Lane line denoising method and system, electronic equipment and storage medium

Abstract

The invention provides a lane line denoising method and system, electronic equipment and a storage medium, and the method comprises the steps: carrying out the fitting of discrete points of a target region, and determining a first lane line set, which comprises at least one first lane line, of the target region; according to at least one of a reference lane line distance, a reference discrete point distance and a reference discrete point angle of the first lane line, screening the first lane line in the first lane line set to obtain a second lane line set of the target area; the second lane line set comprises at least one second lane line, and the at least one second lane line comprises part or all of the at least one first lane line. According to the method, the first lane lines in the first lane line set are screened based on at least one of the reference lane line distance, the reference discrete point distance and the reference discrete point angle of the first lane lines, the lane lines with noise in the first lane line set can be effectively deleted, and the lane line precision of the target area is improved.

Description

Technical Field

[0001] The present application relates to the technical field of lane line detection, and in particular to a lane line denoising method, system, electronic device and storage medium. Background Art

[0002] Lane detection results are required in scenarios such as autonomous driving and building high-precision maps. The accuracy of lane detection results directly affects the safety of autonomous driving and the construction of high-precision maps. Lanes with low accuracy may cause errors in autonomous driving judgment and poor user experience when using high-precision maps.

[0003] Currently, after fitting the lane lines, most of the time they are directly applied to scenarios such as autonomous driving and building high-precision maps, without considering that the fitted lane lines may still contain noise, resulting in poor use of the lane lines.

[0004] How to denoise lane lines and improve lane line accuracy has become an urgent problem to be solved. Summary of the invention

[0005] The main purpose of the embodiments of the present application is to propose a lane line denoising method, system, electronic device and storage medium, aiming to effectively denoise the lane lines and improve the lane line accuracy in the target area.

[0006] The present application provides a lane line denoising method, comprising: fitting discrete points of a target area to determine a first lane line set of the target area, the first lane line set including at least one first lane line; screening the first lane lines in the first lane line set according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line to obtain a second lane line set of the target area; the second lane line set includes at least one second lane line, the at least one second lane line including part or all of the at least one first lane line; wherein the reference lane line distance is determined based on the distance between different first lane lines in a lane line combination, the lane line combination including different first lane lines in the first lane line set that have a common area in a first direction; the reference discrete point distance is determined based on the distance between adjacent discrete points of the first lane line in the second direction; the reference discrete point angle is determined based on the angle of the discrete points of the first lane line relative to the second direction; the first direction is the lane line direction of the target area, and the second direction is perpendicular to the lane line direction of the target area.

[0007] In one embodiment, the screening process of the first lane lines in the first lane line set according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line to obtain a second lane line set of the target area includes: screening the first lane lines in the first lane line set according to a set of angles corresponding to each of the first lane lines in the first lane line set to obtain a second lane line set of the target area; the set of angles corresponding to each of the first lane lines is obtained in the following manner: for each of the first lane lines in the first lane line set, determining the angles between the third direction of each discrete point of the first lane line and the second direction respectively; determining a set consisting of the angles between the third direction of each discrete point of the first lane line and the second direction as the set of angles corresponding to the first lane line; wherein the third direction of the discrete point is the direction of a line connecting the discrete point and the reference origin of the target area.

[0008] In one embodiment, the first lane lines in the first lane line set are screened based on the angle set corresponding to each of the first lane lines in the first lane line set, including: in response to the variance between the angles in the angle set corresponding to the first target lane line being greater than a first variance threshold, deleting the first target lane line from the first lane line set; or in response to the mean between the angles in the angle set corresponding to the first target lane line being outside a preset range, deleting the first target lane line from the first lane line set; wherein, the first target lane line is any one of the first lane lines in the first lane line set.

[0009] In one embodiment, the filtering and processing of the first lane lines in the first lane line set according to the angle set corresponding to each of the first lane lines in the first lane line set to obtain the second lane line set of the target area includes: filtering and processing the first lane lines in the first lane line set according to the first distance set corresponding to each of the lane line combinations to obtain a third lane line set of the target area; the third lane line set includes at least one third lane line; the at least one third lane line includes part or all of the at least one first lane line; filtering and processing the third lane lines in the third lane line set according to the angle set corresponding to each of the third lane lines in the third lane line set to obtain the second lane line set of the target area; or filtering and processing the first lane lines and / or discrete points of the first lane lines in the first lane line set according to the second distance set corresponding to each of the first lane lines in the first lane line set to obtain a fourth lane line set of the target area; the fourth lane line set includes at least one fourth lane line; the at least one fourth lane line includes part or all of the at least one a first lane line; according to the angle set corresponding to each of the fourth lane lines in the fourth lane line set, the fourth lane lines in the fourth lane line set are screened to obtain a second lane line set of the target area; or according to the first distance set corresponding to each of the lane line combinations, the first lane lines in the first lane line set are screened to obtain a third lane line set of the target area; the third lane line set includes at least one third lane line; the at least one third lane line includes part or all of the at least one first lane line; according to the second distance set corresponding to each of the third lane lines in the third lane line set, the third lane lines and / or discrete points of the third lane lines in the third lane line set are screened to obtain a fifth lane line set of the target area; the fifth lane line set includes at least one fifth lane line; the at least one fifth lane line includes part or all of the at least one third lane line; according to the angle set corresponding to each of the fifth lane lines in the fifth lane line set, the fifth lane lines in the fifth lane line set are screened to obtain a second lane line set of the target area.

[0010] In one embodiment, the screening process of the first lane line and / or the discrete points of the first lane line in the first lane line set according to the second distance set corresponding to each of the first lane lines in the first lane line set includes: in response to the variance between the distances in the second distance set corresponding to the second target lane line being greater than the second variance threshold, deleting the second target lane line from the first lane line set; or, in response to the variance between the distances in the second distance set corresponding to the second target lane line being less than or equal to the second variance threshold, and the target distance in the second distance set corresponding to the second target lane line being greater than or equal to the first distance threshold of the second target lane line, deleting the target discrete points from two adjacent discrete points corresponding to the target distance; wherein the second target lane line is any of the first lane lines in the first lane line set; the second distance set corresponding to the second target lane line is obtained in the following manner: sorting the discrete points of the second target lane line according to the distances between the discrete points of the second target lane line and the reference origin of the target area in the first direction to obtain a sorted discrete point set corresponding to the second target lane line; and determining the set of distances between every two adjacent discrete points in the sorted discrete point set of the second target lane line in the second direction as the second distance set corresponding to the second target lane line.

[0011] In one embodiment, the filtering and processing of the first lane lines in the first lane line set according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane lines in the first lane line set to obtain a second lane line set for the target area includes: filtering and processing the first lane lines in the first lane line set according to a second distance set corresponding to each first lane line in the first lane line set to obtain a second lane line set for the target area; the second distance set corresponding to each first lane line is obtained in the following manner: for each first lane line in the first lane line set, sorting the discrete points of the first lane line according to the distance between each discrete point of the first lane line and a reference origin of the target area in the first direction to obtain a sorted discrete point set corresponding to the first lane line; and determining a set of distances between every two adjacent discrete points in the sorted discrete point set of the first lane line as the second distance set corresponding to the first lane line.

