Ground arrow aggregation method, device and equipment and computer readable medium

The method addresses inaccuracies in intelligent driving by standardizing ground arrow representations through angle and size adjustments, enhancing accuracy and safety in intelligent driving systems.

CN119131178APending Publication Date: 2024-12-13CHONGQING CHANGAN AUTOMOBILE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310693963.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-06-12
Publication Date
2024-12-13

AI Technical Summary

Technical Problem

Current intelligent driving technology faces inaccuracies and safety threats due to unevenly sized and skewed ground arrows, leading to incorrect recognition and rendering errors.

Method used

A method for aggregating ground arrows involves calculating angle means, constructing outer rectangular frames, adjusting angles and sizes based on international standards, and using buffer frames to unify and standardize ground arrow representations.

Benefits of technology

The method enhances the accuracy and safety of intelligent driving by ensuring precise construction and representation of ground arrows, reducing errors and improving overall driving safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119131178A_ABST
    Figure CN119131178A_ABST
Patent Text Reader

Abstract

The invention provides a ground arrow aggregation method, device and equipment and a computer readable medium, and belongs to the technical field of high-precision maps, and the method comprises the steps: obtaining a plurality of cluster data in a target road, and calculating an angle mean value; generating a related graph based on the cluster data; obtaining an outsourcing rectangular frame of the related graph and the orientation thereof; based on the outer wrapping rectangular frame and the orientation thereof, adjusting the angle mean value to obtain a first angle of the outer wrapping rectangular frame; constructing a buffer frame, and framing a plurality of wrapping rectangular frames according to the buffer frame; unifying the first angles of the wrapping rectangular frame in the buffer frame to obtain a second angle; and based on the second angle and the international standard size, adjusting the encased rectangular frame to obtain a target encased rectangular frame. The ground arrows constructed through aggregation can have extremely high accuracy, errors such as skewing and inconsistent sizes are avoided, and then the technical effect of improving safety and accuracy in the intelligent driving process is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of high-precision map technology, and in particular to a method, device, equipment and computer-readable medium for aggregating ground arrows. Background Art

[0002] In the era of smart travel, smart driving technology is widely recognized by the public. As an important part of strategic emerging industries, smart driving is the first wonderful chapter in the process from the Internet era to the artificial intelligence era, and it is also one of the strategic commanding heights of the new round of economic and technological development in the world. The comprehensive development of smart driving technology is of great significance to promoting national technology, economy, society, life, safety and comprehensive national strength, and also provides great convenience for the majority of car owners.

[0003] However, today's intelligent driving technology is not perfect. Intelligent driving technology mainly relies on the information returned by the sensors on the vehicle side, and the vehicle uses this information to help judge the surrounding and road information of the vehicle, thereby assisting the driver in driving. Combined with the development of the current era, autonomous driving will gradually become the mainstream mode of travel and become the technology that people rely on for travel and driving.

[0004] Ground arrows are the most important road elements in intelligent driving sections, and the number of ground arrows also ranks first among all elements. Therefore, how to accurately construct ground elements and fully express all the attributes of arrows is an extremely important research topic. If the best expression method can be found, the accuracy of autonomous driving can be greatly improved and driving needs can be better met.

[0005] There are still many problems in the construction of ground arrows, such as uneven size, skewed, redundant, etc. These problems will lead to incorrect intelligent driving recognition and rendering errors, which will lead to intelligent driving errors and pose a great threat to the safety of drivers. Therefore, accurate construction of ground arrows is not only to improve the accuracy of intelligent driving, but also to ensure the safety of drivers. Summary of the invention

[0006] In view of this, the present disclosure aims to propose a method, device, equipment and computer-readable medium for aggregating ground arrows, which can make the aggregated ground arrows have extremely high accuracy, avoid errors such as skewness and inconsistent size, and thus achieve the technical effect of improving the safety and accuracy of the intelligent driving process.

[0007] In a first aspect of an embodiment of the present disclosure, a method for aggregating ground arrows is provided, the method comprising:

[0008] Acquire multiple cluster data in the target road, and calculate the angle mean according to the cluster data;

[0009] generating a related graphic based on the cluster data, wherein the related graphic can include each ground arrow in the cluster data;

[0010] Obtaining an outer rectangular frame of the relevant graphic and its orientation, wherein the outer rectangular frame is a minimum rectangular frame that can contain the relevant graphic, and the angle of the outer rectangular frame is the angle mean;

[0011] Based on the outer rectangular frame and its orientation, the angle mean is adjusted to obtain a first angle of the outer rectangular frame;

[0012] Constructing a buffer frame, and framing the multiple outer rectangular frames according to the buffer frame;

[0013] unifying the first angles of the multiple outer rectangular frames in the buffer frame to obtain a second angle;

[0014] Based on the second angle and the international standard size, the multiple outer rectangular frames are adjusted to obtain a target outer rectangular frame.

[0015] In one embodiment, the cluster data includes at least one ground arrow, and before generating the relevant graphics based on the cluster data, the method further includes:

[0016] Obtaining the geometric centroid coordinates of the centroid point of each ground arrow in the cluster data;

[0017] The geometric centroid coordinates of the fitting centroid point of the cluster data are calculated based on the geometric centroid coordinates and the mean calculation function.

[0018] In one embodiment, generating a related graph based on the cluster data includes:

[0019] Calculate the difference between the geometric centroid coordinates of each ground arrow in the cluster data and the geometric centroid coordinates of the fitted centroid point of the cluster data;

[0020] Based on the difference, the centroid point of each ground arrow is translated to coincide with the fitting centroid point;

[0021] Extract the relevant graphics of the ground arrows when the centroid point of each ground arrow coincides with the fitting centroid point.

[0022] In one embodiment, extracting the relevant graphics of the ground arrows when the centroid of each ground arrow coincides with the fitting centroid includes:

[0023] Using a concave hull function to extract the overlapping area between the ground arrows after the centroid point is translated;

[0024] The correlation graph is generated through the overlapping area.

[0025] In one embodiment, extracting the relevant graphics of the ground arrows when the centroid of each ground arrow coincides with the fitting centroid includes:

[0026] Using a concave hull function to extract the overlapping area between the ground arrows after the centroid point is translated;

[0027] The correlation graph is generated through the overlapping area.

[0028] In one implementation, calculating the angle mean according to the cluster data includes:

[0029] Collecting the angle of each ground arrow in the cluster data;

[0030] Filtering out ground arrows whose angles exceed a first preset angle;

[0031] The angles of the retained ground arrows are averaged, and the average is the angle average of the cluster data.

[0032] In one embodiment, before adjusting the angle mean based on the outer rectangular frame and its orientation to obtain the first angle of the outer rectangular frame, the method further includes:

[0033] Determine whether there is a ground virtual line on the same horizontal plane within the first preset range of the outer rectangular frame;

[0034] When the ground virtual line does not exist, it is determined whether there is a ground solid line on the same horizontal plane within the first preset range of the outer rectangular frame.

[0035] In one embodiment, when it is determined that there is a ground dotted line or a ground solid line on the same horizontal plane within the first preset range of the outer rectangular frame, adjusting the angle mean based on the outer rectangular frame and its orientation to obtain the first angle of the outer rectangular frame includes:

[0036] Determine a projection starting point on the ground imaginary line or the ground solid line;

[0037] Project the fitted centroid of the outer rectangular frame onto the projection starting point, and use the orientation of the outer rectangular frame as the projection direction, and move the projection starting point by a second preset distance in the projection direction on the ground imaginary line or the ground solid line to obtain a projection end point;

[0038] The angle adjustment amount of the outer rectangular frame is calculated according to the projection starting point and the projection end point, and the angle mean is adjusted according to the angle adjustment amount to obtain the first angle of the outer rectangular frame.

