A data processing method and device, electronic equipment and storage medium

Abstract

This invention discloses a data processing method, apparatus, electronic device, and storage medium. The method includes acquiring initial road trajectory data and initial road point cloud data corresponding to a target area; segmenting the initial road trajectory data; determining initial associated trajectory data corresponding to any trajectory data in the obtained multiple trajectory data segments; generating a target trajectory data pair based on any trajectory data in the initial associated trajectory data and the target trajectory data; repairing the multiple trajectory data segments based on the initial position transformation matrix between the corresponding point cloud data of the target trajectory data pair; and performing positioning calculation on the initial road point cloud data based on the repaired trajectory data to obtain the target road point cloud data. Using the embodiments disclosed in this invention, the offset of point cloud data can be automatically corrected, effectively improving the accuracy and efficiency of point cloud offset correction and ensuring the accuracy and consistency of point cloud data.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a data processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the rapid development of artificial intelligence technology, the demand for high-precision maps for autonomous driving is becoming increasingly urgent. Point cloud data is an important data source for producing high-precision maps, and the quality of point clouds directly determines the product quality of high-precision maps. Therefore, it is particularly important to produce point cloud maps that meet the requirements of accuracy, density, texture, and intensity.

[0003] In existing technologies, vehicle-mounted mobile measurement systems have been widely used in geospatial data production and updating, and have gradually become an important method for acquiring high-precision map data for autonomous driving. However, due to various factors such as vehicle mechanical characteristics, road flatness, and complex on-site environments, vehicle-mounted mobile measurement systems are prone to abnormal situations such as jitter. For example, in urban scenarios, the GNSS (Global Navigation Satellite System) signals received by the vehicle-mounted mobile measurement system may be blocked by buildings, trees, and other ground objects, resulting in signal loss or poor signal quality. This leads to data offsets at different times, easily causing inconsistencies in point clouds in the same area, which greatly affects the efficiency and quality of subsequent high-precision map generation. Summary of the Invention

[0004] To address the aforementioned problems in the prior art, this invention discloses a data processing method, apparatus, electronic device, and storage medium capable of automatically correcting the offset of point cloud data, effectively improving the accuracy and efficiency of point cloud offset correction, and ensuring the accuracy and consistency of point cloud data. The technical solution disclosed in this invention is as follows:

[0005] According to one aspect of the embodiments disclosed in this invention, a data processing method is provided, comprising:

[0006] Acquire the initial road trajectory data corresponding to the target area and the initial road point cloud data corresponding to the target area;

[0007] The initial road trajectory data is segmented to obtain multiple trajectory data segments;

[0008] Determine the initial associated trajectory data corresponding to the target trajectory data, wherein the target trajectory data is any trajectory data in the multiple trajectory data segments, and the initial associated trajectory data is the trajectory data in the multiple trajectory data segments whose acquisition time is before the target trajectory data and whose corresponding buffer area overlaps with the buffer area corresponding to the target trajectory data;

[0009] Based on any trajectory data in the initial associated trajectory data and the target trajectory data, a target trajectory data pair is generated;

[0010] Determine the initial position transformation matrix between the first point cloud data, wherein the first point cloud data is the point cloud data corresponding to the target trajectory data pair in the initial road point cloud data;

[0011] Based on the initial position transformation matrix, the multiple trajectory data segments are repaired to obtain repaired trajectory data;

[0012] Based on the repaired trajectory data, the initial road point cloud data is processed for positioning to obtain the target road point cloud data.

[0013] Optionally, the method further includes:

[0014] From the initial associated trajectory data, the trajectory data with the earliest acquisition time is determined as the target associated trajectory data of the target trajectory data;

[0015] Accordingly, generating a target trajectory data pair based on any trajectory data in the initial associated trajectory data and the target trajectory data includes:

[0016] Based on the target-related trajectory data and the target trajectory data, the target trajectory data pair is generated.

[0017] Optionally, the multiple trajectory data segments correspond to multiple target trajectory data pairs, and determining the initial position transformation matrix between the first point cloud data segments includes:

[0018] The first point cloud data is subjected to feature extraction processing to obtain feature information;

[0019] Based on the feature information, the similarity information between the first point cloud data is determined;

[0020] Based on the similarity information, the point cloud data corresponding to the multiple target trajectory data are sorted to obtain sequence information;

[0021] Based on the sequence information, each initial position transformation matrix is ​​determined using the point cloud data corresponding to the first target trajectory data in each target trajectory data pair as a reference, wherein the first target trajectory data is the trajectory data collected earlier.

[0022] Optionally, the multiple trajectory data segments correspond to multiple target trajectory data pairs, and the repair processing of the multiple trajectory data segments based on the initial position transformation matrix to obtain repaired trajectory data includes:

[0023] Based on the initial position transformation matrix corresponding to any two target trajectory data pairs containing the same trajectory data, a target position transformation matrix is ​​generated. The target position transformation matrix represents the position transformation relationship between the earliest point cloud data and the point cloud data corresponding to other trajectory data, with the earliest point cloud data as the reference. The other trajectory data are the trajectory data other than the earliest trajectory data in the two target trajectory data pairs. The earliest trajectory data is the trajectory data with the earliest acquisition time in the two target trajectory data pairs. The earliest point cloud data is the point cloud data corresponding to the earliest trajectory data.

[0024] Based on the target position transformation matrix, the multiple trajectory data segments are repaired to obtain the repaired trajectory data.

[0025] Optionally, the multi-segment trajectory data includes multiple trajectory point data, and the repair processing of the multi-segment trajectory data based on the target position transformation matrix to obtain repaired trajectory data includes:

[0026] Based on the target position transformation matrix, the multiple trajectory point data are corrected to obtain corrected multiple trajectory point data;

[0027] Anomaly repair processing is performed on the corrected multiple trajectory point data to obtain multiple repaired trajectory point data;

[0028] The repair trajectory data is obtained by smoothing the multiple repair trajectory point data.

