Lane Line Annotation Method, Device, Computer Equipment and Storage Medium

Abstract

The present application relates to a lane line annotation method, including: obtaining first lane line sampling points of a target lane and first center line sampling points corresponding to the target lane; respectively performing aggregation processing on the first lane line sampling points of the target lane and the corresponding center line sampling points to respectively obtain a first lane line sampling point sequence of the target lane and a corresponding first center line sampling point sequence; respectively performing curve fitting on the sampling points in the first lane line sampling point sequence and the sampling points in the first center line sampling point sequence to respectively obtain a first fitted lane line and a first fitted center line; obtaining second lane line sampling points of the target lane according to the first fitted lane line; and obtaining second center line sampling points corresponding to the target lane according to the first fitted center line; constructing a target map according to the second lane line sampling points and the second center line sampling points. The present application can improve the accuracy of lane line annotation for constructing a target map.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and particularly to a lane line annotation method, device, computer device, storage medium, and computer program product. Background Art

[0002] A high-precision map, also known as a high-definition map, has accurate vehicle position information and rich road element data information. Road element information such as roads, traffic signs, lane lines, obstacles, and pedestrians in a high-precision map is used to control a vehicle's steering, speed, path planning, lane changing, etc. At the same time, a high-precision map can also be used for positioning and has advantages such as a small map size and fast transmission. Among them, the accuracy of lane lines directly affects many safety behaviors of autonomous vehicles.

[0003] In the scenario of autonomous driving, it is necessary to generate lane lines of a high-precision map for assisting driving. For example, traditional visual three-dimensional reconstruction methods can be used to generate three-dimensional lane lines. However, practice has proved that the accuracy of lane lines obtained by traditional visual three-dimensional reconstruction methods is relatively low. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a lane line annotation method, device, computer device, computer-readable storage medium, and computer program product that can improve the accuracy of lane line annotation.

[0005] In a first aspect, the present application provides a lane line annotation method. The method includes:

[0006] Obtaining first lane line sampling points of a target lane and first center line sampling points corresponding to the target lane;

[0007] Performing aggregation processing on the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; and performing aggregation processing on the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane;

[0008] Performing curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and performing curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line;

[0009] Obtaining second lane line sampling points of the target lane according to the first fitted lane line; and obtaining second center line sampling points corresponding to the target lane according to the first fitted center line;

[0010] Constructing a target map according to the second lane line sampling points and the second center line sampling points.

[0011] In one embodiment, constructing a target map based on the second lane line sampling points and the second center line sampling points includes:

[0012] If the first fitted lane line belongs to the first type of lane line, parallelly constrain the second lane line sampling points and the second center line sampling points to obtain third lane line sampling points and third center line sampling points;

[0013] Construct a target map based on the third lane line sampling points and the third center line sampling points.

[0014] In one embodiment, parallelly constraining the second lane line sampling points and the second center line sampling points to obtain third lane line sampling points and third center line sampling points includes:

[0015] Fix the second center line sampling points, and adjust the second lane line sampling points according to a preset lane distance so that the shortest distance from each second lane line sampling point to the first fitted center line is equal;

[0016] Use the adjusted second lane line sampling points as the third lane line sampling points and the second center line sampling points as the third center line sampling points.

[0017] In one embodiment, the method for determining whether the first fitted lane line belongs to the first type of lane line includes:

[0018] For the first fitted lane line of the target lane, if the first fitted lane line within a preset length does not conform to a preset model, then the first fitted lane line belongs to the first type of lane line.

[0019] In one embodiment, the method further includes:

[0020] If the first fitted lane line within a preset length conforms to a preset model, then the first fitted lane line belongs to the second type of lane line, and the second type of lane line is used to represent the intersection area of the first type of lane line.

[0021] In one embodiment, the preset lane distance is the average value of the shortest distances from all the second lane line sampling points to the first fitted center line.