[0039] In one embodiment, the buffer frame includes a first buffer frame and a second buffer frame, and the buffer frame is constructed to frame the multiple outer rectangular frames according to the buffer frame, including:

[0040] Setting initial values ​​of the first buffer box and the second buffer box, and formulating a target stop zone and an extension threshold;

[0041] Extending both ends of the first buffer frame and the second buffer frame for multiple times, and increasing the extended width according to a third preset distance during each extension process to frame the multiple outer rectangular frames;

[0042] When the first buffer frame or the second buffer frame identifies the target stopping area or reaches the extension threshold during the extension process, the extension is stopped.

[0043] In one implementation, unifying the first angles of the multiple outer rectangular frames in the buffer frame to obtain the second angle includes:

[0044] Constructing a first storage list and a second storage list;

[0045] Calculate a first difference between the orientations of a plurality of outer rectangular frames in the buffer frame and the first angles corresponding thereto;

[0046] When the first difference is greater than or equal to a second preset angle, storing the first angle of the outer rectangular frame in the first storage list;

[0047] When the first difference is smaller than the second preset angle, storing the first angle of the outer rectangular frame in the second storage list;

[0048] Constructing a third storage list and a fourth storage list, and taking the first angle of any one of the plurality of enclosing rectangular frames in the second storage list as an initial value;

[0049] Calculate a second difference between the first angles of the plurality of bounding rectangular frames in the second storage list and the initial value;

[0050] When the second difference is greater than or equal to a third preset angle, storing the first angle of the outer rectangular frame in the third storage list;

[0051] When the second difference is less than the third preset angle, storing the first angle of the outer rectangular frame in the fourth storage list;

[0052] Comparing the number of first angles in the third storage list and the number of first angles in the fourth storage list;

[0053] An average value of the first angles in the storage list having a larger number of first angles in the third storage list and the fourth storage list is calculated, and the average value is used as the second angle.

[0054] In one implementation, adjusting the multiple outer rectangular frames based on the second angle includes:

[0055] According to the second angle, the first angles of the plurality of outer rectangular frames are adjusted to the second angle.

[0056] In one implementation, adjusting the multiple outer rectangular frames based on the international standard size to obtain a target outer rectangular frame includes:

[0057] Selecting the longest outer rectangular frame with the longest rectangular frame length among the multiple outer rectangular frames;

[0058] Selecting a matching template from the international standard size according to the length of the longest outer rectangular frame;

[0059] The multiple outer rectangular frames are reconstructed according to the length and width of the template to obtain the target outer rectangular frame.

[0060] In one embodiment, after obtaining the target outer rectangular frame, the method further includes:

[0061] When the target outer rectangular frame overlaps, the width of the target outer rectangular frame is reduced by a preset ratio;

[0062] When the number of reductions reaches a preset number, the reduction of the target outer rectangular frame is stopped.

[0063] In a second aspect of the embodiments of the present disclosure, a ground arrow aggregation device is provided, the device comprising:

[0064] A first calculation module, used for acquiring a plurality of cluster data in a target road, and calculating an angle mean value according to the cluster data;

[0065] A first generating module, configured to generate a related graphic based on the cluster data, wherein the related graphic can include each ground arrow in the cluster data;

[0066] A first operation module is used to obtain an outer rectangular frame of the relevant graphic and its orientation, wherein the outer rectangular frame is a minimum rectangular frame that can contain the relevant graphic, and the angle of the outer rectangular frame is the angle mean;

[0067] A first adjustment module, configured to adjust the angle mean based on the outer rectangular frame and its orientation to obtain a first angle of the outer rectangular frame;

[0068] A second operating module is used to construct a buffer frame, and to frame the multiple outer rectangular frames according to the buffer frame;

[0069] A third operation module is used to unify the first angles of the multiple outer rectangular frames in the buffer frame to obtain a second angle;

[0070] The second adjustment module is used to adjust the multiple outer rectangular frames based on the second angle and the international standard size to obtain a target outer rectangular frame.

[0071] According to a third aspect of an embodiment of the present disclosure, a device is provided, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method described in the first aspect.

[0072] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable medium is provided, on which a computer program is stored, wherein when the program is executed by a processor, the method described in the first aspect is implemented.

[0073] Beneficial effects of the present disclosure:

[0074] The disclosed embodiment provides a method, device, equipment and computer-readable medium for aggregating ground arrows, by acquiring multiple cluster data in a target road, each cluster data corresponds to a target lane in the target road, each cluster data includes multiple ground arrows on the target lane, and the angle mean of the ground arrows is calculated according to the cluster data; based on the cluster data, a related graphic is generated, wherein the related graphic can contain each ground arrow in the cluster data; an outer rectangular frame and its orientation of the related graphic are acquired, the outer rectangular frame is the smallest rectangular frame that can accommodate the related graphic, and the angle of the outer rectangular frame is the angle mean; based on the outer rectangular frame and its orientation, the angle mean is adjusted to obtain a first angle of the outer rectangular frame; a buffer frame is constructed, and multiple outer rectangular frames are framed according to the buffer frame; the first angles of multiple outer rectangular frames in the buffer frame are unified to obtain a second angle; based on the second angle and the international standard size, multiple outer rectangular frames are adjusted to obtain a target outer rectangular frame. The ground arrows constructed by aggregation can have extremely high accuracy, avoid errors such as skewness and inconsistent size, and thus achieve the technical effect of improving the safety and accuracy of the intelligent driving process. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings:

[0076] Figure 1 It is a flow chart of a method for aggregating ground arrows provided by an embodiment of the present disclosure;

[0077] Figure 2 It is a flow chart of another ground arrow aggregation method provided by an embodiment of the present disclosure;

[0078] Figure 3 It is a schematic framework diagram of a ground arrow aggregation device provided in one embodiment of the present disclosure. DETAILED DESCRIPTION

[0079] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other.

[0080] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0081] In real life, a target road includes multiple target lanes in different directions. In order to provide driving convenience for the majority of drivers, some ground arrows (such as straight, left turn, right turn arrows), ground markings (such as single dashed line, double dashed line, single solid line, double solid line) and other road surface elements are usually drawn on the target lanes. The driver can judge the road surface information of the current target road based on these road surface elements.

[0082] Due to the popularity of intelligent driving, drivers are increasingly relying on existing autonomous driving technology. The construction of high-precision maps in autonomous driving technology needs to rely on current ground arrows. However, the current ground arrows have problems such as different sizes, crooked drawing, and unclear ground arrows. This will cause errors when intelligent driving recognizes these ground arrows to construct maps, resulting in inaccurate construction results, posing a great threat to the safety of drivers. Therefore, it is necessary to propose a ground arrow aggregation method to address these problems.

[0083] Figure 1 is a flow chart of a method for aggregating ground arrows provided by an embodiment of the present disclosure, such as Figure 1 As shown, this embodiment provides a method for aggregating ground arrows, the method comprising:

[0084] In step S101, a plurality of cluster data in a target road is obtained, and an angle mean is calculated according to the cluster data;

[0085] In step S102, a related graphic is generated based on the cluster data, wherein the related graphic can include each ground arrow in the cluster data;

[0086] In step S103, an outer rectangular frame of the relevant graphic and its orientation are obtained, wherein the outer rectangular frame is a minimum rectangular frame that can contain the relevant graphic, and the angle of the outer rectangular frame is the angle mean;

[0087] In step S104, based on the outer rectangular frame and its orientation, the angle mean is adjusted to obtain a first angle of the outer rectangular frame;

[0088] In step S105, a buffer frame is constructed, and the plurality of outer rectangular frames are framed according to the buffer frame;

[0089] In step S106, the first angles of the multiple outer rectangular frames in the buffer frame are unified to obtain a second angle;

[0090] In step S107, the multiple outer rectangular frames are adjusted based on the second angle and the international standard size to obtain a target outer rectangular frame.

