Map processing method, map processing apparatus and computer program product

Abstract

The embodiment of the invention discloses a map processing method, a map processing device and a computer program product, and relates to the technical field of map data. The method comprises the steps that a feature image is generated based on target trajectory data, the target trajectory data comprises multiple trajectories, and the feature image at least represents position information, density information and direction information of the multiple trajectories and can further represent speed information. Then, the feature image is processed, lane line vector data are obtained, and the lane line vector data comprise position information, direction information and type information of all lane lines; and finally, based on the lane line vector data, obtaining a lane level map. According to the method and the device, the lane-level map is obtained through the track data, so that low-cost construction or updating of the lane-level map is realized, a basis is provided for wider coverage and more timely updating of the lane-level map, a better navigation experience is brought to a user using the lane-level map, and meanwhile, the driving safety is improved.

Description

Technical Field

[0001] The present application relates to the field of map data technology, and in particular to a map processing method, a map processing device and a computer program product. Background Art

[0002] Lane-level maps are a key component of navigation systems and autonomous driving systems. Currently, lane-level maps are usually constructed by relying on high-precision LiDAR, cameras, integrated navigation and other equipment. The data collection, transmission and storage costs are high, and it is difficult to construct lane-level maps in large areas. In addition, roads may change due to road repairs, diversions and other reasons. The high-cost construction method makes it difficult to ensure timely updates of maps, resulting in inaccurate road information in lane-level maps, which affects the user's navigation experience and driving safety. Summary of the invention

[0003] In view of this, the present application provides a map processing method, a map processing device and a computer program product to obtain a lane-level map in a low-cost manner.

[0004] This application provides the following solutions: According to a first aspect, a map processing method is provided, the method comprising: generating a feature image based on target trajectory data, the target trajectory data comprising multiple trajectories, the feature image representing at least position information, density information and direction information of the multiple trajectories; processing the feature image to obtain lane line vector data, the lane line vector data comprising position information, direction information and type information of each lane line; and obtaining a lane-level map based on the lane line vector data.

[0005] According to a second aspect, a map processing device is provided, the device comprising: An image generating unit is configured to generate a characteristic image based on target trajectory data, wherein the target trajectory data includes a plurality of trajectories, and the characteristic image at least represents position information, density information, and direction information of the plurality of trajectories; A vector generation unit is configured to process the feature image to obtain lane line vector data, wherein the lane line vector data includes position information, direction information and type information of each lane line; The map processing unit is configured to obtain a lane-level map based on the lane line vector data.

[0006] According to a third aspect, a computer program product is provided, which, when executed by a processor, implements the steps of any one of the methods according to the first aspect.

[0007] According to the specific embodiments provided in this application, this application discloses the following technical effects: This application solves the problem that the construction of lane-level maps usually relies on high-precision equipment. This application can obtain lane-level maps through trajectory data, thereby obtaining lane-level maps in a low-cost manner, providing a basis for wider coverage and more timely updates of lane-level maps, thereby bringing better navigation experience to users using lane-level maps, while improving driving safety.

[0008] In addition, due to the low cost of the solution provided by this application, on the one hand, it makes it possible to build a large-scale lane-level map, and on the other hand, it is also conducive to the timely updating of the map. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0010] Figure 1 A flowchart of a map processing method provided in an embodiment of the present application.

[0011] Figure 2a A schematic diagram of a first image provided in an embodiment of the present application.

[0012] Figure 2b A schematic diagram of a second image provided in an embodiment of the present application.

[0013] Figure 3a A schematic diagram of another first image provided in an embodiment of the present application.

[0014] Figure 3b A schematic diagram of another second image provided in an embodiment of the present application.

[0015] Figure 4a A schematic diagram of a partitioning processing unit provided in an embodiment of the present application.

[0016] Figure 4b A schematic diagram of a grid division provided in an embodiment of the present application.

[0017] Figure 5 Schematic diagram of the process of obtaining lane line vector data in an embodiment of the present application.

[0018] Figure 6a A schematic diagram of lane line vector data in an embodiment of the present application.

[0019] Figure 6b This is a schematic diagram of another lane line vector data in an embodiment of the present application.

[0020] Figure 7 Schematic diagram of the process of processing lane-level maps in an embodiment of the present application.

[0021] Figure 8 A schematic block diagram of a map processing device provided in an embodiment of the present application.

[0022] Fig. 9 A schematic block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0024] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0025] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0026] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0027] In order to reduce the implementation cost while ensuring accuracy, this application provides a new idea. Figure 1 This is a flowchart of a map processing method provided in an embodiment of the present application. The method can be executed by a map processing device, which can be set on a server side or on a computer terminal with strong computing power. Figure 1 As shown in , the method may include the following steps: Step 101: Generate a feature image based on target trajectory data, wherein the target trajectory data includes a plurality of trajectories, and the feature image at least represents position information, density information, and direction information of the plurality of trajectories.

[0028] Step 102: Process the feature image to obtain lane line vector data, where the lane line vector data includes position information, direction information, and type information of each lane line.

[0029] Step 103: Obtain a lane-level map based on the lane line vector data.

[0030] It can be seen from the above process that the present application solves the problem that the construction of lane-level maps usually relies on high-precision equipment, resulting in high production costs. The present application can obtain lane-level maps through trajectory data, thereby obtaining lane-level maps in a low-cost manner, providing a basis for wider coverage and more timely updates of lane-level maps, thereby bringing better navigation experience to users using lane-level maps, while improving driving safety.

[0031] In addition, due to the low cost of the solution provided by the present application, on the one hand, it makes it possible to process lane-level maps over a large area, and on the other hand, it is also conducive to timely updating of maps.

[0032] The following is a detailed description of each step in the above process and the effects that can be further produced in conjunction with the embodiments. It should be noted that the limitations such as "first" and "second" involved in the present disclosure do not have limitations in terms of size, order, and quantity, and are only used to distinguish in name, for example, "first image" and "second image" are used to distinguish two images.