[0022] In a second aspect, the present application further provides a lane line annotation device. The device includes:

[0023] An acquisition module, configured to acquire first lane line sampling points of a target lane and first center line sampling points corresponding to the target lane;

[0024] An aggregation module, configured to aggregate the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; and aggregate the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane;

[0025] A fitting module, configured to perform curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and perform curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line;

[0026] A sampling module, configured to obtain second lane line sampling points of the target lane according to the first fitted lane line; and obtain second center line sampling points corresponding to the target lane according to the first fitted center line;

[0027] A mapping module, configured to construct a target map according to the second lane line sampling points and the second center line sampling points.

[0028] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0029] Obtain the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane;

[0030] Aggregate the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; and aggregate the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane;

[0031] Perform curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and perform curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line;

[0032] Obtain the second lane line sampling points of the target lane according to the first fitted lane line; and obtain the second center line sampling points corresponding to the target lane according to the first fitted center line;

[0033] Construct a target map according to the second lane line sampling points and the second center line sampling points.

[0034] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0035] Obtain the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane;

[0036] Aggregate the first lane line sampling points of the target lane to obtain the first lane line sampling point sequence of the target lane; and aggregate the first center line sampling points corresponding to the target lane to obtain the first center line sampling point sequence of the target lane;

[0037] Perform curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and perform curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line;

[0038] Obtain the second lane line sampling points of the target lane according to the first fitted lane line; and obtain the second center line sampling points corresponding to the target lane according to the first fitted center line;

[0039] Construct a target map according to the second lane line sampling points and the second center line sampling points.

[0040] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0041] Obtain the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane;

[0042] Aggregate the first lane line sampling points of the target lane to obtain the first lane line sampling point sequence of the target lane; and aggregate the first center line sampling points corresponding to the target lane to obtain the first center line sampling point sequence of the target lane;

[0043] Perform curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and perform curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line;

[0044] Obtain the second lane line sampling points of the target lane according to the first fitted lane line; and obtain the second center line sampling points corresponding to the target lane according to the first fitted center line;

[0045] Construct a target map according to the second lane line sampling points and the second center line sampling points.

[0046] The above lane line annotation method, device, computer device, storage medium, and computer program product obtain the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane; aggregate the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; and aggregate the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane; perform curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and perform curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line; obtain second lane line sampling points of the target lane according to the first fitted lane line; and obtain second center line sampling points corresponding to the target lane according to the first fitted center line; construct a target map according to the second lane line sampling points and the second center line sampling points. In this application, by respectively aggregating the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane, a first lane line sampling point series and a first center line sampling point series of the target lane are obtained, then curve fitting is performed on the first lane line sampling point series and the first center line sampling point series, and the sampling points re-obtained according to the fitted curves are used as the sampling points of the final target lane line, and a target map is constructed according to the sampling points of the final target lane line, which can greatly improve the accuracy of lane line annotation of the target lane and also enhance the accuracy of the corresponding constructed target map. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 FIG. is an application environment diagram of the lane line annotation method in an embodiment;

[0048] Figure 2 FIG. is a flowchart of the lane line annotation method in an embodiment;

[0049] Figure 3 FIG. is a flowchart of parallel constraint on the second lane line sampling points and the second center line sampling points in an embodiment;

[0050] Figure 4 FIG. is a flowchart of the lane line annotation method in another embodiment;

[0051] Figure 5 FIG. is a flowchart of the lane line annotation method in another embodiment;

[0052] Figure 6 FIG. is a schematic diagram of the execution result of step 402 in an embodiment;

[0053] Figure 7 FIG. is a schematic diagram of the execution result of step 406 in an embodiment;

[0054] Figure 8 It is a schematic diagram of the execution result of step 408 in an embodiment;

[0055] Figure 9 It is a schematic diagram of the execution result of step 410 in an embodiment;

[0056] Figure 10 It is a schematic diagram of the target map constructed in an embodiment;

[0057] Figure 11 It is a structural block diagram of a lane line marking device in an embodiment;