[0029] Optionally, the initial road trajectory data includes multiple trajectory points sorted according to the collection time, and the segmentation of the initial road trajectory data to obtain multiple trajectory data segments includes:

[0030] Using the first trajectory point as the initial reference point, the trajectory points after the initial reference point are sequentially traversed according to the acquisition time.

[0031] If the current traversed trajectory points meet the preset conditions, the initial road trajectory data is segmented to obtain segmented trajectory data and segmented initial road trajectory data.

[0032] Using the current traversed trajectory point as the initial reference point, and based on the segmented initial trajectory data, repeat the step of sequentially traversing the trajectory points after the initial reference point according to the collection time sorting, until the initial road trajectory data is segmented to obtain segmented trajectory data when the current traversed trajectory point meets the preset conditions, until the traversal ends, and use the multiple segments of segmented trajectory data obtained during the traversal as the multiple segments of trajectory data.

[0033] Optionally, the method further includes:

[0034] Based on the multiple trajectory data, the first point cloud data is processed by positioning calculation to obtain the second point cloud data;

[0035] The determination of the initial position transformation matrix between the target trajectory data pairs and the corresponding first point cloud data includes:

[0036] Determine the initial position transformation matrix between the second point cloud data.

[0037] Optionally, before segmenting the initial road trajectory data to obtain multiple trajectory data segments, the method further includes:

[0038] The initial road trajectory data is deduplicated to obtain the deduplicated trajectory data;

[0039] The segmentation of the initial trajectory data to obtain multiple trajectory data segments includes:

[0040] The deduplicated trajectory data is segmented to obtain the multiple trajectory data segments.

[0041] According to another aspect of the disclosed embodiments of the present invention, a data processing apparatus is provided, comprising:

[0042] The data acquisition module is used to acquire the initial road trajectory data corresponding to the target area and the initial road point cloud data corresponding to the target area;

[0043] The segmentation processing module is used to segment the initial road trajectory data to obtain multiple segments of trajectory data;

[0044] The initial associated trajectory data determination module is used to determine the initial associated trajectory data corresponding to the target trajectory data. The target trajectory data is any trajectory data in the multiple trajectory data segments. The initial associated trajectory data is the trajectory data in the multiple trajectory data segments whose acquisition time is before the target trajectory data and whose corresponding buffer area overlaps with the buffer area corresponding to the target trajectory data.

[0045] The target trajectory data pair generation module is used to generate a target trajectory data pair based on any trajectory data in the initial associated trajectory data and the target trajectory data;

[0046] An initial position transformation matrix determination module is used to determine the initial position transformation matrix between first point cloud data, wherein the first point cloud data is the point cloud data corresponding to the target trajectory data pair in the initial road point cloud data;

[0047] The repair processing module is used to repair the multiple trajectory data based on the initial position transformation matrix to obtain repaired trajectory data;

[0048] The point cloud data generation module is used to perform positioning and calculation processing on the initial road point cloud data based on the repair trajectory data to obtain the target road point cloud data.

[0049] According to another aspect of the embodiments disclosed in this invention, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the data processing method as described in any of the preceding claims.

[0050] According to another aspect of the disclosed embodiments of the present invention, a computer-readable storage medium is provided, wherein when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the data processing method according to any one of the disclosed embodiments of the present invention.

[0051] According to another aspect of the disclosed embodiments of the present invention, a computer program product containing instructions is provided, which, when run on a computer, causes the computer to perform the data processing method described in any one of the disclosed embodiments of the present invention.

[0052] The technical solutions provided by the embodiments disclosed in this invention bring at least the following beneficial effects:

[0053] The data processing method provided by this invention segments the initial road trajectory data corresponding to the target area. Based on the buffer area and acquisition time corresponding to the obtained multiple trajectory data segments, it generates target trajectory data pairs with corresponding relationships. Then, based on the position transformation matrix between the corresponding point cloud data, it repairs the multiple trajectory data segments. Based on the repaired trajectory data, it performs positioning calculation on the initial road point cloud data to obtain the target road point cloud data. It can automatically correct the offset of the point cloud data, effectively improve the accuracy and efficiency of point cloud offset correction, and ensure the accuracy and uniformity of the point cloud data. Attached Figure Description

[0054] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the disclosure of this invention and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit the scope of this disclosure.

[0055] Figure 1 This is a flowchart illustrating a data processing method according to an exemplary embodiment;

[0056] Figure 2 This is a flowchart illustrating an initial position transformation matrix determination method according to an exemplary embodiment;

[0057] Figure 3 This is a flowchart illustrating a trajectory data repair method according to an exemplary embodiment;

[0058] Figure 4 This is a block diagram of a data processing apparatus according to an exemplary embodiment;

[0059] Figure 5 This is a block diagram illustrating a terminal electronic device for data processing according to an exemplary embodiment;

[0060] Figure 6 This is a block diagram illustrating a server electronic device for data processing according to an exemplary embodiment. Detailed Implementation

[0061] To enable those skilled in the art to better understand the technical solutions disclosed in this invention, the technical solutions in the disclosed embodiments will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0062] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention disclosed herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0063] This invention provides a data processing method that can be applied to the generation of map data.

[0064] Figure 1 This is a flowchart illustrating a data processing method according to an exemplary embodiment, such as... Figure 1 As shown, the data processing method includes the following steps.

[0065] S101: Obtain the initial road trajectory data corresponding to the target area and the initial road point cloud data corresponding to the target area.