[0033] First, the above step 101, namely "generating a feature image based on target trajectory data, wherein the target trajectory data includes multiple trajectories, and the feature image at least represents position information, density information, and direction information of the multiple trajectories" is described in detail in conjunction with an embodiment.

[0034] The present application generates a feature image based on target trajectory data, wherein trajectory data refers to data information obtained by sampling the motion process of one or more moving objects in a spatiotemporal environment. In the present application, trajectory data includes multiple trajectories, each trajectory is composed of a set of discrete three-dimensional trajectory points with equal time intervals, and each trajectory point contains the longitude and latitude coordinates and altitude information of the trajectory point. For example, each trajectory point is usually represented by a triple (x, y, h), where x and y represent the longitude and latitude of the trajectory point, respectively, and h represents the altitude information. The feature image generated by the present application at least represents the position information, density information, and direction information of the multiple trajectories contained in the trajectory data.

[0035] As one of the feasible ways, in order to help driving users to navigate or drive automatically to generate lane-level maps, the above-mentioned moving object can be a vehicle, and the collected trajectory data is the trajectory data of the vehicle on the road. In the present application, the trajectory data of the above-mentioned vehicle can be obtained by collecting the location data of the vehicle, or by collecting the location data of the user end such as a mobile phone or a tablet computer used by the driving user, or by collecting the navigation data in the navigation application used by the driving user through the user end as the trajectory data of the vehicle, or by collecting the navigation data in the vehicle's own on-board navigation as the trajectory data of the vehicle, and so on.

[0036] Before generating a feature image based on the target trajectory data, the present application may also filter the target trajectory data. This is because the trajectory data is usually collected from all roads in the area of ​​the map to be processed, and a large amount of initial trajectory data is obtained. In order to enable the generated feature image to accurately characterize the location information, density information, and direction information of multiple trajectories contained in the target trajectory data, the present application may filter the initial trajectory data and eliminate trajectories with abnormal conditions such as jumps and drifts in the initial trajectory data.

[0037] For example, if a trajectory consists of 10 trajectory points, and the time intervals of these 10 trajectory points are the same and known, then the speed of the vehicle at each trajectory point on the trajectory can be calculated based on the latitude and longitude coordinates and altitude information contained in each trajectory point. For example, the distance between two trajectory points can be calculated by the latitude and longitude coordinates and altitude information of two trajectory points, and then the speed of the vehicle at each trajectory point on the trajectory can be calculated using the distance between the two trajectory points and the time interval between the two trajectory points. If the calculated speed is greater than the preset speed threshold (for example, the speed of the vehicle passing through each trajectory point is greater than 200 kilometers per hour), it proves that the trajectory is abnormal and the trajectory is removed from the initial trajectory data. For another example, under normal circumstances, each trajectory in the trajectory data should be a straight line or a smooth curve, and two adjacent trajectory points form a line segment. If the angle between adjacent line segments is greater than the preset angle threshold, it proves that the trajectory has jumped and the trajectory is removed from the initial trajectory data. For another example, if a certain trajectory point suddenly drifts to a certain distance compared to the previous trajectory point, it is proved that the trajectory has drifted, and the trajectory is removed from the initial trajectory data.

[0038] By screening a large amount of initial trajectory data, a certain number of trajectories are screened out for each road. Then, the present application uses the screened trajectory data to obtain target trajectory data, and generates a feature image based on the target trajectory data.

[0039] It should be noted that the feature image generated based on the trajectory data may include only one image, referred to as the first image in the embodiment of the present application, or may include two images, referred to as the first image and the second image, wherein the first image is used to characterize the position information, density information, and direction information of the multiple trajectories included in the trajectory data, and the second image is used to characterize the speed information of the multiple trajectories included in the trajectory data. By introducing the speed information of the multiple trajectories included in the second image, the perception accuracy of stop lines, etc. can be improved, making the generated lane line vector data more accurate and comprehensive, thereby obtaining a more accurate and comprehensive lane-level map.

[0040] In the embodiment of the present application, if it is not necessary to determine lane lines such as stop lines (for example, for a highway, it is not necessary to determine the stop line), the feature image may only include the first image. If it is necessary to determine the stop line, the feature image may include the first image and the second image.

[0041] When generating a feature image based on the target trajectory data, since the target trajectory data represents three-dimensional information (the trajectory points include altitude), the target trajectory data can be first mapped to a two-dimensional plane under a bird's-eye view (BEV, Bird's Eye View), and the trajectory data after mapping is used to perform statistics on the trajectory data of each grid on the two-dimensional plane, and the feature image is rendered based on the statistical results and position information of each grid. Among them, the statistical results of the grid include at least the number and direction information of the trajectories passed by the grid, and the size of the grid can be 0.15m*0.15m. The bird's-eye view can reduce visual occlusion and more clearly display the spatial relationship between different objects. The trajectory data under the bird's-eye view can also more clearly represent the location information, density information, and direction information of the trajectory in the trajectory data. However, in addition to the bird's-eye view, other perspectives can also be used. In addition, the present application refines the feature image of the target trajectory data, further enhancing the ability to capture local features, so that when predicting the lane line vector data, more local detail information can be considered to lay the foundation for obtaining a lane-level map with high accuracy later.

[0042] As one feasible method, the first image corresponding to the target trajectory data may be rendered based on the statistical results and position information of each grid on the two-dimensional plane, wherein the statistical results of the grid include at least the number and direction information of the trajectory passed by the grid.

[0043] As another feasible method, the first image and the second image corresponding to the target trajectory data can be rendered respectively based on the statistical results and position information of each grid on the two-dimensional plane, wherein the statistical results of the grid include the number, direction information and speed information of the trajectories passed through.