[0091] First, in step S101, multiple cluster data in the target road are obtained, and the angle mean is calculated based on the cluster data. Before the aggregation operation of the ground arrows, it is necessary to first obtain the data results collected by the front-end clustering link. The data results include multiple ground arrows and their angles. The collected data results are grouped and clustered by clustering. Clustering is to divide a data set into different classes or clusters according to a specific standard, so that the similarity of data objects in the same cluster is as large as possible, and the difference of data objects in different clusters is as large as possible. Then, the data of the same category of the collected data results are clustered into a cluster, each cluster is a cluster data, and a cluster data corresponds to a target lane. The cluster data includes at least one ground arrow, and each cluster data is assigned a unique identification value to identify the current cluster data: cluster_id, and the angle mean of each cluster data is calculated by the angle of the ground arrow in each cluster data.

[0092] Next, in step S102, a related graph is generated based on the cluster data, wherein the related graph can include each ground arrow in the cluster data. A related graph is obtained, wherein one cluster data corresponds to one related graph, wherein the related graph can include each ground arrow in the cluster data, and the related graph is a graph with the smallest area that can include each ground arrow in the cluster data.

[0093] Further, in step S103, the outer rectangular frame of the relevant graphics and its orientation are obtained, the outer rectangular frame is the minimum rectangular frame that can contain the relevant graphics, and the angle of the outer rectangular frame is the angle mean. As can be seen from step S102, the relevant graphics are graphics with the minimum area that can contain each ground arrow in the cluster data. Therefore, the relevant graphics are generally irregular graphics, and it is necessary to obtain the minimum rectangular frame that can contain the relevant graphics, that is, the outer rectangular frame. The outer rectangular frame is indirectly obtained based on the cluster data, so the angle mean of the cluster data is used as the angle of the outer rectangular frame.

[0094] Further, in step S104, based on the outer rectangular frame and its orientation, the angle mean is adjusted to obtain the first angle of the outer rectangular frame. The outer rectangular frame of the cluster data obtained in step S103 and the orientation of the outer rectangular frame are used to adjust the angle mean of the outer rectangular frame. Since the angle mean is calculated from the collected data, there is a certain error, so it is necessary to calibrate the angle mean according to the orientation of the outer rectangular frame, and use the calibrated angle of the outer rectangular frame as the first angle. Each outer rectangular frame has a first angle corresponding to it.

[0095] Further, in step S105, a buffer frame is constructed, and the multiple outer rectangular frames are framed according to the buffer frame. By constructing the buffer frame, the cluster data of the same-direction lane on the target road is framed. The ground arrows of the same-direction lane are theoretically completely consistent, and the buffer frame includes multiple outer rectangular frames of the cluster data on the same-direction lane.

[0096] Further, in step S106, the first angles of the multiple outer rectangular frames in the buffer frame are unified to obtain a second angle. Since the construction of different outer rectangular frames in the same-direction lane is obtained by the ground arrows in the corresponding cluster data, and the ground arrows of different cluster data have certain deviations in practice and are not completely the same, there are certain differences between the constructed different outer rectangular frames in the same-direction lane, that is, there are certain differences between the first angles of the outer rectangular frames, and since the ground arrows corresponding to different outer rectangular frames in the same-direction lane should be the same in theory, it is necessary to unify the first angles of the outer rectangular frames in the same-direction lane to obtain the second angle, that is, after unification, the angles of each outer rectangular frame in the same-direction lane are the same, which are all the second angles.

[0097] Finally, in step S107, the multiple outer rectangular frames are adjusted based on the second angle and the international standard size to obtain a target outer rectangular frame. The first angle of each outer rectangular frame in the buffer frame is adjusted by the second angle obtained in step S106, so that the angle of each outer rectangular frame in the buffer frame is unified and becomes the second angle. In addition, since the size of each outer rectangular frame in the buffer frame is still different at this time, it is necessary to compare with the international standard size and adjust the length and width of each outer rectangular frame in the buffer frame by the arrow ratio of the international standard size so that it is the same as the arrow ratio of the international standard size.

[0098] Optionally, the cluster data includes at least one ground arrow, and before the above step S102, the method further includes:

[0099] Obtaining the geometric centroid coordinates of the centroid point of each ground arrow in the cluster data;

[0100] The geometric centroid coordinates of the fitting centroid point of the cluster data are calculated based on the geometric centroid coordinates and the mean calculation function.

[0101] First, the geometric centroid coordinates of the centroid point of each ground arrow in the cluster data are obtained. The geometric centroid coordinates (x, y, z) of the ground arrow figure in each cluster data are calculated, where x is called the horizontal coordinate, y is called the vertical coordinate, and z is called the vertical coordinate. The geometric centroid coordinates of the centroid point of the ground arrow are used as the geometric centroid coordinates of the ground arrow.

[0102] Secondly, according to the geometric centroid coordinates and the mean calculation function, the geometric centroid coordinates of the fitted centroid point of the cluster data are calculated. The horizontal coordinate, vertical coordinate, and vertical coordinate of the geometric centroid coordinates of the centroid point of the ground arrow in each cluster data are respectively calculated using the mean function (mean calculation function) to calculate the coordinates of the fitted geometric centroid of each cluster data. The angle of the cluster data can be obtained through the coordinates of the fitted geometric centroid, where the mean function (mean calculation function) can be an arithmetic mean or a geometric mean, and the formula is as follows:

[0103]

[0104] Among them, A is the mean, n is the number of ground arrows in each cluster data, and a i is the value of the horizontal / vertical / vertical coordinate of the center of mass of the i-th ground arrow.

[0105] Optionally, the above step S102 includes:

[0106] Calculate the difference between the geometric centroid coordinates of each ground arrow in the cluster data and the geometric centroid coordinates of the fitted centroid point of the cluster data;

[0107] Based on the difference, the centroid point of each ground arrow is translated to coincide with the fitting centroid point;

[0108] Extract the relevant graphics of the ground arrows when the centroid point of each ground arrow coincides with the fitting centroid point.

[0109] First, the difference between the geometric centroid coordinates of each ground arrow in the cluster data and the geometric centroid coordinates of the fitted centroid point of the cluster data is calculated. The difference between the geometric centroid coordinates of the centroid point of each ground arrow in each cluster data and the fitted geometric centroid coordinates of the cluster data is calculated.

[0110] Secondly, the centroid of each ground arrow is translated to coincide with the fitting centroid based on the difference. The centroid of the ground arrow is translated in the coordinate system by the difference between the geometric centroid coordinates of the centroid of each ground arrow and the fitting geometric centroid coordinates of the cluster data, so that each centroid of each ground arrow is translated to coincide with the fitting geometric centroid of the cluster data.

[0111] Finally, the relevant graph of the ground arrow when the centroid of each ground arrow coincides with the fitting centroid is extracted. The relevant graph when all ground arrows overlap when the centroid of the ground arrow in each cluster data coincides with the fitting geometric centroid of the cluster data is obtained, and the relevant graph includes each ground arrow in the cluster data.

[0112] Optionally, the extracting of the relevant graphics of the ground arrows when the centroid of each ground arrow coincides with the fitting centroid includes:

[0113] Using a concave hull function to extract the overlapping area between the ground arrows after the centroid point is translated;

[0114] The correlation graph is generated through the overlapping area.

[0115] First, the concave hull function is used to extract the overlapping area between the ground arrows after the centroid point is translated, wherein the overlapping area of ​​the relevant graphics can be obtained using the concave hull function of Shaply, and the concave hull is to construct a polygon, and the polygon is used to represent the area occupied by a group of data points.

