Feedback method, device and equipment for lane line recognition error and storage medium

Abstract

The application relates to a lane line identification error feedback method and device, equipment and a storage medium. The method comprises the following steps: acquiring lane line data, track point data and pixel point data generated for a target road surface; based on the track point data, a global road surface calculation area of the target road surface is determined, and lane lines in the lane line data within the global road surface calculation area are regarded as to-be-detected lane lines; positioning points are generated on the to-be-detected lane lines according to a preset rule, and based on the relationship between each positioning point and pixel points in a preset range, an error positioning point and an error lane line corresponding to the error positioning point are determined. The application embodiment can effectively identify error lane lines in the generated lane lines, determine the error positioning point corresponding to the error lane line, and further determine the road surface image corresponding to the error lane line, so that the road surface image is used for subsequent training of a model, and the ability of the model to identify error lane lines is enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of high-definition map, and particularly to a feedback method and device for lane line recognition error, equipment and a storage medium. BACKGROUND

[0002] With the development of automatic driving technology, the automatic degree of automatic driving vehicles is getting higher and higher. For safety and navigation considerations, a high-precision map with sufficient precision needs to be built.

[0003] In a related technical solution, when building a high-definition map, for the production of lane lines in the high-definition map, the lane line recognition program is used to recognize and then calculate pixel points for reference production by operators. The accuracy of the lane line recognition program directly affects the efficiency and quality of map production. The accuracy of the lane line recognition program mainly depends on the improvement of the network model and the calibration of error data samples. Traditional error data sample collection is manually saved after finding that the pixel points do not match the actual situation during data production. This method not only wastes the time of operators but also is prone to omission, resulting in incomplete model error data samples, affecting the recognition accuracy of the model, and ultimately leading to insufficient precision of the high-definition map. SUMMARY

[0004] To solve or partially solve the problems in the related art, the present application provides a feedback method and device for lane line recognition error, equipment and a storage medium, which can accurately identify error data in lane lines.

[0005] The first aspect of the present application provides a feedback method for lane line recognition error, comprising:

[0006] Obtaining lane line data generated for a target road surface, trajectory point data of a collection vehicle driving on the target road surface, and pixel point data used to assist in generating the lane line data;

[0007] Based on the trajectory point data, determining a global road surface calculation area of the target road surface, and regarding lane lines in the lane line data within the global road surface calculation area as to-be-detected lane lines;

[0008] Generating positioning points on the to-be-detected lane lines according to a preset rule, and determining error positioning points and error lane lines corresponding to the error positioning points based on the relationship between each positioning point and pixel points within a preset range.

[0009] As a possible implementation manner of the present application, in this implementation manner, after obtaining the lane line data generated for the target road surface, the method comprises:

[0010] Based on attribute information of each lane line in the lane line data, deleting non-lane lines in the lane line data.

[0011] As a possible implementation of the present application, in this implementation, the determining the global road surface solving area of the target road surface based on the trajectory point data comprises:

[0012] traversing each trajectory point in the trajectory point data, determining a distortion box corresponding to each trajectory point based on the camera internal and external parameters, projecting the distortion box to the target road surface to generate a road surface solving area;

[0013] fusing all the road surface solving areas corresponding to the trajectory points to generate a global road surface solving area.

[0014] As a possible implementation of the present application, in this implementation, before the determining the error positioning point and the error lane line corresponding to the error positioning point based on the relationship between each positioning point and the pixel points within a preset range, the method comprises:

[0015] classifying the to-be-detected lane line based on attribute information of the to-be-detected lane line, wherein the category of the to-be-detected lane line comprises a solid lane line and a dashed lane line.

[0016] As a possible implementation of the present application, in this implementation, when the to-be-detected lane line is a solid lane line, the generating a positioning point on the to-be-detected lane line according to a preset rule and determining an error positioning point and an error lane line corresponding to the error positioning point based on the relationship between each positioning point and the pixel points within a preset range comprises:

[0017] for each solid lane line, generating a positioning point on the solid lane line at every first preset distance based on a starting point of the lane line;

[0018] for each positioning point, when it is determined that there is no pixel point within a first preset range from the positioning point, determining the positioning point as an error positioning point;

[0019] determining the solid lane line within a preset distance before and after the error positioning point as an error lane line.

[0020] As a possible implementation of the present application, in this implementation, when the to-be-detected lane line is a dashed lane line, the generating a positioning point on the to-be-detected lane line according to a preset rule and determining an error positioning point and an error lane line corresponding to the error positioning point based on the relationship between each positioning point and the pixel points within a preset range comprises:

[0021] Step 1, for each dashed lane line, generating a positioning point on the dashed lane line at every second preset distance based on a starting point of the lane line, and setting a counting variable;

[0022] Step 2, set the count variable to zero, and determine whether there is a pixel point in the first preset range of the first positioning point;

[0023] Step 3, when there is a pixel point in the first preset range of the first positioning point, the next positioning point connected with the first positioning point is taken as the first positioning point, and the step 2 is returned, otherwise, the step 4 is executed;

[0024] Step 4, the count variable is added by 1, and it is determined whether the value of the count variable is greater than the count variable threshold value;

[0025] Step 5, when it is determined that the value of the count variable is not greater than the count variable threshold value, the next positioning point connected with the first positioning point is taken as the first positioning point, and the step 3 is returned; otherwise, the step 6 is executed;

[0026] Step 6, the first positioning point is determined as an error positioning point, a positioning point spaced by a preset number of positioning points from the first positioning point after the first positioning point is determined as the first positioning point, and the step 2 is returned until all the positioning points on the dashed lane line are determined.