[0012] In one embodiment, the first lane lines in the first lane line set are screened according to the second distance set corresponding to each of the first lane lines in the first lane line set to obtain the second lane line set of the target area, including: screening the first lane lines in the first lane line set according to the first distance set corresponding to each of the lane line combinations to obtain a third lane line set of the target area; the third lane line set includes at least one third lane line; the at least one third lane line includes part or all of the at least one first lane line; screening the third lane lines in the third lane line set according to the second distance set corresponding to each of the third lane lines in the third lane line set to obtain the second lane line set of the target area.

[0013] In one embodiment, the first lane lines in the first lane line set are screened based on the second distance set corresponding to each of the first lane lines in the first lane line set, including: in response to the variance between the distances in the second distance set corresponding to the second target lane line being greater than a second variance threshold, deleting the second target lane line from the first lane line set; wherein the second target lane line is any one of the first lane lines in the first lane line set.

[0014] In one embodiment, the first lane lines in the first lane line set are screened according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane lines in the first lane line set to obtain a second lane line set for the target area, including: screening the first lane lines in the first lane line set according to a first distance set corresponding to each lane line combination to obtain a second lane line set for the target area; the first distance set corresponding to each lane line combination is obtained in the following manner: for each lane line combination, determining a distance in the second direction of at least one feature point pair in the lane line combination; determining a set of distances of the at least one feature point pair in the second direction as the first distance set corresponding to the lane line combination; wherein each feature point pair in the lane line combination includes two feature points that are equal in distance from a reference origin of the target area in the first direction, and the two feature points are located on different first lane lines in the lane line combination.

[0015] In one embodiment, the first lane lines in the first lane line set are screened based on the first distance set corresponding to each of the lane line combinations, including: in response to the average value of the distances in the first distance set corresponding to the target lane line combination being less than a second distance threshold, deleting the first lane line with a short length in the target lane line combination from the first lane line set; wherein the target lane line combination is any of the lane line combinations.

[0016] The present application also provides a lane line denoising system, comprising a fitting module and a denoising module; the fitting module is used to fit discrete points of a target area to determine a first lane line set of the target area, the first lane line set including at least one first lane line; the denoising module is used to screen the first lane lines in the first lane line set according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line to obtain a second lane line set of the target area; the second lane line set includes at least one second lane line, the at least one second lane line includes part or all of the at least one first lane line; wherein the reference lane line distance is determined based on the distance between different first lane lines in a lane line combination, the lane line combination includes different first lane lines in the first lane line set that have a common area in a first direction; the reference discrete point distance is determined based on the distance between adjacent discrete points of the first lane line in the second direction; the reference discrete point angle is determined based on the angle of the discrete points of the first lane line relative to the second direction; the first direction is the lane line direction of the target area, and the second direction is perpendicular to the lane line direction of the target area.

[0017] The present application also provides an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned lane line denoising method when executing the computer program.

[0018] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the lane line denoising method described above is implemented.

[0019] The present application provides a lane line denoising method, system, electronic device and storage medium, which determines a first lane line set of the target area including at least one first lane line by fitting discrete points of the target area, and screens the first lane lines in the first lane line set according to at least one of the reference lane line distance, reference discrete point distance and reference discrete point angle of the first lane line, thereby effectively deleting noisy lane lines in the first lane line set and improving the lane line accuracy of the target area. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flowchart of a lane line denoising method provided in an embodiment of the present application;

[0021] Figure 2 This is a schematic diagram of the effect of the lane line denoising method provided in the embodiment of the present application. Figure 1 ;

[0022] Figure 3 This is a schematic diagram of the effect of the lane line denoising method provided in the embodiment of the present application. Figure 2 ;

[0023] Figure 4 This is a schematic diagram of the effect of the lane line denoising method provided in the embodiment of the present application. Figure 3 ;

[0024] Figure 5 This is a schematic diagram of the effect of the lane line denoising method provided in the embodiment of the present application. Figure 4 ;

[0025] Figure 6a This is a scene diagram of the lane line denoising method provided in the embodiment of the present application. Figure 1 ;

[0026] Figure 6b This is a scene diagram of the lane line denoising method provided in the embodiment of the present application. Figure 2 ;

[0027] Figure 7 is a schematic diagram of the structure of the lane line denoising system provided in an embodiment of the present application;

[0028] Figure 8 is a schematic diagram of the structure of an embodiment of an electronic device provided in an embodiment of the present application;

[0029] Fig. 9 It is a structural schematic diagram of another embodiment of the electronic device provided by the present application. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0031] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0032] The lane line denoising method provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the lane line denoising method, etc., but is not limited to the above forms.

[0033] The lane line denoising method provided in the embodiment of the present application is described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0034] See also Figure 1 , a lane line denoising method provided in an embodiment of the present application may include:

[0035] Step S101: fitting discrete points of a target area to determine a first lane line set of the target area, where the first lane line set includes at least one first lane line;

[0036] Step S102: Filter the first lane lines in the first lane line set according to at least one of the reference lane line distance, the reference discrete point distance, and the reference discrete point angle of the first lane line to obtain a second lane line set in the target area; the second lane line set includes at least one second lane line, and the at least one second lane line includes part or all of at least one first lane line.

[0037] Among them, the reference lane line distance is determined based on the distance between different first lane lines in the lane line combination, and the lane line combination includes different first lane lines in the first lane line set that have a common area in the first direction; the reference discrete point distance is determined based on the distance between adjacent discrete points of the first lane line in the second direction; the reference discrete point angle is determined based on the angle of the discrete points of the first lane line relative to the second direction; the first direction is the lane line direction of the target area, and the second direction is perpendicular to the lane line direction of the target area.

[0038] A lane line denoising method provided in an embodiment of the present application determines a first lane line set of the target area including at least one first lane line by fitting discrete points of the target area, and screens the first lane lines in the first lane line set based on at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line. The method can effectively delete lane lines with noise in the first lane line set and improve the lane line accuracy of the target area.

[0039] Optionally, the lane line directions of the target area are uniform and pre-set. Figure 2 As shown, an image coordinate system of the target area is established, and the y-axis direction of the image coordinate system can be determined as the lane line direction of the target area, that is, the first direction, and the x-axis direction of the image coordinate system can be determined as the second direction.

[0040] In one embodiment, in the above step S102, the first lane line in the first lane line set is screened according to at least one of the reference lane line distance, the reference discrete point distance, and the reference discrete point angle of the first lane line to obtain the second lane line set of the target area, including:

[0041] The first lane lines in the first lane line set are screened according to the angle set corresponding to each first lane line in the first lane line set to obtain a second lane line set in the target area.

