th point p in the laser point cloud data i , λ is the first adjustment factor of the road marking extraction model, M is the number of center points, p′ is the second adjustment factor of the road marking extraction model, μ is the third adjustment factor of the road marking extraction model, d′(c j ,c k ) is the center point c of the jth road marking j and the center point c of the kth road marking k The distance between them, η is the adjustment factor used to control the influence of distance.

[0123] Step 103, setting a vectorization model for road markings, vectorizing the identified road markings, and generating a vector value for each road marking;

[0124] Specifically, the vectorized model of the road marking includes:

[0125]

[0126] Where V(C) is the vector value of each road marking in the cluster set C of road markings, x i′ is the horizontal coordinate of the i′th node on the road marking, y i′ is the ordinate of the i′th node on the road marking, β is the first adjustment factor of the vector model of the road marking, L i′+1 is the i′+1th node on the road marking, L i′ is the i′th node on the road marking, p″ is the second adjustment factor of the vector model of the road marking, θ is the direction angle of the road marking, and K is the number of nodes on the road marking.

[0127] Step 104, obtaining real road network data, and setting a road network fusion optimization model to optimize the vector value of the road markings so that the vector value of the road markings is closest to the real road network data, thereby generating a road vector map.

[0128] Specifically, the road network fusion optimization model includes:

[0129]

[0130] Among them, M is the number of road markings in the real road network data, R j′ is the vector value of the j′th road marking in the real road network data, Length(V j′ (C)) is the length of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, Curvature (V j′ (C)) is the curvature of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, δ is the adjustment factor of the road network fusion optimization model, V j′ (C) is the vector value of the j′th road marking in the cluster set C of road markings, is the mean of the vector values ​​of all road markings in the cluster set C of road markings.

[0131] Example 4

[0132] An embodiment of the present invention also proposes an electronic device, including a processor and a storage medium connected to the processor, wherein the storage medium stores multiple instructions, and the instructions can be loaded and executed by the processor to enable the processor to execute the automatic road vector map generation method based on laser point cloud data.

[0133] Specifically, the electronic device of this embodiment may be a computer terminal, and the computer terminal may include: one or more processors, and a storage medium.

[0134] Among them, the storage medium can be used to store software programs and modules, such as an automatic road vector map generation method based on laser point cloud data in an embodiment of the present invention, and the corresponding program instructions / modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, realizing the above-mentioned automatic road vector map generation method based on laser point cloud data. The storage medium may include high-speed random storage media, and may also include non-volatile storage media, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include storage media remotely arranged relative to the processor, and these remote storage media may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, corporate intranet, local area network, mobile communication network and combinations thereof.

[0135] The processor can call the information and application stored in the storage medium through the transmission system to perform the following steps: Step 101, obtain all points in the laser point cloud data, set a separation model of ground points and non-ground points, calculate the separation value of each point in the laser point cloud data, thereby separating the ground points from the non-ground points and retaining the ground points;

[0136] Specifically, the separation model of ground points and non-ground points includes:

[0137]

[0138] Where D(p) is the separation value of point p in the laser point cloud data. When the separation value D of point p in the laser point cloud data exceeds the preset separation threshold, point p in the laser point cloud data is a ground point. N is the number of points in the laser point cloud data. a1 is the first adjustment factor of the separation model. b1 is the second adjustment factor of the separation model. z i is the i-th point p in the laser point cloud data i The height eigenvector of , c′ is the third adjustment factor of the separation model, d1 is the fourth adjustment factor of the separation model, N neighbors (p i ) is the i-th point p in the laser point cloud data i The number of neighboring points, α is the fifth adjustment factor of the separation model, p i is the i-th point in the laser point cloud data, p j is the jth point in the neighborhood point set, N′(i) is the i-th point p in the laser point cloud data i is the set of neighborhood points, and σ is the adjustment factor used to control the neighborhood influence range.

[0139] Step 102, extracting feature vectors of all points, setting a road marking extraction model, calculating a cluster set of road markings, and identifying road markings in the laser point cloud data;

[0140] Specifically, the road marking extraction model includes:

[0141]

[0142] Among them, C is the cluster set of road markings, which represents the set of extracted road marking center points, c j is the center point of the jth road marking, F(p i ) is the i-th point p in the laser point cloud data i , λ is the first adjustment factor of the road marking extraction model, M is the number of center points, p′ is the second adjustment factor of the road marking extraction model, μ is the third adjustment factor of the road marking extraction model, d′(c j , c k ) is the center point c of the jth road marking j and the center point c of the kth road marking k The distance between them, η is the adjustment factor used to control the influence of distance.