[0027] As a possible implementation of the present application, in the implementation, the method further comprises:

[0028] determining a target driving track point at which the error positioning point can be photographed when the collection vehicle drives on the target road surface;

[0029] determining a road surface image collected when the vehicle is at the target driving track point as an error road surface image.

[0030] The second aspect of the present application provides a feedback device for lane line recognition error, comprising:

[0031] a data acquisition module, configured to acquire lane line data generated for a target road surface, track point data of a collection vehicle driving on the target road surface, and pixel point data used for assisting in generating the lane line data;

[0032] a lane line determination module, configured to determine a global road surface solving area of the target road surface based on the track point data, and take lane lines in the lane line data within the global road surface solving area as to-be-detected lane lines;

[0033] an error determination module, configured to generate positioning points on the to-be-detected lane lines according to a preset rule, and determine error positioning points and error lane lines corresponding to the error positioning points based on relationships between the positioning points and pixel points in a preset range.

[0034] The third aspect of the present application provides an electronic device, comprising:

[0035] a processor; and

[0036] a memory having stored thereon executable code that, when executed by the processor, is to cause the processor to perform the method as described above.

[0037] The fourth aspect of the present application provides a computer-readable storage medium having stored thereon executable code that, when executed by a processor of an electronic device, is to cause the processor to perform the method as described above.

[0038] The embodiments of the present application acquire lane line data generated for a target road surface, collect trajectory point data of a vehicle driving on the target road surface and pixel point data used for assisting generation of the lane line data, determine a global road surface solving area of the target road surface based on the trajectory point data, determine lane lines in the global solving area as to-be-detected lane lines, determine a positioning point and an error lane line based on a relationship between the positioning point on the to-be-detected lane line and pixel points in a preset range of the positioning point. The error lane line in the generated lane line can be effectively identified, the error positioning point corresponding to the error lane line is determined, and then the road surface image corresponding to the error lane line is determined, which facilitates subsequent training of a model using the road surface image and enhances the ability of the model to identify error lane lines.

[0039] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory and are not restrictive of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0040] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which like reference characters refer to like parts throughout the figures, and wherein:

[0041] Figure 1 is a flowchart of a feedback method for lane line recognition error according to an embodiment of the present application;

[0042] Figure 2 is a schematic diagram of a road surface solving area according to an embodiment of the present application;

[0043] Figure 3 is a flowchart of a method for determining a global road surface solving area according to an embodiment of the present application;

[0044] Figure 4 is a flowchart of a method for detecting a solid lane line according to an embodiment of the present application;

[0045] Figure 5 is a flowchart of a method for detecting a dashed lane line according to an embodiment of the present application;

[0046] Figure 6 is a flowchart of a method for determining an erroneous road image according to an embodiment of the present application;

[0047] Figure 7 is a schematic diagram of determining a target trajectory point according to an embodiment of the present application;

[0048] Figure 8 is a structural schematic diagram of a feedback device for lane line recognition error according to an embodiment of the present application;

[0049] Figure 9 is a structural schematic diagram of an electronic device according to an embodiment of the present application.

[0050] The above and other features, aspects, and advantages of the present disclosure will become more apparent with reference to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals are used to refer to the same or similar elements. It is to be understood that the drawings are schematic, and the proportions of the various elements do not necessarily bear the same scale. DETAILED DESCRIPTION

[0051] Embodiments of the present application will be described in greater detail below with reference to the accompanying drawings. While embodiments of the present application are shown in the drawings, it is understood that the present application can be embodied in various forms without being limited by the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.

[0052] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0053] It is to be understood that the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. It is to be further understood that the terms "comprise", "comprising", "comprises", "including", "comprising", "comprises", "include" and / or "comprising", when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0054] With the development of automatic driving technology, the automatic degree of the automatic driving vehicle is higher and higher, and for the safety and navigation, a high-precision map with sufficient precision needs to be constructed. In the related technical solution, when constructing the high-precision map, for the production of lane lines in the high-precision map, the lane line recognition program is used to recognize and then calculate pixel points for reference production by the operator, the accuracy of the lane line recognition program directly affects the efficiency and quality of the map production, and the accuracy of the lane line recognition program mainly depends on the improvement of the network model and the calibration of the error data sample. The traditional error data sample collection is to manually save the pixel points that do not match the actual situation after finding them in the process of producing data, which not only wastes the time of the operator but also is prone to omission, resulting in incomplete model error data samples and affecting the recognition accuracy of the model, and finally leading to insufficient precision of the high-precision map.

