A method and device for tracing back massive vehicle-mounted point cloud data

Abstract

The present invention provides a method and device for backtracking massive vehicle-mounted point cloud data, which includes the following steps: S1: point cloud data preprocessing, collecting information related to the measuring station and archiving and storing the data file; S2: point cloud data vector line extraction, using the road surface center line of the vehicle-mounted point cloud data as the vector line of the point cloud data; S3: massive point cloud data backtracking base map vector construction, grid cutting, vector integration, and spatial matching of the point cloud data vector line to form a base map vector; S4: massive point cloud data history backtracking, extracting related vector lines by setting a spatial range, then reading the iterative history records stored in the vector lines, and finally querying the related point cloud data according to the iterative history records. Backtracking massive point cloud data based on massive point cloud vector base maps and iterative history solves the problem of point cloud data backtracking in high-precision map production.

Description

Technical Field

[0001] The present invention relates to the technical field of mass point cloud data management for high-precision map making machines, and in particular to a method and device for tracing back mass vehicle-mounted point cloud data. Background Art

[0002] In the field of high-precision map production, data backtracking is extremely important for updating and producing map data. Currently, vehicle-mounted point cloud data is an important input for updating high-precision map data, but there is a lack of backtracking methods for massive vehicle-mounted point cloud data. The present invention is mainly proposed to solve this problem. Summary of the invention

[0003] The present invention provides a method and device for tracing back massive vehicle-mounted point cloud data, which overcomes the above problems or at least partially solves the above problems.

[0004] According to a first aspect of the present invention, a method for backtracking massive vehicle-mounted point cloud data is provided, which comprises the following steps:

[0005] S1: Point cloud data preprocessing, collecting information related to the measuring station and archiving and storing data files;

[0006] S2: Extract vector lines from point cloud data, using the center line of the road surface of the vehicle-mounted point cloud data as the vector line of the point cloud data;

[0007] S3: Back-tracing base map vector construction of massive point cloud data, performing grid cutting, vector integration, and spatial matching on the vector lines of point cloud data to form base map vectors;

[0008] S4: Historical backtracking of massive point cloud data. By setting the spatial range, the associated vector lines are extracted, and then the iteration history records stored in the vector lines are read. Finally, the associated point cloud data is queried based on the iteration history records.

[0009] Based on the above technical solution, the present invention can also make the following improvements.

[0010] Optionally, the step S1 includes:

[0011] S11: Information collection, collecting the station information of the point cloud data collected in the same batch to generate a unique code for the station, which is defined as the station ID;

[0012] S12: Data collection: All data files collected in the same batch are stored in a directory named after the station ID. Different types of data files in the same batch are stored in different fixed directories under the directory, so that data backtracking can quickly obtain the corresponding data files according to the station ID and data type.

[0013] Optionally, step S2 includes:

[0014] S21: Extract the plane from the point cloud using the RANSAC method, and select the plane with the largest area as the initial road surface;

[0015] S22: according to the elevation difference characteristics of the road boundary and the road surface, spreading growth is performed on the edge of the initial road surface and road boundary point information is extracted, and abnormal boundary points in the boundary points are eliminated through threshold filtering and linear filtering;

[0016] S23: According to the linear distribution law of the boundary points of the single-side road, the boundary point information of the branch road is filtered by the angle and length of the boundary points of the set range;

[0017] S24: obtaining a quadratic polynomial expression of the road boundaries on both sides by linear fitting, and obtaining dense road boundary points on both sides by interpolation;

[0018] S25: Traverse the boundary points of the road on one side in turn, find the nearest boundary point among the boundary points of the road on the other side, connect the midpoints of each pair of boundary points in turn to form a point cloud data vector line, record the corresponding relationship between each vector line and the point cloud data, and calculate the maximum value of the distance between each pair of boundary points as the road width record.

[0019] Optionally, step S3 includes:

[0020] S31: To facilitate sequential generation of iteration history, each station information is sorted according to the acquisition date to form a processing station list, and subsequent processing processes each station information in turn according to the processing station list;

[0021] S32: To facilitate rapid positioning during data tracing, the point cloud data vector line is cut into rectangular grids, and each point cloud data vector line after cutting records the corresponding grid number and station ID;

[0022] S33: To ensure a one-to-one correspondence between the vector line and the point cloud data, the point cloud data is segmented according to the starting and ending positions of the vector line after cutting; first, a perpendicular line is drawn with the starting point or the end point of the vector line as the foot of the perpendicular; then, a position point of half the width of the road on both sides of the foot of the perpendicular is taken; secondly, the position points on both sides are sequentially connected to form a segmentation area; finally, the point cloud data is segmented according to the segmentation area, and the association relationship between the vector line and the point cloud is updated at the same time;