[0042] Optionally, for each first lane line, it can be determined whether to delete it from the first lane line set based on the angle set corresponding to the first lane line. In the embodiment of the present application, after traversing the first lane line set, the processed first lane line set can be determined as the second lane line set. It can be understood that the second lane line set includes some or all of the first lane lines in the first lane line set.

[0043] The embodiment of the present application traverses each first lane line in the first lane line set and screens the first lane lines in the first lane line set based on the angle set corresponding to each first lane line, thereby effectively deleting noisy lane lines in the first lane line set and improving the lane line accuracy of the target area.

[0044] In one implementation, the set of angles corresponding to each of the first lane lines is obtained by:

[0045] For each first lane line in the first lane line set, determine the angles between the third direction of each discrete point of the first lane line and the second direction;

[0046] Determine a set of angles between the third direction of each discrete point of the first lane line and the second direction as the angle set corresponding to the first lane line;

[0047] The third direction of the discrete point is the direction of the line connecting the discrete point and the reference origin of the target area.

[0048] Continue reading Figure 2 , the coordinate origin o of the target area image coordinate system can be determined as the reference origin of the target area, and the direction from the reference origin of the target area to the discrete point of the first lane line can be determined as the direction of the line connecting the discrete point and the reference origin of the target area, that is, the third direction of the discrete point. Taking lane line L11 as an example, the third direction of discrete point a of lane line L11 is the direction from the coordinate origin o to discrete point a Its angle with the second direction is equal to the direction The angle α with the positive direction of the x-axis; Taking lane line L12 as an example, the third direction of the discrete point b of lane line L12 is the direction from the coordinate origin o to the discrete point b Its angle with the second direction is equal to the direction The angle β with the positive x-axis.

[0049] It is worth mentioning that the above is only an example. The present application does not limit the method for determining the third direction of the discrete point and the angle between the third direction of the discrete point and the second direction. For each first lane line in the first lane line set, it is sufficient to ensure that the method for determining the third direction of the discrete point and the angle between the third direction of the discrete point and the second direction remain consistent.

[0050] The embodiment of the present application determines a set of angles between the third direction of each discrete point of the first lane line and the second direction as the angle set corresponding to the first lane line, and screens the first lane lines in the first lane line set based on the angle set corresponding to each first lane line in the first lane line set, thereby accurately deleting noisy lane lines in the first lane line set and improving the lane line accuracy of the target area.

[0051] In one embodiment, the screening process of the first lane lines in the first lane line set according to the angle set corresponding to each first lane line in the first lane line set includes:

[0052] In response to the variance between the angles in the angle set corresponding to the first target lane line being greater than a first variance threshold, deleting the first target lane line from the first lane line set; or

[0053] In response to a mean value between angles in the angle set corresponding to the first target lane line being outside a preset range, deleting the first target lane line from the first lane line set;

[0054] The first target lane line is any first lane line in the first lane line set.

[0055] In one embodiment, the relationship between the variance between the angles in the angle set corresponding to the first target lane line and the first variance threshold can be first determined. If the variance between the angles in the angle set corresponding to the first target lane line is greater than the first variance threshold, the first target lane line is deleted from the first target lane line set. If the variance between the angles in the angle set corresponding to the first target lane line is less than or equal to the first variance threshold, it is further determined whether the mean between the angles in the angle set corresponding to the first target lane line is within a preset range. If the mean between the angles in the angle set corresponding to the first target lane line is outside the preset range, the first target lane line is deleted from the first target lane line set. If the mean between the angles in the angle set corresponding to the first target lane line is within the preset range, the first target lane line is retained in the first target lane line set. Optionally, the first variance threshold is 10 degrees, and the preset range is 20 degrees ≤ mean between angles ≤ 160 degrees.

[0056] Continue to refer Figure 2 The first lane line set of the target area includes four first lane lines, namely lane line L11, lane line L12, lane line L13, and lane line L14. In response to the variance between the angles in the angle sets corresponding to lane line L11 and lane line L14 being greater than 10 degrees, lane line L11 and lane line L14 are deleted from the first lane line set to obtain a second lane line set, which includes lane line L12 and lane line L13.

[0057] The embodiments of the present application can accurately delete the first target lane line that has noise in the first lane line set of the target area, thereby improving the lane line accuracy of the target area, by deleting the first target lane line from the first lane line set in response to the variance between the angles in the angle set corresponding to the first target lane line being greater than a first variance threshold, or by deleting the first target lane line from the first lane line set in response to the mean between the angles in the angle set corresponding to the first target lane line being outside a preset range.

[0058] In one embodiment, the first lane lines in the first lane line set are screened according to the angle set corresponding to each first lane line in the first lane line set to obtain the second lane line set in the target area, including:

[0059] According to the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set are screened to obtain a third lane line set in the target area; the third lane line set includes at least one third lane line; the at least one third lane line includes part or all of at least one first lane line; according to the angle set corresponding to each third lane line in the third lane line set, the third lane lines in the third lane line set are screened to obtain a second lane line set in the target area; or

[0060] The first lane lines and / or discrete points of the first lane lines in the first lane line set are screened according to the second distance set corresponding to each first lane line in the first lane line set to obtain a fourth lane line set in the target area; the fourth lane line set includes at least one fourth lane line; the at least one fourth lane line includes part or all of at least one first lane line; the fourth lane lines in the fourth lane line set are screened according to the angle set corresponding to each fourth lane line in the fourth lane line set to obtain a second lane line set in the target area; or

[0061] According to the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set are screened to obtain a third lane line set in the target area; the third lane line set includes at least one third lane line; at least one third lane line includes part or all of at least one first lane line; according to the second distance set corresponding to each third lane line in the third lane line set, the third lane lines and / or discrete points of the third lane lines in the third lane line set are screened to obtain a fifth lane line set in the target area; the fifth lane line set includes at least one fifth lane line; at least one fifth lane line includes part or all of at least one third lane line; according to the angle set corresponding to each fifth lane line in the fifth lane line set, the fifth lane lines in the fifth lane line set are screened to obtain a second lane line set in the target area.