[0044] For each grid, the track passes through once, the number is increased by 1, and the number of tracks passed through is finally counted. Among them, the position of the grid reflects the position information, and the number of tracks passed through the grid reflects the density information. When determining the direction information of the track passed by the grid, the direction value of each track point passed by the grid can be calculated first, and then the direction information of the track passed by the grid can be obtained by averaging the direction values ​​of each track point. Among them, the direction value of track point a can be reflected as the direction value of the previous track point of track point a pointing to track point a.

[0045] It should be noted here that, on the one hand, directional information helps to distinguish the direction of the lane; on the other hand, directional information can reflect the curvature of the road and help predict lane lines in sparse trajectory scenarios; therefore, in the embodiment of the present application, introducing directional information into the feature image can effectively ensure the prediction accuracy of the lane line vector data.

[0046] When determining the speed information of the track through which the grid passes, the speed value of each track point through which the grid passes can be calculated first, and then the speed information of the track through which the grid passes can be obtained by taking the average of the speed values ​​of each track point.

[0047] For the rendering process of the first image, the present application can map each grid to each pixel area of ​​the first image based on the position information of each grid, and map the number of tracks passed by each grid to the first channel of the multi-channel color space used by the first image, and further map the direction information of the tracks passed by each grid to the second channel of the multi-channel color space to obtain the first image. This method allows the different information contained in the target trajectory data to be mapped to different channels in the multi-channel color space to obtain the first image, which is convenient for subsequent analysis and research of the first image corresponding to the target trajectory data.

[0048] The multi-channel color space can be an RGB (red, green, blue) color space or an HSV (hue, saturation, value) color space. This application does not make any specific limitation, and the method in this application can be implemented. This application provides a more preferred implementation scheme, the multi-channel color space can be HSV, and after obtaining a first image using HSV, the first image using HSV is converted to a first image using RGB, and this conversion process is lossless.

[0049] Among them, HSV is a commonly used color representation method, especially suitable for fields such as computer image processing, design and visual art. The name of the HSV color space comes from its three components: hue, saturation and value. Hue indicates the type of color, usually expressed in the form of an angle (ranging from 0 to 360 degrees), for example, red (corresponding to 0 degrees), green (corresponding to 120 degrees), blue (corresponding to 240 degrees), etc. Saturation indicates the purity or intensity of the color, ranging from 0 to 1 (or 0% to 100%), with a saturation of 0 indicating colorless (gray) and a saturation of 1 indicating that the color is fully saturated, that is, a pure color. Increasing saturation makes the color look more vivid, and reducing saturation makes the color more gray or soft. Value indicates the brightness or intensity of the color, ranging from 0 to 1 (or 0% to 100%), with a value of 0 indicating black and a value of 1 indicating the maximum brightness of the color. At intermediate values, the color may show different brightness effects.

[0050] Since the HSV color space is more consistent with human perception of color than the RGB color space, this application first uses HSV to represent the first image. And in the rendering of the first image, using HSV for color adjustment (such as color filtering or color replacement) is easier and more intuitive than operating directly in the RGB color space. When the lane line map needs to be obtained through the first image later, the first image needs to be analyzed. At this time, it is easier to perform visual analysis by converting the first image using HSV into the first image using RGB.

[0051] For the rendering process of the second image, the present application can map each grid to each pixel area of ​​the second image according to the position information of each grid, and map the speed information of the trajectory passed by each grid to the single-channel color space used by the second image to obtain the second image. This method allows the speed information contained in the target trajectory data to be mapped to the single-channel color space to obtain the second image, which is convenient for subsequent analysis and research of the second image corresponding to the target trajectory data.

[0052] Take the HSV color space as a multi-channel color space and the grayscale space as a single-channel color space as an example. The HSV color space is composed of three elements: hue (H), saturation (S) and brightness (V). The grayscale space is a single-channel image with only one grayscale value channel. For the target trajectory data on the highway, the direction information of the trajectory passed by each grid on the two-dimensional plane corresponding to the target trajectory data is mapped to the H channel of the HSV color space, and the number of trajectories passed by each grid on the two-dimensional plane corresponding to the target trajectory data is mapped to the V channel of the HSV color space. The S channel is set to a preset value to obtain the first image corresponding to the target trajectory data in the HSV color space, as shown in FIG. Figure 2a As shown. Since the sensitivity of human eyes to saturation is relatively low compared to chromaticity and brightness, when mapping the direction information and density information (i.e., the number of tracks passed by each grid), only the H channel and the V channel are used, and the S channel adopts the preset value, that is, the direction and density characteristics are more significantly reflected through the H channel and the V channel.

[0053] The speed information of each grid on the two-dimensional plane corresponding to the target trajectory data is filtered, and only the speed information below the preset speed threshold (for example, 20 km / h) is retained, and mapped to the gray value channel of the gray space to obtain the second image corresponding to the target trajectory data in the gray space, such as Figure 2b As shown in Figure 2, since few vehicles on the expressway travel at speeds below 20 km / h, Figure 2b As can be seen, it is almost all white.

[0054] For a target trajectory data containing a crossroad, the direction information of the trajectory of each grid on the two-dimensional plane corresponding to the target trajectory data is mapped to the H channel of the HSV color space, and the number of trajectories of each grid on the two-dimensional plane corresponding to the target trajectory data is mapped to the V channel of the HSV color space, and the S channel is set to a preset value to obtain the first image corresponding to the target trajectory data in the HSV color space, such as Figure 3a As shown; the speed information of the trajectories passed by each grid on the two-dimensional plane corresponding to the target trajectory data is filtered, and only the speed information below the preset speed threshold (for example, 20km / h) is retained, and mapped to the gray value channel of the gray space, and the second image corresponding to the target trajectory data in the gray space is obtained, as shown Figure 3b As shown in the figure, there are some low-speed vehicles at the intersection. Figure 3b It can be seen that there are some pixels with lower gray values.