[0055] To solve the above problems, the embodiment of the present application provides a lane line recognition error feedback method, which can accurately identify error data in the lane line.

[0056] Figure 1 is a flowchart of the lane line recognition error feedback method provided by the embodiment of the present application.

[0057] Referring to Figure 1 , the lane line recognition error feedback method provided by the embodiment of the present application comprises:

[0058] In step S101, lane line data generated for a target road surface, trajectory point data of a collection vehicle driving on the target road surface, and pixel point data used for assisting generation of the lane line data are acquired.

[0059] In the embodiment of the present application, the target road surface refers to a road surface that needs to be added to a high-precision map and generate high-precision lane line data, which is generally achieved by collecting a lane line image of the target road surface by a collection vehicle driving on the target road surface, and then generating lane line data based on the lane line image of the road surface; the trajectory point data is the driving trajectory point when the collection vehicle collects the lane line image of the target road surface; when generating the lane line data, the lane line image collected by the collection vehicle on the target road surface needs to be calculated into pixel point data of the lane line, and then the pixel point data is used to generate lane line data in the high-precision map.

[0060] In the embodiment of the present application, when the lane line data, the trajectory point data and the pixel point data are acquired, the lane line data, the trajectory point data and the pixel point data matched with the lane line data can be acquired from a sample database for the lane line data that has been generated.

[0061] Step S102, determining a global road surface solving region of the target road surface based on the trajectory point data, and taking the lane line in the lane line data within the global road surface solving region as a lane line to be detected.

[0062] In the embodiment of the present application, the global road surface solving region refers to a road surface region that needs to be solved for the road lane line. Due to the influence of the installation angle, the shooting range, the distortion range and other factors of the vehicle image acquisition device, for the lane line in the target road surface image collected by the vehicle image acquisition device, in order to improve the accuracy of lane line recognition, only part of the lane line in the target road surface can be solved, and part of the data needs to be discarded. In order to realize this scheme, as shown in Figure 2 , a road surface solving region is provided. When the vehicle is driving on the target road surface, a distortion box of a preset range is determined for the road surface image collected by the collection vehicle. The lane line image within the distortion box needs to be solved, and the lane line image outside the distortion box does not need to be solved. It should be understood that the above solving process is for the same target road surface image and the same lane line. When the lane line is outside the distortion box in the first target road surface image, the solving of the first target road surface image does not need to be performed on the lane line. If the lane line is within the distortion box in the second target road surface image, the lane line needs to be solved. Specifically, as shown in Figure 3 , the determination of the global road surface solving region of the target road surface based on the trajectory point data comprises:

[0063] Step S301, traversing each trajectory point in the trajectory point data, determining a distortion box corresponding to each trajectory point based on the camera internal and external parameters, projecting the distortion box to the target road surface to generate a road surface solving region.

[0064] In the embodiment of the present application, as described in the foregoing embodiment, for each trajectory point, when the collection vehicle is driving on the trajectory point to collect the target road surface image, a distortion box as shown in Figure 2 is determined based on the camera internal and external parameters, and the distortion box is projected to the target road surface to obtain a road surface solving region corresponding to the trajectory point.

[0065] Step S302, fusing all the road surface solving regions corresponding to the trajectory points to generate a global road surface solving region.

[0066] In the embodiment of the present application, after all the road surface solving regions corresponding to the trajectory points are generated, the all road surface solving regions are fused to obtain a global road surface solving region.

[0067] In the embodiment of the present application, after determining the global road surface solving area, all lane line data needs to be judged to see whether the lane line is in the global road surface solving area, and the lane line in the global solving area is taken as the lane line to be detected. It should be understood that, when judging the lane line, the lane line corresponding to each target road surface image can be determined based on the road surface solving area corresponding to the target road surface image, and then the lane lines of all target road surface images are connected to obtain all lane lines to be detected.

[0068] In step S103, positioning points are generated on the lane line to be detected according to a preset rule, and based on the relationship between each positioning point and the pixel points in the preset range, an error positioning point and an error lane line corresponding to the error positioning point are determined.

[0069] In the embodiment of the present application, after determining the lane line to be detected, positioning points are generated on the lane line to be detected according to a preset rule, and based on the relationship between the positioning points and the pixel points in the preset range, an error positioning point and an error lane line are determined. As a possible implementation manner of the present application, when generating the positioning points on the lane line to be detected, the starting point of the lane line can be taken as the starting point, and one positioning point is generated on the lane line every preset distance, which can be 0.5 meters. For each positioning point, whether the positioning point is correct is judged based on the relationship between the positioning point and the pixel points in the preset range.

[0070] In the embodiment of the present application, after obtaining the lane line data generated for the target road surface, the following steps are included:

[0071] Based on the attribute information of each lane line in the lane line data, the non-lane line in the lane line data is deleted.

[0072] In an embodiment of the present application, after obtaining the lane line data, the lane line data needs to be preliminarily screened, including deleting the non-lane line in the lane line data. It should be understood that the non-lane line refers to the virtual lane line and auxiliary line used for assistance when generating the lane line according to the target road surface image. For example, for the road lacking lane line such as a road intersection, the two sections of the road at the intersection need to be connected by a virtual lane line to facilitate navigation of the vehicle, so the virtual lane line can be deleted when screening the lane line data. In the embodiment of the present application, the non-lane line such as the virtual lane line in the lane line is identified by attribute information, so the non-lane line can be deleted according to the attribute information in the lane line data.