[0058] Figure 12 It is an internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0059] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0060] The lane line marking method provided by the embodiments of the present application can be applied to an application environment as Figure 1 shown. Among them, the terminal 102 communicates with the server 104 through a network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or can be placed in the cloud or other network servers. The server 104 obtains the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane sent by the terminal 102; aggregates the first lane line sampling points of the target lane to obtain a sequence of the first lane line sampling points of the target lane; and aggregates the first center line sampling points corresponding to the target lane to obtain a sequence of the first center line sampling points of the target lane; performs curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and performs curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line; obtains the second lane line sampling points of the target lane according to the first fitted lane line; and obtains the second center line sampling points corresponding to the target lane according to the first fitted center line; constructs a target map according to the second lane line sampling points and the second center line sampling points.

[0061] Among them, the terminal 102 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers.

[0062] It should be noted here that any step in the lane line annotation method disclosed in the embodiments of the present application can be implemented by the interaction between the terminal 102 and the server 104, or can be implemented by the server 104 alone, and no specific limitation is made here.

[0063] In one embodiment, as Figure 2 shown, a lane line annotation method is provided. Taking the method applied to the Figure 1 server as an example, the following steps 202 to 210 are included.

[0064] Step 202, obtain the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane.

[0065] The server can obtain the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane by sending through the terminal. The server can also directly obtain the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane through the interface. Among them, the first lane line sampling points refer to the sampling points obtained by sampling the lane lines in the actual road scene after reducing them according to a certain ratio. They can be directly sampled using the self-trained model, or obtained through manual annotation, or obtained through other means, and no specific limitation is made here. After obtaining the first lane line sampling points of the target lane, it is easy to calculate the first center line sampling points corresponding to the target lane according to the first lane line sampling points of the target lane. Among them, both the first lane line sampling points and the first center line sampling points include the three-dimensional information of the corresponding sampling points, such as the X coordinate and Y coordinate in the horizontal direction and the Z coordinate in the vertical direction in the world coordinate system.

[0066] In a possible implementation, an image of an actual road scene can be acquired in real time by any of laser radar, camera or navigation satellite systems. The image of the actual road scene acquired at the same time includes multiple frames, and the multiple frames of images can usually show the full picture of the actual road. Road environment features are extracted from the multiple frames of images, and a three-dimensional map of the road scene in a three-dimensional coordinate system is formed according to the road environment features. Lane lines in the three-dimensional map of the road scene can be manually annotated so that the three-dimensional map of the road scene has lane line annotation marks. The three-dimensional map of the scene with lane line annotation marks is used as a training sample to train the model so that the trained model can directly output three-dimensional lane line sampling points and corresponding center line sampling points. The specific spacing between the sampling points can be set according to the specific situation.

[0067] It should be noted here that the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane obtained in this embodiment are obtained in units of time. For example, the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane are obtained at the first time, the second time, etc. The lane line sampling points and the corresponding center line sampling points obtained at the same time may include the first lane line sampling points and the first center line sampling points corresponding to multiple lanes. However, when explaining the embodiments of the present application, one of the lanes, namely the target lane, is selected for explanation, and the lane line marking methods of other lanes are similar. It can be understood that the target lane includes the first lane line sampling points and the corresponding first center line sampling points of the left and right lane lines.

[0068] Step 204 , aggregate the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; and aggregate the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane.

[0069] In this embodiment, the server aggregates the first lane line sampling points of the target lane to obtain the first lane line sampling point sequence of the target lane; and aggregates the first center line sampling points corresponding to the target lane to obtain the first center line sampling point sequence of the target lane. In this embodiment, the aggregation process can be understood as aggregating information with common attributes to obtain the overall information of the attribute. Specifically, the first lane line sampling points can be aggregated according to the lane line identification. For example, according to the lane line identification of the target lane, the first lane line of the target lane is aggregated in a specific order. Different lane lines have different lane line identifications. For example, according to the acquisition time, all the first lane line sampling points of the target lane acquired at different times are aggregated in the order of the acquisition time to obtain the first lane line sampling point sequence of the target lane; similarly, the first center line sampling points corresponding to the target lane can be aggregated according to the acquisition time to obtain the first center line sampling point series of the target lane, that is, all the first center line sampling points of the target lane acquired at different times are aggregated in the order of the acquisition time to obtain the first center line sampling point sequence of the target lane.