[0023] S34: In order to reduce the number of vector lines in the grid, data integration is required, including merging and connecting the vector line data connected at the beginning and end of the same station in the same grid; merging the vector lines with a length less than a threshold in the grid into the connected vector lines of the same station. After completing the data integration, it is necessary to merge the point cloud data and update the association between the vector lines and the point cloud at the same time;

[0024] S35: To distinguish the updated data and form an iteration history, spatial matching is performed on the vector lines in the same grid to determine the data update and data new acquisition areas. The matching is performed according to the following rules:

[0025] The directions of vector lines on the same road are similar;

[0026] The road width recorded by vector lines of the same road is similar;

[0027] The distance between vector lines of the same road meets a certain threshold;

[0028] S36: According to the set storage rules, the newly added vector lines are stored in the base map vector line storage structure and the iteration history storage structure.

[0029] Optionally, step S4 includes:

[0030] S41: query the point cloud vector line of the area to be queried, calculate the spatial grids passed by setting the contour points of the spatial range, calculate the spatial grids contained in the spatial range by traversing the grid numbers of each row and column, obtain all the vector line data of the spatial range by querying the base map vector line storage structure, and finally determine the target point cloud vector line through the matching rule of S35;

[0031] S42: According to the point cloud vector lines obtained in S41, the station information related to all vector lines is obtained by querying the iterative history storage structure of each vector line;

[0032] S43: According to the station information obtained in S42, the station data file is obtained by querying the storage path, and the backtracking of the point cloud data is completed.

[0033] Optionally, in step S32, the rectangular grid is a positive direction grid with a side length of 2.4 kilometers, and the far point of the coordinate system is used as the starting position of the grid, and the horizontal position number and the vertical position number of the grid are recorded in the grid number.

[0034] Optionally, the storage rule in step S36 includes:

[0035] If any part of the newly added vector line does not match the existing base map vector line, it will be stored in the vector line base map storage structure, and a new iteration history structure of the base map vector line will be added at the same time, and the update type is new;

[0036] If the newly added vector line completely matches the existing base map vector line, it is stored in the iterative history storage structure of the base map vector line it matches;

[0037] If the newly added vector line partially matches the existing base map vector line, the newly added vector line is divided into two parts: the unmatched part is stored in the vector line base map storage structure, and the data flag is set to the original vector line; the matched part is stored in the iterative history storage structure of the base map vector line it matches, and the update type is set to update;

[0038] If the newly added vector line intersects with the existing base map vector line to form an intersection, the existing base map vector line needs to be interrupted at the intersection, and the base map vector line record is deleted in the vector line base map storage structure. At the same time, a new record is added to its iterative history structure with the update type of deletion. The two interrupted vector lines are stored separately in the base map vector line storage structure, and two new vector history storage structures are created.

[0039] According to a second aspect of the present invention, a massive vehicle-mounted point cloud data backtracking device is provided, which includes: it adopts the above-mentioned base map vector line storage structure and iterative history storage structure, wherein:

[0040] The base map vector line storage structure provides a basic base map vector for subsequent data updates;

[0041] The iteration history storage structure provides data iteration update history for subsequent data backtracking.

[0042] The present invention provides a method and device for tracing back massive vehicle-mounted point cloud data, which realizes tracing back massive point cloud data based on massive point cloud vector base map and iteration history, and solves the problem of point cloud data tracing back in high-precision map production. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of a method for tracing back massive vehicle-mounted point cloud data provided by an embodiment of the present invention;

[0044] Figure 2 Schematic diagram of data structure storage rules for a method for tracing back massive vehicle-mounted point cloud data provided by an embodiment of the present invention Figure 1 ;

[0045] Figure 3 Schematic diagram of data structure storage rules for a method for tracing back massive vehicle-mounted point cloud data provided by an embodiment of the present invention Figure 2 ;

[0046] Figure 4 Schematic diagram of data structure storage rules for a method for tracing back massive vehicle-mounted point cloud data provided by an embodiment of the present invention Figure 3 ;

[0047] Figure 5 Schematic diagram of data structure storage rules for a method for tracing back massive vehicle-mounted point cloud data provided by an embodiment of the present invention Figure 4 ; DETAILED DESCRIPTION

[0048] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0049] Figure 1 A flowchart of a method for tracing back massive vehicle-mounted point cloud data provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method includes:

[0050] S1: Point cloud data preprocessing, collecting information related to the measuring station and archiving and storing data files;

[0051] The step S1 comprises:

[0052] S11: Information collection, collect the station information of the point cloud data collected in the same batch to generate a unique station code, which is defined as the station ID; the station ID is composed of the license plate number of the collection vehicle, collection date, station serial number, regional information, etc., such as "49GT1_20190923_01+001_ruanjianyuan".

