A false perception lane line extraction method

Abstract

The application relates to a misperceived lane line extraction method and system, which splits lane lines into multiple lane line segments seq; adjacent seq on both sides of a lane line to be classified is obtained through screening according to distance and angle difference; when the sum of distances of the nearest adjacent seq on both sides of the lane line to be classified is not more than d and the number of adjacent seq within 0.5d offset is greater than the number of seq that can be connected to the front end or rear end of the lane line to be classified, the lane line to be classified is determined as a misperceived lane line; d represents a lane width; most misperceived lane line data is extracted by using the relative position relationship between lane lines, higher-quality perceived lane line data can be obtained; no other data needs to be input to assist in judgment in addition to the lane line data itself; the accuracy of a finally generated lane level topology is improved, and the finally generated lane level topology better meets a real scene.

Description

Technical Field

[0001] This invention relates to the field of crowdsourced high-precision map production, and in particular to a method and system for extracting misperceived lane lines. Background Technology

[0002] In the process of creating crowdsourced high-precision maps, lane lines are crucial information for generating correct lane topology. Currently, most lane line data comes from perception data, i.e., lane line data is obtained by real-time scanning and image recognition by vehicle-mounted cameras. Due to limitations in image recognition accuracy, ground markings such as printed text and directional arrows are often misidentified as lane lines. These misidentified lane lines can affect the shape of the final generated lane vectors. Summary of the Invention

[0003] This invention addresses the technical problems existing in the prior art by providing a method and system for extracting misperceived lane lines. By utilizing the relative positional relationship between lane lines, it extracts most of the misperceived lane line data, improves the accuracy of the final generated lane-level topology, and better reflects real-world scenarios.

[0004] According to a first aspect of the present invention, a method for extracting false lane lines is provided, comprising: step 1, splitting the lane line into multiple lane line segments seq;

[0005] Step 2: Filter the adjacent sequence numbers on both sides of the lane line to be classified based on distance and angle difference;

[0006] Step 3: If the sum of the distances between the nearest adjacent seqs on both sides of the lane line to be classified does not exceed d and the number of adjacent seqs within an offset of 0.5d is greater than the number of seqs that can be connected to the front or rear end of the lane line to be classified, then the lane line to be classified is determined to be a falsely perceived lane line; d represents the lane width.

[0007] Based on the above technical solution, the present invention can also be improved as follows.

[0008] Optionally, in step 1, the lane lines are split using the line points as dividing points.

[0009] Optionally, the conditions for filtering the adjacent seq in step 2 include:

[0010] The adjacent seq refers to the seq of other lane lines that are on the same road surface as the lane line to be classified.

[0011] The distance between the midpoint of the lane line to be classified and the midpoint of the adjacent seq is less than a set distance threshold;

[0012] The angle difference between the lane line to be classified and the adjacent seq is less than a set angle threshold.

[0013] Optionally, in step 2, if it is determined that the lane line to be classified has at most one side with adjacent seq, in step 3, it is determined whether the lane line to be classified is a falsely perceived lane line only based on whether the number of adjacent seq within half the lane width is greater than the number of seq that can be connected to the front or rear end of the lane line to be classified.

[0014] Optionally, the process in step 3 of determining whether the number of adjacent seqs within a half-lane width offset is greater than the number of seqs that can be connected to the front or rear end of the lane line to be classified includes:

[0015] Step 301: Calculate the line distance dis between the lane line to be classified and each of its adjacent seq numbers;

[0016] Step 302: Record the number of seqs num1 where a≤dis<b and record the number of seqs num2 where dis<c; a and c are set thresholds used to distinguish whether a seq can be connected to the front or back of the lane line to be classified, and b is a set threshold used to indicate that the offset is within half the lane width.

[0017] In step 303, if num1 > num2, then the lane line to be classified is determined to be a falsely perceived lane line.

[0018] Optionally, the process of determining in step 3 that the sum of the distances between the nearest adjacent seq values ​​on both sides of the lane line to be classified does not exceed d includes:

[0019] Step 301': Calculate the line distance dis between the lane line to be classified and each of its adjacent seq numbers;

[0020] Step 302': Filter seqs where dis≥a1, and calculate the minimum line distances dis0 and dis1 between the seqs on both sides of the lane line to be classified; a1 is a set threshold used to limit seqs that are not on the lane line to be classified.