[0116] Among them, the process steps of the concave hull function are:

[0117] (1) Find the polar center P0;

[0118] (2) Find the leftmost point on the screen, which is part of the concave edge P1;

[0119] (3) Select P0 and enter the logic of finding P1;

[0120] (4) Using the rolling ball method, define the radius of the ball (a scanning range);

[0121] (5) Taking P1 as the center, filter out the nearest point within a fixed radius;

[0122] (6) Traverse the nearest point, taking the angle from P0 to P1 as the reference, and rotate counterclockwise to find the nearest point;

[0123] ①Assume that the current traversal reaches P2;

[0124] ② Calculate the angle from P1 to P2, and then compare P0 to P1 and P1 to P2, which is closest to P2 in the counterclockwise direction;

[0125] (7) After the final traversal, the point with the smallest counterclockwise angle is P2;

[0126] (8) Continue executing until you return to the starting point or complete a cycle of traversal.

[0127] Finally, the related graph is generated through the overlapping area. After the process steps of the concave hull function are traversed, the related graph can be obtained.

[0128] Optionally, calculating the angle mean according to the cluster data includes:

[0129] Collecting the angle of each ground arrow in the cluster data;

[0130] Filtering out ground arrows whose angles exceed a first preset angle;

[0131] The angles of the retained ground arrows are averaged, and the average is the angle average of the cluster data.

[0132] First, the angle of each ground arrow in the cluster data is collected. Before the aggregation step, the angle of each ground arrow is returned for the collected ground arrow data, that is, each ground arrow in each cluster data has an angle corresponding to it.

[0133] Secondly, the ground arrows whose angles exceed the first preset angle are filtered out. Before calculating the angle mean of the ground arrows, the angles of the ground arrows in each cluster data need to be filtered out, and the angles of the ground arrows exceeding the first preset angle are filtered out. The first preset angle can be calibrated according to the actual situation. The angles of these filtered ground arrows are represented as values ​​that are too different from the angles of most ground arrows in the cluster data. If they are retained, it will affect the accuracy of the results. Therefore, they need to be filtered out.

[0134] Finally, the angles of the ground arrows that are retained are averaged, and the average is the angle average of the cluster data. The angles of the ground arrows in the cluster data that are retained after filtering are averaged, wherein the average calculation method can refer to the mean function (mean calculation function) used in the above embodiment, and then the angle average of each cluster data is obtained.

[0135] Optionally, before the above step S104, the method further includes:

[0136] Determine whether there is a ground virtual line on the same horizontal plane within the first preset range of the outer rectangular frame;

[0137] When the ground virtual line does not exist, it is determined whether there is a ground solid line on the same horizontal plane within the first preset range of the outer rectangular frame.

[0138] First, determine whether there is a ground imaginary line on the same horizontal plane within the first preset range of the outer rectangular frame. During the aggregation process, the outer rectangular frame needs to be calibrated multiple times, so the ground markings near the outer rectangular frame are needed as auxiliary reference information. When searching for ground markings near the outer rectangular frame, the ground imaginary lines near the outer rectangular frame are preferentially searched to determine whether the ground imaginary lines are found within the first preset range (e.g., 2 meters) of the outer rectangular frame. The first preset range can be calibrated according to actual conditions.

[0139] Secondly, when the ground imaginary line does not exist, it is determined whether there is a ground solid line on the same horizontal plane within the first preset range of the outer rectangular frame. When the ground imaginary line cannot be found within the first preset range (such as 2 meters) of the outer rectangular frame, the ground solid line near the outer rectangular frame is searched again to determine whether the ground solid line is found within the first preset range (such as 2 meters) of the outer rectangular frame.

[0140] Optionally, when it is determined that there is a ground dotted line or a ground solid line on the same horizontal plane within the first preset range of the outer rectangular frame, the step S104 includes:

[0141] Determine a projection starting point on the ground imaginary line or the ground solid line;

[0142] Project the fitted centroid of the outer rectangular frame onto the projection starting point, and use the orientation of the outer rectangular frame as the projection direction, and move the projection starting point by a second preset distance in the projection direction on the ground imaginary line or the ground solid line to obtain a projection end point;

[0143] The angle adjustment amount of the outer rectangular frame is calculated according to the projection starting point and the projection end point, and the angle mean is adjusted according to the angle adjustment amount to obtain the first angle of the outer rectangular frame.

[0144] First, a projection starting point is determined on the ground dotted line or the ground solid line. Since the ground dotted line or the ground solid line is generally located on the left or right of the ground arrow in the actual target road, the ground dotted line or the ground solid line can be used as the reference direction of the outer rectangular frame, and the angle mean of the outer rectangular frame can be adjusted by the reference direction. Therefore, it is necessary to determine a projection starting point of the fitting centroid point of the outer rectangular frame on the ground dotted line or the ground solid line.

[0145] Secondly, the fitting centroid point of the outer rectangular frame is projected onto the projection starting point, and the orientation of the outer rectangular frame is used as the projection direction, and the projection starting point is moved a second preset distance in the projection direction on the ground dotted line or the ground solid line to obtain a projection end point. Through the obtained projection starting point, the fitting centroid point of the outer rectangular frame is projected onto the already determined projection starting point on the ground dotted line or the ground solid line, so that the fitting centroid point of the outer rectangular frame coincides with the projection starting point, and the current approximate azimuth of the outer rectangular frame is used as the projection direction, and the second preset distance (such as 0.1 meter) is moved forward on the ground dotted line or the ground solid line along the azimuth to find a projection end point on the ground dotted line or the ground solid line.

[0146] In other embodiments, the orientation of the outer rectangular box can be determined by setting a point sequence for the four corners of the outer rectangular box, with the lower left corner of the outer rectangular box being point sequence 0, and numbering them in a counterclockwise direction, that is, the point sequence of the lower left corner of the outer rectangular box is 0, the point sequence of the lower right corner is 1, the point sequence of the upper right corner is 2, and the point sequence of the upper left corner is 3, and the direction from point sequence 1 to point sequence 2 is defaulted to the orientation of the outer rectangular box.

[0147] Finally, according to the projection starting point and the projection end point, the angle adjustment amount of the outer rectangular frame is calculated, and the angle mean is adjusted according to the angle adjustment amount to obtain the first angle of the outer rectangular frame. The projection starting point and the projection end point on the ground dotted line or the ground solid line are determined through the above steps, and the angle adjustment amount of the outer rectangular frame can be calculated, and the angle mean of the outer rectangular frame is adjusted according to the angle adjustment amount. The angle of the outer rectangular frame after adjustment is the first angle.

[0148] In other embodiments, if the angle adjustment amount is greater than 180 degrees, it means that the point sequence of the outer rectangular frame is incorrectly labeled and there is a situation where the point sequence is reversed. Therefore, the outer rectangular frame at this time uses point sequence 2 to point sequence 1 as the orientation of the outer rectangular frame.

[0149] In other embodiments, when neither the ground dotted line nor the ground solid line is found within the first preset distance near the outer rectangular frame, the angle of the outer rectangular frame is not adjusted, but the angle mean is maintained as the angle of the outer rectangular frame.

[0150] Optionally, the buffer frame includes a first buffer frame and a second buffer frame, and the above step S105 includes:

[0151] Setting initial values ​​of the first buffer box and the second buffer box, and formulating a target stop zone and an extension threshold;

[0152] Extending both ends of the first buffer frame and the second buffer frame for multiple times, and increasing the extended width according to a third preset distance during each extension process to frame the multiple outer rectangular frames;

[0153] When the first buffer frame or the second buffer frame identifies the target stopping area or reaches the extension threshold during the extension process, the extension is stopped.

[0154] First, set the initial values ​​of the first buffer box and the second buffer box, and formulate the target stop zone and extension threshold. Construct the first buffer box and the second buffer box at the left and right ends of the target road respectively. In other embodiments, the first buffer box and the second buffer box can also be set at any position of the target road. The buffer box is used to frame the same group of outer rectangular frames in the target road. Set initial values ​​for the first buffer box and the second buffer box respectively. The initial values ​​can represent the width of the buffer box before extension or the width of each extension, and formulate the target stop zone and extension threshold. The target stop zone represents the road boundary and diversion area of ​​the buffer box during the extension process. The extension threshold is the maximum range of the buffer box extension. The extension threshold can be calibrated according to the actual road width.

