Road marking information determination method, vehicle, equipment and product
By performing binary classification and contour detection on the road area image, the coordinate information of the target type marking is obtained, which solves the problem that the geometric shape of the road marking in high-precision map is inconsistent with the actual situation, and improves the authenticity of the map and the effect of simulation tests. .
Patent Information
- Application Number
- CN202411496879.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-24
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
When building a high-precision map, preset road marking template information is directly used to replace the real marking information, resulting in the geometric shape of the road markings that does not match the actual situation, affecting the effectiveness of simulation testing.
By acquiring the road area image, performing binary classification segmentation processing and contour detection, the minimum external graphics of the target type marking are determined, thereby obtaining the target coordinate information of the target type marking and building a road map that is consistent with the actual situation.
It improves the consistency between the geometric shape of the target type marking in the road map and the actual situation, ensures the authenticity and accuracy of the map, and improves the effectiveness of simulation testing.
Smart Images

Figure CN118999600A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of image processing technology, and in particular to a method, vehicle, equipment and product for determining road marking information. Background Art
[0002] In the process of vehicle driving, in order to facilitate vehicle planning and control, most of the time, maps are constructed to assist in controlling the vehicle. For high-precision maps, they mainly focus on position accuracy, that is, the absolute position deviation from the actual target.
[0003] In addition to assisting in controlling vehicles, HD maps are mostly used to build large-scale simulation scenarios and for simulation testing of autonomous driving. This requires that in addition to high position accuracy, the geometric shape recognition of road markings on the road surface should also be as consistent as possible with the actual situation, which helps to ensure the authenticity of the simulation results. However, when building HD maps, in order to improve production efficiency, after determining the approximate position of the road markings, the preset road marking template information is directly used to replace the actual road marking information, resulting in the geometric shape of the road markings in the HD map not being consistent with the actual situation, thus affecting the effect of the simulation test. Summary of the invention
[0004] The embodiments of the present disclosure at least provide a method, vehicle, equipment and product for determining road marking information.
[0005] The present disclosure provides a method for determining road marking information, the method comprising: Obtain a road area image containing target type markings; Performing binary classification segmentation processing on the road area image to obtain a segmented image; the segmented image includes a first pixel point having a first pixel value and a second pixel point having a second pixel value, the first pixel value indicating that the corresponding first pixel point belongs to the target type marking, and the second pixel value indicating that the corresponding second pixel point does not belong to the target type marking; Performing contour detection on the segmented image to determine the minimum circumscribed figure of each closed contour; the minimum circumscribed figure is a figure corresponding to the shape of the target type marking line; Based on the minimum circumscribed graph, target coordinate information of each target type marking line is determined; the target coordinate information is used to construct the target type marking line in the road map.
[0006] In an optional implementation manner, the acquiring of a road area image containing target type markings includes: Acquire a road image of a target road section captured by a vehicle camera; The target type marking detection model is used to detect and crop a road area image including the target type marking from the road image.
[0007] In an optional implementation, the adopting the target type marking detection model to detect the road area image from the road image includes: Determine whether the vehicle is currently located within a road range corresponding to a target scene according to the vehicle positioning position information and pre-acquired road map information; the target scene is a scene where target type markings can be deployed; When it is determined that the target scene is located within the road range corresponding to the target scene, the target type marking detection model is used to detect the road area image from the road image.
[0008] In an optional implementation manner, the performing contour detection on the segmented image to determine the minimum circumscribed figure of each closed contour includes: Performing contour detection on the segmented image to obtain each initial closed contour; From the initial closed contours, filter out contours whose corresponding contour areas are smaller than a contour area threshold; Calculate the minimum circumscribed figure of the remaining closed contour.
[0009] In an optional implementation manner, after determining the minimum circumscribed figure of each closed contour and before determining the target coordinate information of each target type marking line based on the minimum circumscribed figure, the method further includes: According to the reference attribute characteristics of the target type marking, from each determined minimum circumscribed graph, the minimum circumscribed graph whose corresponding attribute characteristics do not conform to the reference attribute characteristics is filtered out to obtain a filtered minimum circumscribed graph; the reference attribute characteristics include shape and position characteristics; The step of determining target coordinate information of each target type marking line based on the minimum circumscribed graphic includes: Based on the filtered minimum circumscribed graph, target coordinate information of each target type marking line is determined.
[0010] In an optional implementation manner, after determining the minimum circumscribed figure of each closed contour and before determining the coordinate information of each target type marking line based on the minimum circumscribed figure, the method further includes: Calculating the ratio of pixels corresponding to the first pixel in each minimum circumscribed graph; the pixel ratio is the ratio of the number of the first pixel to the total number of pixels in the segmented image, or the ratio between the number of the first pixel and the number of the second pixel; If the pixel ratio is less than the corresponding ratio threshold, the corresponding minimum circumscribed graph is filtered out to obtain a filtered minimum circumscribed graph; The step of determining target coordinate information of each target type marking line based on the minimum circumscribed graphic includes: Based on the filtered minimum circumscribed graph, target coordinate information of each target type marking line is determined.
[0011] In an optional implementation manner, the method further includes: Acquire road point cloud data scanned by a vehicle radar, and convert the point cloud data into a point cloud image; Using the target type marking detection model, detect the image area where the target type marking is located from the point cloud image, and determine a first coordinate range corresponding to the image area in the world coordinate system; Whether to adopt the determined target coordinate information of each target type marking line is determined based on the overlap between a second coordinate range of the road area image detected from the road image in the world coordinate system and the first coordinate range.