[0053] S12: Data collection: All data files collected in the same batch are stored in a directory named after the station ID. Different types of data files in the same batch are stored in different fixed directories under the directory, so that data backtracking can quickly obtain the corresponding data files according to the station ID and data type.

[0054] S2: Extract vector lines from point cloud data, using the center line of the road surface of the vehicle-mounted point cloud data as the vector line of the point cloud data;

[0055] The step S2 comprises:

[0056] S21: Extract the plane from the point cloud using the RANSAC method, and select the plane with the largest area as the initial road surface;

[0057] S22: according to the elevation difference characteristics of the road boundary and the road surface, spreading growth is performed on the edge of the initial road surface and road boundary point information is extracted, and abnormal boundary points in the boundary points are eliminated through threshold filtering and linear filtering;

[0058] S23: According to the linear distribution law of the boundary points of the single-side road, the boundary point information of the branch road is filtered by the angle and length of the boundary points of the set range;

[0059] S24: obtaining a quadratic polynomial expression of the road boundaries on both sides by linear fitting, and obtaining dense road boundary points on both sides by interpolation;

[0060] S25: Traverse the boundary points of the road on one side in turn, find the nearest boundary point among the boundary points of the road on the other side, connect the midpoints of each pair of boundary points in turn to form a point cloud data vector line, record the corresponding relationship between each vector line and the point cloud data, and calculate the maximum value of the distance between each pair of boundary points as the road width record.

[0061] S3: Back-tracing base map vector construction of massive point cloud data, performing grid cutting, vector integration, and spatial matching on the vector lines of point cloud data to form base map vectors;

[0062] The step S3 comprises:

[0063] S31: To facilitate sequential generation of iteration history, each station information is sorted according to the acquisition date to form a processing station list, and subsequent processing processes each station information in turn according to the processing station list;

[0064] S32: To facilitate rapid positioning during data tracing, the point cloud data vector line is cut into rectangular grids, and each point cloud data vector line after cutting records the corresponding grid number and station ID;

[0065] S33: To ensure a one-to-one correspondence between the vector line and the point cloud data, the point cloud data is segmented according to the starting and ending positions of the vector line after cutting; first, a perpendicular line is drawn with the starting point or the end point of the vector line as the foot of the perpendicular; then, a position point of half the width of the road on both sides of the foot of the perpendicular is taken; secondly, the position points on both sides are sequentially connected to form a segmentation area; finally, the point cloud data is segmented according to the segmentation area, and the association relationship between the vector line and the point cloud is updated at the same time;

[0066] S34: In order to reduce the number of vector lines in the grid, data integration is required, including merging and connecting the vector line data connected at the beginning and end of the same station in the same grid; merging the vector lines with a length less than a threshold in the grid into the connected vector lines of the same station. After completing the data integration, it is necessary to merge the point cloud data and update the association between the vector lines and the point cloud at the same time;

[0067] S35: To distinguish the updated data and form an iteration history, spatial matching is performed on the vector lines in the same grid to determine the data update and data new acquisition areas. The matching is performed according to the following rules:

[0068] The directions of vector lines on the same road are similar;

[0069] The road width recorded by vector lines of the same road is similar;

[0070] The distance between vector lines of the same road meets a certain threshold;

[0071] S36: According to the set storage rules, the newly added vector lines are stored in the base map vector line storage structure and the iteration history storage structure.

[0072] S4: Historical backtracking of massive point cloud data. By setting the spatial range, the associated vector lines are extracted, and then the iteration history records stored in the vector lines are read. Finally, the associated point cloud data is queried based on the iteration history records.

[0073] The step S4 comprises:

[0074] S41: query the point cloud vector line of the area to be queried, calculate the spatial grids passed by setting the contour points of the spatial range, calculate the spatial grids contained in the spatial range by traversing the grid numbers of each row and column, obtain all the vector line data of the spatial range by querying the base map vector line storage structure, and finally determine the target point cloud vector line through the matching rule of S35;

[0075] S42: According to the point cloud vector lines obtained in S41, the station information related to all vector lines is obtained by querying the iterative history storage structure of each vector line;

[0076] S43: According to the station information obtained in S42, the station data file is obtained by querying the storage path, and the backtracking of the point cloud data is completed.