[0070] In a possible implementation, some of the first lane line sampling points of the target lane may be aggregated to obtain a first lane line sampling point sequence of the target lane, or all of the first lane line sampling points of the target lane may be aggregated to obtain a first lane line sampling point sequence of the target lane, which may be selected based on actual conditions. It can be understood that the first lane line sampling points of the target lane acquired at the same time correspond to a section of the road in the actual road scene; the obtained first lane line sampling point series of the target lane corresponds to a longer section of the road or the entire road in the actual road scene.

[0071] Step 206 , curve fitting is performed on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and curve fitting is performed on the sampling points in the first center line sampling point sequence to obtain a first fitted center line.

[0072] The server performs curve fitting on the sampling points in the first lane line sampling point series to obtain the first fitted lane line; and performs curve fitting on the sampling points in the first center line sampling point sequence to obtain the first fitted center line. Curve fitting is a data processing method that uses a continuous curve to approximately characterize or compare the functional relationship between coordinates represented by discrete points. Commonly used curve fitting methods include least squares polynomial fitting, spline curve fitting, Bezier curve fitting, etc. The corresponding curve fitting method can be selected according to specific needs.

[0073] In an optional example, the steps of performing curve fitting on the sampling points in the first lane line sampling point sequence using least squares polynomial fitting to obtain the first fitted lane line are as follows:

[0074] (1) Construct the standard curve equation corresponding to the first fitted lane line as f(x, y, z) = 0, or y = f(x), z = g(x);

[0075] (2) Calculate the sum of the distances from the sampling points in each first lane line sampling series to the standard curve, that is, the sum of squared deviations;

[0076] (3) Construct an equation. Take the partial derivatives of the sum of squared deviations with respect to the coefficients in the standard curve equation on the left side of the equation, and the right side of the equation is 0, so as to obtain the coefficients of the standard curve equation and finally obtain the first fitted lane line.

[0077] Step 208: Obtain the second lane line sampling points of the target lane according to the first fitted lane line; and obtain the second center line sampling points corresponding to the target lane according to the first fitted center line.

[0078] In this embodiment, the server obtains the second lane line sampling points of the target lane according to the first fitted lane line; and obtains the second center line sampling points corresponding to the target lane according to the first fitted center line. In a possible implementation manner, the second lane line sampling points of the target lane are obtained by sampling on the first fitted lane line; and the second center line sampling points corresponding to the target lane are obtained by sampling on the first fitted center line. Specifically, the second lane line sampling points of the target lane can be obtained by equidistant sampling on the first fitted lane line; and the second center line sampling points corresponding to the target lane are obtained by equidistant sampling on the first fitted center line. In addition, the second lane line sampling points of the target lane can also be obtained by non-equidistant sampling on the first fitted lane line; and the second center line sampling points corresponding to the target lane are obtained by non-equidistant sampling on the first fitted center line.

[0079] Step 210: Construct a target map according to the second lane line sampling points and the second center line sampling points.

[0080] After obtaining the second lane line sampling points and the second center line sampling points of the target lane, a three-dimensional target map can be constructed in the corresponding operating environment. Among them, the target map includes the target lane formed by the second lane line sampling points and the corresponding second center line sampling points; the target map can also include traffic signs, obstacles and other road elements.