[0012] In an optional implementation manner, after determining the target coordinate information of each target type marking line, the method further includes: Based on the target coordinate information of each target type marking line, as well as the collected road image and vehicle positioning data of the target road section, a target map for simulation testing is constructed; Based on the target map, a simulation test of autonomous driving is performed on the vehicle.
[0013] The present disclosure also provides a vehicle, including a controller, wherein the controller includes: a memory for storing machine-readable instructions; and The processor is used to call and execute the machine-readable instructions stored in the memory to implement the steps in any possible implementation of the above-mentioned road marking information determination method.
[0014] An embodiment of the present disclosure also provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the machine-readable instructions are executed by the processor to perform steps in any possible implementation of the above-mentioned road marking information determination method.
[0015] The embodiments of the present disclosure further provide a computer program product, on which a computer program is stored. When the computer program is executed by a processor, the steps in any possible implementation of the above-mentioned method for determining road marking information are executed.
[0016] The road marking information determination method, vehicle, equipment and product provided by the embodiments of the present disclosure can obtain a road area image containing a target type marking; then perform binary segmentation processing on the road area image to obtain a segmented image; wherein the segmented image includes a first pixel point having a first pixel value and a second pixel point having a second pixel value, the first pixel value indicating that the corresponding first pixel point belongs to the target type marking, and the second pixel value indicating that the corresponding second pixel point does not belong to the target type marking; then perform contour detection on the segmented image to determine the minimum circumscribed figure of each closed contour; and then determine the target coordinate information of each target type marking based on the minimum circumscribed figure; here, the target coordinate information is used to construct the target type marking in the road map.
[0017] In this way, for the road area image containing the target type marking, a binary segmentation process is adopted to obtain a segmented image, so that the pixel points belonging to the target type marking and the pixel points not belonging to the target type marking can be distinguished by different pixel values, and then the segmented image is subjected to contour detection and the minimum circumscribed figure of each closed contour is calculated, so that the target coordinate information of the target type marking can be obtained. Obtaining the target coordinate information in the above manner helps to improve the accuracy of the target coordinate information. Furthermore, the target type marking in the road map is constructed through the target coordinate information, so that the geometric shape of the target type marking in the road map is highly consistent with the actual situation, which helps to ensure the authenticity and accuracy of the road map.
[0018] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and are not intended to limit the technical solutions of the present disclosure.
[0019] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.
[0021] Figure 1 A flow chart of a method for determining road marking information provided by an embodiment of the present disclosure is shown; Figure 2A schematic diagram of a road area image provided by an embodiment of the present disclosure is shown; Figure 3 Shows Figure 2 A schematic diagram of a segmented image corresponding to the road area image shown; Figure 4 Shows Figure 3 Schematic diagram of each minimum circumscribed figure in the segmented image shown; Figure 5 A flow chart of a road marking information verification method provided by an embodiment of the present disclosure is shown; Figure 6 A flow chart of a method for determining and simulating road marking information provided by an embodiment of the present disclosure is shown; Figure 7 A schematic diagram of a process of determining and simulating road marking information provided by an embodiment of the present disclosure is shown; Figure 8 One of the schematic diagrams of a road marking line information determination device provided by an embodiment of the present disclosure is shown; Fig. 9 A second schematic diagram of a road marking line information determination device provided by an embodiment of the present disclosure is shown; Fig.10 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0022] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the present disclosure for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.
[0023] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0024] The term "and / or" herein only describes an association relationship, indicating that three relationships may exist. For example, A and / or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.
[0025] Research has found that in addition to assisting in controlling vehicles, HD maps are mostly used to build large-scale simulation scenarios and for simulation testing of autonomous driving. This requires that in addition to high position accuracy, the elements of HD maps also need to identify the geometric shape of road markings on the road surface as much as possible in a way that is consistent with the actual situation, which helps to ensure the authenticity of the simulation results. However, when building HD maps, in order to improve production efficiency, after determining the approximate position of road markings, most of them use template matching to directly replace the actual road marking information with the preset road marking template information. In real scenarios, there are often multiple forms for a certain type of marking, and the number and spacing of the lines are different in different forms. This will cause the geometric shape of the road markings in the HD map to be inconsistent with the actual situation, thus affecting the effect of the simulation test.
[0026] Based on the above research, the present disclosure provides a method for determining road marking information, which can accurately extract the target coordinate information of the target type marking, and then when constructing the target type marking in the road map through the target coordinate information, the geometric shape of the target type marking in the road map is highly consistent with the actual situation, which helps to ensure the authenticity and accuracy of the road map.
[0027] To facilitate understanding of this embodiment, a method for determining road marking information disclosed in the embodiment of the present disclosure is first introduced in detail. The execution subject of the method for determining road marking information provided in the embodiment of the present disclosure is generally an electronic device with certain computing capabilities. The electronic device may be a server or a terminal device or other device. If it is a server, the server may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud storage, big data and artificial intelligence platforms. If it is a terminal device or other device, the terminal device may be a user equipment (UserEquipment, UE), a mobile device, a terminal, a computing device, and a vehicle-mounted device. Other processing devices may be devices including a processor and a memory, which are not limited here. In some possible implementations, the method for determining road marking information may be implemented by a processor calling computer-readable instructions stored in a memory.