[0143] Step 103, setting a vectorization model for road markings, vectorizing the identified road markings, and generating a vector value for each road marking;

[0144] Specifically, the vectorized model of the road marking includes:

[0145]

[0146] Where V(C) is the vector value of each road marking in the cluster set C of road markings, x i′ is the horizontal coordinate of the i′th node on the road marking, y i′ is the ordinate of the i′th node on the road marking, β is the first adjustment factor of the vector model of the road marking, L i′+1 is the i′+1th node on the road marking, L i′ is the i′th node on the road marking, p″ is the second adjustment factor of the vector model of the road marking, θ is the direction angle of the road marking, and K is the number of nodes on the road marking.

[0147] Step 104, obtaining real road network data, and setting a road network fusion optimization model to optimize the vector value of the road markings so that the vector value of the road markings is closest to the real road network data, thereby generating a road vector map.

[0148] Specifically, the road network fusion optimization model includes:

[0149]

[0150] Among them, M is the number of road markings in the real road network data, R j′ is the vector value of the j′th road marking in the real road network data, Length(V j′ (C)) is the length of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, Curvature (V j′ (C)) is the curvature of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, δ is the adjustment factor of the road network fusion optimization model, V j′ (C) is the vector value of the j′th road marking in the cluster set C of road markings, is the mean of the vector values ​​of all road markings in the cluster set C of road markings.

[0151] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0152] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0153] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0154] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0155] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0156] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0157] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A method for automatically generating road vector maps based on laser point cloud data, characterized in that: include: Get all the points in the laser point cloud data, set the separation model of ground points and non-ground points, calculate the separation value of each point in the laser point cloud data, so as to separate the ground points from the non-ground points and retain the ground points; Extract the feature vectors of all points, set up a road marking extraction model, calculate the clustering set of road markings, and identify the road markings in the laser point cloud data; Setting a vectorization model for road markings, vectorizing the identified road markings, and generating a vector value for each road marking; The real road network data is obtained, and a road network fusion optimization model is set to optimize the vector value of the road marking so that the vector value of the road marking is closest to the real road network data, thereby generating a road vector map.

2. The method for generating an automatic road vector map based on laser point cloud data according to claim 1, characterized in that: The separation model of ground points and non-ground points includes: Where D(p) is the separation value of point p in the laser point cloud data. When the separation value D of point p in the laser point cloud data exceeds the preset separation threshold, point p in the laser point cloud data is a ground point. N is the number of points in the laser point cloud data. a1 is the first adjustment factor of the separation model. b1 is the second adjustment factor of the separation model. z i is the i-th point p in the laser point cloud data i The height eigenvector of , c′ is the third adjustment factor of the separation model, d1 is the fourth adjustment factor of the separation model, N neighbors (p i ) is the i-th point p in the laser point cloud data i The number of neighboring points, α is the fifth adjustment factor of the separation model, p i is the i-th point in the laser point cloud data, p j is the ,th point in the neighborhood point set, N′(i) is the ith point p in the laser point cloud data i is the set of neighborhood points, and σ is the adjustment factor used to control the neighborhood influence range.

3. The method for generating an automatic road vector map based on laser point cloud data according to claim 1, characterized in that: The road marking extraction model includes: Among them, C is the cluster set of road markings, which represents the set of extracted road marking center points, c j is the center point of the jth road marking, F(p i ) is the i-th point p in the laser point cloud data i , λ is the first adjustment factor of the road marking extraction model, M is the number of center points, p′ is the second adjustment factor of the road marking extraction model, μ is the third adjustment factor of the road marking extraction model, d′(c j , c k ) is the center point c of the th road marking j and the center point c of the kth road marking k The distance between them, η is the adjustment factor used to control the influence of distance.

4. The method for generating an automatic road vector map based on laser point cloud data according to claim 3, characterized in that: The vectorized model of the road marking includes: Where V(C) is the vector value of each road marking in the cluster set C of road markings, x i′ is the horizontal coordinate of the i′th node on the road marking, y i′ is the ordinate of the i′th node on the road marking, β is the first adjustment factor of the vector model of the road marking, L i′+1 is the i′+1th node on the road marking, L i′ is the i′th node on the road marking, p″ is the second adjustment factor of the vector model of the road marking, θ is the direction angle of the road marking, and K is the number of nodes on the road marking.