[0155] Secondly, the two ends of the first buffer frame and the second buffer frame are extended multiple times, and in each extension process, the extended width is increased according to the third preset distance to frame the multiple outer rectangular frames. The left and right ends of the first buffer frame and the left and right ends of the second buffer frame can be extended to both sides, and the buffer frames are gradually expanded according to the third preset distance (such as 1 meter), and the outer rectangular frames near the first buffer frame and the second buffer frame are cyclically searched, and multiple outer rectangular frames containing ground arrows in the same direction are framed.

[0156] Finally, when the first buffer box or the second buffer box identifies the target stopping area or reaches the extension threshold during the extension process, the extension is stopped. During the extension process of the first buffer box or the second buffer box, when the left and right ends of the first buffer box or the second buffer box identify the road boundary or the diversion area, it means that if the first buffer box or the second buffer box is continued to be extended, it may be framed to the outer rectangular frame on other roads in the same direction or the outer rectangular frame of the opposite lane, wherein the direction of the lane can be extracted through the driver's trajectory data to obtain the trajectory direction of each lane; when the extension distance of the left and right ends of the first buffer box or the second buffer box reaches the extension threshold (such as 20 meters), regardless of whether the first buffer box or the second buffer box frames the outer rectangular frame, the extension process of the first buffer box or the second buffer box is stopped.

[0157] Optionally, the above step S106 includes:

[0158] Constructing a first storage list and a second storage list;

[0159] Calculate a first difference between the orientations of a plurality of outer rectangular frames in the buffer frame and the first angles corresponding thereto;

[0160] When the first difference is greater than or equal to a second preset angle, storing the first angle of the outer rectangular frame in the first storage list;

[0161] When the first difference is smaller than the second preset angle, storing the first angle of the outer rectangular frame in the second storage list;

[0162] Constructing a third storage list and a fourth storage list, and taking the first angle of any one of the plurality of enclosing rectangular frames in the second storage list as an initial value;

[0163] Calculate a second difference between the first angles of the plurality of bounding rectangular frames in the second storage list and the initial value;

[0164] When the second difference is greater than or equal to a third preset angle, storing the first angle of the outer rectangular frame in the third storage list;

[0165] When the second difference is less than the third preset angle, storing the first angle of the outer rectangular frame in the fourth storage list;

[0166] Comparing the number of first angles in the third storage list and the number of first angles in the fourth storage list;

[0167] An average value of the first angles in the storage list having a larger number of first angles in the third storage list and the fourth storage list is calculated, and the average value is used as the second angle.

[0168] First, a first storage list and a second storage list are constructed. Two storage lists are constructed to filter the outer rectangular frame framed by the buffer frame, wherein the first storage list is used to store the first angle of the outer rectangular frame whose first difference between the orientation of the outer rectangular frame in the buffer frame and the corresponding first angle is greater than or equal to the second preset angle; the second storage list is used to store the first angle of the outer rectangular frame whose first difference between the orientation of the outer rectangular frame in the buffer frame and the corresponding first angle is less than the second preset angle.

[0169] Further, the first difference between the orientations of the plurality of outer rectangular frames in the buffer frame and their corresponding first angles is calculated. The first difference between the orientation of the outer rectangular frame and the first angle of the outer rectangular frame is calculated by the orientation of the outer rectangular frame determined by the method of the above embodiment.

[0170] Further, when the first difference is greater than or equal to the second preset angle, the first angle of the outer rectangular frame is stored in the first storage list. The first storage list stores the first angle of the outer rectangular frame whose first difference between the orientation of the outer rectangular frame in the buffer frame and the corresponding first angle is greater than or equal to the second preset angle.

[0171] Further, a third storage list and a fourth storage list are constructed, and the first angle of any one of the plurality of outer rectangular frames in the second storage list is used as an initial value. The first angle of the outer rectangular frame whose first difference between the orientation of the outer rectangular frame in the buffer frame and the corresponding first angle is less than the second preset angle is stored in the second storage list.

[0172] Further, a second difference between the first angles of the plurality of outer rectangular boxes in the second storage list and the initial value is calculated. Any first angle is selected from the first angles of the plurality of outer rectangular boxes in the second storage list as the initial value. Since it is not known which first angle of the outer rectangular box in the second storage list is accurate, a first angle of the outer rectangular box can be randomly designated as the initial value, and the first angles of the outer rectangular boxes in the second storage list other than the initial value are compared with the initial value and the second difference is calculated.

[0173] Further, when the second difference is greater than or equal to a third preset angle, the first angle of the outer rectangular frame is stored in the third storage list. The third preset angle is set, and the first angle of the outer rectangular frame whose second difference is greater than or equal to the third preset angle is stored in the third storage list, wherein the third preset angle can be calibrated according to actual conditions.

[0174] Further, when the second difference is less than the third preset angle, the first angle of the outer rectangular frame is stored in the fourth storage list. The fourth preset angle is set, and the first angle of the outer rectangular frame whose second difference is less than the third preset angle is stored in the fourth storage list.

[0175] Further, the number of first angles in the third storage list and the fourth storage list are compared. By comparing the number of first angles of the outer rectangular box in the third storage list and the fourth storage list, it is determined which storage list has more first angles of the outer rectangular box in the third storage list or the fourth storage list.

[0176] Finally, the average value of the first angles in the storage list with more first angles in the third storage list and the fourth storage list is calculated, and the average value is used as the second angle. Since most of the first angles in the multiple outer rectangular frames framed by the buffer frame are accurate, the first angle data in the storage list with more first angles in the outer rectangular frames in the third storage list and the fourth storage list is more accurate, and the average value of the first angles of the outer rectangular frames in the storage list is calculated, and the average value is used as the second angle.

[0177] Optionally, adjusting the multiple outer rectangular frames based on the second angle includes:

[0178] According to the second angle, the first angles of the plurality of outer rectangular frames are adjusted to the second angle. The first angles of the outer rectangular frames in the buffer frame are unified and adjusted to the second angle.

[0179] Optionally, adjusting the multiple outer rectangular frames based on the international standard size to obtain a target outer rectangular frame includes:

[0180] Selecting the longest outer rectangular frame with the longest rectangular frame length among the multiple outer rectangular frames;

[0181] Selecting a matching template from the international standard size according to the length of the longest outer rectangular frame;

[0182] The multiple outer rectangular frames are reconstructed according to the length and width of the template to obtain the target outer rectangular frame.

[0183] First, the longest outer rectangular frame with the longest rectangular frame length is selected from the plurality of outer rectangular frames. In the above embodiment, the angles of the plurality of outer rectangular frames have been uniformly adjusted to the second angle, but the outer rectangular frame is determined based on the relevant graphics obtained by translating the plurality of ground arrows of the target lane, and the lengths and widths of the plurality of outer rectangular frames are not the same. Therefore, the outer rectangular frame with the longest rectangular frame length needs to be selected from the outer rectangular frames.

[0184] Secondly, a matching template is selected in the international standard size according to the rectangular frame length of the longest outer rectangular frame. The length of the longest outer rectangular frame selected in the above step is compared with the lengths of various ground arrow templates in the international standard size, and the length of the ground arrow template with the closest length is matched.

[0185] Finally, the multiple outer rectangular frames are reconstructed according to the length and width of the template to obtain the target outer rectangular frame. The length and width of the multiple outer rectangular frames are reconstructed according to the length and width of the matched ground arrow template, so that the length and width of the multiple outer rectangular frames are the same as the length and width of the matched ground arrow template, and the target outer rectangular frame is obtained.

