Method and device for lane line inspection, electronic equipment and storage medium

Abstract

The application provides a lane line inspection method and device, electronic equipment and a storage medium. The method can include: obtaining a lane line to be inspected and a reference lane line corresponding to a road section to which the lane line to be inspected belongs, wherein the lane line to be inspected is a lane line identified from multiple rounds of collected data, and the reference lane line is a lane line identified from single round of collected data; matching the lane line to be inspected and the reference lane line to obtain at least one lane line matching pair; and for each lane line matching pair, inspecting the lane line to be inspected according to the reference lane line. The lane line identified from single round of collected data can be used as a true value, and the true value can be used to inspect the lane line to be inspected in a map, so that quality problems of the lane line in the map can be found in a timely manner.

Description

Technical Field

[0001] This application relates to the field of high-precision map technology, and in particular to a method, apparatus, electronic device and storage medium for lane line inspection. Background Technology

[0002] Autonomous driving is currently the trend in the development of intelligent vehicles, and high-definition maps are a key infrastructure for realizing autonomous driving. Lane lines are an important map element in high-definition maps. Currently, lane lines in high-definition maps are mainly generated by extracting and editing features from point cloud data collected by mobile measurement systems. After generating lane lines, how to check the generated lane lines to truly determine whether the lane lines in the high-definition map are correct is a problem faced by all high-definition map manufacturers. Summary of the Invention

[0003] This application provides a method, apparatus, electronic device, and storage medium for lane line inspection, so as to inspect the quality of lane lines in high-precision maps.

[0004] In a first aspect, embodiments of this application provide a method for lane line inspection, which may include:

[0005] The lane line to be inspected and the reference lane line corresponding to the road segment to which the lane line to be inspected belong are obtained. The lane line to be inspected is the lane line identified from multiple data collections, and the reference lane line is the lane line identified from a single data collection.

[0006] Match the lane line to be inspected with the reference lane line to obtain at least one lane line matching pair;

[0007] For each lane line matching pair, the lane line to be checked is checked based on the reference lane line.

[0008] Secondly, embodiments of this application provide a method for correcting lane lines in a high-precision map, which may include:

[0009] The lane markings to be inspected are checked using the methods described in the first aspect;

[0010] For lane lines that are identified as abnormal, obtain relevant information about the abnormal lane lines. The relevant information includes at least one of the following: map feature ID and location coordinates.

[0011] The relevant information will be synchronized with the corrector for correction.

[0012] Thirdly, embodiments of this application provide a lane line inspection device, which may include:

[0013] The lane line acquisition module is used to acquire the lane line to be inspected and the reference lane line corresponding to the road segment to which the lane line to be inspected belongs. The lane line to be inspected is the lane line identified from multiple data collections, and the reference lane line is the lane line identified from a single data collection.

[0014] The lane line matching module is used to match the lane line to be checked with the reference lane line to obtain at least one lane line matching pair;

[0015] The lane line inspection module is used to inspect the lane line to be inspected for each lane line matching pair based on the reference lane line.

[0016] Fourthly, embodiments of this application provide a high-precision map lane line correction device, which may include:

[0017] An inspection module is used to inspect the lane lines to be inspected based on the method involved in the first aspect;

[0018] The information acquisition module is used to acquire relevant information about lane lines that are found to be abnormal during inspection. The relevant information includes at least one of the following: map feature ID and location coordinates.

[0019] The correction module is used to synchronize relevant information with the correction party for correction.

[0020] Fifthly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor, when executing the computer program, implements the method described in any of the above-mentioned embodiments.

[0021] Sixthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in any of the preceding claims.

[0022] Compared with the prior art, this application has the following advantages:

[0023] According to the embodiments of this application, when inspecting lane lines in a high-precision map, the lane lines identified from a single pass of data collection can be used as ground truth. Using these ground truths to inspect the lane lines in the high-precision map can promptly identify quality issues with the lane lines. For example, in a smoothness check, non-smooth lane lines in the high-precision map can be detected in a timely manner.

[0024] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application, it can be implemented according to the contents of the specification. In order to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0025] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the various drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments according to this application and should not be construed as limiting the scope of this application.