[0066] In one specific embodiment, the target area can be the area where a map needs to be generated, the initial road trajectory data can be the trajectory data of roads within the target area, and the initial road point cloud data can be the point cloud data of roads within the target area. Both the initial road trajectory data and the initial road point cloud data can be collected by an on-board mobile measurement system. The mobile measurement system may include an IMU (Inertial Measurement Unit), GNSS (Global Navigation Satellite System), wheel speedometer, and lidar, etc. Specifically, the initial road trajectory data can be collected by the inertial measurement unit, the global navigation satellite system, and the wheel speedometer, and the initial road point cloud data can be collected by lidar.

[0067] In one specific embodiment, the point cloud data corresponding to the initial road trajectory data in the initial road point cloud data can be determined based on the acquisition time of the initial road trajectory data.

[0068] S103: The initial road trajectory data is segmented to obtain multiple trajectory data segments.

[0069] In an optional embodiment, the initial road trajectory data includes multiple trajectory points sorted according to the collection time, and the segmentation of the initial road trajectory data to obtain multiple trajectory data segments may include:

[0070] Using the first trajectory point as the initial reference point, the trajectory points after the initial reference point are sequentially traversed according to the acquisition time.

[0071] If the current traversed trajectory points meet the preset conditions, the initial road trajectory data is segmented to obtain segmented trajectory data and segmented initial road trajectory data.

[0072] Using the current traversed trajectory point as the initial reference point, and based on the segmented initial trajectory data, repeat the step of sequentially traversing the trajectory points after the initial reference point according to the collection time sorting, until the initial road trajectory data is segmented to obtain segmented trajectory data when the current traversed trajectory point meets the preset conditions, until the traversal ends, and use the multiple segments of segmented trajectory data obtained during the traversal as the multiple segments of trajectory data.

[0073] In one specific embodiment, the preset conditions can be set according to the actual application situation. Specifically, the preset conditions can be that the sum of the distances corresponding to the current traversed trajectory points is greater than a first preset distance threshold, and the first angle information corresponding to the current traversed trajectory points is greater than a first preset angle threshold, or the sum of the above-mentioned distances is greater than a first preset distance threshold, and the first time interval information corresponding to the current traversed trajectory points is greater than a first preset time threshold.

[0074] Specifically, the aforementioned cumulative distance can be the sum of the distances between adjacent trajectory points between the currently traversed trajectory point and the reference point. The first angle information can be the angle between the first connecting line and the second connecting line. The first connecting line can be the connecting line between the currently traversed trajectory point and the reference point, and the second connecting line can be the connecting line between the reference point and the second trajectory point. The first time interval information can be the acquisition time interval between the currently traversed trajectory point and the reference point. The aforementioned first preset distance threshold, first preset angle threshold, and first preset time threshold can be set according to the actual application situation. Specifically, the first preset distance threshold, first preset angle threshold, and first preset time threshold can be set according to the acquisition time sorting position of the currently traversed trajectory point in the initial trajectory data.

[0075] In an optional embodiment, the step of segmenting the initial road trajectory data to obtain multiple trajectory data segments may further include: when the acquisition time interval between any adjacent trajectory points in the initial road trajectory data is greater than a second preset time threshold, or the distance between any adjacent trajectory points is greater than a second preset distance threshold, the initial road trajectory data is segmented to obtain multiple trajectory data segments.

[0076] In one specific embodiment, any adjacent trajectory points are located in the two trajectory data segments obtained after one segmentation process. The aforementioned second preset distance threshold and second preset time threshold can be set according to the actual application situation. For example, the second preset distance threshold can be 70 meters, and the second preset time threshold can be 10 seconds.

[0077] In the above embodiments, segmenting the initial road trajectory data can segment the point cloud data corresponding to the initial road data. Subsequent positioning calculations and offset corrections can be performed based on the segmented trajectory data and the segmented point cloud data, which can improve processing efficiency.

[0078] In an optional embodiment, before segmenting the initial road trajectory data to obtain multiple trajectory data segments, the method may further include:

[0079] The initial road trajectory data is deduplicated to obtain the deduplicated trajectory data;

[0080] Accordingly, the segmentation of the initial trajectory data to obtain multiple trajectory data segments may include:

[0081] The deduplicated trajectory data is segmented to obtain the multiple trajectory data segments.

[0082] In an optional embodiment, the initial road trajectory data includes multiple trajectory points sorted by collection time. The initial road trajectory data is then deduplicated to obtain the deduplicated trajectory data, which may include:

[0083] The multiple trajectory points are sequentially traversed according to the collection time. Neighboring trajectory points within a preset range of the current traversed trajectory point are determined. If the collection time interval between the current traversed trajectory point and the neighboring trajectory point is greater than a second preset time threshold, and the difference between the second angle information corresponding to the current traversed trajectory point and the third angle information corresponding to the neighboring trajectory point is less than the second preset angle threshold, the neighboring trajectory point is marked as a redundant trajectory point. This process continues until the traversal ends. The redundant trajectory points are then deleted, and the initial road trajectory data after deleting the redundant trajectory points is used as the deduplicated trajectory data.

[0084] In one specific embodiment, the acquisition time of the currently traversed trajectory point is before that of the neighboring trajectory points.

[0085] In one specific embodiment, the second angle information can be the angle between the first direction vector and the second direction vector. The first direction vector can be the direction vector between the first adjacent trajectory point and the currently traversed trajectory point. The second direction vector can be the direction vector between the currently traversed trajectory point and the second adjacent trajectory point. The first adjacent trajectory point can be the previous adjacent trajectory point of the currently traversed trajectory point. The second adjacent trajectory point can be the next adjacent trajectory point of the currently traversed trajectory point. The third angle information can be the angle between the third direction vector and the fourth direction vector. The third direction vector can be the direction vector between the third adjacent trajectory point and the neighboring trajectory point. The second direction vector can be the direction vector between the neighboring trajectory point and the fourth adjacent trajectory point. The third adjacent trajectory point can be the previous adjacent trajectory point of the neighboring trajectory point. The fourth adjacent trajectory point can be the next adjacent trajectory point of the neighboring trajectory point.