[0077] It can be understood that in this embodiment, a massive vehicle-mounted point cloud data backtracking method and device provided in this embodiment realizes the backtracking of massive point cloud data based on the massive point cloud vector base map and iteration history, thereby solving the problem of point cloud data backtracking in high-precision map production.

[0078] In a possible implementation manner, in step S32, the rectangular grid is a positive direction grid with a grid side length of 2.4 kilometers, and the far point of the coordinate system is used as the starting position of the grid, and the horizontal position number and the vertical position number of the grid are recorded in the grid number.

[0079] It can be understood that, in this embodiment, setting in the above manner is more conducive to calculation.

[0080] In a possible embodiment, Figures 2 to 5 As shown, the vector lines below the legend represent the existing stored vector lines, and the vector lines below the existing stored vector lines represent the newly added vector lines; L1, L2, L3, and L4 represent the vector line IDs; the deletion line represents the deletion of the record; the brackets and L1shape represent the shape point data of the L1 vector line.

[0081] The storage rules in step S36 include:

[0082] If any part of the newly added vector line does not match the existing base map vector line, it will be stored in the vector line base map storage structure, and a new iteration history structure of the base map vector line will be added. The update type is new, such as Figure 2As shown;

[0083] If the newly added vector line completely matches the existing base map vector line, it is stored in the iterative history storage structure of the base map vector line it matches;

[0084] If the newly added vector line partially matches the existing base map vector line, the newly added vector line is divided into two parts: the unmatched part is stored in the vector line base map storage structure, and the data flag is set to the original vector line; the matched part is stored in the iterative history storage structure of the matched base map vector line, and the update type is set to update; Figure 3 As shown, L2 is the matching part and L3 is the non-matching part.

[0085] If the newly added vector line intersects with the existing base map vector line to form an intersection, such as Figure 4 As shown in the figure, the existing base map vector line needs to be interrupted at the intersection. The base map vector line record is deleted in the vector line base map storage structure. At the same time, a new record is added to its iterative history structure. The update type is deletion. The two interrupted vector lines are stored in the base map vector line storage structure respectively. At the same time, two new vector history storage structures are created. The status after storage is as follows Figure 5 shown.

[0086] In another embodiment, based on the above technical solution in the embodiment, a massive vehicle-mounted point cloud data backtracking device is provided, which includes: it adopts the above-mentioned base map vector line storage structure and iterative history storage structure, wherein:

[0087] The base map vector line storage structure provides a basic base map vector for subsequent data updates;

[0088] The iteration history storage structure provides data iteration update history for subsequent data backtracking.

[0089] Base map vector line storage structure: mainly records the data information of each vector line belonging to the base map, including the following fields: vector line ID, grid ID, station ID, original vector line ID, shape point data.

[0090] Iteration history storage structure: mainly records each iteration update record of each base map vector line, including the following fields: vector line ID, station ID, update type (new, update, delete), point data, and the iteration update records are sorted according to the data collection time.

[0091] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0092] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0093] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 computer, 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.

[0094] 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.

[0095] 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. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0096] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0097] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for tracing back massive vehicle-mounted point cloud data, characterized in that: It includes the following steps: S1: Point cloud data preprocessing, collecting information related to the measuring station and archiving and storing data files; S2: Extract vector lines from point cloud data, using the center line of the road surface of the vehicle-mounted point cloud data as the vector line of the point cloud data; S3: Back-tracing base map vector construction of massive point cloud data, performing grid cutting, vector integration, and spatial matching on the vector lines of point cloud data to form base map vectors; S4: Historical backtracking of massive point cloud data. By setting the spatial range, the associated vector lines are extracted, and then the iteration history records stored in the vector lines are read. Finally, the associated point cloud data is queried based on the iteration history records. The step S3 comprises: S31: To facilitate sequential generation of iteration history, each station information is sorted according to the acquisition date to form a processing station list, and subsequent processing processes each station information in turn according to the processing station list; S32: To facilitate rapid positioning during data tracing, the point cloud data vector line is cut into rectangular grids, and each point cloud data vector line after cutting records the corresponding grid number and station ID; S33: To ensure a one-to-one correspondence between the vector line and the point cloud data, the point cloud data is segmented according to the starting and ending positions of the vector line after cutting; first, a perpendicular line is drawn with the starting point or the end point of the vector line as the foot of the perpendicular; then, a position point of half the width of the road on both sides of the foot of the perpendicular is taken; secondly, the position points on both sides are sequentially connected to form a segmentation area; finally, the point cloud data is segmented according to the segmentation area, and the association relationship between the vector line and the point cloud is updated at the same time; S34: In order to reduce the number of vector lines in the grid, data integration is required, including merging and connecting the vector line data connected at the beginning and end of the same station in the same grid; merging the vector lines with a length less than a threshold in the grid into the connected vector lines of the same station. After completing the data integration, it is necessary to merge the point cloud data and update the association between the vector lines and the point cloud at the same time; S35: To distinguish the updated data and form an iteration history, spatial matching is performed on the vector lines in the same grid to determine the data update and data new acquisition areas. The matching is performed according to the following rules: The directions of vector lines on the same road are similar; The road width recorded by vector lines of the same road is similar; The distance between vector lines of the same road meets a certain threshold; S36: storing the newly added vector line in the base map vector line storage structure and the iteration history storage structure according to the set storage rule; The step S4 comprises: S41: query the point cloud vector line of the area to be queried, calculate the spatial grids passed by setting the contour points of the spatial range, calculate the spatial grids contained in the spatial range by traversing the grid numbers of each row and column, obtain all the vector line data of the spatial range by querying the base map vector line storage structure, and finally determine the target point cloud vector line through the matching rule of S35; S42: According to the point cloud vector lines obtained in S41, the station information related to all vector lines is obtained by querying the iterative history storage structure of each vector line; S43: According to the station information obtained in S42, the station data file is obtained by querying the storage path, and the backtracking of the point cloud data is completed.