5. The method for generating an automatic road vector map based on laser point cloud data according to claim 4, characterized in that: The road network fusion optimization model includes: Among them, M is the number of road markings in the real road network data, R j′ is the vector value of the j′th road marking in the real road network data, Length(V j ′(C)) is the length of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, Curvature(V j′ (C)) is the curvature of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, δ is the adjustment factor of the road network fusion optimization model, V j′ (C) is the vector value of the j′th road marking in the cluster set C of road markings, is the mean of the vector values ​​of all road markings in the cluster set C of road markings.

6. An automatic road vector map generation system based on laser point cloud data, characterized in that: include: The separation module is used to obtain all points in the laser point cloud data, set the separation model of ground points and non-ground points, calculate the separation value of each point in the laser point cloud data, so as to separate the ground points from the non-ground points and retain the ground points; The road marking recognition module is used to extract the feature vectors of all points, set the road marking extraction model, calculate the clustering set of road markings, and identify the road markings in the laser point cloud data; A vectorization module, used for setting a vectorization model of road markings, vectorizing the identified road markings, and generating a vector value for each road marking; The optimization module is used to obtain real road network data, set up a road network fusion optimization model, optimize the vector value of the road marking, and make the vector value of the road marking closest to the real road network data, thereby generating a road vector map.

7. The automatic road vector map generation system based on laser point cloud data according to claim 6, characterized in that: The separation model of ground points and non-ground points includes: Where D(p) is the separation value of point p in the laser point cloud data. When the separation value D of point p in the laser point cloud data exceeds the preset separation threshold, point p in the laser point cloud data is a ground point. N is the number of points in the laser point cloud data. a1 is the first adjustment factor of the separation model. b1 is the second adjustment factor of the separation model. z i is the i-th point p in the laser point cloud data i The height eigenvector of , c′ is the third adjustment factor of the separation model, d1 is the fourth adjustment factor of the separation model, N neighbors (p i ) is the i-th point p in the laser point cloud data i The number of neighboring points, α is the fifth adjustment factor of the separation model, p i is the i-th point in the laser point cloud data, p j is the jth point in the neighborhood point set, N′(i) is the i-th point p in the laser point cloud data i is the set of neighborhood points, and σ is the adjustment factor used to control the neighborhood influence range.

8. The automatic road vector map generation system based on laser point cloud data according to claim 6, characterized in that: The road marking extraction model includes: Among them, C is the cluster set of road markings, which represents the set of extracted road marking center points, c j is the center point of the jth road marking, F(p i ) is the i-th point p in the laser point cloud data i , λ is the first adjustment factor of the road marking extraction model, M is the number of center points, p′ is the second adjustment factor of the road marking extraction model, μ is the third adjustment factor of the road marking extraction model, d′(c j , c k ) is the center point c of the jth road marking j and the center point c of the kth road marking k The distance between them, η is the adjustment factor used to control the influence of distance.

9. The automatic road vector map generation system based on laser point cloud data according to claim 8, characterized in that: The vectorized model of the road marking includes: Where V(C) is the vector value of each road marking in the cluster set C of road markings, x i′ is the horizontal coordinate of the i′th node on the road marking, y i′ is the ordinate of the i′th node on the road marking, β is the first adjustment factor of the vector model of the road marking, L i′+1 is the i′+1th node on the road marking, L i′ is the i′th node on the road marking, p″ is the second adjustment factor of the vector model of the road marking, θ is the direction angle of the road marking, and K is the number of nodes on the road marking.

10. The automatic road vector map generation system based on laser point cloud data according to claim 9, characterized in that: The road network fusion optimization model includes: Among them, M is the number of road markings in the real road network data, R j′ is the vector value of the j′th road marking in the real road network data, Length(V j′ (C)) is the length of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, Curvature (V j′ (C)) is the curvature of the road marking corresponding to the vector value of the j′th road marking in the cluster set C of road markings, δ is the adjustment factor of the road network fusion optimization model, V j′ (C) is the vector value of the j′th road marking in the cluster set C of road markings, is the mean of the vector values ​​of all road markings in the cluster set C of road markings.

Images