[0086] In one specific embodiment, the aforementioned preset range, second preset time threshold, and second preset angle threshold can be set according to the actual application situation. For example, the second preset angle threshold can be 5°.

[0087] In the above embodiments, if any trajectory point and its neighboring trajectory points in the initial road trajectory data meet the preset conditions, the neighboring trajectory points are deleted as redundant trajectory points. This can effectively delete redundant trajectory data generated by repeated collection during the data acquisition process, improve the effectiveness of trajectory data, and thus improve the accuracy and efficiency of subsequent positioning and calculation processing based on trajectory data.

[0088] In an optional embodiment, the initial road trajectory data includes multiple trajectory points sorted by acquisition time and the speed information corresponding to each trajectory point. The initial road trajectory data is then deduplicated to obtain the deduplicated trajectory data, which may further include:

[0089] If the velocity information corresponding to any trajectory point is less than the velocity threshold, and the acquisition time interval between any trajectory point and the third trajectory point is greater than the third preset time threshold, the trajectory points between any trajectory point and the third trajectory point are marked as redundant trajectory points; the redundant trajectory points are deleted to obtain the deduplicated trajectory data.

[0090] In one specific embodiment, the third trajectory point can be the first trajectory point after any of the above trajectory points and whose speed information is greater than the above speed threshold. Specifically, the above speed threshold and the third preset time threshold can be set according to the actual application situation. For example, the speed threshold can be 0.01 m / s and the third preset time threshold can be 5 seconds.

[0091] In the above embodiments, when the trajectory points in the initial road trajectory data meet the preset conditions, redundant trajectory points are determined and deleted. This can effectively delete redundant trajectory data generated by parking during the data collection process, improve the effectiveness of the trajectory data, and thus improve the accuracy and efficiency of subsequent positioning and calculation processing based on the trajectory data.

[0092] S105: Determine the initial associated trajectory data corresponding to the target trajectory data.

[0093] In one specific embodiment, the target trajectory data is any trajectory data among the multiple trajectory data segments, and the initial associated trajectory data is the trajectory data among the multiple trajectory data segments whose acquisition time is before the target trajectory data and whose corresponding buffer area overlaps with the buffer area corresponding to the target trajectory data. Specifically, the buffer area corresponding to the trajectory data can be a polygonal area enclosed by a line connecting the trajectory points in the trajectory data as the central axis and a preset distance from the central axis. Specifically, the preset distance can be set according to the actual application situation.

[0094] S107: Generate a target trajectory data pair based on any trajectory data in the initial associated trajectory data and the target trajectory data.

[0095] In an optional embodiment, the method may further include:

[0096] From the initial associated trajectory data, the trajectory data with the earliest acquisition time is determined as the target associated trajectory data of the target trajectory data;

[0097] Accordingly, generating a target trajectory data pair based on any trajectory data in the initial associated trajectory data and the target trajectory data may include:

[0098] Based on the target-related trajectory data and the target trajectory data, the target trajectory data pair is generated.

[0099] In one specific embodiment, the target trajectory data pair can be determined by constructing a directed spanning tree. The target trajectory data pair is used as an edge of the directed spanning tree, and the reciprocal of the timestamp of the trajectory data with the earlier acquisition time in the target trajectory data pair is used as the weight of the edge. When the directed spanning tree is a directed minimum spanning tree, the target trajectory data pair is the trajectory data pair generated by the target trajectory data and the initial associated trajectory data with the earliest acquisition time.

[0100] In the above embodiments, the trajectory data collected before the target trajectory data in the multiple trajectory data is used as the initial associated trajectory data, and the target trajectory data is generated by combining any one of the trajectory data and the target trajectory data. This can obtain the continuous correspondence of multiple trajectory data based on at least one trajectory data, thereby enabling the subsequent determination of the position transformation relationship based on the point cloud data corresponding to at least one trajectory data, thus improving processing efficiency.

[0101] S109: Determine the initial position transformation matrix between the first point cloud data.

[0102] In one specific embodiment, the first point cloud data can be the point cloud data corresponding to the target trajectory data pair in the initial road point cloud data, and the initial position transformation matrix can characterize the position transformation relationship between the first point cloud data.

[0103] In an optional embodiment, determining the initial position transformation matrix between the first point cloud data may include: determining the initial rotation transformation matrix between the first point cloud data; and determining the initial position transformation matrix based on the initial rotation transformation matrix.

[0104] In one specific embodiment, the initial rotation transformation matrix can characterize the rotation transformation relationship between the first point cloud data. The initial rotation transformation matrix can be determined according to the first point cloud registration algorithm. Specifically, the first point cloud registration algorithm can be a robust registration algorithm based on the truncated least squares cost function. The initial position transformation matrix can be determined according to the second point cloud registration algorithm. For example, the second point cloud registration algorithm can be the Iterative Closest Point (ICP) algorithm.

[0105] In an optional embodiment, Figure 2 This is a flowchart illustrating an initial position transformation matrix determination method according to an exemplary embodiment, such as... Figure 2 As shown, the multiple trajectory data segments correspond to multiple target trajectory data pairs, and determining the initial position transformation matrix between the first point cloud data segments may include:

[0106] S201: Perform feature extraction processing on the first point cloud data to obtain feature information.

[0107] In one specific embodiment, feature information can characterize the density of point cloud data.

[0108] In an optional embodiment, feature extraction processing of the first point cloud data to obtain feature information may include:

[0109] Based on the first preset algorithm, multiple feature points of the first point cloud data are extracted; the initial feature information corresponding to each feature point is calculated; based on the second preset algorithm, the multiple feature points are clustered to obtain multiple target feature points; based on the initial feature information of the multiple target feature points, the feature information of the first point cloud data is determined.

