Lane center line identification method and device, equipment, storage medium and product

Abstract

The invention discloses a lane center line identification method and device, equipment, a storage medium and a product, relates to the technical field of automatic driving, and discloses the lane center line identification method. Comprising the following steps: acquiring a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame; determining a first coordinate point within the first boundary range and a second coordinate point outside the first boundary range in the target coordinate point set; and performing de-duplication processing on the lane line corresponding to the first coordinate point, and performing splicing of the target lane line based on the second coordinate point to obtain spliced lane line data. According to the method, the repeated coordinate points of the target lane line in the new road image frame are removed through the first boundary range of the target lane line in the previous road image frame, the situation that the directionality of coordinate sequences of the same lane line in different frames is inconsistent is avoided, and the accuracy of lane line recognition is improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to methods, devices, equipment, storage media, and products for identifying lane centerlines. Background Art

[0002] Accurately detecting lane centerlines in data segments plays a crucial role in vehicle navigation and control.

[0003] Typically, related technologies perform image processing on road images to extract lane features and identify lane lines. However, in practice, due to factors such as vehicle movement and environmental changes, the coordinates of lane lines in previous and subsequent frames of the road image can change. This can cause the coordinate sequences of the same lane line in different frames to have inconsistent directions, which in turn affects the accuracy of lane recognition.

[0004] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a lane centerline recognition method, which aims to solve the technical problem that the coordinate sequences of the same lane line in different frames may have inconsistent directions, resulting in low accuracy of lane line recognition.

[0006] To achieve the above objectives, the present application proposes a lane centerline recognition method, which includes: Obtaining a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame; Determine a first coordinate point in the target coordinate point set that is within the first boundary range and a second coordinate point that is outside the first boundary range; Deduplication processing is performed on the lane lines corresponding to the first coordinate points, and target lane lines are spliced based on the second coordinate points to obtain spliced lane line data.

[0007] Optionally, after the steps of performing deduplication processing on the lane lines corresponding to the first coordinate points and splicing the target lane lines based on the second coordinate points to obtain spliced lane line data, the method includes: Determine the minimum coordinate value of each coordinate system corresponding to each first coordinate point, and determine the maximum coordinate value of each coordinate system corresponding to each second coordinate point; Based on the minimum coordinate value of each coordinate system and the maximum coordinate value of each coordinate system, a second boundary range of the target lane line in the current road image frame is determined, wherein the second boundary range is used to achieve splicing of the target lane line in the next road image frame.

[0008] Optionally, after the step of obtaining a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame, the method includes: Construct a global grid; Mapping each coordinate point in the target coordinate point set to the global grid to determine a perception map corresponding to the current road image frame; The step of determining a first coordinate point within the first boundary range and a second coordinate point outside the first boundary range in the target coordinate point set includes: Based on the perception map, a first coordinate point in the target coordinate point set that is within the first boundary range and a second coordinate point that is outside the first boundary range are determined.

[0009] Optionally, the step of mapping each coordinate point in the target coordinate point set to the global grid to determine a perception map corresponding to the current road image frame includes: Determine a first distance between each coordinate point in the target coordinate point set and the vehicle, and save each coordinate point in a coordinate sequence in ascending order based on the first distance to obtain a coordinate sequence of the target lane line; Mapping the coordinate sequence to corresponding positions in the grid to obtain an initial grid; A thinning operation is performed on the initial grid to obtain a target grid, and a perception map corresponding to the current road image frame is determined based on the target grid.

[0010] Optionally, the step of performing a thinning operation on the initial grid to obtain a target grid includes any one of the following: Performing a first thinning operation on the initial grid to obtain a target grid; A first thinning operation is performed on the initial grid, and a second thinning operation is performed on the network after the first thinning operation to obtain a target grid, wherein the thinning operation includes a floor operation or a rounding operation.

[0011] Optionally, the step of performing a thinning operation on the initial grid to obtain a target grid includes: sequentially calculating second distances between adjacent coordinate points in the coordinate sequence in the initial grid; Determining whether the second distance is less than a preset distance threshold; If the second distance is less than the preset distance threshold, the coordinate points with a larger first distance among the adjacent coordinate points are deleted to obtain a thinned coordinate sequence, and the process returns to the step of sequentially calculating the second distances between adjacent coordinate points in the coordinate sequence in the initial grid until the second distances between adjacent coordinate points in the coordinate sequence in the initial grid are all less than the preset distance threshold, thereby obtaining a target grid.