[0062] The lane line denoising method provided in the embodiment of the present application can, on the one hand, firstly perform screening processing on the first lane lines in the first lane line set based on the first distance set corresponding to each lane line combination to obtain the third lane line set in the target area, and then perform screening processing on the third lane lines in the third lane line set based on the angle set corresponding to each third lane line in the third lane line set, so as to effectively delete the repeated first lane lines in the first lane line set, thereby improving the efficiency and effect of screening processing on the first lane lines in the first lane line set based on the angle set corresponding to the first lane lines in the first lane line set, and improving the lane line accuracy in the target area;

[0063] On the other hand, based on the second distance set corresponding to each first lane line in the first lane line set, the first lane lines and / or discrete points of the first lane lines in the first lane line set can be screened to obtain a fourth lane line set in the target area, and then the fourth lane lines in the fourth lane line set can be screened according to the angle set corresponding to each fourth lane line in the fourth lane line set, so that the first lane lines and / or abnormal discrete points with abnormal discrete point distances in the first lane line set can be effectively deleted, thereby improving the efficiency and effect of screening the first lane lines in the first lane line set based on the angle set corresponding to the first lane lines in the first lane line set, and improving the lane line accuracy in the target area;

[0064] On the other hand, based on the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set can be screened to obtain the third lane line set in the target area, and then the third lane lines and / or discrete points of the third lane lines in the third lane line set can be screened according to the second distance set corresponding to each third lane line in the third lane line set to obtain the fifth lane line set in the target area. Finally, according to the angle set corresponding to each fifth lane line in the fifth lane line set, the fifth lane lines in the fifth lane line set can be screened, which can effectively delete the repeated first lane lines in the first lane line set, as well as the first lane lines and / or abnormal discrete points with abnormal discrete point distances, thereby improving the efficiency and effect of screening the first lane lines in the first lane line set based on the angle set corresponding to the first lane lines in the first lane line set, and improving the lane line accuracy of the target area.

[0065] In one embodiment, the screening process of the first lane lines and / or the discrete points of the first lane lines in the first lane line set according to the second distance set corresponding to each first lane line in the first lane line set includes:

[0066] In response to the variance between the distances in the second distance set corresponding to the second target lane line being greater than a second variance threshold, deleting the second target lane line from the first lane line set; or,

[0067] In response to the variance between the distances in the second distance set corresponding to the second target lane line being less than or equal to the second variance threshold, and the target distance in the second distance set corresponding to the second target lane line being greater than or equal to the first distance threshold of the second target lane line, the target discrete point among the two adjacent discrete points corresponding to the target distance is deleted.

[0068] The second target lane line is any first lane line in the first lane line set.

[0069] Optionally, the second distance set corresponding to the second target lane line is obtained in the following manner:

[0070] Sort the discrete points of the second target lane line according to the distance between the discrete points of the second target lane line and the reference origin of the target area in the first direction to obtain a sorted set of discrete points corresponding to the second target lane line;

[0071] A set of distances between every two adjacent discrete points in the sorted set of discrete points of the second target lane line in the second direction is determined as a second distance set corresponding to the second target lane line.

[0072] Optionally, the two adjacent discrete points are two adjacent discrete points whose sorting values ​​differ by 1 in the sorted set of discrete points corresponding to the first lane line.

[0073] like Figure 3 As shown, the first direction can be set as the y-axis direction of the target area image coordinate system, the second direction can be set as the x-axis direction of the target area image coordinate system, and the first lane line set of the target area includes 5 first lane lines, namely lane line L21, lane line L22, lane line L23, lane line L24, and lane line L25.

[0074] Among them, the discrete points of lane line L24 are discrete points 1 to discrete points 5. The discrete points 1 to discrete points 5 of lane line L24 can be sorted according to the order of the distance between discrete points 1 to discrete points 5 and the coordinate origin o of the target area image coordinate system in the y-axis direction of the target area image coordinate system from large to small, that is, the order of the vertical coordinates of discrete points 1 to discrete points 5 in the target area image coordinate system from large to small, and the discrete point sorting set of lane line L24 is obtained as (discrete point 1, discrete point 5, discrete point 4, discrete point 2, discrete point 3), among which the sorting value of discrete point 1 is the smallest and the sorting value of discrete point 3 is the largest.

[0075] Furthermore, according to the distance between every two adjacent discrete points in the discrete point sorting set of lane line L24 in the x-axis direction of the target area image coordinate system, that is, the absolute value of the difference between the horizontal coordinates of every two adjacent discrete points in the target area image coordinate system, it is obtained that the second distance set corresponding to lane line L24 includes 5 distances, namely, distance 1 between discrete point 1 and discrete point 5, distance 2 between discrete point 5 and discrete point 4, distance 3 between discrete point 4 and discrete point 2, and distance 4 between discrete point 2 and discrete point 3.

[0076] Optionally, the second variance threshold is 30 pixels. Figure 3 In response to the variance between distance 1 and distance 4 in the second distance set corresponding to lane line L24 being greater than 30 pixels, lane line L24 may be deleted from the first lane line set. In addition, while lane line L24 is deleted from the first lane line set, discrete points 1 to 4 of lane line L24 may also be deleted.

[0077] Optionally, the first distance threshold of the second target lane line is determined by the following formula:

[0078] well_dis=median(X)+2*mad+5

[0079] mad=median(|X i -median(X)|)

[0080] Wherein, well_dis represents the first distance threshold of the second target lane line, median(X) represents the median of the distance in the first distance set corresponding to the second target lane line, mad represents the median absolute deviation of the distance in the first distance set corresponding to the second target lane line, and X i Represents the distance i in the first distance set corresponding to the second target lane line.

[0081] Optionally, if the number of distances in the second distance set corresponding to the second target lane line is an even number, the average of the two distances whose distance values ​​are located in the middle of the second distance set is selected as the median. For example, if the number of distances in the second distance set corresponding to lane line L23 is 10, and the distance values ​​of distance 5 and distance 6 are located in the middle of the second distance set corresponding to lane line L23, the average of distance 5 and distance 6 can be selected as the median of the distances in the second distance set corresponding to lane line L23.

[0082] Continue to refer Figure 3, the discrete points of lane line L21 are discrete point 6 to discrete point 10. According to the order of the distances between discrete point 6 to discrete point 10 on lane line L21 and the coordinate origin o of the target area image coordinate system in the y-axis direction of the target area image coordinate system from large to small, that is, the order of the ordinates of discrete point 6 to discrete point 10 in the target area image coordinate system from large to small, the discrete points 6 to discrete point 10 in the discrete point set of lane line L21 can be sorted, and the discrete point sorting set of lane line L21 is obtained as (discrete point 9, discrete point 10, discrete point 6, discrete point 8, discrete point 7), among which the sorting value of discrete point 9 is the smallest and the sorting value of discrete point 7 is the largest.

[0083] Furthermore, according to the distance between every two adjacent discrete points in the discrete point sorting set of lane line L21 in the x-axis direction of the target area image coordinate system, that is, the absolute value of the difference between the horizontal coordinates of every two adjacent discrete points in the target area image coordinate system, a second distance set corresponding to each lane line L21 can be obtained, including the distance 5 between discrete point 9 and discrete point 10, the distance 6 between discrete point 10 and discrete point 6, the distance 7 between discrete point 6 and discrete point 8, and the distance 8 between discrete point 8 and discrete point 7.

[0084] Further, in response to the variance of distance 5 to distance 8 in the second distance set corresponding to lane line L21 being less than or equal to 30 pixels, and distance 5 and distance 7 being greater than the first distance threshold of lane line L21, discrete point 9 corresponding to distance 5 and discrete point 10 with a larger ranking value, as well as discrete point 6 and discrete point 8 with a larger ranking value, corresponding to distance 7, are taken as target discrete points, and discrete point 8 and discrete point 10 are deleted.