[0081] In this embodiment, the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane obtained are respectively subjected to aggregation processing to obtain the first lane line sampling point series and the first center line sampling point series of the target lane. Then, the first lane line sampling point series and the first center line sampling point series are subjected to curve fitting, and the sampling points re-obtained according to the fitted curve are used as the sampling points of the final target lane line, and the target map is constructed according to the sampling points of the final target lane line. This embodiment does not need to process each frame of image separately, shortening the running processing time. At the same time, the accuracy of the lane line annotation of the target lane can be greatly improved, and the accuracy of the corresponding constructed target map is also enhanced.

[0082] In one embodiment, step 210 of constructing a target map according to the second lane line sampling points and the second center line sampling points includes:

[0083] If the first fitted lane line belongs to the first type of lane line, the second lane line sampling points and the second center line sampling points are subjected to parallel constraint to obtain the third lane line sampling points and the third center line sampling points;

[0084] Construct a target map according to the third lane line sampling points and the third center line sampling points.

[0085] In this embodiment, the first type of lane line can be understood as the lane lines on the left and right sides with a relatively high degree of parallelism and a small curvature. In order to further improve the accuracy of the labeled lane lines and make the lane lines with a relatively high degree of parallelism more parallel, the second lane line sampling points and the second center line sampling points are subjected to parallel constraint to obtain the third lane line sampling points and the third center line sampling points of the target lane, and a target map is constructed according to the third lane line sampling points and the third center line sampling points. When performing parallel constraint on the second lane line sampling points and the second center line sampling points, any one of the second lane line sampling points and the first center line sampling points can be adjusted so that the shortest distance from each second lane line sampling point to the first fitted center line is equal. For example, fix the second center line sampling points and adjust the second lane line sampling points; or fix the second lane line sampling points and adjust the second center line sampling points.

[0086] In one embodiment, as Figure 3 shown, subjecting the second lane line sampling points and the second center line sampling points to parallel constraint to obtain the third lane line sampling points and the third center line sampling points includes the following steps 302 to 304.

[0087] Step 302, fix the second center line sampling points, and adjust the second lane line sampling points according to the preset lane distance so that the shortest distance from each second lane line sampling point to the first fitted center line is equal.

[0088] In this embodiment, the second centerline sampling points are fixed, and the second lane line sampling points are adjusted according to the preset lane distance, so that the shortest distances from each second lane line sampling point to the first fitted centerline are equal. Herein, the preset lane distance refers to the preset lane width, which can be set according to road standards or according to the shortest distance from the second lane line sampling point to the first fitted centerline.

[0089] In a possible implementation, the second centerline sampling points are fixed, and the second lane line sampling points on both sides are adjusted simultaneously, so that the shortest distances from the second lane line sampling points on both sides to the first fitted centerline are the same and meet the preset lane distance, that is, the shortest distance from the second lane line sampling point to the first fitted centerline is half of the preset lane distance.

[0090] Step 304: Use the adjusted second lane line sampling points as the third lane line sampling points, and use the second centerline sampling points as the third centerline sampling points.

[0091] In this embodiment, the adjusted second lane line sampling points are used as the third lane line sampling points. Since the second centerline sampling points are fixed, they are directly used as the third centerline sampling points.

[0092] In this embodiment, by fixing the second centerline sampling points and setting the preset lane distance to adjust the second lane line sampling points, the lane line where the obtained third lane line sampling points are located is made more parallel, further improving the accuracy of lane line marking.

[0093] In an embodiment, the method for determining whether the first fitted lane line belongs to the first type of lane line includes:

[0094] For the first fitted lane line of the target lane, if the first fitted lane line within the preset length does not conform to the preset model, the first fitted lane line belongs to the first type of lane line.

[0095] In this embodiment, for the first fitted lane line of the target lane, taking the preset length as a unit, that is, comparing and matching each preset length of the first fitted lane line with the preset model. If the matching is unsuccessful, it means that the first fitted lane line within the preset length does not conform to the preset model, and it is determined that the first fitted lane line belongs to the first type of lane line. Herein, the preset model includes a road model formed by special areas such as a lane turning area, an on-ramp area, or an area from the on-ramp to the main road.