[0028] A method for determining road marking information provided by an embodiment of the present disclosure is described below in conjunction with the accompanying drawings.
[0029] See also Figure 1 As shown, it is a flow chart of a method for determining road marking information provided by an embodiment of the present disclosure, such as Figure 1 As shown in , the road marking information determination method provided by the embodiment of the present disclosure includes steps S101 to S104, wherein: S101: Acquire a road area image containing target type markings.
[0030] Here, the target type marking line is located on the road surface of the target road section, and is used to assist the vehicle in traveling through the target road section.
[0031] The road marking determination method of the embodiment of the present disclosure can be applicable to road markings of various shapes. The shape of the target type marking is not limited in the embodiment of the present disclosure. For example, the target type marking can be a road marking of a rectangular shape or other shapes.
[0032] In some embodiments, the target type marking may be a rectangular road marking, such as a lateral deceleration marking, a crosswalk marking, a stop line, a stop and yield line, etc.
[0033] In other embodiments, the target type marking may also be road markings of other shapes, such as diamond-shaped pedestrian crossing warning markings, arrow-shaped direction indication lines, and the like.
[0034] In practical applications, a vehicle is often provided with a camera, and the image of the road area can be acquired through the vehicle camera.
[0035] Specifically, a road image of a target road section captured by a vehicle camera may be acquired; and a target type marking detection model may be used to detect and crop a road area image including the target type marking from the road image.
[0036] In the above steps, the road image of the target road section can be first obtained by photographing with a vehicle camera, and then the road area image including the target type markings can be detected from the road image by a target type marking detection model, and then the road area image can be cropped from the road image.
[0037] It can be understood that the road area image is an image corresponding to the area including the target type marking line in the road image.
[0038] The target type marking detection model is used to detect whether the target type marking exists in the road image. It can be understood that if it exists, the detection result is the road area image, and if it does not exist, the detection result is empty.
[0039] Here, the target type marking detection model can be pre-trained. Specifically, the target type marking detection model can be trained through the following steps: Acquire multiple sample road images and at least one annotated detection frame information corresponding to each sample road image, wherein the annotated detection frame information includes the detection frame position in the sample road image of the detection frame for annotating the target type line marking detection in the sample road image. Use the multiple sample road images to perform neural network training. Specifically, the sample road image can be used as the input of the neural network and the annotated detection frame information can be used as the output of the neural network to train the neural network. Alternatively, the sample road image can be used as the input of the neural network, and the sample road image can be detected by the neural network to obtain predicted annotated detection frame information. The neural network can be converged by the annotated detection frame information and the predicted annotated detection frame information to complete the training and obtain the target type line marking detection model.
[0040] In this way, the road image of the target road section can be obtained with the help of the camera set up on the vehicle itself, and then the road area image where the target type marking is located can be detected from the road image through the deep learning detection model, which helps to improve the accuracy of determining the road area image.
[0041] In other implementations, the road area image where the target type marking line is located may be detected from the road image by other means, such as a target positioning algorithm.
[0042] It should be noted that vehicles are often equipped with various sensor components such as cameras, radars, global positioning systems (GPS), inertial measurement units (IMU), etc. Before taking road images through the vehicle camera, the various sensor components can be calibrated and registered in advance to obtain calibration and registration parameters. The calibration and registration parameters include the conversion relationship between the camera coordinate system and the vehicle coordinate system, and the conversion relationship between the vehicle coordinate system and the world coordinate system.
[0043] In some possible implementations, the acquired road image may be a bird's-eye view, that is, an image under a bird's-eye-view (BEV) perspective.
[0044] In actual applications, vehicles mostly capture the road image from a forward perspective, and target type markings are mostly printed on the ground. Therefore, the target type markings may be presented at an inclined angle in the road image. By converting to the BEV perspective, the target type markings are presented more regularly in the road image from the BEV perspective.
[0045] Specifically, when the perspective conversion is performed, the road image under the perspective of the BEV can be obtained based on the correspondence between the shooting perspective of the vehicle camera and the perspective of the BEV.
[0046] In this way, considering that the target type markings are printed signs on the ground, obtaining the road image from the BEV perspective can help focus more on the target type markings and effectively improve the accuracy of road area image recognition.
[0047] Correspondingly, when the target type road marking detection model is obtained through training, the sample road image may specifically be a sample road image from a BEV perspective.
[0048] When the target type line marking detection model is used to detect the road area image from the road image, specifically, based on the vehicle positioning position information and the pre-acquired road map information, it is determined whether the vehicle is currently located within the road range corresponding to the target scene; the target scene is a scene where target type line markings can be deployed; when it is determined that the vehicle is currently located within the road range corresponding to the target scene, the target type line marking detection model is used to detect the road area image from the road image.
[0049] In actual applications, target type markings are generally located in some specific scenes. For example, for lateral deceleration markings, the target scenes are generally a certain distance before the intersection, at bends, near special areas (such as schools, hospitals, etc. where vehicles need to slow down), downhill sections, etc., and are generally not located within the scope of intersections; for stop lines, the target scenes are generally at the entrance of intersections, and are generally not located within the scope of intersections; for example, for pedestrian crossing warning markings, the target scenes are generally a certain distance before the pedestrian crossing, and are generally not located within the scope of intersections.