[0012] In addition, to achieve the above-mentioned purpose, the present application also proposes a lane centerline recognition device, the lane centerline recognition device comprising: An acquisition module, configured to acquire a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame; A determination module, configured to determine a first coordinate point in the target coordinate point set that is within the first boundary range and a second coordinate point that is outside the first boundary range; The deduplication module is used to perform deduplication processing on the lane lines corresponding to the first coordinate points, and to splice the target lane lines based on the second coordinate points to obtain spliced lane line data.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a lane centerline recognition device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the computer program is configured to implement the steps of the lane centerline recognition method as described above.

[0014] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the lane centerline recognition method as described above are implemented.

[0015] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the lane centerline recognition method as described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: Compared with the related art, which usually uses image processing and machine learning algorithms to extract lane line features in road images, identify and splice complete lane lines. However, since the lane line coordinate points in each frame of the road image will change with the movement of the vehicle and changes in the environment, the coordinate sequence of the same lane line in different frames will have inconsistent directions, which in turn leads to low lane line recognition accuracy. In comparison, the present application obtains the target coordinate point set of the target lane line in the current road image frame and the first boundary range of the target lane line in the previous road image frame; determines the first coordinate point within the first boundary range and the second coordinate point outside the first boundary range in the target coordinate point set; deduplicates the lane line corresponding to the first coordinate point, and splices the target lane line based on the second coordinate point to obtain the spliced lane line data. It can be understood that the present application needs to obtain the first boundary range of the target lane line in the previous road image frame, where the boundary range is the boundary range determined based on the coordinate point of the target lane line in the previous road image frame. Each time this application receives a new road image frame, it determines the overlapping coordinate points (i.e., first coordinate points) and non-overlapping coordinate points (i.e., second coordinate points) between the target coordinate point set of the target lane line in the road image frame and the first boundary range. Since the overlapping coordinate points can cause anomalies in the coordinate sequence of the lane centerline, they are deduplicated and then the target lane line is spliced based on the second coordinate points. This can avoid directional inconsistencies in the coordinate sequences of the same lane line in different frames. In other words, by removing duplicate coordinate points of the target lane line in the new road image frame using the first boundary range of the target lane line, it ensures that the coordinate sequence of the target lane line does not have reverse points due to repeated data, thereby improving the accuracy of lane line recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0019] Figure 1 A schematic flow chart of the first embodiment of the lane centerline recognition method provided in this application; Figure 2 A schematic flow chart of a second embodiment of the lane centerline recognition method of the present application; Figure 3This is a schematic diagram of the module structure of the lane centerline recognition device according to an embodiment of the present application; Figure 4 Schematic diagram of the device structure of the hardware operating environment involved in the lane centerline recognition method in the embodiment of the present application.

[0020] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0022] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0023] The main solution of the embodiment of the present application is: obtaining the target coordinate point set of the target lane line in the current road image frame and the first boundary range of the target lane line in the previous road image frame; determining the first coordinate point in the target coordinate point set that is within the first boundary range and the second coordinate point that is outside the first boundary range; deduplicating the lane line corresponding to the first coordinate point, and splicing the target lane line based on the second coordinate point to obtain the spliced lane line data.

[0024] In this embodiment, the lane centerline recognition device is used as the execution body. For the convenience of description, it will be explained as "device" hereinafter.

[0025] Related technologies typically use image processing and machine learning algorithms to extract lane features from road images, identify, and stitch together complete lane lines. However, because the coordinates of lane lines in each frame of a road image change with vehicle movement and environmental changes, the coordinate sequences of the same lane line in different frames may have inconsistent directions, resulting in low lane recognition accuracy.

[0026] This application provides a solution to achieve lane centerline splicing, thereby improving the accuracy of lane line recognition.

[0027] It can be seen from the above embodiments that the present application removes the repeated coordinate points of the target lane line in the new road image frame through the cumulative boundary range of the target lane line, thereby avoiding the directional inconsistency of the coordinate sequence of the same lane line in different frames, thereby improving the accuracy of lane line recognition.

[0028] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication, and program execution capabilities, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device or terminal system capable of implementing the aforementioned functions. This embodiment and the following embodiments will be described below using a lane centerline recognition device as an example.

[0029] Based on this, the embodiment of the present application provides a method for identifying the center line of a lane, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the lane centerline recognition method of the present application.

[0030] In this embodiment, the lane centerline recognition method includes steps S100 to S300: Step S100, obtaining a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame; It should be noted that the target coordinate point set refers to the set of coordinate points on the lane line identified by the lane line detection algorithm in the current road image frame, where the target coordinate point can be the endpoint, inflection point or point sampled according to certain rules of the lane line, which together define the position and shape of the lane line in the image; the first boundary range refers to the bounding box determined based on the target lane line coordinate point in the previous road image frame of the current road image frame, and the boundary range is also the horizontal and vertical range of the lane line in the image, where the cumulative boundary range can be understood as the minimum circumscribed rectangle or polygon of the lane line in the image, which includes all parts of the lane line and can be used to represent the global position and coverage of the lane line.