[0110] In one specific embodiment, the first preset algorithm can be an algorithm for extracting feature points, such as principal component analysis; the second preset algorithm can be an algorithm for clustering, specifically, the second preset algorithm can be a density-based clustering algorithm, such as DBSCAN algorithm (Density-Based Spatial Clustering of Applications with Noise); the value of the feature information of the first point cloud data can be the sum of the residuals of the initial feature information of multiple target feature points.

[0111] In one specific embodiment, the above calculation of the initial feature information corresponding to each feature point may include: establishing a local coordinate system corresponding to each feature point with each feature point as the origin; and performing normalization and binarization processing on the coordinate information of other points in the local coordinate system besides the feature points of the first point cloud data to obtain the initial feature information.

[0112] In an optional embodiment, the method may further include: performing sparsification processing on the first point cloud data to obtain the third point cloud data;

[0113] Accordingly, performing feature extraction processing on the first point cloud data to obtain feature information may include performing feature extraction processing on the third point cloud data to obtain feature information.

[0114] In one specific embodiment, sparse processing of the first point cloud data to obtain the third point cloud data may include resampling and denoising of the first point cloud data to obtain the third point cloud data.

[0115] In the above embodiments, sparsification of point cloud data can improve the accuracy of point cloud data, thereby improving the accuracy of subsequent offset correction processing, while reducing the amount of computation in the offset correction process and improving efficiency.

[0116] S203: Based on the feature information, determine the similarity information between the first point cloud data.

[0117] In one specific embodiment, similarity information can characterize the degree of similarity between the first point cloud data. The value of similarity information can be the reciprocal of the difference between the aforementioned feature information of the first point cloud data. Specifically, the larger the value of similarity information, the higher the degree of similarity between the first point cloud data.

[0118] S205: Based on the similarity information, sort the point cloud data corresponding to the multiple target trajectory data pairs to obtain sequence information.

[0119] In one specific embodiment, the point cloud data corresponding to multiple target trajectory data pairs are sorted according to the similarity information values ​​from largest to smallest to obtain sequence information.

[0120] S207: Based on the sequence information, and taking the point cloud data corresponding to the first target trajectory data in each target trajectory data pair as a reference, determine each initial position transformation matrix.

[0121] In one specific embodiment, the first target trajectory data is trajectory data collected earlier.

[0122] In the above embodiments, the data is sorted according to the similarity between the first point cloud data, and the initial position transformation matrix is ​​determined based on the sequence information to ensure the orderliness of the processing and improve the efficiency of determining the initial position transformation matrix.

[0123] In an optional embodiment, the method may further include:

[0124] Based on the multiple trajectory data, the first point cloud data is processed by positioning calculation to obtain the second point cloud data;

[0125] Accordingly, determining the initial position transformation matrix between the target trajectory data pair and the corresponding first point cloud data may include:

[0126] Determine the initial position transformation matrix between the second point cloud data.

[0127] In one specific embodiment, based on the repaired trajectory data and the multiple trajectory segments, performing positioning calculations on the first point cloud data to obtain the second point cloud data may include: performing positioning calculations on the first point cloud data based on the multiple trajectory segments and preset calibration parameters to obtain the second point cloud data. Specifically, the preset calibration parameters may be the calibration parameters between the inertial measurement unit and the lidar in the vehicle-mounted mobile measurement system.

[0128] S111: Based on the initial position transformation matrix, the multiple trajectory data segments are repaired to obtain repaired trajectory data.

[0129] In an optional embodiment, the multiple trajectory data segments correspond to multiple target trajectory data pairs, and the repair processing of the multiple trajectory data segments based on the initial position transformation matrix to obtain repaired trajectory data may include:

[0130] Generate the target position transformation matrix based on the initial position transformation matrix corresponding to any two target trajectory data pairs containing the same trajectory data;

[0131] Based on the target position transformation matrix, the multiple trajectory data segments are repaired to obtain the repaired trajectory data.

[0132] In one specific embodiment, the target position transformation matrix can characterize the position transformation relationship between the earliest point cloud data and the point cloud data corresponding to other trajectory data, based on the earliest point cloud data. The other trajectory data can be the trajectory data other than the earliest trajectory data in any two target trajectory data pairs. The earliest trajectory data can be the trajectory data with the earliest acquisition time in any two target trajectory data pairs. The earliest point cloud data can be the point cloud data corresponding to the earliest trajectory data.

[0133] In one specific embodiment, the target position transformation matrix can be obtained by multiplying the corresponding initial position transformation matrices of any two target trajectory data containing the same trajectory data sequentially, based on the earliest point cloud data.

[0134] In the above embodiments, when any two target trajectory data pairs contain the same trajectory data, the target position transformation matrix is ​​determined based on the point cloud data corresponding to the trajectory data with the earliest acquisition time. This can unify the spatial reference of the point cloud data, thereby improving the accuracy and efficiency of subsequent offset correction, and also enhancing applicability.

[0135] In an optional embodiment, Figure 3 This is a flowchart illustrating a trajectory data repair method according to an exemplary embodiment, such as... Figure 3 As shown, the multi-segment trajectory data includes multiple trajectory point data. The repair processing of the multi-segment trajectory data based on the target position transformation matrix to obtain repaired trajectory data may include:

[0136] S301: Based on the target position transformation matrix, the multiple trajectory point data are corrected to obtain corrected multiple trajectory point data.

[0137] In one specific embodiment, the above-mentioned correction processing of the plurality of trajectory point data based on the target position transformation matrix to obtain the corrected plurality of trajectory point data may include performing corresponding translation and rotation on the plurality of trajectory point data based on the target position transformation matrix to obtain the corrected plurality of trajectory point data.