2. A method for tracing back massive vehicle-mounted point cloud data according to claim 1, characterized in that: The step S1 comprises: S11: Information collection, collecting the station information of the point cloud data collected in the same batch to generate a unique code for the station, which is defined as the station ID; S12: Data collection: All data files collected in the same batch are stored in a directory named after the station ID. Different types of data files in the same batch are stored in different fixed directories under the directory, so that data backtracking can quickly obtain the corresponding data files according to the station ID and data type.

3. The method for tracing back massive vehicle-mounted point cloud data according to claim 1, characterized in that: The step S2 comprises: S21: Extract the plane from the point cloud using the RANSAC method, and select the plane with the largest area as the initial road surface; S22: according to the elevation difference characteristics of the road boundary and the road surface, spreading growth is performed on the edge of the initial road surface and road boundary point information is extracted, and abnormal boundary points in the boundary points are eliminated through threshold filtering and linear filtering; S23: According to the linear distribution law of the boundary points of the single-side road, the boundary point information of the branch road is filtered by the angle and length of the boundary points of the set range; S24: obtaining a quadratic polynomial expression of the road boundaries on both sides by linear fitting, and obtaining dense road boundary points on both sides by interpolation; S25: Traverse the boundary points of the road on one side in turn, find the nearest boundary point among the boundary points of the road on the other side, connect the midpoints of each pair of boundary points in turn to form a point cloud data vector line, record the corresponding relationship between each vector line and the point cloud data, and calculate the maximum value of the distance between each pair of boundary points as the road width record.

4. The method for tracing back massive vehicle-mounted point cloud data according to claim 1, characterized in that: In step S32, the rectangular grid is a positive direction grid, the grid side length is 2.4 kilometers, and the far point of the coordinate system is used as the starting position of the grid, and the horizontal position number and the vertical position number of the grid are recorded in the grid number.

5. The method for tracing back massive vehicle-mounted point cloud data according to claim 1, characterized in that: The storage rules in step S36 include: If any part of the newly added vector line does not match the existing base map vector line, it will be stored in the vector line base map storage structure, and a new iteration history structure of the base map vector line will be added at the same time, and the update type is new; If the newly added vector line completely matches the existing base map vector line, it is stored in the iterative history storage structure of the base map vector line it matches; If the newly added vector line partially matches the existing base map vector line, the newly added vector line is divided into two parts: the unmatched part is stored in the vector line base map storage structure, and the data flag is set to the original vector line; the matched part is stored in the iterative history storage structure of the base map vector line it matches, and the update type is set to update; If the newly added vector line intersects with the existing base map vector line to form an intersection, the existing base map vector line needs to be interrupted at the intersection, and the base map vector line record is deleted in the vector line base map storage structure. At the same time, a new record is added to its iterative history structure with the update type of deletion. The two interrupted vector lines are stored separately in the base map vector line storage structure, and two new vector history storage structures are created.

6. A massive vehicle-mounted point cloud data backtracking device, characterized in that: It adopts the massive vehicle-mounted point cloud data backtracking method described in claim 1, 4 or 5, wherein: The base map vector line storage structure provides a basic base map vector for subsequent data updates; The iteration history storage structure provides data iteration update history for subsequent data backtracking.

Images