[0050] Taking into account that the vehicle may not have reached the target scene when capturing the road image, the embodiment of the present disclosure determines whether the vehicle is located within the road range corresponding to the target scene when capturing the road image (i.e., the vehicle reaches the target scene where target type markings can be deployed) based on the vehicle positioning position information and pre-acquired road map information. When it is determined that the vehicle is within the road range corresponding to the target scene, a target type marking detection model is used to detect the road area image from the road image.
[0051] The vehicle positioning position information refers to the position of the vehicle when the corresponding road image is captured, which can be specifically determined by data collected by the GPS and IMU set on the vehicle.
[0052] In this way, when it is determined that the vehicle is located within the road range corresponding to the target scene when taking the road image, the target type marking detection model is used to perform road area image detection, avoiding the extra workload caused by detecting each road image.
[0053] For example, please refer to Figure 2 , Figure 2 Schematic diagram of a road area image provided by an embodiment of the present disclosure. This example is described by taking the target type marking as a rectangular horizontal deceleration marking as an example. Figure 2 As shown in , the road area image 200 includes a plurality of transverse speed reduction markings.
[0054] S102: Perform binary classification segmentation processing on the road area image to obtain a segmented image; the segmented image includes a first pixel point with a first pixel value and a second pixel point with a second pixel value, the first pixel value indicates that the corresponding first pixel point belongs to the target type marking, and the second pixel value indicates that the corresponding second pixel point does not belong to the target type marking.
[0055] In this step, for each pixel point in the road area image, determine whether the pixel point belongs to the target type marking; if it does, adjust the pixel value of the pixel point to the first pixel value; if it does not, adjust the pixel value of the pixel point to the second pixel value; based on the adjusted pixel values of each pixel point, obtain the segmented image.
[0056] Optionally, a trained binary classification segmentation model may be used to perform binary classification segmentation processing on the road area image to obtain a segmented image.
[0057] In practical applications, the road surface is generally a dark asphalt road surface, and the target type marking line is generally white or yellow. In order to make the segmented image better reflect the real scene, the first pixel value can be set to be greater than the second pixel value.
[0058] Here, the greater the pixel value deviation between the first pixel value and the second pixel value, the more obvious the deviation between the first pixel point and the second pixel point presented in the segmented image.
[0059] Exemplarily, the first pixel value may be set to 255 (ie, white) and the second pixel value may be set to 0 (ie, black), which helps to distinguish the first pixel point from the second pixel point.
[0060] Following the above example, you can also refer to Figure 3 , Figure 3 for Figure 2 Schematic diagram of the segmented image corresponding to the road area image shown in FIG. Take the first pixel value as 255 (ie, white) and the second pixel value as 0 (ie, black) as an example for explanation. Figure 3 As shown in , the black part in the segmented image 300 is actually the road surface, and the white part may actually be the target type marking line.
[0061] S103: Performing contour detection on the segmented image to determine the minimum circumscribed figure of each closed contour; the minimum circumscribed figure is a figure corresponding to the shape of the target type marking line.
[0062] In this step, contour detection may be performed on the segmented image to obtain each closed contour in the segmented image, and then the minimum circumscribed figure of each closed contour may be calculated.
[0063] Here, the minimum circumscribed figure is a figure corresponding to the shape of the target type marking. For example, if the target type marking is a rectangular shape, then the minimum circumscribed rectangle of the target type marking can be determined accordingly; if the target type marking is a diamond shape, then the minimum circumscribed diamond of the target type marking can be determined accordingly; if the target type marking is an arrow shape, then the minimum circumscribed arrow of the target type marking can be determined accordingly.
[0064] Specifically, contour detection is performed on the segmented image to obtain initial closed contours; contours whose corresponding contour areas are smaller than a contour area threshold are filtered out from the initial closed contours; and the minimum circumscribed graph of the remaining closed contours is calculated.
[0065] It can be understood that here contour detection is performed on each area formed by the convergence of multiple first pixel points in the segmented image to obtain each initial closed contour, which may be the contour of a target type marking line or the contour of other graphics.
[0066] In practical applications, due to weather, lighting and other reasons, there may be light spots in the road image taken by the vehicle camera. The area of these light spots is generally small. For example, refer again to Figure 3 It can be seen that there are multiple white dots in the segmented image 300, such as 311, 312, etc. It can be seen that although these white dots are also judged as the first pixel points, they do not actually belong to the target type marking line. Therefore, by filtering out the contours whose corresponding contour areas are smaller than the contour area threshold, it is helpful to eliminate the influence of other factors in the shooting process.
[0067] The specific value of the contour area threshold may be determined according to actual contour screening needs and is not specifically limited here.
[0068] Then, see Figure 4 , Figure 4 for Figure 3 Schematic diagram of each minimum circumscribed graph in the segmented image shown in FIG. Figure 4 As shown in , after filtering out contours whose corresponding contour areas are smaller than the contour area threshold, 7 closed contours remain in the segmented image 300 , and thus the minimum circumscribed rectangles 321 - 327 of the 7 closed contours can be obtained respectively.
[0069] In this way, for each initial closed contour, by filtering out contours whose corresponding contour areas are smaller than the contour area threshold, and then calculating the minimum circumscribed figure of the remaining closed contours, it is helpful to eliminate the influence of weather environment and other factors during the shooting process of the vehicle camera, and improve the accuracy of the target coordinate information of the target type marking.