[0073] In the embodiment of the present application, before the step of generating the positioning points on the to-be-detected lane line according to the preset rule, determining the error positioning points and the error lane lines corresponding to the error positioning points based on the relationship between each positioning point and the pixel points within the preset range, the method further comprises:

[0074] classifying the to-be-detected lane line based on the attribute information of the to-be-detected lane line, wherein the category of the to-be-detected lane line comprises a solid lane line and a dashed lane line.

[0075] As a possible implementation manner of the present application, in this implementation manner, before detecting the to-be-detected lane line, the to-be-detected lane line needs to be classified, wherein the to-be-detected lane line comprises a solid lane line and a dashed lane line, and when classifying the to-be-detected lane line, the attribute information of the to-be-detected lane line can be used to determine whether the to-be-detected lane line is a solid lane line or a dashed lane line. Optionally, the attribute information of each lane line has a type marking identifier for indicating the type of the lane line, and the type of the lane line is determined based on the type marking identifier.

[0076] As a possible implementation manner of the present application, in this implementation manner, as shown in Figure 4 when the to-be-detected lane line is a solid lane line, the step of generating the positioning points on the to-be-detected lane line according to the preset rule, determining the error positioning points and the error lane lines corresponding to the error positioning points based on the relationship between each positioning point and the pixel points within the preset range, comprises:

[0077] In step S401, for each solid lane line, a positioning point is generated on the solid lane line every first preset distance based on the starting point of the lane line.

[0078] In the embodiment of the present application, when detecting the solid lane line, a positioning point is generated on the solid lane line every first preset distance based on the starting point of the solid lane line, wherein the first preset distance can be determined according to actual conditions, such as 0.5 meters, and starting from the starting positioning point of the solid lane line, a positioning point is generated every 0.5 meters on the solid lane line.

[0079] In step S402, for each positioning point, when it is determined that there is no pixel point within the first preset range from the positioning point, the positioning point is determined as an error positioning point.

[0080] In the embodiment of the present application, for each positioning point, it is needed to determine whether there is a pixel point in the first preset range thereof. When it is determined that there is a pixel point in the first preset range of the positioning point, it is indicated that the lane line at the positioning point is generated according to the pixel point, and it is indicated that the lane line at the positioning point is a correct lane line. When it is determined that there is no pixel point in the first preset range of the positioning point, it is indicated that the lane line corresponding to the positioning point is not generated according to the pixel point, and it is an incorrect lane line. The positioning point is an incorrect positioning point. In the embodiment of the present application, the first preset range can be a range within 0.2 meters around the positioning point.

[0081] In step S403, the solid lane line within the preset distance before and after the incorrect positioning point is determined as an incorrect lane line.

[0082] In the embodiment of the present application, after the incorrect positioning point is determined, the corresponding incorrect lane line can be determined according to the lane line information near the incorrect positioning point. Optionally, the solid lane line within the preset distance before and after the incorrect positioning point can be selected on the solid lane line, and the solid lane line is determined as an incorrect lane line. In the embodiment of the present application, when an incorrect positioning point is determined when the positioning points on the solid lane line are traversed and detected, to prevent too much invalid work, the detection of the positioning points within the preset distance after the incorrect positioning point can be cancelled. For example, the first positioning point 5 meters after the incorrect positioning point is selected as the next positioning point to be detected.

[0083] As a possible implementation manner of the present application, in the implementation manner, as shown in FIG. 6, when the lane line to be detected is a solid lane line, the step of generating the positioning points on the lane line to be detected according to the preset rule, determining the incorrect positioning points and the incorrect lane lines corresponding to the incorrect positioning points based on the relationship between each positioning point and the pixel points in the preset range includes the following steps. Figure 5

[0084] In step 1, for each solid lane line, a positioning point is generated on the solid lane line every second preset distance based on the starting point of the lane line, and a counting variable is set.

[0085] ​In the embodiment of the present application, when detecting the to-be-detected dashed lane line, based on the starting point of the dashed lane line, a positioning point is generated every second preset distance on the dashed lane line, wherein the second preset distance can be determined according to actual conditions, such as 0.5 meters, and on the dashed lane line, a positioning point is generated every 0.5 meters starting from the starting positioning point of the dashed lane line, at the same time, a counting variable, such as end_num=0, needs to be set, and based on the lane line level where the dashed lane line is located, the threshold value, i.e., the maximum value, of the counting variable needs to be set, wherein the threshold value of the counting variable is determined based on the distance of the two lane line segments connected by the dashed lane line, for example, for the dashed lane line on the highway, the threshold value can be set to 20, for the dashed lane line on the urban road, the threshold value can be set to 14, and for the dashed lane line on the rural road, the threshold value can be set to 10, of course, the above embodiment is only one possible implementation manner of the present application, and the present application is not limited thereto.