[0021] Step 303', if dis0 + dis1 ≤ b1, then the lane line to be classified is determined to be a falsely perceived lane line; b1 is a set threshold used to represent the lane width.

[0022] Optionally, the calculation process of the line-to-line distance dis includes:

[0023] Step 30101: Generate the minimum bounding rectangle from adjacent seq as the reference road surface, and generate the perpendicular bisector of the road surface based on the reference road surface;

[0024] Step 30102: Calculate the distance between the projection point of the midpoint of the lane line to be classified onto the perpendicular bisector of the adjacent seq and the midpoint of the adjacent seq, denoted as proj_dis1; calculate the distance between the projection point of the midpoint of the lane line to be classified and the midpoint of the adjacent seq onto the perpendicular bisector of the reference road surface, denoted as proj_dis2;

[0025] Step 30103: Calculate the line distance dis as min(proj_dis1, proj_dis2).

[0026] According to a second aspect of the present invention, a system for extracting false lane lines is provided, comprising: a lane line splitting module, an adjacent seq filtering module, and a false lane line determination module;

[0027] The lane line splitting module is used to split the lane line into multiple lane line segments seq;

[0028] The adjacent seq filtering module is used to filter adjacent seq on both sides of the lane line to be classified based on distance and angle difference;

[0029] The misperceived lane line determination module is used to determine whether the lane line to be classified is a misperceived lane line if the sum of the distances between the nearest adjacent seqs on both sides of the lane line to be classified does not exceed d and the number of adjacent seqs within an offset of 0.5d is greater than the number of seqs that can be connected to the front or rear end of the lane line to be classified; d represents the lane width.

[0030] According to a third aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the processor is configured to implement the steps of a method for extracting false-perceived lane lines when executing a computer management program stored in the memory.

[0031] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer management class program stored thereon, which, when executed by a processor, implements the steps of the method for extracting false-perceived lane lines.

[0032] This invention provides a method, system, electronic device, and storage medium for extracting falsely perceived lane lines. By utilizing the relative positional relationship between lane lines, it extracts most of the falsely perceived lane line data, obtaining higher-quality perceived lane line data. Apart from the lane line data itself, no other data is required to assist in the judgment. This improves the accuracy of the final generated lane-level topology and better matches real-world scenarios. Attached Figure Description

[0033] Figure 1 A flowchart of a method for extracting false lane lines provided by the present invention;

[0034] Figure 2 A flowchart illustrating an embodiment of a method for extracting false lane lines provided by the present invention;

[0035] Figure 3 A structural block diagram for extracting false lane lines provided by the present invention;

[0036] Figure 4 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0037] Figure 5 This is a schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. Detailed Implementation

[0038] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0039] Because misperceived lane lines exhibit the following geographical characteristics: at least one lane on either side of the lane line is incomplete; the number of lane lines offset by half a lane from the current lane line is greater than the number of lane lines that can be matched with the current lane line. This invention provides a method for extracting misperceived lane lines. Figure 1 A flowchart of a method for extracting false lane lines provided by the present invention is shown below. Figure 1 As shown, the extraction method includes:

[0040] Step 1: Divide the lane lines into multiple lane line segments (seq).

[0041] Step 2: Filter the adjacent sequence numbers on both sides of the lane line to be classified based on distance and angle difference;

[0042] Step 3: If the sum of the distances between the nearest adjacent seqs on both sides of the lane line to be classified does not exceed d and the number of adjacent seqs within an offset of 0.5d is greater than the number of seqs that can be connected to the front or back end of the lane line to be classified, then the lane line to be classified is determined to be a falsely perceived lane line; d represents the lane width.

[0043] The present invention provides a method for extracting false lane lines, which can obtain higher quality perceived lane line data; apart from the lane line data itself, no other data is required to assist in the judgment.

[0044] Example 1

[0045] Embodiment 1 provided by this invention is an embodiment of a method for extracting false lane lines provided by this invention, such as... Figure 2 The diagram shown is a flowchart of an embodiment of a method for extracting false lane lines provided by the present invention. Figure 1 and Figure 2 It can be seen that embodiments of this extraction method include:

[0046] Step 1: Divide the lane lines into multiple lane line segments (seq).