[0070] S104: Determine target coordinate information of each target type marking line based on the minimum circumscribed graph; the target coordinate information is used to construct the target type marking line in the road map.
[0071] In this step, the pixel coordinates of each vertex of the minimum circumscribed figure in the road area image may be determined as the target coordinate information of the target type marking line corresponding to the minimum circumscribed figure.
[0072] Exemplarily, for the target type marking line in a rectangular shape, the pixel coordinates of the four vertices of the corresponding minimum circumscribed rectangle can be determined; for the target type marking line in an arrow shape, the pixel coordinates of each vertex of the corresponding minimum circumscribed arrow can be determined.
[0073] Here, the target type markings in the road map can be subsequently constructed through the target coordinate information. In actual applications, there are often multiple target type markings in the road map, so it is often necessary to use multiple road area images to determine the target coordinate information of all target type markings in the road map.
[0074] Therefore, in order to facilitate the subsequent processing of the target coordinate information of each target type marking line determined by multiple road area images, the pixel point coordinates of each vertex of the minimum circumscribed figure can be converted to a unified coordinate system, such as the world coordinate system, to obtain absolute precision coordinates.
[0075] Specifically, according to the above content, the calibration and registration parameters are obtained by pre-calibration and registration, and the calibration and registration parameters include the conversion relationship between the camera coordinate system and the vehicle coordinate system, and the conversion relationship between the vehicle coordinate system and the world coordinate system. Accordingly, the pixel point coordinates of each vertex of the minimum circumscribed figure can be converted into absolute precision coordinates based on the calibration and registration parameters and the vehicle positioning position information.
[0076] The vehicle positioning position information refers to the position of the vehicle when the corresponding road image is captured, which can be specifically determined by data collected by the GPS and IMU set on the vehicle.
[0077] The absolute precision coordinates refer to the coordinates of each vertex of the minimum circumscribed figure in the world coordinate system. Optionally, the absolute precision coordinates are specifically in the form of longitude and latitude.
[0078] When the pixel point coordinate values of each vertex of the minimum circumscribed figure are converted into absolute precision coordinate values based on the calibration registration parameters and the vehicle positioning position information, specifically, the conversion relationship between the camera coordinate system and the vehicle coordinate system can be obtained based on the calibrated camera parameter matrix, so that the pixel point coordinates can be converted to the vehicle coordinate system (such as a coordinate system with the center of the vehicle body as the coordinate origin) to obtain the intermediate coordinates in the vehicle coordinate system. Combined with the vehicle positioning position information, the intermediate coordinates can be converted from the vehicle coordinate system to the world coordinate system.
[0079] In some possible implementations, after determining the minimum circumscribed figure of each closed contour and before determining the target coordinate information of each target type marking line based on the minimum circumscribed figure, the method further includes: According to the reference attribute characteristics of the target type marking, the minimum circumscribed figures whose corresponding attribute characteristics do not conform to the reference attribute characteristics are filtered out from the determined minimum circumscribed figures to obtain the filtered minimum circumscribed figures; the reference attribute characteristics include shape and position characteristics.
[0080] Correspondingly, determining the target coordinate information of each target type marking line based on the minimum circumscribed figure includes: determining the target coordinate information of each target type marking line based on the filtered minimum circumscribed figure.
[0081] The reference attribute feature corresponds to the shape feature of the target type marking line.
[0082] Exemplarily, taking the target type marking as a horizontal deceleration marking in a rectangular shape as an example, the shape feature indicated by the reference attribute feature is that the rectangle corresponding to the long side of the minimum circumscribed rectangle is longitudinally parallel to the driving direction of the vehicle, and the position feature indicated by the reference attribute feature is that the minimum circumscribed rectangle is not located at the edge of the segmented image.
[0083] Here again you can see Figure 4 It can be seen that the rectangles corresponding to the long sides of the minimum enclosing rectangle 321 and the minimum enclosing rectangle 327 are vertically perpendicular to the driving direction of the vehicle and are located on both sides of the segmented image. It can be considered that the minimum enclosing rectangle 321 and the minimum enclosing rectangle 327 do not meet the reference attribute characteristics, so the minimum enclosing rectangle 321 and the minimum enclosing rectangle 327 can be filtered out.
[0084] In this way, the position and shape features can be combined to filter out the minimum circumscribed graphics that do not meet the reference attribute features, which helps to exclude the minimum circumscribed graphics that obviously do not belong to the target type marking, and then determine the coordinates based on the filtered minimum circumscribed graphics, which helps to improve the accuracy of determining the target coordinate information of each target type marking based on the minimum circumscribed graphics.
[0085] In some other possible implementations, after determining the minimum circumscribed figure of each closed contour and before determining the coordinate information of each target type marking line based on the minimum circumscribed figure, the method further includes: Calculating the ratio of pixels corresponding to the first pixel in each minimum circumscribed graph; the pixel ratio is the ratio of the number of the first pixel to the total number of pixels in the segmented image, or the ratio between the number of the first pixel and the number of the second pixel; If the pixel ratio is smaller than the corresponding ratio threshold, the corresponding minimum circumscribed graphic is filtered out.
[0086] Correspondingly, determining the target coordinate information of each target type marking line based on the minimum circumscribed figure includes: determining the target coordinate information of each target type marking line based on the filtered minimum circumscribed figure.