[0138] S303: Perform anomaly repair processing on the corrected multiple trajectory point data to obtain multiple repaired trajectory point data.

[0139] In an optional embodiment, the above-described anomaly repair processing of the corrected multiple trajectory point data to obtain multiple repaired trajectory point data may include:

[0140] If any trajectory point data has a horizontal anomaly, the average value of the coordinate information of the trajectory points within the preset repair interval is used as the repair coordinate information of any trajectory point, thus obtaining multiple repair trajectory point data.

[0141] If any trajectory point data has an elevation direction anomaly, determine the vector distance information from multiple corrected trajectory points to adjacent trajectory points. If the vector distance information exceeds the preset vector distance threshold, determine the elevation information of each repaired trajectory point based on the distance information corresponding to each corrected trajectory point and the elevation information of adjacent trajectory points, thus obtaining multiple repaired trajectory point data.

[0142] In one specific embodiment, a horizontal anomaly in the data of any trajectory point can be defined as the distance between adjacent trajectory points of any trajectory point being greater than a third preset distance threshold, and the fourth angle information corresponding to any trajectory point being less than a third preset angle threshold. The fourth angle information can be the angle between any trajectory point and the line connecting adjacent trajectory points. Specifically, the preset repair interval range, the third preset distance threshold, and the third preset angle threshold can be set according to the actual application situation. For example, the third preset angle threshold can be 150°.

[0143] In one specific embodiment, the elevation direction can be the direction along the normal of the ellipsoid of the WGS-84 coordinate system (World Geodetic System-1984 Coordinate System), and the distance information can include first distance information and second distance information. The first distance information can be the distance between the fifth adjacent trajectory point of the corrected trajectory point and the fifth adjacent trajectory point can be the previous adjacent trajectory point of the corrected trajectory point. The second distance information can be the distance between the corrected trajectory point and the sixth adjacent trajectory point and the sixth adjacent trajectory point can be the next adjacent trajectory point of the corrected trajectory point.

[0144] In one specific embodiment, the elevation information of the repaired trajectory points can be represented as:

[0145]

[0146] Wherein, H represents the elevation information of the repaired trajectory point, H1 represents the elevation information of the fifth adjacent trajectory point, D1 represents the first distance information, H2 represents the elevation information of the sixth adjacent trajectory point, and D2 represents the second distance information.

[0147] In one specific embodiment, the preset vector distance threshold can be set according to the actual application situation. For example, the preset vector distance threshold can be 0.015 meters.

[0148] S305: Smooth the multiple repair trajectory point data to obtain the repair trajectory data.

[0149] In one specific embodiment, the multiple repair trajectory point data are fitted to obtain a fitted curve; the fitted curve is then smoothed to obtain the repair trajectory data.

[0150] S113: Based on the repaired trajectory data, the initial road point cloud data is processed for positioning to obtain the target road point cloud data.

[0151] In one specific embodiment, the process of performing localization calculations on the initial road point cloud data based on the repaired trajectory data to obtain the target road point cloud data may include: performing localization calculations on the initial road point cloud data based on the repaired trajectory data and preset calibration parameters to obtain the target road point cloud data. Specifically, the preset calibration parameters may be the calibration parameters between the inertial measurement unit and the lidar in the vehicle-mounted mobile measurement system.

[0152] As can be seen from the technical solutions provided in the embodiments of this specification above, the initial road trajectory data corresponding to the target area is segmented. Based on the buffer areas and acquisition times corresponding to the obtained multiple trajectory data segments, target trajectory data pairs with corresponding relationships are generated. Then, based on the position transformation matrix between the corresponding point cloud data of the target trajectory data pairs, the multiple trajectory data segments are repaired. Based on the repaired trajectory data, the initial road point cloud data is used for positioning calculation to obtain the target road point cloud data. This can automatically correct the offset of the point cloud data, effectively improving the accuracy and efficiency of point cloud offset correction and ensuring the accuracy and uniformity of the point cloud data. Furthermore, deduplication of the initial road trajectory data can improve the effectiveness of the trajectory data, thereby improving the accuracy and efficiency of positioning calculation based on the trajectory data. In addition, when any two target trajectory data pairs contain the same trajectory data, the position transformation matrix is ​​determined based on the point cloud data corresponding to the trajectory data with the earliest acquisition time. This can unify the spatial reference of the point cloud data, thereby improving the accuracy and efficiency of offset correction and enhancing applicability.

[0153] Figure 4 This is a block diagram of a data processing apparatus according to an exemplary embodiment. (Refer to...) Figure 4 The device may include:

[0154] The data acquisition module 410 is used to acquire the initial road trajectory data corresponding to the target area and the initial road point cloud data corresponding to the target area;

[0155] The segmentation processing module 420 is used to segment the initial road trajectory data to obtain multiple segments of trajectory data;

[0156] The initial associated trajectory data determination module 430 is used to determine the initial associated trajectory data corresponding to the target trajectory data;

[0157] The target trajectory data pair generation module 440 is used to generate a target trajectory data pair based on any trajectory data in the initial associated trajectory data and the target trajectory data;

[0158] The initial position transformation matrix determination module 450 is used to determine the initial position transformation matrix between the first point cloud data.

[0159] The repair processing module 460 is used to repair the multiple trajectory data based on the initial position transformation matrix to obtain repaired trajectory data;

[0160] The point cloud data generation module 470 is used to perform positioning calculation on the initial road point cloud data based on the repair trajectory data to obtain the target road point cloud data.

[0161] Optionally, the device may further include:

[0162] The target associated trajectory data determination module is used to determine the trajectory data with the earliest acquisition time from the initial associated trajectory data as the target associated trajectory data of the target trajectory data;

[0163] Accordingly, the target trajectory data generation module 440 may include:

[0164] The target trajectory data pair generation unit is used to generate the target trajectory data pair based on the target associated trajectory data and the target trajectory data.