[0186] Optionally, after the above step S107, the method further includes:

[0187] When the target outer rectangular frame overlaps, the width of the target outer rectangular frame is reduced by a preset ratio;

[0188] When the number of reductions reaches a preset number, the reduction of the target outer rectangular frame is stopped.

[0189] First, when the target outer rectangular frame overlaps, the width of the target outer rectangular frame is reduced by a preset ratio. When the target outer rectangular frame obtained in the above embodiment is set on the target lane, if the target outer rectangular frame overlaps, that is, the target outer rectangular frame covers the ground markings of the target lane, it is necessary to adjust the width of the target outer rectangular frame, with the length direction of the target outer rectangular frame passing through the center as the axis, and the two ends are reduced toward the center according to a preset ratio, so that the width of the target outer rectangular frame becomes narrower, while the length remains unchanged, thereby avoiding the overlap.

[0190] Finally, when the number of shrinking times reaches the preset number, the shrinking of the target outer rectangular frame stops. During the width shrinking process, the two ends will repeatedly shrink toward the center according to the preset ratio, but will not continue to shrink indefinitely. When the number of shrinking times reaches the preset number, regardless of whether there is an overlap, the shrinking of the width of the target outer rectangular frame stops, so as not to affect the final effect of the target outer rectangular frame.

[0191] Figure 2 is a flow chart of another method for aggregating ground arrows provided by an embodiment of the present disclosure, such as Figure 2 As shown, another ground arrow aggregation method provided by this embodiment includes:

[0192] In practical applications, when aggregating ground arrows to build a high-precision map, the following steps are required:

[0193] (1) Obtain cluster data after ground arrow clustering:

[0194] First, we select multiple target roads where we need to aggregate ground arrows. One target road usually contains multiple target lanes with different or same directions. We collect ground arrow data for multiple target lanes, collect multiple ground arrows contained in each target lane, and cluster the ground arrows in each target lane to obtain cluster data through clustering method. Each target lane corresponds to one cluster data, and one cluster data includes graphics of all ground arrows contained in the corresponding target lane.

[0195] (2) Calculate the angle mean of the cluster data obtained by clustering:

[0196] The angles of the ground arrows contained in the multiple cluster data of the target road obtained in the above steps are collected to obtain the angle of each ground arrow, and a first preset angle is set for the target lane in the same direction, and the ground arrows in the cluster data whose angles exceed the first preset angle are filtered out, and the angles of the ground arrows retained in the cluster data are averaged to obtain the angle mean of each cluster data.

[0197] (3) Calculate the centroid and the fitting centroid, and move the centroid to the fitting centroid:

[0198] Get the geometric coordinates of the centroid of the ground arrow in each cluster data, and calculate the coordinates of the fitted centroid of each cluster data through the mean calculation function, translate the centroid of the ground arrow in each cluster data to the calculated fitted centroid, so that all the centroids in each cluster data coincide with the fitted centroid.

[0199] (4) Extract the overlapping area and output related graphics:

[0200] The concave hull function is used to extract the overlapping area when all the centroid points in each cluster data coincide with the fitted centroid point and the ground arrow graphics in each cluster data coincide with each other, and the relevant graphics when the ground arrow graphics in each cluster data coincide with each other are output.

[0201] (5) Output the outer rectangle:

[0202] Since the related graphics when the ground arrow graphics in each cluster data overlap are irregular graphics, it is necessary to obtain the minimum rectangular frame that can contain the irregular related graphics. The rectangular frame is a quadrilateral, and the outer rectangular frame of each cluster data is obtained. The angle mean of the cluster data is used as the angle of the outer rectangular frame at this time, and the fitted centroid point of the cluster data is used as the centroid point of the outer rectangular frame.

[0203] (6) Adjust the angle mean of the outer rectangular frame:

[0204] Determine whether there is a ground dotted line or a ground solid line on the same horizontal plane within a first preset range (such as 2 meters) for each outer rectangular frame (preferentially determine whether there is a ground dotted line, and if there is no ground dotted line, determine whether there is a ground solid line). When it is determined that there is a ground dotted line or a ground solid line on the outer rectangular frame within the first preset range (such as 2 meters), determine a projection starting point on the ground dotted line or the ground solid line, and project the fitting centroid point of the outer rectangular frame onto the projection starting point, and project along the direction of the outer rectangular frame (the direction from point sequence 1 to point sequence 2), and move the projection starting point forward along the projection direction by a second preset distance (such as 0.1 meter) to obtain a projection end point, calculate the angle between the projection starting point and the projection end point in the coordinate system, and adjust the angle mean of the outer rectangular frame according to the angle, so that the angle of the outer rectangular frame at this time is the first angle. When there is no ground dotted line or a ground solid line within the first preset range (such as 2 meters) for the outer rectangular frame, continue to use the angle mean of the outer rectangular frame as the angle of the outer rectangular frame.

[0205] (7) Construct a buffer frame to define the outer rectangular frame of the same group:

[0206] Two buffer frames are constructed, namely a first buffer frame and a second buffer frame, and the first buffer frame and the second buffer frame are set at both ends of the target road. Initial data, extension thresholds and target stopping areas of the first buffer frame and the second buffer frame are set, so that the first buffer frame and the second buffer frame are extended by the width of the boundaries of the first buffer frame and the second buffer frame according to a third preset distance each time, until the boundaries of the first buffer frame and the second buffer frame overlap and are spliced ​​into one buffer frame, so as to frame an outer rectangular frame of the same group (indicating an outer rectangular frame corresponding to a target lane in the same direction). When the boundary of the first buffer frame or the second buffer frame recognizes the target stopping area (such as a road boundary and a diversion area) during the extension process or when the extended length reaches the extension threshold (such as 20 meters), the extension of the first buffer frame or the second buffer frame is stopped.

[0207] (8) The first angle of the outer rectangular frame in the unified buffer frame:

[0208] Construct a first storage list and a second storage list, the first storage list and the second storage list are used to store incorrect angles and correct angles respectively, determine a first difference between the orientation of the outer rectangular box and the first angle of the outer rectangular box, if the first difference is greater than or equal to a second preset angle (such as 8 degrees), determine that the first angle of the outer rectangular box is an incorrect angle, and store the first angle of the outer rectangular box in the first storage list, if the first difference is less than the second preset angle (such as 8 degrees), determine that the first angle of the outer rectangular box is a correct angle, and store the first angle of the outer rectangular box in the second storage list.

[0209] Construct a third storage list and a fourth storage list, and use any first angle among the correct angles in the second storage list as an initial value. Since it is not known in the second storage list which outer rectangular box's first angle is accurate, a first angle of an outer rectangular box can be randomly specified as the initial value, and the first angles of the outer rectangular boxes in the second storage list except the initial value are compared with the initial value and the second difference is calculated.

[0210] Most of the first angles of the multiple outer rectangular boxes framed by the buffer box are accurate. Therefore, a larger number of first angles of the outer rectangular boxes in the third storage list and the fourth storage list indicates that the first angle data in the storage list is more accurate. The average value of the first angles of the outer rectangular boxes in the storage list is calculated and used as the second angle.

[0211] The first angles of the plurality of outer rectangular frames are adjusted according to the second angle, so that the angles of all the outer rectangular frames in the buffer frame are unified to the second angle.

[0212] (9) Reconstruct the outer rectangular frame:

[0213] Since the length and width of the outer rectangular frame after the angle adjustment have not been unified, the longest outer rectangular frame with the longest rectangular frame length in the buffer frame is selected, and the outer rectangular frame is matched with the international standard size. The template with the length closest to the length of the longest outer rectangular frame in the international standard size is found, and the length and width of all the outer rectangular frames in the buffer frame are reconstructed according to the length and width of the template to obtain the target outer rectangular frame.