[0087] Here, the first pixel point belongs to the target type marking line, and the second pixel point does not belong to the target type marking line. Therefore, in the minimum circumscribed figure belonging to the target type marking line, the number of first pixel points is often greater than the number of second pixel points. Therefore, the proportion of pixels corresponding to the minimum circumscribed figure belonging to the target type marking line is often greater than the corresponding proportion threshold.
[0088] In practical applications, there is often a situation where a certain target type marking line is not fully captured, or the target type marking line itself is damaged, resulting in the pixel point ratio corresponding to the minimum circumscribed pattern being less than the corresponding ratio threshold.
[0089] The specific value of the ratio threshold may be determined according to the actual minimum external graphic screening requirement, and is not specifically limited here.
[0090] Here again you can see Figure 4 It can be seen that the lateral speed reduction marking corresponding to the minimum circumscribed rectangle 326 has a certain wear in the actual environment, resulting in the corresponding pixel point ratio being less than the corresponding ratio threshold, so the minimum circumscribed rectangle 326 can be filtered out.
[0091] In this way, the minimum bounding box graphics corresponding to the target type markings that are not fully photographed or are damaged can be filtered out according to the ratio of pixel points corresponding to the first pixel point in the minimum bounding box graphics, thereby ensuring the consistency of the screened minimum bounding box graphics with the actual situation, and then determining the coordinates based on the filtered minimum bounding box graphics, which helps to improve the accuracy of determining the target coordinate information of each target type marking based on the minimum bounding box graphics.
[0092] See again Figure 4 Through the above two filterings, the minimum bounding rectangle 321, the minimum bounding rectangle 327 and the minimum bounding rectangle 326 in the segmented image 300 are filtered out, and the remaining minimum bounding rectangles are 322~325. Then, based on the minimum bounding rectangles 322~325, the target coordinate information of each corresponding target type marking line can be determined.
[0093] After obtaining the target coordinate information of each target type marking, the target coordinate information can be verified to determine whether to adopt the determined target coordinate information of each target type marking for subsequent map construction. Specifically, road point cloud data scanned by a vehicle radar can be obtained, and the target coordinate information can be verified based on the road point cloud data.
[0094] Optionally, manual verification may be performed by processing the road point cloud data and the road area image through a target tool, thereby mapping the point cloud onto the road area image, so that it can be intuitively seen whether there is a deviation between the target coordinate information and the road point cloud data.
[0095] Optionally, you can also perform automatic verification. For details, see Figure 5 , is a flow chart of a road marking information verification method provided by an embodiment of the present disclosure, such as Figure 5 As shown in , it includes steps S501 to S503, wherein: S501: Acquire road point cloud data scanned by a vehicle radar, and convert the point cloud data into a point cloud image.
[0096] Here, the road point cloud data includes a plurality of sampling points, each of which carries reflection intensity.
[0097] Specifically, based on the reflection intensity carried by each sampling point and the mapping relationship between the reflection intensity and the pixel value, the road point cloud data is subjected to overhead projection processing to obtain a grayscale image corresponding to the road point cloud data as the point cloud image.
[0098] The vehicle radar may be a laser radar or other radar capable of collecting point cloud data, which is not specifically limited here.
[0099] S502: Using the target type marking detection model, detect an image area where the target type marking is located from the point cloud image, and determine a first coordinate range corresponding to the image area in a world coordinate system.
[0100] Here, the specific process of using the target type marking detection model to detect the image area where the target type marking is located from the point cloud image is similar to the process of using the target type marking detection model to detect the road area image from the road image in the aforementioned embodiment. The specific description can be referred to the aforementioned embodiment and will not be repeated here.
[0101] S503: Determine whether to adopt the determined target coordinate information of each target type marking line according to the overlap between the second coordinate range of the road area image detected from the road image in the world coordinate system and the first coordinate range.
[0102] In this step, when the degree of overlap between the second coordinate range and the first coordinate range is greater than a preset overlap threshold, it can be determined that the target coordinate information of each target type marking line can be adopted; otherwise, it is determined that the target coordinate information of each target type marking line cannot be adopted.
[0103] Here, since the point cloud image is obtained by mapping the reflection intensity to the pixel value, this will cause a large deviation in the reflection intensity of two adjacent pixel points themselves, but the deviation is small after conversion to pixel values, which will result in lower boundary clarity in the point cloud image. Therefore, when the degree of overlap between the second coordinate range and the first coordinate range is greater than the preset overlap threshold, the target coordinate information determined by the road area image shall prevail, and the target coordinate information shall be adopted subsequently.
[0104] The specific value of the preset overlap threshold can be determined according to actual information usage needs and is not specifically limited here.
[0105] In this way, the accuracy of the target coordinate information can be verified through the road point cloud data to determine whether the target coordinate information of each target type marking should be adopted in the future, which helps to ensure the authenticity and accuracy of the road map when constructing the target type markings in the road map through the target coordinate information.
[0106] The method for determining road marking information provided by the embodiment of the present disclosure adopts a binary segmentation process to obtain a segmented image for a road area image containing a target type marking, so that pixel points belonging to the target type marking and pixel points not belonging to the target type marking can be distinguished by different pixel values, and then contour detection is performed on the segmented image and the minimum circumscribed figure of each closed contour is calculated, so that target coordinate information of the target type marking can be obtained. Obtaining the target coordinate information in the above manner helps to improve the accuracy of the target coordinate information. Furthermore, the target type marking in the road map is constructed by the target coordinate information, so that the geometric shape of the target type marking in the road map is highly consistent with the actual situation, which helps to ensure the authenticity and accuracy of the road map.