[0031] In a specific implementation, this application proposes to define a dictionary screen_lane_center_polyline to store the splicing information of different lane centerlines, where the dictionary includes key values and value values. The key value is the lane centerline index, and the value value is a list of coordinate points where the lane centerline has appeared in the scene segment, and reverse points are eliminated based on the lane grid corresponding to the previous frame. Furthermore, this application proposes to define grid_scope, which is used to record the boundary information dictionary of the most recent appearance of the same lane centerline in the historical frame, where the key value is the lane centerline index, and the value value is the coordinate space range boundary of the same lane centerline in the historical frame at the closest moment to the current frame, represented by x_min, x_max, y_min, and y_max.

[0032] In a specific implementation, the device obtains a road image by receiving an information acquisition module on a vehicle, where the road image includes multiple image frames. In one embodiment, the device can use a deep learning model such as a convolutional neural network to identify the target lane line in the image frame and determine the coordinate point and the first boundary range of the target lane line.

[0033] In a specific implementation, the step of obtaining a target coordinate point set of a target lane line in a current road image frame includes: Acquire a current road image frame; extract vehicle position information and an initial coordinate point set of a target lane line from the current road image frame; and based on the vehicle position information, convert the initial coordinate point set of the target lane line from a global coordinate system to a local coordinate system with the vehicle as the origin to obtain a target coordinate point set of the target lane line.

[0034] In a specific implementation, the device obtains an image of the road the vehicle is currently traveling on from an onboard camera. This image contains visual information such as the road, lane markings, and the vehicle's surroundings. Furthermore, the device can extract the vehicle's precise position and posture (including position coordinates and orientation) from the current road image frame using the vehicle's sensors (such as GPS, IMU, wheel speed sensors, etc.) or visual algorithms (such as feature matching, SLAM, etc.). The device can also use a lane detection algorithm (such as a deep learning-based model) to identify the target lane from the image and extract a series of coordinate points on the target lane to form an initial set of coordinate points for the target lane, which define the position of the lane in the image coordinate system.

[0035] In a specific implementation, the device converts the initial coordinate point set of the target lane line from the global coordinate system to a local coordinate system with the vehicle as the origin. Specifically, this step involves coordinate transformation, the purpose of which is to transform the points on the lane line from the global coordinate system (such as the road coordinate system) to the local coordinate system centered on the vehicle. The purpose of doing this is to simplify subsequent processing, because in the local coordinate system, the position of the vehicle is fixed (origin), and the relative position relationship between the position of the lane line and the vehicle becomes more intuitive. Furthermore, the above conversion process requires knowing the position (position and orientation) of the vehicle relative to the global coordinate system, and then applying geometric transformations (such as translation and rotation) to map the points on the lane line from the global coordinate system to the local coordinate system. This transformation can be achieved through matrix operations, including rotation matrices and translation vectors.

[0036] In a specific implementation, after the conversion is completed, the set of coordinate points obtained by the device is the lane line coordinate point (also known as the target coordinate point) in the vehicle's local coordinate system. The target coordinate point describes the specific position and shape of the lane line relative to the vehicle.

[0037] In a specific implementation, the device converts the initial coordinate point set of the target lane line from a global coordinate system to a local coordinate system with the vehicle as the origin based on the vehicle position information to obtain the target coordinate point set of the target lane line, including: Extract the vehicle coordinate value and vehicle heading angle from the vehicle position information; subtract the corresponding coordinate value of each initial coordinate point in the initial coordinate point set of the target lane line from the vehicle coordinate value to calculate the coordinate translation value of each initial coordinate point; calculate the rotation angle of the target lane line based on the vehicle heading angle; based on the coordinate translation value and the rotation angle, convert the initial coordinate point set of the target lane line from the global coordinate system to the local coordinate system with the vehicle as the origin to obtain the target coordinate point set of the target lane line.

[0038] It should be noted that the vehicle coordinates are the position of the vehicle in the global coordinate system, usually expressed as (x, y) coordinates. The vehicle coordinate value is the value of the vehicle in the x-axis coordinate system and the value of the vehicle in the y-axis coordinate system; the vehicle heading angle is the orientation of the vehicle relative to the global coordinate system, expressed as an angle value, used to describe the direction of the vehicle.