[0086] Step 2, set the counting variable to zero, and determine whether there is a pixel point in the first preset range of the first positioning point.

[0087] In the embodiment of the present application, when the first first positioning point is selected, the starting point of the dashed lane line can be taken as the first first positioning point, the counting variable is set to zero, the positioning points on the dashed lane line are traversed, and it is determined whether there is a pixel point in the first preset range of each positioning point, wherein the first preset range can be 0.2 meters.

[0088] Step 3, when there is a pixel point in the first preset range of the first positioning point, the next positioning point connected with the first positioning point is taken as the first positioning point, and step 2 is returned, otherwise, step 4 is executed.

[0089] In the embodiment of the present application, when it is determined that there is a pixel point within 0.2 meters around the first positioning point, it is indicated that the lane line corresponding to the positioning point is generated according to the pixel point, which is a correct lane line, and the positioning point is a correct positioning point, the positioning point is not processed, the next positioning point connected with the first positioning point is taken as the first positioning point, and step 2 is returned, otherwise, step 4 is executed.

[0090] Step 4, the counting variable is incremented by 1, and it is determined whether the value of the counting variable is greater than the threshold value of the counting variable.

[0091] In the embodiment of the present application, when it is determined that there is no pixel point within 0.2 meters around the first positioning point, the counting variable is incremented by 1, and it is determined whether the value of the counting variable is greater than the threshold value of the counting variable.

[0092] Step 5, when determining that the value of the count variable is not greater than the count variable threshold value, taking the next positioning point connected with the first positioning point as the first positioning point, and returning to step 3; otherwise, step 6 is executed.

[0093] In the embodiment of the present application, since there is an intermittent blank on the dashed lane line, even if it is determined that there is no pixel point within 0.2 meters near the first positioning point, it is possible that the positioning point is just at the blank of the dashed lane line, and the count variable needs to be determined. When determining that the value of the count variable is not greater than the count variable threshold value, it is indicated that the positioning point is indeed just at the blank of the dashed lane line, and the lane line is still normal. The next positioning point connected with the first positioning point is taken as the first positioning point, and step 3 is returned.

[0094] Step 6, determining the first positioning point as an error positioning point, determining a positioning point spaced by a preset number of positioning points from the first positioning point after the first positioning point as the first positioning point, and returning to step 2 until all positioning points on the dashed lane line are determined.

[0095] In the embodiment of the present application, when determining that the value of the count variable is greater than the count variable threshold value, it is indicated that the positioning point should be at the white line segment of the dashed lane line at this time, but there is no pixel point within 0.2 meters near the positioning point, which indicates that the lane line at this position is an error lane line. The positioning point is determined as an error positioning point, and step 2 is returned until all positioning points on the dashed lane line are determined.

[0096] As a possible implementation manner of the present application, in the implementation manner, as described above, the method further includes: Figure 6

[0097] Step S601, determining a target driving track point at which the error positioning point is located when the collection vehicle drives on the target road surface.

[0098] In the embodiment of the present application, after the error lane line and the error positioning point are determined, the road surface image collected by the collection vehicle corresponding to the error positioning point and the error lane line needs to be found, so as to facilitate subsequent research on why the error image in the road surface image collected by the collection vehicle is recognized as a lane line. For example, after the positioning point is determined, a target driving track point at which the error positioning point is located when the collection vehicle drives on the target road surface is determined. Specifically, the method for determining the target driving track point can be a geometric graphing method, as described above. Figure 7 ​As shown, the target road surface has two lanes composed of three lane lines, and the driving track point of the collection vehicle is in the middle of the right lane. When determining the target driving track point corresponding to the error positioning point, a line segment with an angle of 30° between the positioning point and the lane line and a distance of 15 m can be taken, the closest point to the line segment in the track point is found, and whether the track direction of the point is consistent with the position of the lane line is determined. If not, the second closest point is found, until a track point with a consistent direction with the position of the lane line is found, and the track point is determined as the target track point.

[0099] In step S602, the road surface image collected when the vehicle is at the target driving track point is determined as the error road surface image.

[0100] In the embodiment of the present application, after the target track point is determined, the target road surface image collected when the collection vehicle drives at the target track point is obtained, and the target road surface image is recorded as the error road surface image for subsequent analysis.

[0101] The embodiment of the present application determines the global road surface solving area of the target road surface based on the track point data, and determines the lane line in the global solving area as the to-be-detected lane line. Based on the relationship between the positioning point on the to-be-detected lane line and the pixel points in the preset range thereof, the positioning point and the error lane line are determined. The error lane line in the generated lane line can be effectively identified, and the error positioning point corresponding to the error lane line is determined, and then the road surface image corresponding to the error lane line is determined, which facilitates subsequent training of the model using the road surface image and enhances the ability of the model to identify error lane lines.

[0102] Corresponding to the foregoing application function implementation method embodiment, the present application also provides a lane line error feedback device, an electronic device, and corresponding embodiments.

[0103] Figure 8 FIG. 1 is a structural schematic diagram of a lane line error feedback device according to an embodiment of the present application.