[0055] It should be noted that before rendering the image based on the statistical results and position information of each grid on the two-dimensional plane, in order to enable the obtained feature image to better represent the statistical results, the statistical results can be converted to the same scale for better comparison and analysis. Specifically, the present application can normalize the statistical results of each grid (that is, the number of tracks passed by each grid, direction information, and speed information are scaled to the range of 0 to 1 respectively), and then the normalized statistical results are mapped to the interval of 0 to 255 to obtain the mapped statistical results, and the feature image is rendered based on the mapped statistical results. In this way, when the feature extraction network is subsequently required to extract the features of the feature image, the obtained feature representation is more accurate, and the stability and capability of the feature extraction network can be improved.

[0056] It can be seen that the present application maps the number and direction information of the tracks passed by each grid into a color space, and maps the speed information into another color space, and obtains the first image and the second image respectively, thereby mapping the expression of the target track data in the geographic space into the color space, which becomes an expression mode that is easier for the model to understand, so as to facilitate the use of the model in the subsequent process to extract more accurate lane lines from the target track data.

[0057] As one possible implementation, a complete road can be taken as a whole to generate a feature image, and then generate lane line vector data of the road. For example, a road corresponds to a first image, that is, a first image is generated using the trajectory data on the road. For another example, a road corresponds to a first image and a second image.

[0058] However, the size and data volume of the feature image generated by this feasible method are large, which is not conducive to the subsequent generation of lane line vector data. Therefore, in order to reduce the processing granularity of the feature image and obtain a more accurate lane-level map, a feasible method is provided, that is, the road corresponding to the trajectory data to be processed can be segmented in advance. Specifically, the trajectory data to be processed is obtained, and the road corresponding to the trajectory data to be processed is segmented to obtain multiple processing units. The trajectory data corresponding to each processing unit is used as the target trajectory data, and the feature image of the trajectory data corresponding to the processing unit is generated, and then the lane line vector data of each processing unit is generated.

[0059] The embodiment of the present application can obtain feature images corresponding to N processing units respectively based on the statistical results of the trajectory data corresponding to N processing units on the road, wherein the N processing units are obtained by segmenting the road according to a preset first size, and N is a positive integer. For example, the first size can be 120m*180m. The present application selects this size as the first size for ease of calculation. One processing unit can cover the width of a normal road (such as a two-way highway with multiple lanes). The entire road can be covered by a series of processing units. The target trajectory data in the present application corresponds to the trajectory data in one of the processing units. Other first sizes can also be used, such as using two side-by-side processing units to cover the width of the road.

[0060] like Figure 4a As shown, the purple line shows the road, the black line shows the track on the road, and the blue rectangle shows multiple processing units. The multiple processing units essentially divide the road. The present application uses the track data in each processing unit as the target track data. There may be partial overlap between adjacent processing units.

[0061] The present application divides the road according to a preset first size to obtain a feature image corresponding to each processing unit. The local feature image obtained in this way can more clearly represent the trajectory characteristics in each processing unit, thereby improving the prediction accuracy of subsequent lane line vector data, and further improving the accuracy of lane-level maps. In addition, multiple processing units can also be processed in parallel, thereby improving computing efficiency.

[0062] Furthermore, as a preferred embodiment, trajectory data statistics can be performed on each grid in N processing units, wherein the grid is obtained by dividing the processing unit according to a preset second size, and the second size is smaller than the first size. For example, the second size can be 0.15m*0.15m. Figure 4b To intercept Figure 4a The grid diagram obtained by one of the processing units is as follows: Figure 4b As shown in , the yellow squares represent the grids in the processing unit. It should be noted that the number of grids in the figure is only schematic.

[0063] Then, as one of the feasible methods, the first images corresponding to the N processing units can be rendered respectively according to the statistical results and position information of each grid in the N processing units, wherein the statistical results of the grid include at least the number and direction information of the tracks passed by the grid.

[0064] As another feasible method, the first image and the second image corresponding to the N processing units can be rendered respectively based on the statistical results and position information of each grid in the N processing units, wherein the statistical results of the grid include the number, direction information and speed information of the trajectories passed through.

[0065] The present application further divides each processing unit according to the second size to obtain each grid, and obtains the feature image corresponding to each processing unit based on the statistical results and position information of each grid of each processing unit. In this way, the present application further refines the feature image of the trajectory data in the road, further enhancing the ability to capture local features, so that when predicting the lane line vector data, more local detail information can be considered, laying the foundation for obtaining a lane-level map with high accuracy later.

[0066] For the rendering process of the first image, the present application maps each grid in each processing unit to each pixel area of ​​the first image based on the position information of each grid in the processing unit, maps the number of trajectories passed by each grid in the processing unit to the first channel of the multi-channel color space used by the first image, and maps the direction information of the trajectories passed by each grid in the processing unit to the second channel of the multi-channel color space used by the first image, so as to obtain the first image corresponding to the processing unit in the multi-channel color space.

[0067] For the rendering process of the second image, the present application maps each grid in each processing unit to each pixel area of ​​the second image based on the position information of each grid in the processing unit, and maps the speed information of the trajectory passed by each grid in the processing unit to the single-channel color space used by the second image, so as to obtain the second image corresponding to N processing units under the single-channel color space.

[0068] The above step 102, i.e., "processing the feature image to generate lane line vector data, wherein the lane line vector data includes position information, direction information and type information of each lane line" is described in detail below in conjunction with an embodiment.

[0069] The present application can process the feature image to generate lane line vector data, wherein the lane line vector data includes the position information, direction information and type information of multiple lane lines. The type information here is used to characterize the type of each lane line in the lane line vector data. The lane line is mainly used to distinguish the area of ​​each lane, which can be divided into a left boundary line (usually a yellow solid line in the actual road scene, used to separate the motor vehicle lanes traveling in the opposite direction), an outer boundary line (usually a white solid line in the actual road scene), a lane separator line (usually a white dashed line or a white solid line in the actual road scene, used to separate the motor vehicle lanes traveling in the same direction), and can further include stop lines and other types. Each lane line is composed of geometric elements such as points and lines. The lane line is essentially composed of continuous position points, and its direction can be reflected by the order of the position points. For example, assuming that a lane line contains position points a, b, c, ..., z, then the coordinates of each position point can reflect the position information of the lane line. The order of the position points on the lane line, such as the order from a to z or from z to a, reflects the direction information of the lane line. In the embodiment of the present application, a vector data format may be used to represent the shape and position of a geometric element.