[0047] In one possible embodiment, in order to improve the retrieval efficiency of adjacent lines, the lane lines are split in step 1 using the row points as the dividing points.

[0048] In practice, after inputting the perceived lane line data, the lane line is split. If the number of lane line points is n, then the lane line is split into n-1 seq segments, and the midpoint of the segment represents the position of the segment.

[0049] Step 2: Filter the adjacent sequence numbers on both sides of the lane line to be classified based on distance and angle difference;

[0050] In one possible embodiment, the conditions for filtering adjacent seq in step 2 include:

[0051] Adjacent seq refers to the seq of other lane lines on the same road surface as the lane line to be classified.

[0052] The distance between the midpoint of the lane line to be classified and the midpoint of the adjacent seq is less than the set distance threshold.

[0053] In practice, this distance threshold can be 40m.

[0054] The angle difference between the lane line to be classified and the adjacent seq is less than the set angle threshold.

[0055] In practice, this angle threshold can be 20°.

[0056] Step 3: If the sum of the distances between the nearest adjacent seqs on both sides of the lane line to be classified does not exceed d and the number of adjacent seqs within an offset of 0.5d is greater than the number of seqs that can be connected to the front or back end of the lane line to be classified, then the lane line to be classified is determined to be a falsely perceived lane line; d represents the lane width.

[0057] In one possible implementation, if in step 2 it is determined that there are at most one side of the lane line to be classified as an adjacent seq, in step 3 it is determined whether the lane line to be classified is a falsely perceived lane line only based on whether the number of adjacent seq within half the lane width is greater than the number of seq that can be connected to the front or rear end of the lane line to be classified.

[0058] In one possible embodiment, step 3, determining whether the number of adjacent seqs within half a lane width offset is greater than the number of seqs that can be connected to the front or rear end of the lane line to be classified, includes:

[0059] Step 301: Calculate the line distance dis between the lane line to be classified and each of its adjacent seq numbers.

[0060] Step 302: Record the number of seqs num1 where a≤dis<b and record the number of seqs num2 where dis<c; a and c are set thresholds used to distinguish whether a seq can be connected to the front or back of the lane line to be classified, and b is a set threshold used to indicate that the offset is within half the lane width.

[0061] In practice, a can be 1.25m, b can be 2m, and c can be 0.75m. That is, when the offset distance is less than 0.75m, it is determined that the seq can be connected to the front or back end of the lane line to be classified.

[0062] In step 303, if num1 > num2, then the lane line to be classified is determined to be a falsely perceived lane line.

[0063] In one possible embodiment, step 3, determining that the sum of the distances between the nearest adjacent seq values ​​on both sides of the lane line to be classified does not exceed d, includes:

[0064] Step 301': Calculate the line distance dis between the lane line to be classified and the number of adjacent seqs.

[0065] Step 302': Filter seqs where dis≥a1, and calculate the minimum line distances dis0 and dis1 between seqs on both sides of the lane line to be classified; a1 is a set threshold used to limit seqs that are not on the lane line to be classified.

[0066] Step 303', if dis0+dis1≤b1, then the lane line to be classified is determined to be a falsely perceived lane line; b1 is a set threshold used to represent the lane width.

[0067] In practice, a1 can be 1m and b1 can be 5.5m.

[0068] In one possible embodiment, the line-to-line distance *dis* is converted into the point spacing, and the calculation process for the line-to-line distance *dis* includes:

[0069] Step 30101: Generate the minimum bounding rectangle from adjacent seq as the reference road surface, and generate the perpendicular line of the road surface based on the reference road surface.

[0070] Step 30102: Calculate the distance between the midpoint of the lane line to be classified and the projection point on the perpendicular bisector of the adjacent seq, and the midpoint of the adjacent seq, denoted as proj_dis1; calculate the distance between the midpoint of the lane line to be classified and the projection point on the perpendicular bisector of the reference road surface, denoted as proj_dis2.

[0071] Step 30103: Calculate the line distance dis as min(proj_dis1, proj_dis2).