[0085] It is worth mentioning that the above are only examples, and this application does not limit the selection rules of target discrete points.

[0086] In one embodiment, the screening process of the third lane lines and / or the discrete points of the third lane lines in the third lane line set according to the second distance set corresponding to each third lane line in the third lane line set includes:

[0087] In response to the variance between the distances in the second distance set corresponding to the third target lane line being greater than the second variance threshold, deleting the third target lane line from the third lane line set; or,

[0088] In response to the variance between the distances in the second distance set corresponding to the third target lane line being less than or equal to the second variance threshold, and the target distance in the second distance set corresponding to the third target lane line being greater than or equal to the first distance threshold of the second target lane line, the target discrete point among two adjacent discrete points of the third target lane line corresponding to the target distance is deleted.

[0089] Among them, the third target lane line is any third lane line in the third lane line set, and the method for obtaining the second distance set corresponding to the third target lane line is the same as the method for obtaining the second distance set corresponding to the first target lane line, which will not be repeated here.

[0090] The embodiment of the present application can accurately determine every two adjacent discrete points of the target lane line and the second distance set corresponding to the target lane line based on the distance between each discrete point of the target lane line and the reference origin of the target area in the first direction, and further delete the target lane line from the target lane line set in response to the variance between the distances in the second distance set corresponding to the target lane line being greater than the second variance threshold; or, in response to the variance between the distances in the second distance set corresponding to the target lane line being less than or equal to the second variance threshold, and the target distance in the second distance set corresponding to the target lane line being greater than or equal to the first distance threshold of the target lane line, delete the target discrete point in the two adjacent discrete points corresponding to the target distance, which can ensure the accuracy of deleting the target lane line with abnormal discrete point distance in the target lane line set and / or the abnormal discrete point of the target lane line, thereby improving the accuracy of screening the first lane line in the first lane line set based on the angle set corresponding to the first lane line in the first lane line set, and improving the lane line accuracy of the target area. Wherein, the target lane line refers to the second target lane line or the third target lane line, and the target lane line set refers to the first lane line set or the third lane line set.

[0091] In one embodiment, in the above step S102, the first lane line in the first lane line set is screened according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line in the first lane line set to obtain a second lane line set in the target area, including:

[0092] The first lane lines in the first lane line set are screened according to the second distance set corresponding to each first lane line in the first lane line set to obtain a second lane line set in the target area.

[0093] Optionally, for each first lane line, it can be determined whether to delete it from the first lane line set based on the second distance set corresponding to the first lane line. In the embodiment of the present application, after traversing the first lane line set, the processed first lane line set can be determined as the second lane line set. It can be understood that the second lane line set includes some or all of the first lane lines in the first lane line set.

[0094] The embodiment of the present application traverses each first lane line in the first lane line set, and screens the first lane lines in the first lane line set based on the second distance set corresponding to each first lane line in the first lane line set, thereby effectively deleting noisy lane lines in the first lane line set and improving the lane line accuracy of the target area.

[0095] Optionally, the second distance set corresponding to each first lane line is obtained in the following manner:

[0096] For each first lane line in the first lane line set, sorting the discrete points of the first lane line according to the distance between the discrete points of the first lane line and the reference origin of the target area in the first direction to obtain a sorted set of discrete points corresponding to the first lane line;

[0097] A set of distances between every two adjacent discrete points in the sorted set of discrete points of the first lane line in the second direction is determined as a second distance set corresponding to the first lane line.

[0098] The embodiment of the present application determines every two adjacent discrete points of the first lane line based on the distance between each discrete point of the first lane line and the reference origin of the target area in the first direction, and obtains a second distance set corresponding to the first lane line based on the distance between every two adjacent discrete points of the first lane line in the second direction. This can improve the accuracy of screening the first lane lines in the first lane line set based on the second distance set corresponding to each first lane line in the first lane line set, and improve the lane line accuracy of the target area.

[0099] In one embodiment, the first lane lines in the first lane line set are screened according to the second distance set corresponding to each first lane line in the first lane line set to obtain the second lane line set of the target area, including:

[0100] According to the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set are screened to obtain a third lane line set in the target area; the third lane line set includes at least one third lane line; and at least one third lane line includes part or all of at least one first lane line;

[0101] The third lane lines in the third lane line set are screened according to the second distance set corresponding to each third lane line in the third lane line set to obtain a second lane line set in the target area.

[0102] The lane line denoising method provided in the embodiment of the present application can first screen the first lane lines in the first lane line set based on the first distance set corresponding to each lane line combination to obtain a third lane line set in the target area, and then screen the third lane lines in the third lane line set according to the second distance set corresponding to each third lane line in the third lane line set, which can effectively delete the repeated first lane lines in the first lane line set, thereby improving the efficiency and effect of screening the first lane lines in the first lane line set based on the second distance set corresponding to each first lane line in the first lane line set, and improving the lane line accuracy of the target area.

[0103] In one embodiment, the screening process of the first lane lines in the first lane line set according to the second distance set corresponding to each first lane line in the first lane line set includes:

[0104] In response to a variance between distances in a second distance set corresponding to the second target lane line being greater than a second variance threshold, deleting the second target lane line from the first lane line set;

[0105] The second target lane line is any first lane line in the first lane line set.

[0106] Optionally, the second variance threshold is 30 pixels. Figure 3 In response to the variance between distance 1 to distance 4 in the second distance set corresponding to lane line L24 being greater than 30 pixels, lane line L24 may be deleted from the first lane line set.

[0107] The embodiment of the present application deletes the second target lane line from the first lane line set in response to the variance between the distances in the second distance set corresponding to the second target lane line being greater than the second variance threshold. This can accurately delete the second target lane line that has noise in the first lane line set, thereby improving the lane line accuracy of the target area.

[0108] In one embodiment, in the above step S102, the first lane line in the first lane line set is screened according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line in the first lane line set to obtain a second lane line set in the target area, including:

[0109] According to the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set are screened to obtain a second lane line set in the target area.

[0110] The lane line combination includes different first lane lines in the first lane line set that have a common area in the first direction.

[0111] Optionally, for each lane line combination, it can be determined whether there are repeated first lane lines in the lane line combination based on the first distance set corresponding to the lane line combination, and the repeated first lane lines can be deleted from the first lane line set. In the embodiment of the present application, after traversing each lane line combination, the processed first lane line set can be determined as the second lane line set. It can be understood that the second lane line set includes some or all of the first lane lines in the first lane line set.

[0112] The embodiment of the present application traverses each lane line combination and screens the first lane lines in the first lane line set based on the first distance set corresponding to each lane line combination, thereby effectively deleting the noisy lane lines in the first lane line set and improving the lane line accuracy of the target area.