[0070] After obtaining the feature image, image processing methods such as image segmentation and boundary recognition can be used to extract the lane centerline, and then the lane line can be further extracted based on the lane centerline, and then the type of lane boundary line can be identified based on the relative position relationship between the lane line and the lane centerline. However, this method involves multiple steps and processing results, resulting in the accumulation of errors, and the accuracy of the obtained lane line vector data is not high. Therefore, this application provides a better way to use a model based on the Transformer network to achieve prediction from feature images to lane line vector data.

[0071] Specifically, if the feature image only includes the first image, the present application can use a feature extraction network (i.e., the feature extraction model mentioned in the above process) to extract features from the first image to obtain a feature representation of the first image, where the feature representation of the first image integrates the position information, density information, and direction information of multiple trajectories included in the trajectory data. Among them, the feature extraction network can be a convolutional neural network (CNN), the core of which is to use convolution operations and pooling operations to extract features and reduce dimensions. It can also be a residual network (Residual Networks, ResNet), the core of which is residual learning (Residual Learning), that is, by learning the residual function instead of directly learning the output, the residual network can directly pass the features of the previous layer to the subsequent layer, thereby avoiding the problem of gradient disappearance and gradient explosion. Furthermore, the present application uses a Transformer-based decoder (i.e., the generation model mentioned in the above process) to decode the feature representation of the first image to obtain lane line vector data.

[0072] If the feature image includes a first image and a second image, such as Figure 5 As shown, Figure 5 A schematic diagram of the process of obtaining lane line vector data provided for an embodiment of the present application. The present application can use a feature extraction network to extract features from a first image and a second image respectively to obtain a feature representation of the first image and a feature representation of the second image, and then fuse the feature representation of the first image and the feature representation of the second image to obtain a fused feature representation. At this time, the fused feature representation integrates the position information, density information, direction information, and speed information of multiple trajectories included in the trajectory data, and then uses the Transformer decoder to decode the fused feature representation to obtain lane line vector data.

[0073] Among them, when fusing the feature representation of the first image and the feature representation of the second image, a multi-head attention mechanism based on Transformer can be adopted. Through the multi-head attention mechanism, the association and weight between the feature representations of the two images can be learned to achieve effective fusion between them.

[0074] When decoding using a Transformer-based decoder, the fused feature representation is processed by the Transformer layer, such as self-attention, and the latent vector output by the Transformer layer is mapped to the position and direction of the lane line, for example, using regression to predict the position and direction of the lane line. For the type information of the lane line, it can be understood as mapping the latent vector output by the Transformer layer to the lane line type space, obtaining the confidence of the lane line in each type, and outputting the type with the highest confidence as the type of the lane line.

[0075] In addition, in addition to using the Transformer-based decoder to achieve the above-mentioned lane line type prediction, other methods can also be used, such as a classifier based on Softmax (an activation layer).

[0076] In execution Figure 1 Before the process shown, the end-to-end model can be pre-trained. For example, the first image and the second image can be obtained in advance using the method shown in step 101 for some roads with known lane line vector data. The first image, the second image and the corresponding lane line vector data of the road are used as training samples, and training data is obtained by constructing multiple training samples, and the above-mentioned end-to-end model is trained using the training data. That is, the first image and the second image in the training sample are used as model inputs to obtain the lane line vector data predicted by the model. In each round of iteration, the loss function is obtained by using the difference between the predicted lane line vector data and the lane line vector data in the training sample. The value of the loss function is used to update the model parameters by a method such as gradient descent until the training end condition is reached. The training end condition may be, for example, that the loss function is less than or equal to a preset loss function threshold, or that the number of iterations reaches a preset round number threshold.

[0077] This application uses an end-to-end model based on the Transformer network to directly predict lane line vector data based on feature images, avoiding the error accumulation that may be caused by multiple steps and processing results. In addition, the Transformer network has a strong ability to capture global and local features, thereby improving the prediction accuracy of lane line vector data. In addition, this end-to-end model implementation method does not require complex post-processing and has higher processing efficiency.

[0078] The above step 103, namely "obtaining a lane-level map based on the lane line vector data", is described in detail below in conjunction with an embodiment.

[0079] The present application can obtain a lane-level map based on lane line vector data. The lane line vector data is in a two-dimensional plane, so the obtained lane-level map can be a two-dimensional map. It can also be further combined with the height value to obtain a three-dimensional lane-level map.

[0080] For example, if in the embodiment of the present application, the feature image generated in step 101 is from a bird's-eye view, and based on this, the lane-level vector data generated in step 102 is from a bird's-eye view, then in step 102, the height information of each lane line can be further determined by using the height information of the neighboring trajectory point set of each lane line in the trajectory data, wherein the neighboring trajectory point set of the lane line includes the trajectory points within a preset distance range from the position of the lane line; then, according to the height information of each lane line and the lane line vector data from a bird's-eye view, the three-dimensional lane line vector data is obtained. Then, in step 103, according to the three-dimensional lane line vector data, a lane-level map in three-dimensional space is obtained.

[0081] This application provides two methods for determining the height information of each lane line: One way is to use the Ransac algorithm to fit the height value of each lane line.

[0082] Specifically, for each lane line, a set of neighboring trajectory points is found in the trajectory data, and a mathematical model is assumed to represent the height of the lane line in three-dimensional space. Then, a set of trajectory points is randomly selected from the set of neighboring trajectory points as inliers (i.e., it is assumed that this set of inliers conforms to the mathematical model), and this set of inliers is used to fit the assumed mathematical model to obtain the parameters of the mathematical model, and then the fitting model is determined. Then, the distance from all trajectory points in all the neighboring trajectory points to the fitting model is calculated, and which trajectory points are inliers (conforming to the fitting model) and which are outliers (not conforming to the fitting model) are determined based on the preset threshold.