[0039] In a specific implementation, the device first obtains the corresponding map data information (map_data_item) from the road image frame information, and then extracts the ego vehicle's position information, including the X coordinate (ego_x), Y coordinate (ego_y), and heading angle (ego_theta).

[0040] In the specific implementation, the coordinate point P on the target lane line lane_center is (global_x, global_y). The coordinate position of any point P in the vehicle coordinate system is calculated as follows: x_diff=global_x-ego_x; y_diff=global_y-ego_y; angle_cos=math.cos(ego_theta); angle_sin=math.sin(ego_theta); local_x=x_diff*angle_cos-y_diff*angle_sin; local_y=x_diff*angle_sin+y_diff*angle_cos.

[0041] Among them, x_diff is the translation value of point P in the x direction relative to the ego vehicle, specifically, the difference between the global x coordinate of the calculated point P and the global x coordinate of the ego vehicle; y_diff is the translation value of point P in the y direction relative to the ego vehicle, specifically, the difference between the global y coordinate of the calculated point P and the global y coordinate of the ego vehicle; angle_cos refers to the cosine value of the ego vehicle's heading angle; angle_sin refers to the sine value of the ego vehicle's heading angle; local_x refers to the x coordinate of point P in the local coordinate system, specifically, the x-direction translation value multiplied by the cosine value, the y-direction translation value multiplied by the sine value, and then subtracted; local_y refers to the y coordinate of point P in the local coordinate system, specifically, the x-direction translation value multiplied by the sine value, the y-direction translation value multiplied by the cosine value, and then added.

[0042] In the specific implementation, the above conversion process is to first translate point P to the position of the ego vehicle, then rotate it according to the orientation angle of the ego vehicle, and finally obtain the position of point P in the local coordinate system with the ego vehicle as the origin.

[0043] Step S200, determining a first coordinate point within the first boundary range and a second coordinate point outside the first boundary range in the target coordinate point set; In a specific implementation, the first coordinate point refers to a coordinate point in the target coordinate point set that is within the first boundary range. The first coordinate point can be understood as a coordinate point that appears in the current image frame and is repeated in the previous image frame. Since repeated coordinate points will be recorded repeatedly in the coordinate sequence, directional confusion will occur, thereby affecting the directionality of the entire coordinate sequence and generating reverse points. That is, repeated coordinate points will cause abnormalities in the coordinate sequence of the lane centerline. By deduplicating repeated coordinate points, it can be ensured that the coordinate sequence of the target lane line will not have reverse points due to repeated data.

[0044] Furthermore, the second coordinate point refers to a coordinate point in the target coordinate point set that is outside the first boundary range, and the second coordinate point can be understood as an extension of the lane line in the image.

[0045] Step S300: Deduplication processing is performed on the lane lines corresponding to the first coordinate points, and target lane lines are spliced based on the second coordinate points to obtain spliced lane line data.

[0046] In a specific implementation, the device performs deduplication processing on the lane lines corresponding to the first coordinate points, that is, the device removes or merges duplicate or redundant coordinate points to ensure that the coordinate sequence of the target lane line does not contain reverse points due to repeated data. The deduplication process can be implemented by checking the distance, angle, or other geometric characteristics between consecutive points to ensure the smoothness and continuity of the lane lines.

[0047] Furthermore, the device performs splicing of the target lane line based on the second coordinate points, that is, the device incorporates these points outside the boundary range into the representation of the lane line to expand the lane line data, wherein the splicing process involves coordinate transformation, interpolation or other geometric algorithms to ensure that the newly added points are geometrically consistent with the existing lane line data, thereby obtaining a continuous and complete lane line representation.

[0048] In a specific implementation, after the device performs deduplication processing on the lane lines corresponding to the first coordinate points and splices the target lane lines based on the second coordinate points to obtain the spliced lane line data, the method includes: Determine the minimum coordinate value of each coordinate system corresponding to each first coordinate point, and determine the maximum coordinate value of each coordinate system corresponding to each second coordinate point; based on the minimum coordinate value of each coordinate system and the maximum coordinate value of each coordinate system, determine the second boundary range of the target lane line in the current road image frame, wherein the second boundary range is used to achieve splicing of the target lane line in the next road image frame.

[0049] In a specific implementation, the device determines the minimum coordinate value of each coordinate system corresponding to each first coordinate point. That is, during the lane line detection process, for the second coordinate points outside the first boundary range (that is, the points that do not exceed the boundary of the target lane line in the previous frame image), the device needs to determine their minimum coordinate value in their respective coordinate systems. The device determines the maximum coordinate value of each coordinate system corresponding to each second coordinate point. That is, during the lane line detection process, for the second coordinate points outside the first boundary range (that is, those points that exceed the boundary of the current lane line), the device needs to determine their maximum coordinate value in their respective coordinate systems. These coordinate values include the minimum x value, minimum y value, maximum x value, maximum y value, etc., which represent the boundary range of the lane line in each direction.