[0026] Figure 1 A schematic diagram illustrating a scenario for the lane line inspection method provided in this application;

[0027] Figure 2 This is a flowchart of a lane line inspection method according to an embodiment of this application;

[0028] Figure 3 This is one of the lane line diagrams according to an embodiment of this application;

[0029] Figure 4 This is a second schematic diagram of lane lines according to an embodiment of this application;

[0030] Figure 5 This is a flowchart of a high-precision map lane line correction method according to an embodiment of this application;

[0031] Figure 6 This is a structural block diagram of a lane line inspection device according to an embodiment of this application;

[0032] Figure 7 This is a structural block diagram of a high-precision map lane line correction device according to an embodiment of this application; and

[0033] Figure 8 This is a block diagram of an electronic device used to implement embodiments of this application. Detailed Implementation

[0034] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the concept or scope of this application. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0035] To facilitate understanding of the technical solutions of the embodiments of this application, the relevant technologies of the embodiments of this application are described below. The following relevant technologies are optional solutions and can be combined with the technical solutions of the embodiments of this application in any way, and all of them fall within the protection scope of the embodiments of this application.

[0036] Figure 1This is a schematic diagram illustrating an application scenario for implementing the method of this application embodiment. A high-precision map database stores different lane lines, which are used to distinguish or determine lanes. Each lane contains at least two lane lines. The geographic coordinates of a lane line can be expressed in a geographic coordinate system by sampling points on the lane line. Based on the geographic coordinates of the lane line, a reference lane line for the corresponding location can be obtained. That is, based on the geographic coordinates, data corresponding to the geographic coordinates is selected. The data can include point cloud data or image data, etc., and the data is then identified to obtain the reference lane line. To improve the accuracy of the inspection, the data can be recently collected. For example, if the inspection time for the lane line is January 1, 2023, then the data collection time can be the most recently collected data within one year, the most recently collected data within six months, etc. The lane line identified from a single trip of collected data can be used as a reference lane line. The lane lines in the high-precision map database and the reference lane lines are matched together. That is, the lane lines in the matching pair correspond to the same lane line in both the high-precision map and the real environment. By using reference lane lines in the matching pairs to check the lane lines in the high-precision map database, results such as lane line smoothness and integrity can be obtained. Lane lines in the high-precision map database that fail the checks can be reported. The purpose of the report can be to notify staff to annotate or correct them, or to inform the high-precision map correction program to automatically correct the map. This improves the consistency between lane lines in the high-precision map database and lane lines in the real environment, thus enhancing the quality of high-precision map delivery.

[0037] This application provides a method for lane line inspection, such as... Figure 2 The diagram shown is a flowchart of a lane line inspection method according to an embodiment of this application, corresponding to the first implementation. The first implementation may include:

[0038] Step S201: Obtain the lane line to be inspected and the reference lane line corresponding to the road segment to which the lane line to be inspected belongs. The lane line to be inspected is the lane line identified from multiple data collections, and the reference lane line is the lane line identified from a single data collection.

[0039] The implementing entity of this application may be a map service platform. Lane lines in high-definition maps serve as map elements. Unchecked lane lines can be used as lane lines to be checked. In high-definition maps, each lane line corresponds to location information, which can be represented by the start and end positions of the lane line, or as... Figure 3 As shown, the positions of multiple sampling points in the lane lines are represented.

[0040] Based on the location representation of the lane line to be inspected, a reference lane line corresponding to the road segment to which the lane line to be inspected belongs can be obtained. The reference lane line can be obtained by identifying data from a single pass of real-world data. This data can include image data, point cloud data, etc. Through semantic recognition of the data, different objects in the data can be identified. For example, buildings, vehicles, and lane lines can be identified. For point cloud data, the position information of the lane line can be directly obtained. For image data, the position information of the lane line can be obtained based on the intrinsic and extrinsic parameters of the image acquisition device and the position information corresponding to the image acquired. Since the reference lane line is identified from a single pass of acquired data, the result is relatively more reliable, so it is used as a reference. The lane line to be inspected, however, is identified from multiple passes of acquired data. Lane lines identified from multiple passes of acquired data require a series of operations such as alignment. Due to issues with acquisition accuracy and alignment algorithms, lane lines identified from multiple passes of acquired data may exhibit unevenness, thus requiring inspection.

[0041] Step S202: Match the lane line to be checked with the reference lane line to obtain at least one lane line matching pair.

[0042] For the same road segment, there are usually multiple lane lines to be inspected, and correspondingly, multiple reference lane lines. Therefore, each lane line to be inspected needs to be paired with one reference lane line to form a lane line matching pair. Matching methods can include overlap matching, coordinate matching (of lane line sampling points), and traffic marking category matching. Traffic marking categories can include solid lines, dashed lines, etc.

[0043] Step S203: For each lane line matching pair, check the lane line to be checked according to the reference lane line.