[0083] Repeat the above process multiple times. After each iteration, evaluate the quality of the fitting model based on the number of inliers corresponding to the fitting model, and select the fitting model as the final model based on the evaluation results. Finally, for each lane line, calculate its corresponding height value based on the final model.

[0084] Another way is to map each lane line included in the lane line vector data to a grid in a two-dimensional plane, and determine the height information of each lane line according to the height value corresponding to the trajectory data in each grid.

[0085] Specifically, each trajectory included in the trajectory data is projected into a grid in a two-dimensional plane (for example, the size of the grid can be 2m*2m), and the height value of the trajectory data contained in each grid is determined through the trajectory data. Assuming that there are 10 trajectory points in a grid, the height value corresponding to the grid is determined based on the average value of the altitude information of these 10 trajectory points. Then, each lane line included in the lane line vector data is also mapped to the grid of the two-dimensional plane, and the height value of the lane line is determined based on the height value corresponding to the mapped grid. In other words, the trajectory points mapped to the same grid as the lane line are used as the neighboring trajectory point set of the lane line, and the height average of the neighboring trajectory point set is used as the height value of the lane line.

[0086] It can be seen that the application first determines the lane line vector data from a bird's-eye view, and then determines the height information of each lane line included in the lane line vector data from a bird's-eye view based on the trajectory data, and then obtains the three-dimensional lane line vector data. This method can obtain three-dimensional lane line vector data in three-dimensional space, and then obtain a three-dimensional lane-level map, and because the height of the lane line is determined by the actual height of the trajectory data, it is closer to the road conditions in the actual scene, which can bring a better navigation experience to users.

[0087] In step 101, if the road corresponding to the trajectory data to be processed is divided into N processing units, when N is equal to 1, the lane-level map can be processed directly according to the lane line vector data corresponding to the processing unit.

[0088] If N is an integer greater than 1 and there is no overlapping area between adjacent processing units, the lane line vector data corresponding to the N processing units can be directly spliced ​​to obtain global lane line vector data, and then the lane-level map can be processed based on the global lane line vector data.

[0089] As a preferred implementation, if N is an integer greater than 1, and there is an overlapping area between adjacent processing units, the lane line vector data corresponding to the N processing units can be fused to obtain a lane-level map of the road.

[0090] For the fusion of lane line vector data between different processing units, the present application provides a method, when the road is segmented according to the first size, a certain overlap area is left between adjacent processing units, which is to facilitate the fusion of lane line vector data between adjacent processing units to obtain a more complete and accurate lane-level map. Specifically, for adjacent processing units with overlapping areas, Hungarian matching is used to match the lane line vector data of the two processing units, and the matched lane lines are fused, or a more complex weighted fusion strategy can be used.

[0091] This application fuses the lane line vector data corresponding to N processing units to achieve alignment between adjacent processing units and obtain more complete and accurate lane line vector data, thereby providing a basis for the construction and update of lane-level maps.

[0092] Figure 2a is a schematic diagram of a first image corresponding to trajectory data on a road, Figure 2b is a schematic diagram of a second image corresponding to the trajectory data on the same road, Figure 2b Only the speed information below the preset speed threshold is retained based on the preset speed threshold. Figure 2a and Figure 2b The corresponding lane line vector data is as follows Figure 6a As shown in . For another example, Figure 3a is a schematic diagram of a first image corresponding to trajectory data on a road corresponding to an intersection area, Figure 3b is a schematic diagram of a second image corresponding to the trajectory data on the road corresponding to the area, Figure 3a and Figure 3b The corresponding lane line vector data is as follows Figure 6b As shown. Figure 6a and Figure 6b Different colors are used to distinguish different types of lane lines. The outer boundary line can be displayed in red, the left boundary line can be displayed in blue, the stop line can be displayed in green, and the guide lane line can be displayed in black. In fact, the lane line vector data is a collection of position information, direction information, and type information of each lane line.

[0093] It can be seen from the above embodiments that the target trajectory data is composed of trajectory points, and the trajectory points include the latitude, longitude and altitude information of the trajectory points. The target trajectory data is usually in a global coordinate system, for example, the global coordinate system is a coordinate system with the center of the earth as the origin, the geographic North Pole as the positive direction of the Z axis, the geographic equatorial plane as the XOY plane, and the X axis pointing to the geographic prime meridian. However, the expression of the global coordinate system is not convenient for data processing when generating the feature images of each processing unit. Therefore, the target trajectory data in N processing units can be converted to the local coordinate systems corresponding to the N processing units respectively. For each processing unit, the local coordinate system corresponding to the processing unit takes the geometric center of the processing unit as the origin. Then, the feature image and lane line vector diagram are generated based on the expression of the target trajectory data in the local coordinate system.

[0094] Among them, when the target trajectory data in the N processing units are converted to the local coordinate systems corresponding to the N processing units, the SD (Standard Definition Map) map can be used as a reference. For example, the longitude and latitude coordinates of each trajectory point and the longitude and latitude coordinates of the geometric center of each processing unit are determined using the SD map, and the geometric center of each processing unit is used as the origin of the local coordinate system, and then the target trajectory data in the N processing units are converted to the local coordinate systems corresponding to the N processing units.

[0095] Furthermore, after the lane line vector data corresponding to N processing units are generated, for each processing unit, the lane line vector data corresponding to the processing unit is converted from the local coordinate system corresponding to the processing unit to the global coordinate system, and a lane-level vector map is obtained based on the lane line vector data corresponding to the N processing units in the global coordinate system.