[0107] See also Figure 6 As shown, Figure 6 A flowchart of a method for determining and simulating road marking information provided by an embodiment of the present disclosure is shown in FIG. Figure 6 As shown in , it includes steps S601 to S606, wherein: S601: Acquire a road area image containing target type markings.
[0108] S602: Perform binary classification segmentation processing on the road area image to obtain a segmented image; the segmented image includes a first pixel point with a first pixel value and a second pixel point with a second pixel value, the first pixel value indicates that the corresponding first pixel point belongs to the target type marking, and the second pixel value indicates that the corresponding second pixel point does not belong to the target type marking.
[0109] S603: Perform contour detection on the segmented image to determine the minimum circumscribed figure of each closed contour; the minimum circumscribed figure is a figure corresponding to the shape of the target type marking line.
[0110] S604: Determine target coordinate information of each target type marking line based on the minimum circumscribed graph; the target coordinate information is used to construct the target type marking line in the road map.
[0111] Among them, the description of step S601 to step S604 can refer to the description of step S101 to step S104, and can achieve the same technical effects and solve the same technical problems, and will not be repeated here.
[0112] S605: Based on the target coordinate information of each target type marking, and the collected road image and vehicle positioning data of the target road section, a target map for simulation testing is constructed.
[0113] In this step, the corresponding target type markings can be constructed based on the target coordinate information of each target type marking, and then combined with the collected road image and vehicle positioning data of the target section to construct a target map for simulation testing.
[0114] S606: Based on the target map, a simulation test of autonomous driving is performed on the vehicle.
[0115] Here, a three-dimensional simulation scene corresponding to the real scene can be constructed based on the target map and other scene elements in the real scene, so that the vehicle can be simulated tested for autonomous driving in the three-dimensional simulation scene.
[0116] In order to more clearly demonstrate the process of road marking information determination and simulation testing, you can also refer to Figure 7 . Figure 7 A schematic diagram of a simulation test process provided by an embodiment of the present disclosure. Figure 7 As shown in, a road image of a target road section taken by a vehicle camera is obtained; a target type marking detection model is used to detect and crop a road area image where the target type marking is located from the road image; and a binary classification segmentation process is performed on the road area image to obtain a segmented image.
[0117] Then, contour detection is performed on the segmented image to obtain initial closed contours; contours whose corresponding contour areas are smaller than a contour area threshold are filtered out from the initial closed contours; and the minimum circumscribed graph of the remaining closed contours is calculated.
[0118] Next, each minimum bounding box is filtered. Specifically, the minimum bounding box whose corresponding attribute characteristics do not conform to the reference attribute characteristics can be filtered out, and the minimum bounding box whose corresponding pixel point ratio is less than the corresponding ratio threshold can also be filtered out. Then, based on the filtered minimum bounding box, the target coordinate information of each target type marking line can be determined.
[0119] To ensure the accuracy of the target coordinate information, the road point cloud data scanned by the vehicle radar can also be obtained and converted into a point cloud image, based on which it is determined whether to adopt the target coordinate information of each target type marking.
[0120] In the case where the target coordinate information of each target type marking can be used, a target map for simulation testing is constructed based on the target coordinate information of each target type marking, and the collected road image and vehicle positioning data of the target road section; based on the target map, a simulation test of autonomous driving is performed on the vehicle. The description of the specific steps can be found in the above embodiments, which will not be repeated here.
[0121] The above road marking information determination and simulation test method, for the road area image containing the target type marking, adopts a binary segmentation process to obtain a segmented image, so that the pixel points belonging to the target type marking and the pixel points not belonging to the target type marking can be distinguished by different pixel values, and then the segmented image is contour detected and the minimum circumscribed figure of each closed contour is calculated, so that the target coordinate information of the target type marking can be obtained, and then based on the target coordinate information and the collected road image and vehicle positioning data of the target road section, a target map for simulation testing is constructed, and based on the target map, a simulation test of automatic driving of the vehicle is performed. Obtaining the target coordinate information in the above manner helps to improve the accuracy of the target coordinate information. Further, the target type marking in the road map is constructed by the target coordinate information, so that the geometric shape of the target type marking in the road map is highly consistent with the actual situation, which helps to ensure the authenticity and accuracy of the road map. Moreover, the simulation test of automatic driving of the vehicle using the target map obtained in the above manner helps to ensure the consistency of the target type marking in the simulation scene with the real scene, thereby improving the simulation test effect and ensuring the reliability of the simulation test results.
[0122] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.
[0123] Based on the same inventive concept, the embodiment of the present disclosure also provides a road marking information determination device corresponding to the road marking information determination method. Since the principle of solving the problem by the road marking information determination device in the embodiment of the present disclosure is similar to the above-mentioned road marking information determination method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
[0124] See also Figure 8 and Fig. 9 , Figure 8 This is one of the schematic diagrams of a device for determining road marking information provided by an embodiment of the present disclosure, Fig. 9 This is a second schematic diagram of a road marking information determination device provided by an embodiment of the present disclosure. Figure 8 As shown in , the road marking information determination device 800 provided by the embodiment of the present disclosure includes: An image acquisition module 810 is used to acquire an image of a road area containing target type markings; The image segmentation module 820 is used to perform a binary classification segmentation process on the road area image to obtain a segmented image; the segmented image includes a first pixel point having a first pixel value and a second pixel point having a second pixel value, wherein the first pixel value indicates that the corresponding first pixel point belongs to the target type marking line, and the second pixel value indicates that the corresponding second pixel point does not belong to the target type marking line; The contour detection module 830 is used to perform contour detection on the segmented image to determine the minimum circumscribed figure of each closed contour; the minimum circumscribed figure is a figure corresponding to the shape of the target type marking line; The information determination module 840 is used to determine the target coordinate information of each target type marking line based on the minimum circumscribed figure; the target coordinate information is used to construct the target type marking line in the road map.