[0096] In an embodiment, the method further includes:

[0097] If the first fitted lane line within the preset length conforms to the preset model, the first fitted lane line belongs to the second type of lane line, and the second type of lane line is used to represent the intersection area of the first type of lane line.

[0098] In this embodiment, for the first fitted lane line of the target lane, the first fitted lane line within a preset length is compared and matched with a preset model. If the matching is successful, it indicates that the first fitted lane line conforms to the preset model, and it is determined that the first fitted lane line within the preset length belongs to the second type of lane line. Among them, the second type of lane line is used to represent the intersection area of the first type of lane line, including the lane turning area, the ramp entry area, or the area from the ramp to the main road, etc. The size of the specific corresponding area can be set according to the actual situation.

[0099] In one embodiment, the lane preset distance is the average value of the shortest distances from all the sampling points of the second lane lines to the first fitted center line.

[0100] In this embodiment, the average value of the shortest distances from all the sampling points of the second lane lines to the first fitted center line is used as the lane preset distance. Among them, the average value can be the arithmetic average or the weighted average of the shortest distances from all the sampling points of the second lane lines to the first fitted center line, and no specific limitation is made here.

[0101] In one embodiment, as Figure 4 shown, a lane line annotation method is applied to a server and includes the following steps 402 to 412, which will be described in conjunction with Figure 5 this.

[0102] Step 402, obtain the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane through a self-trained model.

[0103] The self-trained model is trained through a three-dimensional scene map with lane line annotation marks. Inputting the image of the actual road scene obtained by a sensor or a camera into the self-trained model, the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane can be obtained, as specifically shown in Figure 6 this.

[0104] Step 404, according to the lane line identifier of the target lane, perform aggregation processing on the first lane line sampling points to obtain the first lane line sampling point sequence of the target lane; and according to the center line identifier corresponding to the target lane, perform aggregation processing on the first center line sampling points to obtain the first center line sampling point sequence of the target lane.

[0105] In this embodiment, by identifying lane lines, for example, by assigning different numbers, the lane line identifiers corresponding to different lane lines are made different. After selecting the target lane, according to the lane line identifier corresponding to the target lane, the first lane line sampling points of the target lane are aggregated in the order of the obtained time to obtain the first lane line sampling points of the target lane; and, according to the center line identifier corresponding to the target lane, the first center line sampling points are aggregated in the order of the obtained time to obtain the first center line sampling point sequence of the target lane.

[0106] Step 406: Curve fit the sampling points in the first lane line sampling point sequence to obtain the first fitted lane line; and curve fit the sampling points in the first center line sampling point sequence to obtain the first fitted center line.

[0107] In a specific example, curve fit the sampling points in the first lane line sampling point sequence to obtain the first fitted lane line; and curve fit the sampling points in the first center line sampling point sequence to obtain the first fitted center line, as Figure 7 shown.

[0108] Step 408: Perform equidistant sampling on the first fitted lane line to obtain the second lane line sampling points of the target lane; and perform equidistant sampling on the first fitted center line to obtain the second center line sampling points corresponding to the target lane.

[0109] After obtaining the first fitted lane line, perform equidistant sampling on the first fitted lane line according to the first preset distance to obtain the second lane line sampling points of the target lane; after obtaining the first fitted center line, perform equidistant sampling on the first fitted center line according to the second preset distance to obtain the second center line sampling points corresponding to the target lane, as specifically Figure 8 shown. Among them, the first preset distance and the second preset distance can be equal.

[0110] Step 410: If the first fitted lane line belongs to the first type of lane line, parallelly constrain the second lane line sampling points and the second center line sampling points to obtain the third lane line sampling points and the third center line sampling points, and construct the target map according to the third lane line sampling points and the third center line sampling points.

[0111] In this embodiment, when parallelly constraining the second lane line sampling points and the second center line sampling points, any one of the second lane line sampling points and the first center line sampling points can be adjusted so that the shortest distance from each second lane line sampling point to the first fitted center line is equal, as specifically shown in Figure 9 shown, such that di = dj.