[0050] Furthermore, the minimum coordinate value of each coordinate system and the maximum coordinate value of each coordinate system are used to determine a second boundary range of the target lane line in the current road image frame, wherein the second boundary range is used to achieve splicing of the target lane line in the next road image frame.

[0051] It is understandable that the lane line stitching method preferably uses a scene where the lane is turning (i.e., the lane line changes greatly). By only using the boundary range of the previous image frame, the lane center line stitching of the vehicle in the scene where the lane line changes greatly is achieved.

[0052] Compared with the related art, which usually uses image processing and machine learning algorithms to extract lane line features in road images, identify and splice complete lane lines. However, since the lane line coordinate points in each frame of the road image will change with the movement of the vehicle and changes in the environment, the coordinate sequence of the same lane line in different frames will have inconsistent directions, which in turn leads to low lane line recognition accuracy. In comparison, the present application obtains the target coordinate point set of the target lane line in the current road image frame and the first boundary range of the target lane line in the previous road image frame; determines the first coordinate point within the first boundary range and the second coordinate point outside the first boundary range in the target coordinate point set; deduplicates the lane line corresponding to the first coordinate point, and splices the target lane line based on the second coordinate point to obtain the spliced lane line data. It can be understood that the present application needs to obtain the first boundary range of the target lane line in the previous road image frame, where the boundary range is the boundary range determined based on the coordinate point of the target lane line in the previous road image frame. Each time this application receives a new road image frame, it determines the overlapping coordinate points (i.e., first coordinate points) and non-overlapping coordinate points (i.e., second coordinate points) between the target coordinate point set of the target lane line in the road image frame and the first boundary range. Since the overlapping coordinate points can cause anomalies in the coordinate sequence of the lane centerline, they are deduplicated and then the target lane line is spliced based on the second coordinate points. This can avoid directional inconsistencies in the coordinate sequences of the same lane line in different frames. In other words, by removing duplicate coordinate points of the target lane line in the new road image frame using the first boundary range of the target lane line, it ensures that the coordinate sequence of the target lane line does not have reverse points due to repeated data, thereby improving the accuracy of lane line recognition.

[0053] This application also proposes another embodiment based on the above first embodiment, referring to Figure 2 , the lane centerline recognition method includes: Step A100, obtaining a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame; Step A100 is equivalent to the above-mentioned step S100 and will not be described again here.

[0054] Step A200, constructing a global grid; In practice, building a global grid involves creating a global grid model that covers the entire road environment during the perception map construction process. The global grid serves as the underlying framework for the map, integrating and aligning data from different sensors and at different points in time to create a continuous, consistent map representation.

[0055] Step A300, mapping each coordinate point in the target coordinate point set to the global grid to determine a perception map corresponding to the current road image frame; In a specific implementation, the device maps each point in the target point set to the global grid and determines a perceptual map corresponding to the current road image frame. Specifically, the device converts each point on the lane line from its local coordinates in the image to its global coordinates in the global grid. Specifically, the device converts points from the local coordinate system to the global coordinate system so that these points can be understood and processed within a unified reference frame.

[0056] Furthermore, by mapping these points onto a global grid, a perception map of the current road image frame can be constructed, which includes the location and shape of lane lines and other relevant road information.

[0057] In a specific implementation, the device maps each coordinate point in the target coordinate point set to the global grid, and determines the perception map corresponding to the current road image frame, including: Determine a first distance between each coordinate point in the target coordinate point set and the vehicle, and save each coordinate point in a coordinate sequence in ascending order based on the first distance to obtain a coordinate sequence of the target lane line; map the coordinate sequence to a corresponding position in a grid to obtain an initial grid; perform a thinning operation on the initial grid to obtain a target grid, and determine a perception map corresponding to the current road image frame based on the target grid.

[0058] In a specific implementation, the device determines the first distance between each coordinate point in the target coordinate set and the vehicle. Specifically, the device calculates the distance from the vehicle's current position to each coordinate point on the target lane line. This distance, the straight-line distance from the vehicle to each coordinate point, is used for subsequent sorting and grid construction.

[0059] In a specific implementation, the device saves each coordinate point in ascending order based on the first distance into a coordinate sequence to obtain the coordinate sequence of the target lane line. That is, based on the calculated distance, the device sorts the coordinate points in ascending order of their distance from the vehicle and saves them into a sequence. This produces a distance-sorted coordinate sequence that can be used for subsequent lane line splicing and map construction.