[0044] The reference lane line in the lane line matching pair can be used as the ground truth. The lane line to be inspected can be inspected based on the differences between the lane line to be inspected and the reference lane line. For example, the smoothness of the lane line to be inspected can be determined based on the distance from each sampling point in the lane line to the reference lane line. The position and integrity of the lane line to be inspected can be determined based on the degree of overlap between the lane line to be inspected and the reference lane line.

[0045] In related technologies, lane line inspection in high-precision maps typically employs inspection algorithms to examine the lane lines themselves. For example, taking smoothness inspection as an example, related technologies often use smoothness inspection algorithms to check the smoothness of lane lines. If the inspection result shows non-smoothness, correction is required. However, there may be... Figure 4 In a real-world environment, this lane marking might indeed be non-smooth, leading to incorrect checks. For example, Figure 4The lane lines on both sides of the central lane are serrated, indicating that the central lane is a reversible lane. A reversible lane means that the driving situation in that lane can be adjusted according to road usage needs, such as straight-ahead, left-turn, or right-turn. If the lane lines are checked directly without using reference points, the jagged edges may result in an uneven judgment, leading to inconsistencies between the inspection results and the actual environment. Through the process described in this embodiment, when checking lane lines in a high-precision map, lane lines identified from a single-trip data collection can be used as ground truth. These ground truth values ​​are then used to check the lane lines in the high-precision map, allowing for the correction or reporting of lane lines in the high-precision map that differ from those identified from the single-trip data collection. The reported information may include the lane line identifier, lane line location, the high-precision map page number where the lane line is located, the city or district where the lane line is located, and the inspection results, facilitating rapid lane line location. The above inspection process can improve the consistency between lane lines in high-precision maps and lane lines in the real environment, thereby improving the delivery quality of high-precision maps.

[0046] In one implementation, the step S201 involving obtaining the reference lane line corresponding to the road segment to which the lane line to be inspected belongs may include:

[0047] Step S2011: Determine the information of the trajectory points corresponding to the lane line to be detected.

[0048] The trajectory points corresponding to the lane line to be detected can be the start and end points, or multiple trajectory points between the start and end points. The information of the trajectory points can include the coordinates of the trajectory points, the distance between the current trajectory point and adjacent trajectory points, etc.

[0049] Step S2012: Based on the information of the trajectory points, determine the reference lane line corresponding to the lane line to be inspected in the database; the database stores the reference lane lines.

[0050] Based on the information from the trajectory points, a database can be searched to determine the reference lane line corresponding to the lane line to be inspected. For example, the average coordinates of the trajectory points can be used as a basis, thereby accelerating the query efficiency in the database. Alternatively, information from each trajectory point can be used as a basis, thereby improving the query accuracy in the database.

[0051] In one implementation, the matching of the lane line to be inspected and the reference lane line in step S202 may include:

[0052] Step S2021: Segment the lane line to be inspected and the reference lane line to obtain lane lines to be inspected and reference lane lines with the same number of segments.

[0053] Segmentation can be done using the same distance. For example, each segment could be 50 meters or 100 meters. Ultimately, the lane lines to be inspected and the reference lane lines can be divided into multiple segments.

[0054] When segmenting, either direct segmentation or sliding window segmentation can be used. When using sliding window segmentation, the overlap of the sliding windows can be controlled. For example, if each segment is 100 meters long and the overlap is 50%, it means that the last 50 meters of the k-th segment and the first 50 meters of the (k+1)-th segment overlap. Here, k is a positive integer.

[0055] Step S2022: Within each segment, match the lane lines to be checked with the reference lane lines.

[0056] Within each segment, the matching of the lane line to be checked and the reference lane line can be based on predetermined rules. For example, the predetermined rules could be based on the traffic marking category of the lane lines, the distance between lane lines, etc.

[0057] In one implementation, the segmentation of the lane line to be inspected and the reference lane line involved in step S2021 includes:

[0058] Step S20211: Use a window of preset size to perform sliding window processing on the lane line to be inspected and the reference lane line to obtain multiple segments.

[0059] Before performing the sliding window operation, a coordinate system transformation step can be included. For example, the coordinate systems of the lane line to be inspected and the reference lane line are latitude and longitude coordinate systems. The coordinate system transformation can be performed using the midpoint of the lane line to be inspected or the midpoint of the reference lane line as the origin. This transforms the latitude and longitude coordinate system into a two-dimensional rectangular coordinate system. Therefore, in subsequent calculations, the values ​​represented by coordinate points (x, y) in the two-dimensional rectangular coordinate system can be used. Here, x corresponds to the x-axis coordinate in the two-dimensional rectangular coordinate system, and y corresponds to the y-axis coordinate in the two-dimensional rectangular coordinate system.