[0214] (10) Solve the problem of the target outer rectangular frame covering:

[0215] When the target outer rectangular box is displayed on the high-precision map, if there is a situation of overlapping the nearby markings, the width of the target outer rectangular box is reduced by a preset ratio (such as reducing it to 0.75 times the current width) until the overlapping situation is resolved. When the number of width reductions reaches a preset number (such as 5 times), regardless of whether there is still an overlapping situation, the reduction of the width of the target outer rectangular box is stopped, and the width of the outer rectangular box after the last reduction is maintained.

[0216] Figure 3 is a schematic diagram of a framework of a ground arrow aggregation device provided by an embodiment of the present disclosure, such as Figure 3 As shown, this embodiment provides a ground arrow aggregation device, the device comprising:

[0217] A first calculation module 11 is used to obtain a plurality of cluster data in a target road and calculate an angle mean value according to the cluster data;

[0218] A first generating module 12 is used to generate a related graphic based on the cluster data, wherein the related graphic can include each ground arrow in the cluster data;

[0219] A first operating module 13 is used to obtain an outer rectangular frame of the relevant graphic and its orientation, wherein the outer rectangular frame is a minimum rectangular frame that can contain the relevant graphic, and the angle of the outer rectangular frame is the angle mean;

[0220] A first adjustment module 14, configured to adjust the angle mean based on the outer rectangular frame and its orientation to obtain a first angle of the outer rectangular frame;

[0221] The second operating module 15 is used to construct a buffer frame, and to frame the multiple outer rectangular frames according to the buffer frame;

[0222] A third operating module 16 is used to unify the first angles of the multiple outer rectangular frames in the buffer frame to obtain a second angle;

[0223] The second adjustment module 17 is used to adjust the multiple outer rectangular frames based on the second angle and the international standard size to obtain a target outer rectangular frame.

[0224] Optionally, the device further comprises:

[0225] A first acquisition unit is used to acquire the geometric centroid coordinates of the centroid point of each ground arrow in the cluster data before generating the relevant graphics based on the cluster data;

[0226] The first calculation unit is used to calculate the geometric centroid coordinates of the fitting centroid point of the cluster data according to the geometric centroid coordinates and the mean calculation function.

[0227] Optionally, the first generating module 12 includes:

[0228] A second calculation unit is used to calculate the difference between the geometric centroid coordinates of each ground arrow in the cluster data and the geometric centroid coordinates of the fitting centroid point of the cluster data;

[0229] A first translation unit, configured to translate the centroid of each ground arrow to coincide with the fitting centroid based on the difference;

[0230] The first extraction unit is used to extract the relevant graphics of the ground arrows when the centroid point of each ground arrow coincides with the fitting centroid point.

[0231] Optionally, the first calculation module 11 includes:

[0232] A first acquisition unit is used to acquire the angle of each ground arrow in the cluster data;

[0233] A first filtering unit, used to filter out ground arrows whose angles exceed a first preset angle;

[0234] The third calculation unit is used to calculate the mean of the angles of the retained ground arrows, where the mean is the angle mean of the cluster data.

[0235] Optionally, the device further comprises:

[0236] A first judging unit is used for judging whether there is a ground virtual line on the same horizontal plane within a first preset range of the outer rectangular frame before adjusting the angle mean based on the outer rectangular frame and its orientation to obtain a first angle of the outer rectangular frame;

[0237] The second judgment unit is used to judge whether there is a solid ground line on the same horizontal plane within the first preset range of the outer rectangular frame when the virtual ground line does not exist.

[0238] Optionally, the first adjustment module 14 includes:

[0239] A first determining unit is used to determine a projection starting point on the ground dotted line or the ground solid line when it is determined that the outer rectangular frame has a ground dotted line or a ground solid line on the same horizontal plane within the first preset range;

[0240] A first projection unit is used to project the fitting centroid point of the outer rectangular frame onto the projection starting point, and use the orientation of the outer rectangular frame as the projection direction, and move the projection starting point by a second preset distance in the projection direction on the ground imaginary line or the ground solid line to obtain a projection end point;

[0241] The fourth calculation unit is used to calculate the angle adjustment amount of the outer rectangular frame according to the projection starting point and the projection end point, and adjust the angle mean according to the angle adjustment amount to obtain the first angle of the outer rectangular frame.

[0242] Optionally, the buffer frame includes a first buffer frame and a second buffer frame, and the second operating module 15 includes:

[0243] A first setting unit, used to set initial values ​​of the first buffer frame and the second buffer frame, and to formulate a target stop zone and an extension threshold;

[0244] A first extension unit, used to extend the two ends of the first buffer frame and the second buffer frame for multiple times, and in each extension process, increase the extended width according to a third preset distance to frame the multiple outer rectangular frames;

[0245] The first recognition unit is configured to stop extending when the first buffer frame or the second buffer frame recognizes the target stopping area or reaches the extension threshold during the extension process.

[0246] Optionally, the third operating module 16 includes:

[0247] A first construction unit, configured to construct a first storage list and a second storage list;

[0248] A fifth calculation unit, configured to calculate a first difference between the orientations of a plurality of outer rectangular frames in the buffer frame and the first angles corresponding thereto;

[0249] A first operating unit, configured to store the first angle of the outer rectangular frame in the first storage list when the first difference is greater than or equal to a second preset angle;

[0250] A second operating unit, configured to store the first angle of the outer rectangular frame in the second storage list when the first difference is smaller than the second preset angle;

[0251] A second construction unit is used to construct a third storage list and a fourth storage list, and use the first angle of any one of the multiple enclosing rectangular frames in the second storage list as an initial value;

[0252] A sixth calculation unit, configured to calculate a second difference between the first angles of the plurality of enclosing rectangular frames in the second storage list and the initial value;

[0253] A third operating unit, configured to store the first angle of the outer rectangular frame in the third storage list when the second difference is greater than or equal to a third preset angle;

[0254] A fourth operating unit, configured to store the first angle of the outer rectangular frame in the fourth storage list when the second difference is smaller than the third preset angle;

[0255] A first comparing unit, used for comparing the number of first angles in the third storage list and the number of first angles in the fourth storage list;

[0256] The seventh calculation unit is used to calculate the average value of the first angles in the storage list with more first angles in the third storage list and the fourth storage list, and use the average value as the second angle.

[0257] Optionally, the second adjustment module 17 includes:

[0258] The first adjustment unit is used to adjust the first angle of the plurality of outer rectangular frames to the second angle according to the second angle.

[0259] Optionally, the second adjustment module 17 includes:

[0260] A first selection unit is used to select the longest outer rectangular frame with the longest rectangular frame length among the multiple outer rectangular frames;

[0261] A second selection unit, configured to select a matching template in the international standard size according to the length of the longest outer rectangular frame;

[0262] The third construction unit is used to reconstruct the multiple outer rectangular frames according to the length and width of the template to obtain the target outer rectangular frame.

[0263] Optionally, the device further comprises:

[0264] A fifth operating unit is used for reducing the width of the target outer rectangular frame at a preset ratio after the target outer rectangular frame is obtained and when the target outer rectangular frame overlaps;

[0265] The sixth operating unit is used to stop shrinking the target outer rectangular frame when the number of shrinking times reaches a preset number.

[0266] Based on the same inventive concept, another embodiment of the present disclosure further provides a device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement any of the methods described above.

[0267] Based on the same inventive concept, another embodiment of the present disclosure further provides a computer-readable medium on which a computer program is stored, wherein when the program is executed by a processor, any of the methods described above is implemented.

[0268] As for the device, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0269] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0270] Those skilled in the art will appreciate that the embodiments of the disclosed embodiments may be provided as methods, devices, or computer program products. Therefore, the disclosed embodiments may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the disclosed embodiments may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0271] The embodiments of the present disclosure are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0272] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0273] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1The steps for the functions specified in one or more boxes.

[0274] Although the preferred embodiments of the present disclosure have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present disclosure.