[0104] Referring to FIG. 1, Figure 8 The lane line error feedback device 80 provided by the embodiment of the present application includes a data acquisition module 810, a lane line determination module 820, and an error judgment module 830, wherein:

[0105] The data acquisition module 810 is configured to acquire lane line data generated for a target road surface, track point data of a collection vehicle driving on the target road surface, and pixel point data used to assist in generating the lane line data.

[0106] The lane line determination module 820 is configured to determine a global road solving area of the target road surface based on the trajectory point data, and take lane lines in the lane line data within the global road solving area as to-be-detected lane lines.

[0107] The error determination module 830 is configured to generate positioning points on the to-be-detected lane lines according to a preset rule, determine error positioning points and error lane lines corresponding to the error positioning points based on a relationship between each positioning point and a pixel point within a preset range.

[0108] As a possible implementation form of the present application, after the lane line data generated for the target road surface is obtained, the following steps are included:

[0109] Based on attribute information of each lane line in the lane line data, non-lane lines in the lane line data are deleted.

[0110] As a possible implementation form of the present application, the determination of the global road solving area of the target road surface based on the trajectory point data includes:

[0111] Each trajectory point in the trajectory point data is traversed, a distortion box corresponding to each trajectory point is determined based on camera internal and external parameters, the distortion box is projected to the target road surface to generate a road solving area;

[0112] All road solving areas corresponding to the trajectory points are fused to generate the global road solving area.

[0113] As a possible implementation form of the present application, before the positioning points are generated on the to-be-detected lane lines according to a preset rule, and the error positioning points and the error lane lines corresponding to the error positioning points are determined based on a relationship between each positioning point and a pixel point within a preset range, the following steps are included:

[0114] Based on attribute information of the to-be-detected lane lines, the to-be-detected lane lines are classified, and the classes of the to-be-detected lane lines include solid lane lines and dashed lane lines.

[0115] As a possible implementation form of the present application, when the to-be-detected lane line is a solid lane line, the positioning points are generated on the to-be-detected lane line according to a preset rule, and the error positioning points and the error lane lines corresponding to the error positioning points are determined based on a relationship between each positioning point and a pixel point within a preset range, including:

[0116] For each solid lane line, a positioning point is generated on the solid lane line at every first preset distance based on a starting point of the lane line.

[0117] For each positioning point, when it is judged that there is no pixel point within a first preset range from the positioning point, the positioning point is determined as an error positioning point;

[0118] The solid lane line within a preset distance before and after the error positioning point is determined as an error lane line.

[0119] As a possible implementation form of the present application, in the implementation form, when the lane line to be detected is a dashed lane line, the generating positioning points on the lane line to be detected according to a preset rule, determining error positioning points and error lane lines corresponding to the error positioning points based on the relationship between each positioning point and pixel points within a preset range, comprises:

[0120] Step 1, for each dashed lane line, a positioning point is generated on the dashed lane line every second preset distance based on a lane line starting point, and a counting variable is set;

[0121] Step 2, the counting variable is set to zero, and it is judged whether there is a pixel point within a first preset range of a first positioning point;

[0122] Step 3, when there is a pixel point within the first preset range of the first positioning point, the next positioning point connected with the first positioning point is taken as the first positioning point, and the step 2 is returned, otherwise, the step 4 is executed;

[0123] Step 4, the counting variable is increased by 1, and it is judged whether the value of the counting variable is greater than a counting variable threshold value;

[0124] Step 5, when it is judged that the value of the counting variable is not greater than the counting variable threshold value, the next positioning point connected with the first positioning point is taken as the first positioning point, and the step 3 is returned; otherwise, the step 6 is executed;

[0125] Step 6, the first positioning point is determined as an error positioning point, a positioning point spaced from the first positioning point by a preset number of positioning points after the first positioning point is determined as the first positioning point, the step 2 is returned, and all positioning points on the dashed lane line are judged until the step 2 is returned.

[0126] As a possible implementation form of the present application, in the implementation form, the method further comprises:

[0127] Determining a target driving track point at which the error positioning point can be photographed when the collection vehicle drives on the target road surface;

[0128] Determining a road surface image collected when the vehicle is at the target driving track point as an error road surface image.

[0129] With regard to the apparatus in the above-described embodiments, in which the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, no detailed elaboration will be made here.

[0130] The embodiment of the present application can effectively identify the error lane line in the generated lane line, and determine the error positioning point corresponding to the error lane line, and then determine the road surface image corresponding to the error lane line, so as to facilitate subsequent training of the model using the road surface image, and enhance the ability of the model to identify the error lane line.

[0131] Reference will be made to the following Figure 9 , which shows a structural schematic diagram of an electronic device 900 suitable for implementing the embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure can include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Personal Computers), PMPs (Portable Multimedia Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. Figure 9 The electronic device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present disclosure.