[0060] The preset window size can be 50 meters, 100 meters, 200 meters, etc. Taking 100 meters as an example, using a preset window size to perform sliding window processing on the lane line to be inspected and the reference lane line can mean dividing it into segments of 100 meters. Ultimately, you can obtain t segments of the lane line to be inspected and t segments of the reference lane line. The length of each segment is 100 meters. t can be a positive integer.

[0061] When segmenting the lane lines to be inspected and the reference lane lines using a sliding window, segments can be directly divided into 100-meter segments. The end point of the k-th segment is the start point of the (k+1)-th segment, where k can be a positive integer and k < t. Furthermore, an overlap area threshold can be set. For example, the specified overlap area threshold could be 30%, 50%, etc. Taking a segment length of 100 meters and an overlap area of ​​50% as an example, the last 50 meters of the k-th segment and the first 50 meters of the (k+1)-th segment are the same.

[0062] By using the above process, the lane lines to be inspected and the reference lane lines are segmented using a sliding window method, which can avoid missed calls at the segmentation points.

[0063] Step S20212: Within each segment, match the lane lines to be checked with the reference lane lines.

[0064] Typically, lane lines in high-precision maps include both up and down directions. For example, the up and down directions of the lane line to be inspected can be determined by reading map information. Based on this, after determining the up and down directions of the lane line to be inspected, reference lane lines can be filtered based on these directions. For instance, if the lane line to be inspected is a lane line that allows vehicles to travel from south to north, only reference lane lines that allow vehicles to travel from south to north can be retained. Therefore, filtering reference lane lines based on the up and down directions corresponding to the lane line to be inspected can effectively reduce the time spent or complexity of the subsequent matching process. Afterwards, matching can be performed based on attributes such as the distance and type between the lane line to be inspected and the reference lane lines.

[0065] In one implementation, the matching of the lane line to be inspected and the reference lane line involved in step S202 and step S2022 may include:

[0066] Step S2023: Determine the traffic marking category of the lane line to be inspected and the traffic marking category of the reference lane line.

[0067] Traffic markings can be categorized into double solid lines, single solid lines, and dashed lines. Corresponding functions include distinguishing between opposing lanes, distinguishing lanes traveling in the same direction, and defining road edges. The category of traffic markings can be determined by identifying their location and alignment.

[0068] Taking image data as an example, image data is typically acquired by image acquisition equipment installed on a data acquisition vehicle. Therefore, for image data, elements such as up / down lane dividers and lane boundaries generally exhibit better clarity, distortion, and completeness compared to road edge lines. Thus, it can be assumed that the confidence level of reference lane lines such as up / down lane dividers and lane boundaries is relatively high.

[0069] Step S2024: Determine the positional relationship between the lane line to be inspected and the reference lane line.

[0070] The positional relationship between the lane line to be inspected and the reference lane line can be determined using sampling points on the lane lines. Furthermore, the positional relationship between the lane line to be inspected and the reference lane line can also be determined using overlap. The positional relationship between the lane line to be inspected and each reference lane line can be determined separately.

[0071] Taking the determination using sampling points as an example, we can select three points: the start point, the midpoint, and the end point of the lane line to be inspected. Correspondingly, we select three points: the start point, the midpoint, and the end point of the reference lane line for which distance calculations are to be performed, where m is a positive integer. We calculate the first distance between the start point of the lane line to be inspected and the start point of the reference lane line, the second distance between the midpoint of the lane line to be inspected and the midpoint of the reference lane line, and the third distance between the end point of the lane line to be inspected and the end point of the reference lane line. The average of the first, second, and third distances determines the positional relationship. For example, a relatively small average value indicates that the lane lines to be inspected and the reference lane lines are relatively close, while a relatively large average value indicates that the lane lines to be inspected and the reference lane lines are relatively far apart.

[0072] Taking the use of overlap as an example, the starting and ending coordinates of the lane line to be inspected in the real environment can be calculated. Similarly, the starting and ending coordinates of the reference lane line in the real environment can be calculated. Thus, the lane line to be inspected and the reference lane line are placed in the same space for comparison to determine their overlap, and the positional relationship can be determined based on the overlap.

[0073] Step S2025: Match according to traffic marking category and position relationship to obtain at least one lane line matching pair.