[0060] In a specific implementation, the device maps the coordinate sequence to corresponding positions in a grid to obtain an initial grid. Specifically, the device maps each coordinate point in the sorted coordinate sequence to a predefined grid. This grid can be a two-dimensional grid system, with each grid representing an area in the map. In this way, an initial grid containing all lane line information can be constructed.

[0061] It should be noted that thinning refers to reducing the number of points in a grid to simplify it and improve processing efficiency. This is usually achieved by removing redundant points or merging similar points. Thinning can include flooring or rounding, as well as other algorithms such as Voxel Grid filtering.

[0062] In a specific implementation, the device determines a perception map corresponding to the current road image frame based on the target grid. Specifically, the device uses the thinned target grid to determine the perception map for the current road image frame. The perception map is a representation of the road environment within the autonomous driving system, encompassing the location and shape of lane markings and other relevant road information.

[0063] In a specific implementation, the device performs a thinning operation on the initial grid to obtain a target grid, including any one of the following: Performing a first thinning operation on the initial grid to obtain a target grid; A first thinning operation is performed on the initial grid, and a second thinning operation is performed on the network after the first thinning operation to obtain a target grid, wherein the thinning operation includes a floor operation or a rounding operation.

[0064] In a specific implementation, the device performs a first thinning operation on the initial grid. Thinning reduces the number of points in the grid to simplify the grid and improve processing efficiency. This first thinning operation can be performed by rounding down or rounding up, depending on the specific application requirements and algorithm design, to obtain the target grid. The target grid is the grid obtained after the first thinning operation. This grid is sparser than the initial grid, but still retains the key lane information.

[0065] Furthermore, the device can perform a second thinning operation based on the first thinning operation to achieve higher processing efficiency or meet specific application requirements. Specifically, the thinning operation includes flooring or rounding up. That is, the first thinning operation and the second thinning operation are different. The first thinning operation can be a flooring operation and the second thinning operation can be a rounding up operation; or the first thinning operation can be a rounding up operation and the second thinning operation can be a flooring operation.

[0066] In a specific implementation, the step of performing a thinning operation on the initial grid to obtain a target grid includes: The second distances between adjacent coordinate points in the coordinate sequence in the initial grid are calculated sequentially; whether the second distance is less than a preset distance threshold is determined; if the second distance is less than the preset distance threshold, coordinate points with a larger first distance among the adjacent coordinate points are deleted to obtain a thinned coordinate sequence, and the process returns to the step of sequentially calculating the second distances between adjacent coordinate points in the coordinate sequence in the initial grid, until the second distances between adjacent coordinate points in the coordinate sequence in the initial grid are all less than the preset distance threshold, thereby obtaining a target grid.

[0067] First, in the initial grid, the device calculates the second distance between adjacent coordinate points in the coordinate sequence, where this distance is the straight-line distance between two adjacent coordinate points. Second, the device compares this calculated second distance with a preset distance threshold. This threshold, determined based on specific application requirements and algorithm design, controls the density of the thinning process. If the second distance is less than the preset distance threshold, indicating that the two adjacent coordinate points are too close, one of the two adjacent coordinate points can be deleted to reduce redundancy. Specifically, the point deleted is the one farther from the vehicle (i.e., the one with the greater first distance). After deleting the redundant points, a thinned-out coordinate sequence is obtained. The process then returns to the first step and recalculates the second distances between adjacent coordinate points in the thinned-out coordinate sequence. This process is repeated until the second distances between all adjacent coordinate points in the coordinate sequence are greater than or equal to the preset distance threshold. At this point, the resulting coordinate sequence is the thinned-out target grid.

[0068] In a specific implementation, the above process thins out the coordinate sequence in the initial grid to reduce redundant coordinate points and improve processing efficiency. Thinning is achieved by calculating the distance between adjacent coordinate points and comparing them with a preset distance threshold.

[0069] Step A400: Based on the perception map, determine a first coordinate point in the target coordinate point set that is within the first boundary range and a second coordinate point that is outside the first boundary range.

[0070] In a specific implementation, the device determines, on the perception map, a first coordinate point in the set of coordinate points that is within the cumulative boundary range and a second coordinate point that is outside the cumulative boundary range.

[0071] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the lane centerline identification method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0072] This application also provides a lane centerline recognition device, referring to Figure 3 , the lane centerline recognition device includes: An acquisition module 10 is configured to acquire a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame; A determination module 20 is configured to determine a first coordinate point within the first boundary range and a second coordinate point outside the first boundary range in the target coordinate point set; The deduplication module 30 is configured to perform deduplication processing on the lane lines corresponding to the first coordinate points, and to perform splicing of the target lane lines based on the second coordinate points to obtain spliced lane line data.