[0113] Optionally, the first distance set corresponding to each lane line combination is obtained in the following manner:

[0114] For each lane line combination, determining a distance in a second direction between at least one feature point pair in the lane line combination;

[0115] Determine a set of distances of at least one feature point pair in the second direction as a first distance set corresponding to the lane line combination;

[0116] Each feature point pair in the lane line combination includes two feature points that are equidistant from a reference origin of the target area in a first direction, and the two feature points are located on different first lane lines in the lane line combination.

[0117] like Figure 4 As shown, the reference origin of the target area can be set as the coordinate origin o of the target area image coordinate system, the first direction is the y-axis direction of the target area image coordinate system, and the second direction is the x-axis direction of the target area image coordinate system. Lane line L31 ( Figure 4Not drawn), lane line L32 is two different first lane lines in the first lane line set that have a common area S in the y-axis direction of the target area image coordinate system, that is, lane line L31 and lane line L32 are a lane line combination. Among them, the common area S can be determined in the following manner: according to the distance between each discrete point of lane line L31 and the coordinate origin o of the target area image coordinate system in the y-axis direction of the target area image coordinate system, the lane line area A of lane line L31 in the y-axis direction of the target area image coordinate system is determined, and according to the distance between each discrete point of lane line L32 and the coordinate origin o of the target area image coordinate system in the y-axis direction of the target area image coordinate system, the lane line area B of lane line L32 in the y-axis direction of the target area image coordinate system is determined. Further, the overlapping area of ​​lane line area A and lane line area B in the y-axis direction of the target area image coordinate system can be determined as the common area S of lane line L31 and lane line L32 in the y-axis direction of the target area image coordinate system.

[0118] Further, for the lane line combination of lane line L31 and lane line L32, feature point 1, feature point 2, and feature point 3 can be selected at the position where lane line L31 is located in the public area S according to the same preset interval along the positive direction of the y-axis of the target area image coordinate system, and feature point 4, feature point 5, and feature point 6 can be selected correspondingly at the position where lane line L32 is located in the public area S. In the y-axis direction of the target area image coordinate system, the distances between feature point 1 and feature point 4 and the coordinate origin o of the target area image coordinate system are equal, the distances between feature point 2 and feature point 5 and the coordinate origin o of the target area image coordinate system are equal, and the distances between feature point 3 and feature point 6 and the coordinate origin o of the target area image coordinate system are equal. Feature point 1 and feature point 4 can be regarded as a feature point pair, feature point 2 and feature point 5 can be regarded as a feature point pair, and feature point 3 and feature point 6 can be regarded as a feature point pair, respectively, to obtain three feature point pairs of the lane line combination of lane line L31 and lane line L32: feature point 1 of lane line L31 and feature point 4 of lane line L32, feature point 2 of lane line L31 and feature point 5 of lane line L32, and feature point 3 of lane line L31 and feature point 6 of lane line L32.

[0119] Further, the first distance set corresponding to the lane line combination of lane line L31 and lane line L32 includes the distance 9 between feature point 1 of lane line L31 and feature point 4 of lane line L32 in the x-axis direction of the target area image coordinate system, the distance 10 between feature point 2 of lane line L31 and feature point 5 of lane line L32 in the x-axis direction of the target area image coordinate system, and the distance 11 between feature point 3 of lane line L31 and feature point 6 of lane line L32 in the x-axis direction of the target area image coordinate system.

[0120] In the embodiment of the present application, for each lane line combination, a first distance set corresponding to the lane line combination is determined by the distance in the second direction of at least one feature point pair in the lane line combination, and based on the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set are screened and processed, so that duplicate lane lines in the first lane line set can be effectively deleted, the efficiency and effect of lane line denoising can be improved, and the lane line accuracy in the target area can be improved.

[0121] In one embodiment, the screening process of the first lane line in the first lane line set according to the first distance set corresponding to each lane line combination includes:

[0122] In response to an average value of distances in a first distance set corresponding to the target lane line combination being less than a second distance threshold, deleting a first lane line with a short length in the target lane line combination from the first lane line set;

[0123] The target lane line combination is any lane line combination.

[0124] Optionally, the second distance threshold is 30 pixels. Figure 4 In response to the average value of the distances in the first distance set corresponding to the lane line combination of lane line L31 and lane line L32 being less than 30 pixels, lane line L31 with a shorter length between lane line L31 and lane line L32 may be deleted from the first lane line set.

[0125] The embodiment of the present application deletes the first lane line with a short length in the target lane line combination from the first lane line set in response to the average value of the distances in the first distance set corresponding to the target lane line combination being less than the second distance threshold, thereby ensuring the accuracy of deleting duplicate lane lines in the first lane line set, improving the efficiency and effect of lane line denoising, and improving the lane line accuracy in the target area.

[0126] like Figure 5 As shown, in a specific embodiment, first, the discrete points identified in the target area can be fitted to obtain a first lane line set of the target area, and the first lane line set includes 6 first lane lines, namely lane line L41 ( Figure 5 Not shown), lane line L42 ( Figure 5 Not shown), lane line L43, lane line L44, lane line L45 ( Figure 5 Not drawn), lane line L46 ( Figure 5 not shown).

[0127] Secondly, among the six first lane lines in the first lane line set, lane line L42 and lane line L43 have a common area in the y-axis direction of the target area image coordinate system, and the two are a lane line combination. Then, according to the distance of at least one feature point pair of the lane line combination of lane line L42 and lane line L43 in the x-axis direction of the target area image coordinate system, the first distance set corresponding to the lane line combination of lane line L42 and lane line L43 is obtained. Finally, in response to the average value of the distances in the first distance set corresponding to the lane line combination of lane line L42 and lane line L43 being less than 30 pixels, the shorter lane line L42 of lane line L42 and lane line L43 is deleted to obtain the third lane line set of the target area, which includes five third lane lines, namely lane line L41, lane line L43, lane line L44, lane line L45, and lane line L46.

[0128] Then, for each third lane line in the third lane line set, the discrete points of the third lane line are sorted according to the distance between each discrete point in the discrete point set of the third lane line and the coordinate origin o of the target area image coordinate system in the y-axis direction of the target area image coordinate system, to obtain a discrete point sorting set of the third lane line. Then, according to the distance between every two adjacent discrete points in the discrete point sorting set of the third lane line in the x-axis direction of the target area image coordinate system, a second distance set corresponding to each third lane line is obtained. Finally, in response to the variance between distances in the second distance set corresponding to lane line L41 being less than or equal to 30 pixels, but the first target distance in the second distance set corresponding to lane line L41 being greater than the first distance threshold of lane line L41, the target discrete points in the two adjacent discrete points corresponding to the first target distance of lane line L41 are deleted; in response to the variance between distances in the second distance set corresponding to lane line L45 being greater than 30 pixels, lane line L45 and all discrete points of lane line L45 are deleted; in response to the variance between distances in the second distance set corresponding to lane line L46 being less than or equal to 30 pixels, but the second target distance in the second distance set corresponding to lane line L46 being greater than the first distance threshold of lane line L46, the target discrete points in the two adjacent discrete points corresponding to the second target distance of lane line L46 are deleted, and a fifth lane line set of the target area is obtained, and the fifth lane line set includes four fifth lane lines, namely lane line L41, lane line L43, lane line L44, and lane line L46.