[0074] Based on traffic marking categories, lane markings of the same category and their reference lane markings can be grouped into candidate matching pairs. Next, based on the positional relationship between the lane markings and reference lane markings in the candidate matching pairs, the closest reference lane marking is determined as the target reference lane marking. This forms a matching pair between the lane marking to be inspected and the target reference lane marking. If there are multiple lane markings to be inspected, they can be matched sequentially during the matching process. Typically, for lane markings to be inspected, priority can be given to checking up / down lane dividers and lane boundaries based on their importance or the confidence level of the reference lane markings.

[0075] In one implementation method, matching is performed based on traffic marking categories and positional relationships to obtain at least one lane line matching pair, which may specifically include:

[0076] The lane lines to be inspected that are of the same traffic marking category and whose positional relationship meets the specified requirements are matched with the reference lane lines.

[0077] Traffic markings of the same category can mean that the lane line to be inspected and the reference lane line both belong to the same category, such as a solid line or a dashed line. That is, only lanes of the same category can be compared in terms of positional relationship. Specified requirements may include an overlap degree not less than the corresponding overlap degree threshold and a distance between trajectory points not greater than the corresponding distance threshold. Based on this, lane lines to be inspected and reference lane lines that are of the same category and whose positional relationship meets the specified requirements can be paired.

[0078] In one implementation, the step S203, which involves checking the lane line to be checked based on the reference lane line, may include:

[0079] Step S2031: Determine multiple first sampling points of the lane line to be checked in the lane line matching pair, and determine multiple second sampling points of the reference lane line; the first sampling points and the second sampling points correspond one-to-one.

[0080] There can be multiple first sampling points. There can also be multiple second sampling points. The number of first and second sampling points can be the same. Taking the first sampling point as an example, the selection of sampling points will be explained. Starting from a starting point, sampling points can be selected at regular intervals until the end point. For example, the interval distance could be 20 cm, 50 cm, etc. Alternatively, the number of sampling points can be preset, such as 100. Then, based on the length of the lane line to be inspected and the preset number of sampling points, the interval between adjacent sampling points is obtained. The method for determining the second sampling point is the same as that for determining the first sampling point.

[0081] Step S2032: Calculate the distance between each first sampling point of the lane line to be inspected and the second sampling point corresponding to the reference lane line.

[0082] Since the first and second sampling points are in one-to-one correspondence, the distance can be calculated using the coordinates of the sampling points. The distance can be expressed as ds = {d1, d2, ..., d...} n}. Here, ds can represent the distance set. In the distance set, d1 can represent the distance between the first sampling point and the first sampling point. Similarly, d... n It can be represented as the distance between the nth first sampling point and the nth second sampling point. n is a positive integer.

[0083] Step S2033: Determine the inspection result using distance.

[0084] After obtaining the distance between the sampling points of the lane line to be inspected and the corresponding sampling points of the reference lane line, the mean of the distance can be calculated first. The mean of the distance is expressed as... In the formula d i This can be expressed as the distance between the i-th first sampling point and the i-th second sampling point. n can represent the number of sampling points. In one implementation, the average distance can be used as the inspection result to determine whether the lane lines to be inspected in the high-precision map are smooth.

[0085] Furthermore, the mean of the distances can be used for further calculations to inspect the lane markings. For example, the standard deviation can be calculated using the distances and their mean. The standard deviation can be represented as σ. Therefore, the standard deviation can be used as the inspection result to determine whether the lane lines to be inspected in the high-precision map are smooth. For checking the smoothness of the lane lines, a difference threshold can be preset. If the calculated standard deviation is not greater than the difference threshold, it indicates that the lane lines are smooth; conversely, if the calculated standard deviation is greater than the difference threshold, it indicates that the lane lines are not smooth. Similarly, if the mean distance is used for smoothness checking, the result can be obtained by comparing the mean distance with a mean inspection threshold.

[0086] Through the above process, if the standard deviation is used to inspect the lane lines to be inspected, the standard deviation can reflect the dispersion of a dataset. By comparing the dispersion of the lane lines to be inspected with that of the reference lane lines, the inspection results of the lane lines to be inspected can be obtained.

[0087] In one implementation, the step S2033 involving determining the inspection result using distance may include:

[0088] Step S20331: Based on the curvature of the lane line to be inspected, dynamically adjust the preset difference threshold to obtain the adjustment result.

[0089] In high-precision maps, lane lines to be inspected can include straight lane lines and curved lane lines. For curved lane lines, the curvature of the curved lane line can be determined first. Curvature represents the degree of bending of the curved lane line; the greater the curvature, the greater the degree of bending. The curvature of the lane line to be inspected can be determined based on the coordinates of the sampling points on the curved lane line and the curvature calculation formula.