[0165] Optionally, the multiple trajectory data segments correspond to multiple target trajectory data pairs, and the initial position transformation matrix determination module 450 may include:

[0166] The feature extraction and processing unit is used to perform feature extraction processing on the first point cloud data to obtain feature information;

[0167] A similarity information determination unit is used to determine the similarity information between the first point cloud data based on the feature information;

[0168] The sorting unit is used to sort the corresponding point cloud data of multiple target trajectory data based on the similarity information to obtain sequence information;

[0169] The first initial position transformation matrix determination unit is used to determine each initial position transformation matrix based on the sequence information and using the point cloud data corresponding to the first target trajectory data in each target trajectory data pair as a reference, wherein the first target trajectory data is the trajectory data collected earlier.

[0170] Optionally, the multiple trajectory data segments correspond to multiple target trajectory data pairs, and the repair processing module 460 may include:

[0171] The target position transformation matrix generation unit is used to generate a target position transformation matrix based on the initial position transformation matrix corresponding to any two target trajectory data pairs containing the same trajectory data. The target position transformation matrix represents the position transformation relationship between the earliest point cloud data and the point cloud data corresponding to other trajectory data, with the earliest point cloud data as the reference. The other trajectory data are the trajectory data other than the earliest trajectory data in any two target trajectory data pairs. The earliest trajectory data is the trajectory data with the earliest acquisition time in any two target trajectory data pairs. The earliest point cloud data is the point cloud data corresponding to the earliest trajectory data.

[0172] The repair processing unit is used to repair the multiple trajectory data based on the target position transformation matrix to obtain the repaired trajectory data.

[0173] Optionally, the multi-segment trajectory data includes multiple trajectory point data, and the repair processing unit may include:

[0174] The correction processing unit is used to perform correction processing on the multiple trajectory point data based on the target position transformation matrix to obtain corrected multiple trajectory point data.

[0175] An anomaly repair processing unit is used to perform anomaly repair processing on the corrected multiple trajectory point data to obtain multiple repaired trajectory point data.

[0176] A smoothing processing unit is used to smooth the multiple repair trajectory point data to obtain the repair trajectory data.

[0177] Optionally, the initial road trajectory data includes multiple trajectory points sorted by acquisition time, and the segmentation processing module 420 may include:

[0178] The traversal unit is used to traverse the multiple trajectory points after the initial reference point in sequence according to the acquisition time, with the first trajectory point as the initial reference point.

[0179] The segmentation processing unit is used to segment the initial road trajectory data when the current traversed trajectory point meets the preset conditions, so as to obtain segmented trajectory data and segmented initial road trajectory data.

[0180] The repeating unit is used to take the current traversed trajectory point as the initial reference point, and based on the segmented initial trajectory data, repeat the step of sequentially traversing the trajectory points after the initial reference point according to the acquisition time, until the initial road trajectory data is segmented to obtain segmented trajectory data when the current traversed trajectory point meets the preset conditions, until the traversal ends, and the multiple segments of segmented trajectory data obtained during the traversal are taken as the multiple segments of trajectory data.

[0181] Optionally, the device may further include:

[0182] The second point cloud data generation module is used to perform positioning and calculation processing on the first point cloud data based on the multiple trajectory data to obtain the second point cloud data;

[0183] Correspondingly, the initial position transformation matrix determination module 450 may also include:

[0184] The second initial position transformation matrix determination unit is used to determine the initial position transformation matrix between the second point cloud data.

[0185] Optionally, prior to the segmentation processing module 420, the apparatus may further include:

[0186] The deduplication module is used to deduplicatize the initial road trajectory data to obtain deduplicated trajectory data.

[0187] Correspondingly, the segmentation processing module 420 may also include:

[0188] The segmentation processing unit is used to segment the deduplicated trajectory data to obtain the multi-segment trajectory data.

[0189] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0190] Figure 5 This is a block diagram illustrating an electronic device for data processing according to an exemplary embodiment. The electronic device may be a terminal, and its internal structure diagram may be as follows: Figure 5As shown, the electronic device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a data processing method. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0191] Figure 6 This is a block diagram illustrating an electronic device for data processing according to an exemplary embodiment. The electronic device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, this electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a data processing method.

[0192] Those skilled in the art will understand that Figure 5 or Figure 6 The structures shown are merely block diagrams of some structures related to the disclosed solutions of this invention, and do not constitute a limitation on the electronic devices to which the disclosed solutions of this invention are applied. Specific electronic devices may include more or fewer components than those shown in the figures, or combine certain components, or have different component arrangements.

[0193] In an exemplary embodiment, an electronic device is also provided, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement a data processing method as disclosed in the embodiments of the present invention.

[0194] In an exemplary embodiment, a computer-readable storage medium is also provided, wherein when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the data processing method disclosed in this invention.

[0195] In an exemplary embodiment, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform the data processing method disclosed in this invention.

[0196] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0197] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles disclosed herein and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0198] It should be understood that the present invention is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A data processing method, characterized in that, include: Acquire the initial road trajectory data corresponding to the target area and the initial road point cloud data corresponding to the target area; The initial road trajectory data is segmented to obtain multiple trajectory data segments; Determine the initial associated trajectory data corresponding to the target trajectory data, wherein the target trajectory data is any trajectory data in the multiple trajectory data segments, and the initial associated trajectory data is the trajectory data in the multiple trajectory data segments whose acquisition time is before the target trajectory data and whose corresponding buffer area overlaps with the buffer area corresponding to the target trajectory data; Based on any trajectory data in the initial associated trajectory data and the target trajectory data, a target trajectory data pair is generated; Determine the initial position transformation matrix between the first point cloud data, wherein the first point cloud data is the point cloud data corresponding to the target trajectory data pair in the initial road point cloud data; Based on the initial position transformation matrix, the multiple trajectory data segments are repaired to obtain repaired trajectory data; Based on the repaired trajectory data, the initial road point cloud data is processed for positioning to obtain the target road point cloud data.