[0129] Next, for each fifth lane line in the fifth lane line set, determine the directions of the coordinate origin o of the target area image coordinate system pointing to the remaining discrete points of the fifth lane line, and the angles with the positive direction of the x-axis of the target area image coordinate system, respectively, to obtain the angle set corresponding to each fifth lane line. In response to the variance between the angles in the angle set corresponding to lane line L41 being greater than 10 degrees, lane line L41 is deleted from the fifth lane line set, and in response to the mean value between the angles in the angle set corresponding to lane line L46 being greater than 160 degrees, or the mean value being less than 20 degrees, lane line L46 is deleted from the fifth lane line set, and finally the second lane line set of the target area is determined, and the second lane line set includes two first target lane lines, namely lane line L43 and lane line L44.

[0130] Take the intersection lane line detection scenario as an example. Figure 6a As shown in Figure 2, the discrete points identified in the intersection area are fitted, and the resulting lane lines contain a lot of noise, such as Figure 6b As shown, using the lane line denoising method provided in the present application, the noise of the lane lines obtained by fitting the discrete points identified in the intersection area is effectively deleted, and two high-precision lane lines in the intersection area are obtained.

[0131] The lane line denoising method provided in the embodiment of the present application screens the first lane lines in the first lane line set based on the first distance set corresponding to each lane line combination in the first lane line set, and can effectively delete repeated first lane lines in the first lane line set; screens the first lane lines and / or discrete points of the first lane lines in the first lane line set based on the second distance set corresponding to each first lane line in the first lane line set, and can effectively delete first lane lines and / or abnormal discrete points with abnormal discrete point distances in the first lane line set; screens the first lane lines in the first lane line set based on the angle set corresponding to each first lane line in the first lane line set, and can effectively delete first lane lines whose relationship between the angle between the third direction of each discrete point and the second direction does not meet the requirements; the above three denoising methods are used alone to effectively delete the lane lines with noise in the first lane line set and improve the lane line accuracy of the target area; the above three denoising methods are used in combination to improve the lane line denoising efficiency and denoising effect, and further improve the lane line accuracy of the target area.

[0132] See also Figure 7 The embodiment of the present application also provides a lane line denoising system 700, which can implement the above lane line denoising method. The system includes: a fitting module 701 and a denoising module 702.

[0133] The fitting module 701 is used to fit the discrete points of the target area to determine a first lane line set of the target area, where the first lane line set includes at least one first lane line;

[0134] The denoising module 702 is used to filter the first lane lines in the first lane line set according to at least one of the reference lane line distance, the reference discrete point distance, and the reference discrete point angle of the first lane line to obtain a second lane line set in the target area; the second lane line set includes at least one second lane line, and the at least one second lane line includes part or all of the at least one first lane line;

[0135] Among them, the reference lane line distance is determined based on the distance between different first lane lines in the lane line combination, and the lane line combination includes different first lane lines in the first lane line set that have a common area in the first direction; the reference discrete point distance is determined based on the distance between adjacent discrete points of the first lane line in the second direction; the reference discrete point angle is determined based on the angle of the discrete points of the first lane line relative to the second direction; the first direction is the lane line direction of the target area, and the second direction is perpendicular to the lane line direction of the target area.

[0136] The lane line denoising system provided in the embodiment of the present application can implement each step of the above-mentioned lane line denoising method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0137] Alternatively, if Figure 8 As shown, the embodiment of the present application further provides an electronic device 800, including a processor 801 and a memory 802, and the memory 802 stores a program or instruction that can be run on the processor 801, and when the program or instruction is executed by the processor 801, each step of the above lane line denoising method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. It should be noted that the electronic device in the embodiment of the present application includes the above-mentioned mobile electronic device and non-mobile electronic device.

[0138] Fig. 9 A schematic diagram of the hardware structure of an electronic device for implementing an embodiment of the present application is provided, wherein the electronic device includes:

[0139] The processor 901 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0140] The memory 902 may be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 may store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 902, and the processor 901 calls and executes the lane line denoising method of the embodiment of this application;

[0141] Input / output interface 903, used to implement information input and output;

[0142] Communication interface 904, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0143] A bus 905 that transmits information between the various components of the device (e.g., the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);

[0144] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0145] The electronic device provided in the embodiment of the present application can implement each step of the above-mentioned lane line denoising method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0146] An embodiment of the present application also provides a computer-readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various steps of the above-mentioned lane line denoising method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0147] The processor is the processor in the electronic device described in the above embodiment. The computer-readable storage medium includes a computer-readable storage medium such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0148] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the various steps of the above-mentioned lane line denoising method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0149] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0150] An embodiment of the present application provides a computer program product, which is stored in a storage medium. The program product is executed by at least one processor to implement the various steps of the lane line denoising method embodiment as described above, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0151] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not delete the existence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0152] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0153] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A lane line denoising method, characterized in that: include: Fitting discrete points of a target area to determine a first lane line set of the target area, wherein the first lane line set includes at least one first lane line; According to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line, the first lane line in the first lane line set is screened to obtain a second lane line set of the target area; the second lane line set includes at least one second lane line, and the at least one second lane line includes part or all of the at least one first lane line; Among them, the reference lane line distance is determined based on the distance between different first lane lines in the lane line combination, and the lane line combination includes different first lane lines in the first lane line set that have a common area in the first direction; the reference discrete point distance is determined based on the distance between adjacent discrete points of the first lane line in the second direction; the reference discrete point angle is determined based on the angle of the discrete points of the first lane line relative to the second direction; the first direction is the lane line direction of the target area, and the second direction is perpendicular to the lane line direction of the target area.

2. The lane line denoising method according to claim 1, characterized in that: The filtering of the first lane line in the first lane line set according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line to obtain a second lane line set in the target area includes: Filtering the first lane lines in the first lane line set according to the angle set corresponding to each first lane line in the first lane line set to obtain a second lane line set in the target area; The set of angles corresponding to each of the first lane lines is obtained in the following manner: For each of the first lane lines in the first lane line set, determining an angle between the third direction of each discrete point of the first lane line and the second direction; Determine a set of angles between the third direction of each discrete point of the first lane line and the second direction as a set of angles corresponding to the first lane line; The third direction of the discrete point is the direction of a line connecting the discrete point and a reference origin of the target area.