[0090] The difference threshold can be dynamically adjusted based on the curvature. For example, for straight lane lines, the difference threshold can be a preset value, i.e., the adjustment is zero. For curved lane lines, the difference threshold can be adjusted based on the curvature to obtain the adjusted result. For example, a linear model can be set, where the horizontal axis can be curvature and the vertical axis can be the difference threshold. When the curvature is 0, the difference threshold is the preset value. As the curvature increases, the difference threshold increases accordingly.

[0091] The reason for setting a difference threshold is that, compared to straight lane lines, curved lane lines have slightly lower recognition accuracy during inspection or identification. This leads to a decrease in the accuracy of curved lane lines generated in high-precision maps or identified based on single-pass data collection from the real environment. Therefore, the difference threshold can be dynamically adjusted appropriately to offset the inspection error caused by the decreased accuracy of curve recognition to some extent.

[0092] Step S20332: Calculate the difference using distance. The difference is used to represent the degree of overlap between the reference lane line and the lane line to be inspected.

[0093] The process of calculating the degree of difference using distance can be the same as the steps described above; that is, the mean of the distance or the standard deviation of the distance can be used as the degree of difference. Both the mean and the standard deviation of the distance can represent the degree of overlap between the target reference lane line and the lane line to be inspected.

[0094] Step S20333: Determine the inspection result by comparing the difference with the adjustment result.

[0095] The difference is compared with the adjustment result. If the difference is no greater than the adjustment result, it indicates that the lane lines to be inspected in the high-precision map are relatively smooth. Conversely, if the difference is greater than the adjustment result, it indicates that the lane lines to be inspected in the high-precision map are not smooth.

[0096] This application provides a method for correcting lane lines in high-precision maps, such as... Figure 5 The diagram shown is a flowchart of a high-precision map lane line correction method according to an embodiment of this application, corresponding to the second embodiment. The second embodiment may include:

[0097] Step S501: Inspect the lane lines to be inspected using the method described in the first embodiment.

[0098] The inspection can be a regular check, such as weekly, monthly, or quarterly. Alternatively, the inspection can be triggered upon receiving a notification that lane markings have been adjusted or changed. The process for inspecting lane markings can be the same as in the first embodiment. By inspecting the lane markings in the high-precision map, inspection results can be obtained. The inspection results include whether correction is needed or not. Lane markings that do not require correction can be left unprocessed.

[0099] Step S502: For lane lines that are detected as abnormal, obtain relevant information about the abnormal lane lines. The relevant information includes at least one of the following: map feature ID and location coordinates.

[0100] For lane markings that require correction according to the inspection results, the relevant information of the lane markings that need correction can be determined first. This includes the map feature ID corresponding to the lane marking, the location of the lane marking, and the inspection results.

[0101] Step S503: Synchronize the relevant information with the corrector for correction.

[0102] The correction process for lane markings that require adjustment can include synchronizing relevant information about the lane markings to the correction party based on the inspection results. The correction party can be a third party or the entity executing the lane marking inspection method. The corrected high-precision map data can then be published as update data. Publishing the corrected high-precision map allows map clients to receive it and complete the update process.

[0103] Corresponding to the application scenarios and methods provided in the embodiments of this application, the embodiments of this application also provide a device for lane line inspection. For example... Figure 6 The diagram shown is a structural block diagram of a lane line inspection device according to an embodiment of this application, corresponding to the third embodiment. The lane line inspection device may include:

[0104] The lane line acquisition module 601 is used to acquire the lane line to be inspected and the reference lane line corresponding to the road segment to which the lane line to be inspected belongs. The lane line to be inspected is the lane line identified from multiple data collections, and the reference lane line is the lane line identified from a single data collection.

[0105] Lane matching module 602 is used to match the lane line to be checked with the reference lane line to obtain at least one lane line matching pair;

[0106] The lane line inspection module 603 is used to inspect the lane line to be inspected for each lane line matching pair based on the reference lane line.

[0107] In one embodiment, the lane line acquisition module 601 may include:

[0108] The trajectory point information determination submodule is used to determine the information of the trajectory points corresponding to the lane lines to be detected;

[0109] The reference lane line determination submodule is used to determine the reference lane line corresponding to the lane line to be inspected in the database based on the information of the trajectory points; the database stores the reference lane lines.