[0096] In order to more clearly illustrate the method provided by the present application, a specific example is used here to explain, as a preferred embodiment, Figure 7 As shown, Figure 7 FIG. 1 is a flow chart of obtaining a lane-level map in an embodiment of the present application. Figure 7 It can be seen that the present application first needs to collect trajectory data on the road and pre-process the trajectory data, including mapping the trajectory data to a two-dimensional plane from a bird's-eye view. Then the road is divided into N processing units, each processing unit including each grid. For each processing unit, the first image and the second image corresponding to the processing unit are rendered respectively according to the statistical results of each grid in the processing unit. The lane-level vector data corresponding to each processing unit is generated using an end-to-end model. The lane-level vector data is the lane line vector data in a two-dimensional plane from a bird's-eye view. Then, based on the collected original trajectory data, the height information of each lane line in the lane line vector data corresponding to the N processing units is determined, and then the three-dimensional lane line vector data corresponding to the N processing units is obtained, and the three-dimensional lane line vector data corresponding to the N processing units is fused, and a lane-level map is obtained based on the fused global lane line vector data.

[0097] The above method provided in the embodiment of the present application can be applied to a variety of application scenarios, including but not limited to: for navigation on urban roads. In urban roads, road conditions are usually more complicated, such as temporary widening or narrowing of lanes, continuous intersections, bus lanes and other special situations. The lane-level map obtained based on the method of the present application can more accurately display the actual road conditions and help users make lane change decisions in advance; for automatic driving assistance, in some advanced automatic driving systems, the lane-level map is combined with vehicle sensor data to realize the vehicle's autonomous navigation and obstacle avoidance functions.

[0098] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0099] According to an embodiment of another aspect, a map processing device is provided. Figure 8 FIG. 4 is a schematic block diagram of a map processing device according to an embodiment. Figure 8 As shown, the device 800 includes: an image generation unit 801, a vector generation unit 802 and a map processing unit 803, and may further include a road segmentation unit 804. The main functions of each component unit are as follows: The image generating unit 801 is configured to generate a feature image based on target trajectory data, wherein the target trajectory data includes a plurality of trajectories, and the feature image at least represents position information, density information, and direction information of the plurality of trajectories.

[0100] The vector generation unit 802 is configured to process the feature image to obtain lane line vector data, where the lane line vector data includes position information, direction information, and type information of each lane line.

[0101] The map processing unit 803 is configured to obtain a lane-level map based on the lane line vector data.

[0102] As one possible implementation method, the feature image includes a first image and a second image; wherein the first image is used to represent the position information, density information and direction information of the multiple tracks; and the second image is used to represent the speed information of the multiple tracks.

[0103] As one of the feasible ways, when the image generation unit 801 generates a feature image based on the target trajectory data, it can be configured as follows: mapping the target trajectory data to a two-dimensional plane from a bird's-eye view; based on the mapped trajectory data, performing statistics on the trajectory data of each grid on the two-dimensional plane; rendering a feature image based on the statistical results and position information of each grid, wherein the statistical results of the grid include at least the number and direction information of the trajectory passed by the grid.

[0104] As one of the feasible ways, the feature image includes a first image. When the image generation unit 801 renders the feature image based on the statistical results and position information of each grid, it can be configured as follows: mapping each grid to each pixel area of ​​the first image according to the position information of each grid, mapping the number of trajectories passed by each grid to the first channel of the multi-channel color space used by the first image, and mapping the direction information of the trajectories passed by each grid to the second channel of the multi-channel color space to obtain the first image.

[0105] As one of the possible implementations, the multi-channel color space is HSV.

[0106] When obtaining the first image, the image generating unit 801 may also be configured to: obtain the first image using HSV, and convert the first image using HSV into a first image using RGB.

[0107] As one of the feasible ways, the feature image also includes a second image. When the image generation unit 801 renders the feature image based on the statistical results and position information of each grid, it can also be configured as follows: mapping each grid to each pixel area of ​​the second image according to the position information of each grid, mapping the speed information of the trajectory passed by each grid to the single-channel color space used by the second image, so as to obtain the second image.

[0108] As one of the feasible ways, when the vector generation unit 802 processes the feature image to obtain the lane line vector data, it can be configured as follows: using the feature extraction network to extract features from the first image and the second image respectively to obtain the feature representation of the first image and the feature representation of the second image; fusing the feature representation of the first image and the feature representation of the second image to obtain the fused feature representation; using the decoder of the Transformer network to decode the fused feature representation to obtain the lane line vector data.

[0109] As one of the feasible ways, when obtaining a lane-level map based on lane line vector data, the map processing unit 803 can be configured as follows: determining the height information of each lane line by utilizing the height information of a neighboring trajectory point set of each lane line in the target trajectory data, wherein the neighboring trajectory point set of the lane line includes trajectory points within a preset distance range from the position of the lane line; obtaining three-dimensional lane line vector data based on the height information and the lane line vector data from a bird's-eye view; and obtaining a three-dimensional lane-level map based on the three-dimensional lane line vector data.

[0110] Furthermore, based on the target trajectory data, before generating the feature image, the road segmentation unit 804 can be configured to: obtain the trajectory data to be processed; divide the road corresponding to the trajectory data to be processed into multiple processing units according to the first size; and use the trajectory data corresponding to each processing unit as the target trajectory data.

[0111] As one of the feasible ways, when obtaining a lane-level map based on lane line vector data, the map processing unit 803 can be configured as follows: fusing the lane line vector data corresponding to adjacent processing units to obtain the lane line vector data of the road; and obtaining a lane-level map based on the lane line vector data of the road.

[0112] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

[0113] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0114] In addition, the present application also provides a computer program product, including a computer program, which implements the steps of any one of the methods in the aforementioned method embodiments when executed by a processor.

[0115] In addition, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the methods in the aforementioned method embodiments are implemented.

[0116] And an electronic device, comprising: one or more processors; and A memory associated with the one or more processors, the memory being used to store program instructions, wherein the program instructions, when read and executed by the one or more processors, execute the steps of the method described in any one of the aforementioned method embodiments.