3. The lane line denoising method according to claim 2, characterized in that: The screening process of the first lane lines in the first lane line set according to the angle set corresponding to each first lane line in the first lane line set includes: In response to a variance between angles in the angle set corresponding to the first target lane line being greater than a first variance threshold, deleting the first target lane line from the first lane line set; or In response to a mean value between angles in the angle set corresponding to the first target lane line being outside a preset range, deleting the first target lane line from the first lane line set; The first target lane line is any of the first lane lines in the first lane line set.

4. The lane line denoising method according to claim 2, characterized in that: The filtering of the first lane lines in the first lane line set according to the angle set corresponding to each first lane line in the first lane line set to obtain the second lane line set in the target area includes: According to the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set are screened to obtain a third lane line set of the target area; the third lane line set includes at least one third lane line; the at least one third lane line includes part or all of the at least one first lane line; according to the angle set corresponding to each third lane line in the third lane line set, the third lane lines in the third lane line set are screened to obtain a second lane line set of the target area; or The first lane lines and / or discrete points of the first lane lines in the first lane line set are screened according to the second distance set corresponding to each of the first lane lines in the first lane line set to obtain a fourth lane line set for the target area; the fourth lane line set includes at least one fourth lane line; the at least one fourth lane line includes part or all of the at least one first lane line; the fourth lane lines in the fourth lane line set are screened according to the angle set corresponding to each of the fourth lane lines in the fourth lane line set to obtain a second lane line set for the target area; or According to the first distance set corresponding to each of the lane line combinations, the first lane lines in the first lane line set are screened to obtain a third lane line set for the target area; the third lane line set includes at least one third lane line; the at least one third lane line includes part or all of the at least one first lane line; according to the second distance set corresponding to each of the third lane lines in the third lane line set, the third lane lines and / or discrete points of the third lane lines in the third lane line set are screened to obtain a fifth lane line set for the target area; the fifth lane line set includes at least one fifth lane line; the at least one fifth lane line includes part or all of the at least one third lane line; according to the angle set corresponding to each of the fifth lane lines in the fifth lane line set, the fifth lane lines in the fifth lane line set are screened to obtain a second lane line set for the target area.

5. The lane line denoising method according to claim 4, characterized in that: The screening process of the first lane lines and / or the discrete points of the first lane lines in the first lane line set according to the second distance set corresponding to each first lane line in the first lane line set includes: In response to the variance between the distances in the second distance set corresponding to the second target lane line being greater than a second variance threshold, deleting the second target lane line from the first lane line set; or, In response to the variance between the distances in the second distance set corresponding to the second target lane line being less than or equal to the second variance threshold, and the target distance in the second distance set corresponding to the second target lane line being greater than or equal to the first distance threshold of the second target lane line, deleting the target discrete point from two adjacent discrete points corresponding to the target distance; The second target lane line is any of the first lane lines in the first lane line set; the second distance set corresponding to the second target lane line is obtained in the following manner: Sort the discrete points of the second target lane line according to the distance between the discrete points of the second target lane line and the reference origin of the target area in the first direction to obtain a sorted set of discrete points corresponding to the second target lane line; A set of distances between every two adjacent discrete points in the sorted set of discrete points of the second target lane line in the second direction is determined as a second distance set corresponding to the second target lane line.

6. The lane line denoising method according to claim 1, characterized in that: The filtering of the first lane lines in the first lane line set according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane lines in the first lane line set to obtain a second lane line set in the target area includes: Filtering the first lane lines in the first lane line set according to the second distance set corresponding to each first lane line in the first lane line set to obtain a second lane line set in the target area; The second distance set corresponding to each first lane line is obtained in the following manner: For each first lane line in the first lane line set, sorting the discrete points of the first lane line according to the distance between the discrete points of the first lane line and the reference origin of the target area in the first direction to obtain a sorted set of discrete points corresponding to the first lane line; A set of distances between every two adjacent discrete points in the sorted set of discrete points of the first lane line in the second direction is determined as a second distance set corresponding to the first lane line.

7. The lane line denoising method according to claim 6, characterized in that: The filtering the first lane lines in the first lane line set according to the second distance set corresponding to each first lane line in the first lane line set to obtain the second lane line set in the target area includes: According to the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set are screened to obtain a third lane line set of the target area; the third lane line set includes at least one third lane line; the at least one third lane line includes part or all of the at least one first lane line; The third lane lines in the third lane line set are screened according to the second distance set corresponding to each third lane line in the third lane line set to obtain a second lane line set in the target area.

8. The lane line denoising method according to claim 6, characterized in that: The screening process of the first lane lines in the first lane line set according to the second distance set corresponding to each first lane line in the first lane line set includes: In response to a variance between distances in the second distance set corresponding to a second target lane line being greater than a second variance threshold, deleting the second target lane line from the first lane line set; The second target lane line is any of the first lane lines in the first lane line set.

9. The lane line denoising method according to claim 1, characterized in that: The filtering of the first lane lines in the first lane line set according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane lines in the first lane line set to obtain a second lane line set in the target area includes: According to the first distance set corresponding to each lane line combination, the first lane lines in the first lane line set are screened to obtain a second lane line set in the target area; The first distance set corresponding to each lane line combination is obtained in the following manner: For each lane line combination, determining a distance between at least one feature point pair in the lane line combination in the second direction; Determine a set of distances of the at least one feature point pair in the second direction as a first distance set corresponding to the lane line combination; Each of the feature point pairs in the lane line combination includes two feature points that are equidistant from a reference origin of the target area in the first direction, and the two feature points are located on different first lane lines in the lane line combination.

10. The lane line denoising method according to claim 9, characterized in that: The screening process of the first lane lines in the first lane line set according to the first distance set corresponding to each lane line combination includes: In response to an average value of distances in a first distance set corresponding to the target lane line combination being less than a second distance threshold, deleting a first lane line with a short length in the target lane line combination from the first lane line set; Wherein, the target lane line combination is any of the lane line combinations.

11. A lane line denoising system, characterized in that: Including fitting module and denoising module; The fitting module is used to fit the discrete points of the target area to determine a first lane line set of the target area, wherein the first lane line set includes at least one first lane line; The denoising module is used to filter the first lane line in the first lane line set according to at least one of a reference lane line distance, a reference discrete point distance, and a reference discrete point angle of the first lane line to obtain a second lane line set in the target area; the second lane line set includes at least one second lane line, and the at least one second lane line includes part or all of the at least one first lane line; Among them, the reference lane line distance is determined based on the distance between different first lane lines in the lane line combination, and the lane line combination includes different first lane lines in the first lane line set that have a common area in the first direction; the reference discrete point distance is determined based on the distance between adjacent discrete points of the first lane line in the second direction; the reference discrete point angle is determined based on the angle of the discrete points of the first lane line relative to the second direction; the first direction is the lane line direction of the target area, and the second direction is perpendicular to the lane line direction of the target area.

12. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the lane line denoising method according to any one of claims 1 to 10 when executing the computer program.

13. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the lane line denoising method according to any one of claims 1 to 10 is implemented.

Images