[0073] Optionally, the lane centerline recognition device further includes: A coordinate value determination module, configured to determine the minimum coordinate value of each coordinate system corresponding to each first coordinate point, and to determine the maximum coordinate value of each coordinate system corresponding to each second coordinate point; A boundary range determination module is used to determine a second boundary range of the target lane line in the current road image frame based on the minimum coordinate value of each coordinate system and the maximum coordinate value of each coordinate system, wherein the second boundary range is used to achieve the splicing of the target lane line in the next road image frame.

[0074] Optionally, the lane centerline recognition device further includes: Construction module, used to build the global grid; a mapping module, configured to map each coordinate point in the target coordinate point set to the global grid, and determine a perception map corresponding to the current road image frame; Optionally, the determining module 20 includes: A coordinate point determination module is configured to determine, based on the perception map, a first coordinate point in the target coordinate point set that is within the first boundary range and a second coordinate point that is outside the first boundary range.

[0075] Optionally, the mapping module includes: a coordinate sequence determination module, configured to determine a first distance between each coordinate point in the target coordinate point set and the vehicle, and save each coordinate point in a coordinate sequence in ascending order based on the first distance to obtain a coordinate sequence of the target lane line; A coordinate mapping module, used for mapping the coordinate sequence to corresponding positions in the grid to obtain an initial grid; The thinning module is configured to perform a thinning operation on the initial grid to obtain a target grid, and determine a perception map corresponding to the current road image frame based on the target grid.

[0076] Optionally, the thinning module includes: A primary thinning module, configured to perform a first thinning operation on the initial grid to obtain a target grid; The secondary thinning module is used to perform a first thinning operation on the initial grid and a second thinning operation on the network after the first thinning operation to obtain a target grid, wherein the thinning operation includes a floor operation or a rounding operation.

[0077] Optionally, the thinning module includes: a distance calculation module, configured to sequentially calculate second distances between adjacent coordinate points in the coordinate sequence in the initial grid; A judging module, configured to judge whether the second distance is less than a preset distance threshold; The iterative thinning module is configured to delete the coordinate points with a larger first distance among the adjacent coordinate points if the second distance is less than a preset distance threshold, thereby obtaining a thinned coordinate sequence, and return to the step of sequentially calculating the second distances between adjacent coordinate points in the coordinate sequence in the initial grid, until the second distances between adjacent coordinate points in the coordinate sequence in the initial grid are all less than the preset distance threshold, thereby obtaining a target grid.

[0078] The lane centerline recognition device provided in this application utilizes the lane centerline recognition method described in the aforementioned embodiment to address the technical issues surrounding lane centerline recognition. Compared to the prior art, the lane centerline recognition device provided in this application achieves the same beneficial effects as the lane centerline recognition method described in the aforementioned embodiment. Other technical features of the lane centerline recognition device are the same as those disclosed in the aforementioned embodiment and are not further detailed here.

[0079] The present application provides a lane centerline recognition device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the lane centerline recognition method in the above-mentioned embodiment one.

[0080] Reference below Figure 4, which shows a schematic structural diagram of a lane centerline recognition device suitable for implementing an embodiment of the present application. The lane centerline recognition device in the embodiment of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The lane centerline recognition device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0081] like Figure 4 As shown, the lane centerline recognition device may include a processing device 1001 (e.g., a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the lane centerline recognition device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to I / O interface 1006: input device 1007, such as a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output device 1008, such as a liquid crystal display (LCD), speaker, vibrator, etc.; storage device 1003, such as a magnetic tape or hard disk; and communication device 1009. Communication device 1009 can allow the lane centerline recognition device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows a lane centerline recognition device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented or have alternatively.

[0082] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0083] The lane centerline recognition device provided in this application utilizes the lane centerline recognition method described in the aforementioned embodiment to address the technical issues surrounding lane centerline recognition. Compared to the prior art, the lane centerline recognition device provided in this application achieves the same beneficial effects as the lane centerline recognition method described in the aforementioned embodiment. Other technical features of this lane centerline recognition device are the same as those disclosed in the aforementioned embodiment and are not further elaborated upon here.

[0084] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0085] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0086] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, a computer program) stored thereon, the computer-readable program instructions being used to execute the lane centerline recognition method in the above-mentioned embodiment.

[0087] The computer-readable storage medium provided herein may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including, but not limited to, wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0088] The computer-readable storage medium may be included in the lane centerline recognition device; or may exist independently without being assembled into the lane centerline recognition device.