[0110] In one embodiment, the lane line matching module 602 may include:

[0111] The segmentation submodule is used to segment the lane lines to be inspected and the reference lane lines to obtain lane lines to be inspected and reference lane lines with the same number of segments.

[0112] The matching execution submodule is used to match the lane lines to be checked with the reference lane lines within each segment.

[0113] In one implementation, the segmented submodule can be specifically used for:

[0114] Using a window of a preset size, slide the window to be inspected and the reference lane line to obtain multiple segments;

[0115] The sliding window processing ensures that the overlap area between the current segment and the adjacent segment is not less than a specified overlap area threshold.

[0116] In one embodiment, the lane line matching module 602 may include:

[0117] The traffic marking category determination submodule is used to determine the traffic marking category of the lane line to be inspected and the traffic marking category of the reference lane line.

[0118] The positional relationship determination submodule is used to determine the positional relationship between the lane line to be inspected and the reference lane line;

[0119] The lane line matching execution submodule is used to perform matching based on traffic marking category and positional relationship to obtain at least one lane line matching pair.

[0120] In one embodiment, the lane line inspection module 603 may include:

[0121] The sampling point determination submodule is used to determine multiple first sampling points of the lane line to be checked in the lane line matching pair, and to determine multiple second sampling points of the reference lane line; the first sampling points and the second sampling points correspond one-to-one.

[0122] The distance calculation submodule is used to calculate the distance between each first sampling point of the lane line to be inspected and the corresponding second sampling point of the reference lane line.

[0123] The inspection result determination submodule is used to determine the inspection result using distance.

[0124] In one implementation, the distance calculation submodule may include:

[0125] The difference threshold determination unit is used to dynamically adjust the pre-set difference threshold according to the curvature of the lane line to be inspected, and obtain the adjustment result;

[0126] The difference calculation unit is used to calculate the difference using distance. The difference is used to represent the degree of overlap between the reference lane line and the lane line to be inspected.

[0127] The inspection result determination unit is used to determine the inspection result by comparing the difference with the adjustment result.

[0128] Corresponding to the application scenarios and methods provided in the embodiments of this application, the embodiments of this application also provide a high-precision map lane line correction device. For example... Figure 7 The diagram shown is a structural block diagram of a high-precision map lane line correction device according to an embodiment of this application, corresponding to the fourth embodiment. It may include:

[0129] Inspection module 701 is used to inspect the lane lines to be inspected based on the method involved in the first embodiment;

[0130] The information acquisition module 702 is used to acquire relevant information about the lane line to be inspected that is found to be abnormal, wherein the relevant information includes at least one of the following: map feature ID and location coordinates.

[0131] The correction module 703 is used to synchronize relevant information to the correction party for correction.

[0132] The functions of each module in each device in the embodiments of this application can be found in the corresponding description in the above method, and they have corresponding beneficial effects, which will not be repeated here.

[0133] Figure 8 This is a block diagram of an electronic device used to implement embodiments of this application. Figure 8 As shown, the electronic device includes a memory 810 and a processor 820. The memory 810 stores a computer program that can run on the processor 820. When the processor 820 executes the computer program, it implements the method described in the above embodiments. The number of memories 810 and processors 820 can be one or more.

[0134] The electronic device also includes:

[0135] The communication interface 830 is used to communicate with external devices and exchange and transmit data.

[0136] If the memory 810, processor 820, and communication interface 830 are implemented independently, they can be interconnected via a bus to communicate with each other. This bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 8 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0137] Optionally, in a specific implementation, if the memory 810, processor 820, and communication interface 830 are integrated on a single chip, then the memory 810, processor 820, and communication interface 830 can communicate with each other through an internal interface.

[0138] This application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method provided in this application.

[0139] This application also provides a chip including a processor for calling and executing instructions stored in a memory, causing a communication device with the chip installed to perform the method provided in this application.

[0140] This application also provides a chip, including: an input interface, an output interface, a processor, and a memory. The input interface, output interface, processor, and memory are connected through an internal connection path. The processor is used to execute code in the memory. When the code is executed, the processor is used to execute the method provided in the application embodiment.

[0141] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. General-purpose processors can be microprocessors or any conventional processor. It is worth noting that the processor can be a processor supporting Advanced Reduced Instruction Set Machines (ARM) architecture.

[0142] Further, optionally, the aforementioned memory may include read-only memory and random access memory. The memory may be volatile memory or non-volatile memory, or may include both. Non-volatile memory may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may include random access memory (RAM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available. Examples include Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Sync Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM).

[0143] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0144] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0145] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0146] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process. Furthermore, the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functionality involved.