2. The data processing method according to claim 1, characterized in that, The method further includes: From the initial associated trajectory data, the trajectory data with the earliest acquisition time is determined as the target associated trajectory data of the target trajectory data; The step of generating a target trajectory data pair based on any trajectory data in the initial associated trajectory data and the target trajectory data includes: Based on the target-related trajectory data and the target trajectory data, the target trajectory data pair is generated.

3. The data processing method according to claim 1, characterized in that, The multiple trajectory data segments correspond to multiple target trajectory data pairs. Determining the initial position transformation matrix between the first point cloud data includes: The first point cloud data is subjected to feature extraction processing to obtain feature information; Based on the feature information, the similarity information between the first point cloud data is determined; Based on the similarity information, the point cloud data corresponding to the multiple target trajectory data are sorted to obtain sequence information; Based on the sequence information, each initial position transformation matrix is ​​determined using the point cloud data corresponding to the first target trajectory data in each target trajectory data pair as a reference, wherein the first target trajectory data is the trajectory data collected earlier.

4. The data processing method according to claim 1, characterized in that, The multiple trajectory data segments correspond to multiple target trajectory data pairs. The repair processing of the multiple trajectory data segments based on the initial position transformation matrix to obtain repaired trajectory data includes: Based on the initial position transformation matrix corresponding to any two target trajectory data pairs containing the same trajectory data, a target position transformation matrix is ​​generated. The target position transformation matrix represents the position transformation relationship between the earliest point cloud data and the point cloud data corresponding to other trajectory data, with the earliest point cloud data as the reference. The other trajectory data are the trajectory data other than the earliest trajectory data in the two target trajectory data pairs. The earliest trajectory data is the trajectory data with the earliest acquisition time in the two target trajectory data pairs. The earliest point cloud data is the point cloud data corresponding to the earliest trajectory data. Based on the target position transformation matrix, the multiple trajectory data segments are repaired to obtain the repaired trajectory data.

5. The data processing method according to claim 4, characterized in that, The multi-segment trajectory data includes multiple trajectory point data. The repair processing of the multi-segment trajectory data based on the target position transformation matrix yields repaired trajectory data, including: Based on the target position transformation matrix, the multiple trajectory point data are corrected to obtain corrected multiple trajectory point data. Anomaly repair processing is performed on the corrected multiple trajectory point data to obtain multiple repaired trajectory point data; The repair trajectory data is obtained by smoothing the multiple repair trajectory point data.

6. A data processing method according to any one of claims 1-5, characterized in that, The initial road trajectory data includes multiple trajectory points sorted by collection time. The segmentation of the initial road trajectory data to obtain multiple trajectory segments includes: Using the first trajectory point as the initial reference point, the trajectory points after the initial reference point are sequentially traversed according to the acquisition time. If the current traversed trajectory points meet the preset conditions, the initial road trajectory data is segmented to obtain segmented trajectory data and segmented initial road trajectory data. Using the current traversed trajectory point as the initial reference point, and based on the segmented initial trajectory data, repeat the step of sequentially traversing the trajectory points after the initial reference point according to the collection time sorting, until the initial road trajectory data is segmented to obtain segmented trajectory data when the current traversed trajectory point meets the preset conditions, until the traversal ends, and use the multiple segments of segmented trajectory data obtained during the traversal as the multiple segments of trajectory data.

7. A data processing method according to any one of claims 1-5, characterized in that, The method further includes: Based on the multiple trajectory data, the first point cloud data is processed by positioning calculation to obtain the second point cloud data; The step of determining the initial position transformation matrix between the target trajectory data pairs and the corresponding first point cloud data includes: Determine the initial position transformation matrix between the second point cloud data.

8. A data processing method according to any one of claims 1-5, characterized in that, Before segmenting the initial road trajectory data to obtain multiple trajectory data segments, the method further includes: The initial road trajectory data is deduplicated to obtain the deduplicated trajectory data; Accordingly, the segmentation of the initial road trajectory data to obtain multiple trajectory data segments includes: The deduplicated trajectory data is segmented to obtain the multiple trajectory data segments.

9. A data processing apparatus, characterized in that, include: The data acquisition module is used to acquire the initial road trajectory data corresponding to the target area and the initial road point cloud data corresponding to the target area; The segmentation processing module is used to segment the initial road trajectory data to obtain multiple segments of trajectory data; The initial associated trajectory data determination module is used to determine the initial associated trajectory data corresponding to the target trajectory data. The target trajectory data is any trajectory data in the multiple trajectory data segments. The initial associated trajectory data is the trajectory data in the multiple trajectory data segments whose acquisition time is before the target trajectory data and whose corresponding buffer area overlaps with the buffer area corresponding to the target trajectory data. The target trajectory data pair generation module is used to generate a target trajectory data pair based on any trajectory data in the initial associated trajectory data and the target trajectory data; An initial position transformation matrix determination module is used to determine the initial position transformation matrix between first point cloud data, wherein the first point cloud data is the point cloud data corresponding to the target trajectory data pair in the initial road point cloud data; The repair processing module is used to repair the multiple trajectory data based on the initial position transformation matrix to obtain repaired trajectory data; The point cloud data generation module is used to perform positioning and calculation processing on the initial road point cloud data based on the repair trajectory data to obtain the target road point cloud data.

10. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the data processing method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is able to perform the data processing method as described in any one of claims 1 to 8.

Images