[0132] The electronic device includes a memory and a processor, where the processor here can be referred to as the processing device 901 described below, and the memory can include at least one of a read-only memory (ROM) 902, a random access memory (RAM) 903, and a storage device 908, as shown below:

[0133] As shown in Figure 9 , the electronic device 900 can include a processing device (such as a central processor, a graphics processor, etc.) 901, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 902 or programs loaded from a storage device 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the electronic device 900 are also stored. The processing device 901, the ROM 902, and the RAM 903 are connected to each other through a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0134] In general, the following devices can be connected to the I / O interface 905: input devices 906, including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, and the like; output devices 907, including, for example, a liquid crystal display (LCD), a speaker, a vibrator, and the like; storage devices 908, including, for example, a magnetic tape, a hard disk, and the like; and communication devices 909. The communication devices 909 can allow the electronic device 900 to communicate wirelessly or through a wire with other devices to exchange data. Although Figure 9 The electronic device 900 is shown with various devices, but it is understood that all of the illustrated devices are not required to implement or be present. More or fewer devices can alternatively be implemented or present.

[0135] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication devices 909, or installed from the storage devices 908, or installed from the ROM 902. When the computer program is executed by the processing devices 901, the above-described functions defined in the methods of the embodiments of the present disclosure are performed.

[0136] It should be noted that the computer-readable medium or media used to provide the communications of the present disclosure can be located in the storage medium, access port, or any combination thereof. The computer-readable medium or media used to provide the communications of the present disclosure can be of any type suitable to the relevant application, including but not limited to any type of disk including floppy disks, optical disks, magnetic disks, magneto-optical disks, compact disk read-only memories (CD-ROMs), compact disk rewritables (CD-RWs), and any type of solid-state drive (SSDs) including flash, solid state drives (SSDs), or any suitable combination of the foregoing. The computer-readable medium can be a computer program product that includes a computer program tangibly embodied in the medium for use with a processor. The computer program product can or can not be encoded. When encoded, the computer program is not solely comprised of instructions since physical implementation of the program can not be solely comprised of instructions.

[0137] In some embodiments, the client, server, or both can communicate using any known or future developed network protocols, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (for example, a communication network). Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), the Internet, and peer-to-peer networks (for example, ad hoc peer-to-peer networks), as well as any current or future developed network.

[0138] The computer readable medium described above carries one or more programs, when the one or more programs are executed by the electronic device, cause the electronic device to: acquire lane line data generated for a target road surface, track point data collected by a vehicle driving on the target road surface, and pixel point data used to assist in generating the lane line data; determine a global road solving area of the target road surface based on the track point data, and take lane lines in the lane line data within the global road solving area as lane lines to be detected; and generate positioning points on the lane lines to be detected according to a preset rule, and determine an error positioning point and an error lane line corresponding to the error positioning point based on a relationship between each positioning point and pixel points within a preset range.

[0139] Computer program code for carrying out operations of the present disclosure can be written in one or more programming languages or combinations of languages including object oriented programming languages such as Java, Smalltalk, C++ as well as conventional procedural programming languages such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0140] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of present disclosure. In this regard, each block in the flow diagrams or block diagrams can represent a module, a procedure, or a portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or in the reverse order, depending on the functionality involved. It is also noted that each block of the block diagrams and / or flow diagrams and combinations of blocks in the block diagrams and / or flow diagrams can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.

[0141] The modules or units involved in embodiments of the present disclosure can be implemented by software, or by hardware, or by a combination of software and hardware. In some cases, the names of the modules or units do not constitute a limitation on the modules or units themselves.

[0142] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, non-limiting examples of exemplary types of hardware logic components that can be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SOCs), complex programmable logic devices (CPLDs), etc.

[0143] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include a lined paper, an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0144] The above description is merely exemplary of preferred embodiments of the present disclosure and of the application of the principles thereof, and the scope of the disclosure is not limited to those embodiments. Those skilled in the art will readily devise their own methods and embodiments within the scope of the present disclosure, which is defined by the appended claims, and their equivalents. For example, the above-described features can be interchanged between embodiments, or with other features disclosed herein (but not limited to) that perform similar functions.

[0145] Furthermore, while operations are depicted in a particular, sequential order, this should not be understood as requiring or implying that the operations are performed in the order depicted or in sequential order, and that other operations were not intervening, including partially or fully in parallel with other operations. In certain circumstances, multitasking and parallel processing can be advantageous. Also, while a number of specific implementation details are discussed herein, these should not be construed as limiting the scope of the disclosure, but merely as describing examples of implementations. Certain features that are described in the context of separate embodiments can also be implemented together in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented separately or in any suitable subcombination.