[0147] The logic and / or steps described in the flowchart or otherwise herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus or device (such as a computer-based system, a processor-included system or other system that can fetch and execute instructions from, an instruction execution system, apparatus or device).

[0148] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. All or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware, the program being stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiments.

[0149] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. This storage medium can be a read-only memory, a disk, or an optical disk, etc.

[0150] The above description is merely an exemplary embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope described in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for lane line inspection, characterized in that, include: Obtain the lane line to be inspected and the reference lane line corresponding to the road segment to which the lane line to be inspected belongs, wherein the lane line to be inspected is a lane line identified from multiple sets of collected data, and the reference lane line is a lane line identified from a single set of collected data. The lane line to be inspected and the reference lane line are matched to obtain at least one lane line matching pair; For each lane line matching pair, multiple first sampling points of the lane line to be inspected in the lane line matching pair are determined, and multiple second sampling points of the reference lane line are determined; the first sampling points and the second sampling points correspond one-to-one. Calculate the distance between each first sampling point of the lane line to be inspected and the second sampling point corresponding to the reference lane line; Based on the curvature of the lane line to be inspected, the pre-set difference threshold is dynamically adjusted to obtain the adjustment result; The difference is calculated using the distance, and the difference is used to represent the degree of overlap between the reference lane line and the lane line to be inspected. The inspection result is determined by comparing the difference with the adjustment result.

2. The method according to claim 1, characterized in that, Obtaining the reference lane line corresponding to the road segment to which the lane line to be inspected belongs includes: Determine the information of the trajectory points corresponding to the lane line to be inspected; Based on the information of the trajectory points, a reference lane line corresponding to the lane line to be inspected is determined in the database; the database stores reference lane lines.

3. The method according to claim 1, characterized in that, The step of matching the lane line to be inspected with the reference lane line includes: The lane line to be inspected and the reference lane line are segmented to obtain lane lines to be inspected and reference lane lines with the same number of segments. Within each segment, the lane line to be inspected and the reference lane line are matched.

4. The method according to claim 3, wherein segmenting the lane line to be inspected and the reference lane line includes: The lane line to be inspected and the reference lane line are processed by sliding a window of a preset size to obtain multiple segments; The sliding window processing satisfies the condition that the overlap area between the current segment and the adjacent segment is not less than a specified overlap area threshold.

5. The method according to claim 1 or 3, characterized in that, The step of matching the lane line to be inspected with the reference lane line includes: Determine the traffic marking category of the lane line to be inspected and the traffic marking category of the reference lane line; Determine the positional relationship between the lane line to be inspected and the reference lane line; Matching is performed based on the traffic marking category and the positional relationship to obtain at least one lane line matching pair.

6. The method according to claim 5, characterized in that, The step of matching based on the traffic marking category and the positional relationship to obtain at least one lane line matching pair includes: The lane lines to be inspected that are of the same traffic marking category and whose positional relationship meets the specified requirements are matched with the reference lane lines.

7. A method for correcting lane lines in a high-precision map, characterized in that, include: The lane markings to be inspected are performed based on the method described in any one of claims 1-6; For lane lines that are found to be abnormal during inspection, relevant information about the abnormal lane lines is obtained, wherein the relevant information includes at least one of map feature ID and location coordinates; The relevant information will be synchronized with the corrector for correction.

8. A device for lane line inspection, characterized in that, include: The lane line acquisition module is used to acquire the lane line to be inspected and the reference lane line corresponding to the road segment to which the lane line to be inspected belongs. The lane line to be inspected is a lane line identified from multiple sets of collected data, and the reference lane line is a lane line identified from a single set of collected data. The lane line matching module is used to match the lane line to be inspected with the reference lane line to obtain at least one lane line matching pair; The lane line inspection module is used to determine, for each lane line matching pair, multiple first sampling points of the lane line to be inspected in the lane line matching pair, and multiple second sampling points of the reference lane line; the first sampling points and the second sampling points correspond one-to-one; calculate the distance between each first sampling point of the lane line to be inspected and the corresponding second sampling point of the reference lane line; dynamically adjust a preset difference threshold according to the curvature of the lane line to be inspected to obtain an adjustment result; calculate the difference degree using the distance, the difference degree being used to represent the degree of overlap between the reference lane line and the lane line to be inspected; and determine the inspection result by comparing the difference degree with the adjustment result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory, wherein the processor, when executing the computer program, implements the method of any one of claims 1-7.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of any one of claims 1-7.

Images