[0072] Example 2

[0073] Embodiment 2 provided by the present invention is an embodiment of a false lane line extraction system provided by the present invention. Figure 3 This invention provides a structural diagram of a lane line extraction system for accidental detection, combined with... Figure 3 It can be seen that this embodiment includes: a lane line splitting module, an adjacent seq filtering module, and a misperceived lane line determination module.

[0074] The lane line splitting module is used to split lane lines into multiple lane line segments (seq).

[0075] The adjacent seq filtering module is used to filter the adjacent seq on both sides of the lane line to be classified based on distance and angle difference;

[0076] The misperceived lane line determination module is used to determine whether the lane line to be classified is a misperceived lane line if the sum of the distances between the nearest adjacent seq on both sides of the lane line to be classified does not exceed d and the number of adjacent seq within an offset of 0.5d is greater than the number of seq that can be connected to the front or back end of the lane line to be classified; d represents the lane width.

[0077] It is understood that the false lane line extraction system provided by the present invention corresponds to the false lane line extraction method provided in the foregoing embodiments. The relevant technical features of the false lane line extraction system can be referred to the relevant technical features of the false lane line extraction method, and will not be repeated here.

[0078] Please see Figure 4 , Figure 4 This is a schematic diagram illustrating an embodiment of the electronic device provided in this invention. For example... Figure 4 As shown, this embodiment of the invention provides an electronic device, including a memory 1310, a processor 1320, and a computer program 1311 stored in the memory 1310 and executable on the processor 1320. When the processor 1320 executes the computer program 1311, it performs the following steps: splitting the lane line into multiple lane line segments seq; filtering the adjacent seq on both sides of the lane line to be classified based on distance and angle difference; determining that the lane line to be classified is a falsely perceived lane line when the sum of the distances of the nearest adjacent seq on both sides of the lane line to be classified does not exceed d and the number of adjacent seq within an offset of 0.5d is greater than the number of seq that can be connected to the front or rear end of the lane line to be classified; d represents the lane width.

[0079] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating an embodiment of a computer-readable storage medium provided by the present invention. (See diagram below.) Figure 5As shown, this embodiment provides a computer-readable storage medium 1400, on which a computer program 1411 is stored. When the computer program 1411 is executed by a processor, it performs the following steps: splitting the lane line into multiple lane line segments seq; filtering the adjacent seq on both sides of the lane line to be classified according to the distance and angle difference; determining that the lane line to be classified is a falsely perceived lane line when the sum of the distances of the nearest adjacent seq on both sides of the lane line to be classified does not exceed d and the number of adjacent seq within an offset of 0.5d is greater than the number of seq that can be connected to the front or rear end of the lane line to be classified; d represents the lane width.

[0080] This invention provides a method, system, electronic device, and storage medium for extracting falsely perceived lane lines. By utilizing the relative positional relationship between lane lines, it extracts most of the falsely perceived lane line data, resulting in higher quality perceived lane line data. No additional data is required for judgment besides the lane line data itself. This improves the accuracy of the final generated lane-level topology, better reflecting real-world scenarios.

[0081] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0082] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0083] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0087] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for extracting falsely perceived lane lines, characterized in that, The extraction method includes: Step 1: Split the lane line into multiple lane line segments seq; Step 2: Screen and obtain the adjacent seq on both sides of the lane line to be classified according to the distance and angle difference; Step 3: When the sum of the distances of the nearest adjacent seq on both sides of the lane line to be classified does not exceed d and the number of adjacent seq within an offset of 0.5d is greater than the number of seq that can be connected to the front end or the rear end of the lane line to be classified, determine that the lane line to be classified is a misperceived lane line; d represents the lane width; When it is determined in Step 2 that there is at most one side of the lane line to be classified with adjacent seq, in Step 3, only determine whether the lane line to be classified is a misperceived lane line based on whether the number of adjacent seq within an offset of half the lane width is greater than the number of seq that can be connected to the front end or the rear end of the lane line to be classified; The process of determining in Step 3 whether the number of adjacent seq within an offset of half the lane width is greater than the number of seq that can be connected to the front end or the rear end of the lane line to be classified includes: Step 301: Calculate the line-line distance dis between the lane line to be classified and each of its adjacent seq; Step 302: Record the number num1 of seq where a ≤ dis < b, and record the number num2 of seq where dis < c; a and c are set thresholds used to distinguish whether a seq can be connected to the front end or the rear end of the lane line to be classified, and b is a set threshold used to represent being within an offset of half the lane width; Step 303: When num1 > num2, determine that the lane line to be classified is a misperceived lane line.