[0112] Step 412: If the first fitted lane line belongs to the second type of lane line, construct a target map based on the second lane line sampling points and the second center line sampling points.

[0113] In this embodiment, the first type of lane line can be understood as the lane lines on the left and right sides with a relatively high degree of parallelism and a small curvature. The second type of lane line is used to represent the intersection area of the first type of lane line, including the lane turning area, the ramp entry area, or the area from the ramp to the main road, etc. The specific size of the corresponding area can be set according to the actual situation. Among them, the same target lane may include the first type of lane line and the second type of lane line.

[0114] In a specific example, the constructed target map is as Figure 10 shown.

[0115] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps in other steps.

[0116] Based on the same inventive concept, an embodiment of the present application also provides a lane line annotation device for implementing the above-mentioned lane line annotation method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the lane line annotation device provided below can refer to the limitations on the lane line annotation method in the above text, and will not be repeated here.

[0117] In one embodiment, as Figure 11 shown, a lane line annotation device is provided, including: an acquisition module 1102, an aggregation module 1104, a fitting module 1106, a sampling module 1108, and a mapping module 1110, where:

[0118] The acquisition module 1102 is configured to acquire the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane;

[0119] An aggregation module 1104, configured to aggregate the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; and aggregate the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane;

[0120] A fitting module 1106, configured to perform curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and perform curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line;

[0121] A sampling module 1108, configured to obtain second lane line sampling points of the target lane according to the first fitted lane line; and obtain second center line sampling points corresponding to the target lane according to the first fitted center line;

[0122] A mapping module 1110, configured to construct a target map according to the second lane line sampling points and the second center line sampling points.

[0123] In one embodiment, the mapping module 1110 is further configured to:

[0124] If the first fitted lane line belongs to the first type of lane line, perform parallel constraint on the second lane line sampling points and the second center line sampling points to obtain third lane line sampling points and third center line sampling points;

[0125] Construct a target map according to the third lane line sampling points and the third center line sampling points.

[0126] In one embodiment, the mapping module 1110 is further configured to:

[0127] Fix the second center line sampling points, and adjust the second lane line sampling points according to a preset lane distance, so that the shortest distance from each of the second lane line sampling points to the first fitted center line is equal;

[0128] Use the adjusted second lane line sampling points as the third lane line sampling points, and the second center line sampling points as the third center line sampling points.

[0129] In one embodiment, the device further includes a lane line type determination module, configured to determine whether the first fitted lane line belongs to the first type of lane line, including:

[0130] For the first fitted lane line of the target lane, if the first fitted lane line within a preset length does not conform to a preset model, then the first fitted lane line belongs to the first type of lane line.

[0131] In one embodiment, the lane line type determination module is further configured to:

[0132] If the first fitted lane line within a preset length conforms to a preset model, then the first fitted lane line belongs to the second type of lane line, and the second type of lane line is used to represent the intersection area of the first type of lane line.

[0133] Each module in the above lane line annotation device can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in the processor of the computer device in hardware form or be independent of it, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0134] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 12 shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store lane line sampling point data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a lane line annotation method.

[0135] Those skilled in the art can understand that Figure 12 the structure shown in

[0136] is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0137] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, it implements the steps of the lane line annotation method in the above embodiments.

[0138] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, it implements the steps of the lane line annotation method in the above embodiments.

[0139] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0140] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0141] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A lane line annotation method, characterized in that, The method includes: Obtaining first lane line sampling points of a target lane and first center line sampling points corresponding to the target lane; the first lane line sampling points include sampling points of the left and right lane lines of the target lane; both the first lane line sampling points and the first center line sampling points include three-dimensional information of the corresponding sampling points; Performing an aggregation process on the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; and performing an aggregation process on the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane; Performing curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and performing curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line; Obtaining second lane line sampling points of the target lane according to the first fitted lane line; and obtaining second center line sampling points corresponding to the target lane according to the first fitted center line; Constructing a three-dimensional target map according to the second lane line sampling points and the second center line sampling points.