[0275] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0276] The above is a detailed introduction to the construction method, device, equipment and computer-readable medium of a ground arrow provided by the present disclosure. Specific examples are used in this article to illustrate the principles and implementation methods of the present disclosure. The description of the above embodiments is only used to help understand the method of the present disclosure and its core idea; at the same time, for general technical personnel in this field, according to the ideas of the present disclosure, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present disclosure.

Claims

1. A method for aggregating ground arrows, characterized in that: The method comprises: Acquire multiple cluster data in the target road, and calculate the angle mean according to the cluster data; generating a related graphic based on the cluster data, wherein the related graphic can include each ground arrow in the cluster data; Obtaining an outer rectangular frame of the relevant graphic and its orientation, wherein the outer rectangular frame is a minimum rectangular frame that can contain the relevant graphic, and the angle of the outer rectangular frame is the angle mean; Based on the outer rectangular frame and its orientation, the angle mean is adjusted to obtain a first angle of the outer rectangular frame; Constructing a buffer frame, and framing the multiple outer rectangular frames according to the buffer frame; unifying the first angles of the multiple outer rectangular frames in the buffer frame to obtain a second angle; Based on the second angle and the international standard size, the multiple outer rectangular frames are adjusted to obtain a target outer rectangular frame.

2. The method for aggregating ground arrows according to claim 1, characterized in that: The cluster data includes at least one ground arrow, and before generating a related graphic based on the cluster data, the method further includes: Obtaining the geometric centroid coordinates of the centroid point of each ground arrow in the cluster data; The geometric centroid coordinates of the fitting centroid point of the cluster data are calculated based on the geometric centroid coordinates and the mean calculation function.

3. The method for aggregating ground arrows according to claim 2, characterized in that: The step of generating a related graph based on the cluster data comprises: Calculate the difference between the geometric centroid coordinates of each ground arrow in the cluster data and the geometric centroid coordinates of the fitting centroid point of the cluster data; Based on the difference, the centroid point of each ground arrow is translated to coincide with the fitting centroid point; Extract the relevant graphics of the ground arrows when the centroid point of each ground arrow coincides with the fitting centroid point.

4. The method for aggregating ground arrows according to claim 3, characterized in that: The extracting of the relevant graphics of the ground arrow when the centroid point of each ground arrow coincides with the fitting centroid point includes: Using a concave hull function to extract the overlapping area between the ground arrows after the centroid point is translated; The correlation graph is generated through the overlapping area.

5. The method for aggregating ground arrows according to claim 1, characterized in that: The calculating the angle mean according to the cluster data comprises: Collecting the angle of each ground arrow in the cluster data; Filtering out ground arrows whose angles exceed a first preset angle; The angles of the retained ground arrows are averaged, and the average is the angle average of the cluster data.

6. The method for aggregating ground arrows according to claim 1, characterized in that: Before adjusting the angle mean based on the outer rectangular frame and its orientation to obtain the first angle of the outer rectangular frame, the method further includes: Determine whether there is a ground virtual line on the same horizontal plane within the first preset range of the outer rectangular frame; When the ground virtual line does not exist, it is determined whether there is a ground solid line on the same horizontal plane within the first preset range of the outer rectangular frame.

7. The method for aggregating ground arrows according to claim 6, characterized in that: When it is determined that there is a ground dotted line or a ground solid line on the same horizontal plane within the first preset range of the outer rectangular frame, the adjusting the angle mean based on the outer rectangular frame and its orientation to obtain the first angle of the outer rectangular frame includes: Determine a projection starting point on the ground imaginary line or the ground solid line; Project the fitted centroid of the outer rectangular frame onto the projection starting point, and use the orientation of the outer rectangular frame as the projection direction, and move the projection starting point by a second preset distance in the projection direction on the ground imaginary line or the ground solid line to obtain a projection end point; The angle adjustment amount of the outer rectangular frame is calculated according to the projection starting point and the projection end point, and the angle mean is adjusted according to the angle adjustment amount to obtain the first angle of the outer rectangular frame.

8. The method for aggregating ground arrows according to claim 1, characterized in that: The buffer frame includes a first buffer frame and a second buffer frame, and the construction of the buffer frame, based on the buffer frame, frames the multiple outer rectangular frames, including: Setting initial values ​​of the first buffer box and the second buffer box, and formulating a target stop zone and an extension threshold; Extending both ends of the first buffer frame and the second buffer frame for multiple times, and increasing the extended width according to a third preset distance during each extension process to frame the multiple outer rectangular frames; When the first buffer frame or the second buffer frame identifies the target stopping area or reaches the extension threshold during the extension process, the extension is stopped.

9. The method for aggregating ground arrows according to claim 1, characterized in that: The step of unifying the first angles of the plurality of outer rectangular frames in the buffer frame to obtain the second angle includes: Constructing a first storage list and a second storage list; Calculate a first difference between the orientations of a plurality of outer rectangular frames in the buffer frame and the first angles corresponding thereto; When the first difference is greater than or equal to a second preset angle, storing the first angle of the outer rectangular frame in the first storage list; When the first difference is smaller than the second preset angle, storing the first angle of the outer rectangular frame in the second storage list; Constructing a third storage list and a fourth storage list, and taking the first angle of any one of the plurality of enclosing rectangular frames in the second storage list as an initial value; Calculate a second difference between the first angles of the plurality of bounding rectangular frames in the second storage list and the initial value; When the second difference is greater than or equal to a third preset angle, storing the first angle of the outer rectangular frame in the third storage list; When the second difference is less than the third preset angle, storing the first angle of the outer rectangular frame in the fourth storage list; Comparing the number of first angles in the third storage list and the number of first angles in the fourth storage list; An average value of the first angles in the storage list having a larger number of first angles in the third storage list and the fourth storage list is calculated, and the average value is used as the second angle.

10. The method for aggregating ground arrows according to claim 9, characterized in that: The adjusting the multiple outer rectangular frames based on the second angle includes: According to the second angle, the first angles of the plurality of outer rectangular frames are adjusted to the second angle.

11. The method for aggregating ground arrows according to claim 1, characterized in that: The step of adjusting the plurality of outer rectangular frames based on the international standard size to obtain a target outer rectangular frame includes: Selecting the longest outer rectangular frame with the longest rectangular frame length among the multiple outer rectangular frames; Selecting a matching template from the international standard size according to the length of the longest outer rectangular frame; The multiple outer rectangular frames are reconstructed according to the length and width of the template to obtain the target outer rectangular frame.

12. The method for aggregating ground arrows according to claim 1, characterized in that: After obtaining the target outer rectangular frame, the method further includes: When the target outer rectangular frame overlaps, the width of the target outer rectangular frame is reduced by a preset ratio; When the number of reductions reaches a preset number, the reduction of the target outer rectangular frame is stopped.

13. A ground arrow aggregation device, characterized in that: The device comprises: A first calculation module, used for acquiring a plurality of cluster data in a target road, and calculating an angle mean value according to the cluster data; A first generating module, configured to generate a related graphic based on the cluster data, wherein the related graphic can include each ground arrow in the cluster data; A first operation module is used to obtain an outer rectangular frame of the relevant graphic and its orientation, wherein the outer rectangular frame is a minimum rectangular frame that can contain the relevant graphic, and the angle of the outer rectangular frame is the angle mean; A first adjustment module, configured to adjust the angle mean based on the outer rectangular frame and its orientation to obtain a first angle of the outer rectangular frame; A second operating module is used to construct a buffer frame, and to frame the multiple outer rectangular frames according to the buffer frame; A third operation module is used to unify the first angles of the multiple outer rectangular frames in the buffer frame to obtain a second angle; The second adjustment module is used to adjust the multiple outer rectangular frames based on the second angle and the international standard size to obtain a target outer rectangular frame.

14. A device, characterized in that The method comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 12.

15. A computer readable medium, characterized in that A computer program is stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 12 is implemented.