[0146] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

1. A method for providing feedback on lane line recognition errors, characterized in that, include: Acquire lane line data generated for the target road surface, collect trajectory point data of the vehicle traveling on the target road surface, and pixel point data used to assist in generating the lane line data, and delete non-lane lines in the lane line data; Based on the trajectory point data, the global road surface solution area of ​​the target road surface is determined, and the lane lines in the lane line data within the global road surface solution area are taken as the lane lines to be detected. On the lane line to be detected, positioning points are generated according to preset rules. Based on the relationship between each positioning point and pixels within a preset range, erroneous positioning points and the erroneous lane lines corresponding to the erroneous positioning points are determined, including: when the lane line to be detected is a dashed lane line, step 1, for each dashed lane line, based on the starting point of the lane line, a positioning point is generated every second preset distance on the dashed lane line, and a counter variable is set; step 2, the counter variable is set to zero, and it is determined whether there are pixels within the first preset range of the first positioning point; step 3, when there are pixels within the first preset range of the first positioning point, the next positioning point connected to the first positioning point is taken as the first positioning point, and the process returns to step 2; otherwise, step 4 is executed; step 4, the counter variable is incremented by 1, and it is determined whether the value of the counter variable is greater than the counter variable threshold; step 5, when it is determined that the value of the counter variable is not greater than the counter variable threshold, the next positioning point connected to the first positioning point is taken as the first positioning point, and the process returns to step 3; Otherwise, proceed to step 6; in step 6, the first positioning point is determined as an incorrect positioning point, and it is determined whether the positioning points have been traversed. If yes, the set of incorrect positioning points is output. At this time, it is considered that all positioning points on the dashed lane line have been judged. If not, the positioning point after the first positioning point and spaced a preset number of positioning points away from the first positioning point is determined as the first positioning point, and the process returns to step 2.

2. The method for feedback of lane line recognition errors according to claim 1, characterized in that, After acquiring the lane line data generated for the target road surface, the process includes: Based on the attribute information of each lane line in the lane line data, delete the non-lane lines in the lane line data.

3. The method for feedback of lane line recognition errors according to claim 1, characterized in that, The step of determining the global road surface solution area of ​​the target road surface based on the trajectory point data includes: Traverse each trajectory point in the trajectory point data, determine the distortion box corresponding to each trajectory point based on the camera intrinsic and extrinsic parameters, project the distortion box onto the target road surface, and generate a road surface solution area; The road surface solution regions corresponding to all the trajectory points are merged to generate a global road surface solution region.

4. The method for feedback of lane line recognition errors according to claim 1, characterized in that, Before generating positioning points on the lane to be detected according to preset rules, and determining erroneous positioning points and the corresponding erroneous lane lines based on the relationship between each positioning point and pixels within a preset range, the process includes: Based on the attribute information of the lane lines to be detected, the lane lines to be detected are classified, wherein the categories of the lane lines to be detected include solid lane lines and dashed lane lines.

5. The method for feedback of lane line recognition errors according to claim 4, characterized in that, When the lane line to be detected is a solid lane line, the process of generating positioning points on the lane line to be detected according to preset rules, and determining erroneous positioning points and the erroneous lane lines corresponding to the erroneous positioning points based on the relationship between each positioning point and pixels within a preset range, includes: For each solid lane line, a positioning point is generated at a first preset distance along the solid lane line based on the starting point of the lane line; For each positioning point, if it is determined that there are no pixels within a first preset range from the positioning point, the positioning point is identified as an incorrect positioning point. The solid lane lines within a preset distance before and after the erroneous location point are identified as erroneous lane lines.

6. The method for feedback of lane line recognition errors according to claim 5, characterized in that, The method further includes: It is determined that when the data collection vehicle is traveling on the target road surface, it is able to capture the target travel trajectory point where the erroneous positioning point is located; The road surface image collected when the vehicle is at the target driving trajectory point is identified as an erroneous road surface image.

7. A feedback device for lane line recognition errors, characterized in that, include: The data acquisition module is used to acquire lane line data generated for the target road surface, collect trajectory point data of the vehicle traveling on the target road surface, and pixel point data used to assist in generating the lane line data, and delete non-lane lines in the lane line data. The lane line determination module is used to determine the global road surface solution area of ​​the target road surface based on the trajectory point data, and to take the lane lines in the lane line data within the global road surface solution area as the lane lines to be detected. An error detection module is used to generate positioning points on the lane line to be detected according to preset rules, and to determine the erroneous positioning points and the erroneous lane lines corresponding to the erroneous positioning points based on the relationship between each positioning point and the pixels within a preset range. This includes: when the lane line to be detected is a dashed lane line, step 1: for each dashed lane line, based on the starting point of the lane line, generate a positioning point every second preset distance on the dashed lane line, and set a counter variable; step 2: set the counter variable to zero, and determine whether there are pixels within a first preset range of the first positioning point; step 3: when there are pixels within the first preset range of the first positioning point, take the next positioning point connected to the first positioning point as the first positioning point, and return to step 2; otherwise, execute step 4; step 4: increment the counter variable by 1, and determine whether the value of the counter variable is greater than a counter variable threshold; step 5: when the value of the counter variable is determined not to be greater than the counter variable threshold, take the next positioning point connected to the first positioning point as the first positioning point, and return to step 3. Otherwise, proceed to step 6; in step 6, the first positioning point is determined as an incorrect positioning point, and it is determined whether the positioning points have been traversed. If yes, the set of incorrect positioning points is output. At this time, it is considered that all positioning points on the dashed lane line have been judged. If not, the positioning point after the first positioning point and spaced a preset number of positioning points away from the first positioning point is determined as the first positioning point, and the process returns to step 2.

8. An electronic device, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1-6.

9. A computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the method as described in any one of claims 1-6.

Images