2. The method according to claim 1, wherein The constructing a three-dimensional target map according to the second lane line sampling points and the second center line sampling points includes: If the first fitted lane line belongs to a first type of lane line, performing a parallel constraint on the second lane line sampling points and the second center line sampling points to obtain third lane line sampling points and third center line sampling points; Constructing a three-dimensional target map according to the third lane line sampling points and the third center line sampling points.

3. The method according to claim 2, wherein The performing a parallel constraint on the second lane line sampling points and the second center line sampling points to obtain third lane line sampling points and third center line sampling points includes: Fixing the second center line sampling points, and adjusting the second lane line sampling points according to a preset lane distance so that the shortest distance from each second lane line sampling point to the first fitted center line is equal; Taking the adjusted second lane line sampling points as the third lane line sampling points and the second center line sampling points as the third center line sampling points.

4. The method according to claim 2, wherein The determining method for whether the first fitted lane line belongs to a first type of lane line includes: For the first fitted lane line of the target lane, if the first fitted lane line within a preset length does not conform to a preset model, then the first fitted lane line belongs to a first type of lane line.

5. The method according to claim 4, characterized in that, The method further includes: If the first fitted lane line within a preset length conforms to a preset model, then the first fitted lane line belongs to a second type of lane line, and the second type of lane line is used to represent an intersection area of the first type of lane line.

6. The method according to claim 4, wherein The preset lane distance is the average value of the shortest distances from all the second lane line sampling points to the first fitted center line.

7. The method according to claim 2, wherein The method further includes: If the first fitted lane line belongs to a second type of lane line, constructing a three-dimensional target map according to the second lane line sampling points and the second center line sampling points; the second type of lane line is used to represent an intersection area of the first type of lane line.

8. The method according to claim 1, characterized in that Aggregating the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; And aggregating the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane, including: According to the lane line identifier of the target lane, aggregating the first lane line sampling points of the target lane in the order of acquisition time to obtain a first lane line sampling point sequence of the target lane; and aggregating the first center line sampling points corresponding to the target lane in the order of acquisition time to obtain a first center line sampling point sequence of the target lane.

9. The method according to claim 1, characterized in that, Obtaining the second lane line sampling points of the target lane according to the first fitted lane line; And obtaining the second center line sampling points corresponding to the target lane according to the first fitted center line, including: Performing equidistant sampling on the first fitted lane line according to a first preset distance to obtain the second lane line sampling points of the target lane; And performing equidistant sampling on the first fitted center line according to a second preset distance to obtain the second center line sampling points corresponding to the target lane.

10. The method according to claim 1, wherein Obtaining the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane, including: Obtaining the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane through a self-trained model; the self-trained model is trained by a scene three-dimensional map with lane line annotation identifiers.

11. A lane line marking device, characterized in that, The device includes: An acquisition module, configured to acquire the first lane line sampling points of the target lane and the first center line sampling points corresponding to the target lane; the first lane line sampling points include the sampling points of the left and right lane lines of the target lane; both the first lane line sampling points and the first center line sampling points include the three-dimensional information of the corresponding sampling points; An aggregation module, configured to aggregate the first lane line sampling points of the target lane to obtain a first lane line sampling point sequence of the target lane; and aggregate the first center line sampling points corresponding to the target lane to obtain a first center line sampling point sequence of the target lane; A fitting module, configured to perform curve fitting on the sampling points in the first lane line sampling point sequence to obtain a first fitted lane line; and perform curve fitting on the sampling points in the first center line sampling point sequence to obtain a first fitted center line; A sampling module, configured to obtain the second lane line sampling points of the target lane according to the first fitted lane line; and obtain the second center line sampling points corresponding to the target lane according to the first fitted center line; A mapping module, configured to construct a three-dimensional target map according to the second lane line sampling points and the second center line sampling points.

12. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 10.

14. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

Images