[0089] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the lane centerline recognition device, the lane centerline recognition device is enabled to: recognize the lane centerline.

[0090] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0091] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0092] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0093] The computer-readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the lane centerline identification method described above, thereby resolving the technical problem of lane centerline identification. Compared to the prior art, the computer-readable storage medium provided in this application offers the same beneficial effects as the lane centerline identification method provided in the aforementioned embodiments, and therefore will not be further elaborated upon here.

[0094] The present application also provides a computer program product, comprising a computer program, which implements the steps of the lane centerline recognition method as described above when executed by a processor.

[0095] The computer program product provided in this application can solve the technical problem of lane centerline identification. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the lane centerline identification method provided in the above embodiment, and will not be repeated here.

[0096] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A lane centerline recognition method, characterized in that: The method for identifying the lane centerline includes: Obtaining a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame; Determine a first coordinate point in the target coordinate point set that is within the first boundary range and a second coordinate point that is outside the first boundary range; Deduplication processing is performed on the lane lines corresponding to the first coordinate points, and target lane lines are spliced based on the second coordinate points to obtain spliced lane line data.

2. The lane centerline recognition method according to claim 1, wherein: After the steps of performing deduplication processing on the lane lines corresponding to the first coordinate points and splicing the target lane lines based on the second coordinate points to obtain spliced lane line data, the method includes: Determine the minimum coordinate value of each coordinate system corresponding to each first coordinate point, and determine the maximum coordinate value of each coordinate system corresponding to each second coordinate point; Based on the minimum coordinate value of each coordinate system and the maximum coordinate value of each coordinate system, a second boundary range of the target lane line in the current road image frame is determined, wherein the second boundary range is used to achieve splicing of the target lane line in the next road image frame.

3. The lane centerline recognition method according to claim 1, wherein: After the step of obtaining the target coordinate point set of the target lane line in the current road image frame and the first boundary range of the target lane line in the previous road image frame, the method includes: Construct a global grid; Mapping each coordinate point in the target coordinate point set to the global grid to determine a perception map corresponding to the current road image frame; The step of determining a first coordinate point within the first boundary range and a second coordinate point outside the first boundary range in the target coordinate point set includes: Based on the perception map, a first coordinate point in the target coordinate point set that is within the first boundary range and a second coordinate point that is outside the first boundary range are determined.

4. The lane centerline recognition method according to claim 3, wherein: The step of mapping each coordinate point in the target coordinate point set to the global grid to determine a perception map corresponding to the current road image frame includes: Determine a first distance between each coordinate point in the target coordinate point set and the vehicle, and save each coordinate point in a coordinate sequence in ascending order based on the first distance to obtain a coordinate sequence of the target lane line; Mapping the coordinate sequence to corresponding positions in the grid to obtain an initial grid; A thinning operation is performed on the initial grid to obtain a target grid, and a perception map corresponding to the current road image frame is determined based on the target grid.

5. The lane centerline recognition method according to claim 4, wherein: The step of performing a thinning operation on the initial grid to obtain a target grid includes any one of the following: Performing a first thinning operation on the initial grid to obtain a target grid; A first thinning operation is performed on the initial grid, and a second thinning operation is performed on the network after the first thinning operation to obtain a target grid, wherein the thinning operation includes a floor operation or a rounding operation.

6. The lane centerline recognition method according to claim 4, wherein: The step of performing a thinning operation on the initial grid to obtain a target grid comprises: sequentially calculating second distances between adjacent coordinate points in the coordinate sequence in the initial grid; Determining whether the second distance is less than a preset distance threshold; If the second distance is less than the preset distance threshold, the coordinate points with a larger first distance among the adjacent coordinate points are deleted to obtain a thinned coordinate sequence, and the process returns to the step of sequentially calculating the second distances between adjacent coordinate points in the coordinate sequence in the initial grid until the second distances between adjacent coordinate points in the coordinate sequence in the initial grid are all less than the preset distance threshold, thereby obtaining a target grid.

7. A lane centerline recognition device, characterized in that: The device comprises: An acquisition module, configured to acquire a target coordinate point set of a target lane line in a current road image frame and a first boundary range of the target lane line in a previous road image frame; A determination module, configured to determine a first coordinate point in the target coordinate point set that is within the first boundary range and a second coordinate point that is outside the first boundary range; The deduplication module is used to perform deduplication processing on the lane lines corresponding to the first coordinate points, and to splice the target lane lines based on the second coordinate points to obtain spliced lane line data.

8. A lane centerline recognition device, characterized in that: The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the lane centerline recognition method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the lane centerline recognition method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the lane centerline recognition method according to any one of claims 1 to 6 are implemented.

Images