[0117] in, Fig. 9 The architecture of the electronic device is shown as an example, which may include a processor 910, a video display adapter 911, a disk drive 912, an input / output interface 913, a network interface 914, and a memory 920. The processor 910, the video display adapter 911, the disk drive 912, the input / output interface 913, the network interface 914, and the memory 920 may be communicatively connected via a communication bus 930.

[0118] Among them, the processor 910 can be implemented by a general-purpose CPU, a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., to execute relevant programs to implement the technical solution provided in this application.

[0119] The memory 920 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 920 can store an operating system 921 for controlling the operation of the electronic device 900, and a basic input and output system (BIOS) 922 for controlling the low-level operation of the electronic device 900. In addition, a web browser 923, a data storage management system 924, and a map processing device 800, etc. can also be stored. The above-mentioned map processing device 800 can be an application program that specifically implements the operations of the aforementioned steps in the embodiment of the present application. In short, when the technical solution provided by the present application is implemented by software or firmware, the relevant program code is stored in the memory 920 and is called and executed by the processor 910.

[0120] The input / output interface 913 is used to connect the input / output module to realize information input and output. The input / output module can be configured as a component in the device (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0121] The network interface 914 is used to connect to a communication module (not shown) to achieve communication interaction between the device and other devices. The communication module can achieve communication through a wired method (such as USB, network cable, etc.) or a wireless method (such as mobile network, WIFI, Bluetooth, etc.).

[0122] The bus 930 comprises a pathway for transmitting information between the various components of the device (eg, the processor 910 , the video display adapter 911 , the disk drive 912 , the input / output interface 913 , the network interface 914 , and the memory 920 ).

[0123] It should be noted that, although the above device only shows a processor 910, a video display adapter 911, a disk drive 912, an input / output interface 913, a network interface 914, a memory 920, a bus 930, etc., in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include components necessary for implementing the solution of the present application, and does not necessarily include all the components shown in the figure.

[0124] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a computer program product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0125] The technical solution provided by the present application is described in detail above. The principle and implementation method of the present application are described in detail using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for those skilled in the art, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as limiting the present application.

Claims

1. A map processing method, characterized in that: The method comprises: Generate a feature image based on the target trajectory data, wherein the target trajectory data includes a plurality of trajectories, and the feature image at least represents position information, density information, and direction information of the plurality of trajectories; Processing the feature image to obtain lane line vector data, wherein the lane line vector data includes position information, direction information, and type information of each lane line; A lane-level map is obtained based on the lane line vector data.

2. The method according to claim 1, characterized in that The characteristic image includes a first image and a second image; The first image is used to represent the position information, density information and direction information of the multiple tracks; the second image is used to represent the speed information of the multiple tracks.

3. The method according to claim 1, characterized in that The generating of the feature image based on the target trajectory data comprises: Mapping the target trajectory data to a two-dimensional plane from a bird's-eye view; Based on the mapped trajectory data, statistics are performed on the trajectory data of each grid on the two-dimensional plane; The characteristic image is rendered based on the statistical results and position information of each grid, wherein the statistical results of the grid at least include the number and direction information of the tracks through which the grid is passed.

4. The method according to claim 3, characterized in that The characteristic image includes a first image, and the rendering based on the statistical results and position information of each grid to obtain the characteristic image includes: According to the position information of each grid, each grid is mapped to each pixel area of ​​the first image, the number of tracks passed by each grid is mapped to the first channel of the multi-channel color space used by the first image, and the direction information of the tracks passed by each grid is mapped to the second channel of the multi-channel color space to obtain the first image.

5. The method according to claim 4, characterized in that The multi-channel color space is HSV; The obtaining of the first image includes: obtaining the first image using the HSV, and converting the first image using the HSV into a first image using RGB.

6. The method according to claim 4, characterized in that The characteristic image further includes a second image, and the rendering based on the statistical results and position information of each grid to obtain the characteristic image further includes: According to the position information of each grid, each grid is mapped to each pixel area of ​​the second image, and the speed information of the track passed by each grid is mapped to the single-channel color space adopted by the second image to obtain the second image.

7. The method according to claim 2, characterized in that The processing of the feature image to obtain lane line vector data includes: Using a feature extraction network to extract features from the first image and the second image respectively, to obtain a feature representation of the first image and a feature representation of the second image; fusing the feature representation of the first image and the feature representation of the second image to obtain a fused feature representation; The fused feature representation is decoded using a decoder based on a Transformer network to obtain the lane line vector data.

8. The method according to claim 3, characterized in that The obtaining of a lane-level map based on the lane line vector data further includes: Determine the height information of each lane line by using the height information of a neighboring trajectory point set of each lane line in the target trajectory data, wherein the neighboring trajectory point set of the lane line includes trajectory points within a preset distance range from the position of the lane line; Obtaining three-dimensional lane line vector data according to the height information and the lane line vector data from the bird's-eye view; Based on the three-dimensional lane line vector data, a three-dimensional lane-level map is obtained.

9. The method according to any one of claims 1 to 8, characterized in that Before generating the feature image based on the target trajectory data, the method includes: Get the trajectory data to be processed; Dividing the road corresponding to the trajectory data to be processed into a plurality of processing units according to a first size; The trajectory data corresponding to each of the processing units is used as the target trajectory data.

10. The method according to claim 9, characterized in that The obtaining of a lane-level map based on the lane line vector data includes: Fusing lane line vector data corresponding to adjacent processing units to obtain lane line vector data of the road; A lane-level map is obtained based on the lane line vector data of the road.

11. A map processing device, characterized in that: The device comprises: An image generating unit is configured to generate a characteristic image based on target trajectory data, wherein the target trajectory data includes a plurality of trajectories, and the characteristic image at least represents position information, density information, and direction information of the plurality of trajectories; A vector generation unit is configured to process the feature image to obtain lane line vector data, wherein the lane line vector data includes position information, direction information and type information of each lane line; The map processing unit is configured to obtain a lane-level map based on the lane line vector data.

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

Images