2. The extraction method according to claim 1, characterized in that, In Step 1, the lane line is split with row points as the separation points.

3. The extraction method according to claim 1, characterized in that, The conditions for screening the adjacent seq in Step 2 include: The adjacent seq is the seq of other lane lines on the same road surface as the lane line to be classified; The distance between the midpoint of the lane line to be classified and the midpoint of the adjacent seq is less than a set distance threshold; The angle difference between the lane line to be classified and the adjacent seq is less than a set angle threshold.

4. The extraction method according to claim 1, characterized in that, The process of determining in Step 3 that the sum of the distances of the nearest adjacent seq on both sides of the lane line to be classified does not exceed d includes: Step 301': Calculate the line-line distance dis between the lane line to be classified and each of its adjacent seq; Step 302': Screen the seq where dis ≥ a1, and calculate the minimum line-line distances dis0 and dis1 of the seq on both sides of the lane line to be classified; a1 is a set threshold used to limit that the seq is not on the lane line to be classified; Step 303': When dis0 + dis1 ≤ b1, determine that the lane line to be classified is a misperceived lane line; b1 is a set threshold used to represent not exceeding the lane width.

5. The extraction method according to claim 1 or 4, characterized in that, The calculation process of the line-line distance dis includes: Step 30101: Generate a minimum bounding rectangle from the adjacent seq as a reference road surface, and generate a perpendicular bisector of the road surface based on the reference road surface; Step 30102: Calculate the distance between the projection point of the midpoint of the to-be-classified lane line on the perpendicular bisector of the adjacent seq and the midpoint of the adjacent seq, denoted as proj_dis1; calculate the distance between the projection points of the midpoints of the to-be-classified lane line and the adjacent seq on the perpendicular bisector of the reference road surface, denoted as proj_dis2. Step 30103: Calculate the line-line distance dis as min(proj_dis1, proj_dis2).

6. A system for extracting false lane lines, characterized in that, It includes: A lane line splitting module, an adjacent seq screening module, and a misperceived lane line determination module; The lane line splitting module is used to split the lane line into multiple lane line segments seq; The adjacent seq screening module is used to screen the adjacent seqs on both sides of the to-be-classified lane line according to the distance and the angle difference; The misperceived lane line determination module is used to determine that the to-be-classified lane line is a misperceived lane line when the sum of the distances of the nearest adjacent seqs on both sides of the to-be-classified lane line does not exceed d and the number of adjacent seqs within an offset of 0.5d is greater than the number of seqs that can be connected to the front end or the rear end of the to-be-classified lane line; d represents the lane width; When the adjacent seq screening module determines that there is at most one adjacent seq on one side of the to-be-classified lane line, the misperceived lane line determination module only determines whether the to-be-classified lane line is a misperceived lane line according to whether the number of adjacent seqs within an offset of half of the lane width is greater than the number of seqs that can be connected to the front end or the rear end of the to-be-classified lane line; The process of the misperceived lane line determination module determining whether the number of adjacent seqs within an offset of half of the lane width is greater than the number of seqs that can be connected to the front end or the rear end of the to-be-classified lane line includes: Step 301: Calculate the line-line distance dis between the to-be-classified lane line and each of its adjacent seqs; Step 302: Record the number num1 of seqs where a ≤ dis < b, and record the number num2 of seqs where dis < c; a and c are set thresholds used to distinguish whether a seq can be connected to the front end or the rear end of the to-be-classified lane line, and b is a set threshold used to represent the offset within half of the lane width; When num1 > num2, it is determined that the to-be-classified lane line is a misperceived lane line.

7. An electronic device, characterized in that, It includes a memory and a processor. When the processor executes the computer management program stored in the memory, it implements the steps of the misperceived lane line extraction method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, A computer management program is stored thereon. When the computer management program is executed by the processor, it implements the steps of the misperceived lane line extraction method according to any one of claims 1-5.

Images