[0125] In an optional implementation manner, the image acquisition module 810 is specifically used for: Acquire a road image of a target road section captured by a vehicle camera; The target type marking detection model is used to detect and crop a road area image including the target type marking from the road image.
[0126] In an optional implementation manner, when the image acquisition module 810 is used to detect the road area image from the road image using the target type marking detection model, it is specifically used to: Determine whether the vehicle is currently located within a road range corresponding to a target scene according to the vehicle positioning position information and pre-acquired road map information; the target scene is a scene where target type markings can be deployed; When it is determined that the target scene is located within the road range corresponding to the target scene, the target type marking detection model is used to detect the road area image from the road image.
[0127] In an optional implementation manner, the contour detection module 830 is specifically used for: Performing contour detection on the segmented image to obtain each initial closed contour; From the initial closed contours, filter out contours whose corresponding contour areas are smaller than a contour area threshold; Calculate the minimum circumscribed figure of the remaining closed contour.
[0128] In an optional implementation, Fig. 9 As shown in , the road marking information determination device 800 further includes a first filtering module 850, and the first filtering module 850 is used to: According to the reference attribute characteristics of the target type marking, from each determined minimum circumscribed graph, the minimum circumscribed graph whose corresponding attribute characteristics do not conform to the reference attribute characteristics is filtered out to obtain a filtered minimum circumscribed graph; the reference attribute characteristics include shape and position characteristics; The information determination module 840 is specifically used for: Based on the filtered minimum circumscribed graph, target coordinate information of each target type marking line is determined.
[0129] In an optional implementation, Fig. 9 As shown in , the road marking information determination device 800 further includes a second filtering module 860, and the second filtering module 860 is used to: Calculating the ratio of pixels corresponding to the first pixel in each minimum circumscribed graph; the pixel ratio is the ratio of the number of the first pixel to the total number of pixels in the segmented image, or the ratio between the number of the first pixel and the number of the second pixel; If the pixel ratio is less than the corresponding ratio threshold, the corresponding minimum circumscribed graph is filtered out to obtain a filtered minimum circumscribed graph; The information determination module 840 is specifically used for: Based on the filtered minimum circumscribed graph, target coordinate information of each target type marking line is determined.
[0130] In an optional implementation, Fig. 9 As shown in , the road marking information determination device 800 further includes an information verification module 870, and the information verification module 870 is used to: Acquire road point cloud data scanned by a vehicle radar, and convert the point cloud data into a point cloud image; Using the target type marking detection model, detect the image area where the target type marking is located from the point cloud image, and determine a first coordinate range corresponding to the image area in the world coordinate system; Whether to adopt the determined target coordinate information of each target type marking line is determined based on the overlap between a second coordinate range of the road area image detected from the road image in the world coordinate system and the first coordinate range.
[0131] In an optional implementation, Fig. 9 As shown in , the road marking information determination device 800 further includes a simulation test module 880, and the simulation test module 880 is used to: Based on the target coordinate information of each target type marking line, as well as the collected road image and vehicle positioning data of the target road section, a target map for simulation testing is constructed; Based on the target map, a simulation test of autonomous driving is performed on the vehicle.
[0132] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference may be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.
[0133] The road marking information determination device provided by the embodiment of the present disclosure adopts a binary segmentation process to obtain a segmented image for a road area image containing a target type marking, so that pixel points belonging to the target type marking and pixel points not belonging to the target type marking can be distinguished by different pixel values, and then contour detection is performed on the segmented image and the minimum circumscribed figure of each closed contour is calculated to obtain the target coordinate information of the target type marking. Obtaining the target coordinate information in the above manner helps to improve the accuracy of the target coordinate information. Furthermore, the target type marking in the road map is constructed by the target coordinate information, so that the geometric shape of the target type marking in the road map is highly consistent with the actual situation, which helps to ensure the authenticity and accuracy of the road map.
[0134] The disclosed embodiment also provides a vehicle, including a controller, the controller including: a memory for storing machine-readable instructions; and a processor for calling and executing the machine-readable instructions stored in the memory to implement the steps of the road marking information determination method as described in the above method embodiment. The vehicle may be a vehicle equipped with a variety of sensor components such as a camera, a radar, a global positioning system (GPS), an inertial measurement unit (IMU), etc.
[0135] Corresponding to the above-mentioned road marking information determination method, the embodiment of the present disclosure further provides an electronic device 1000, such as Fig.10 As shown in FIG. 1 , it is a schematic diagram of the structure of an electronic device 1000 provided in an embodiment of the present disclosure, including: Processor 1010, memory 1020, and bus 1030. The memory 1020 is used to store execution instructions, including internal memory 1021 and external memory 1022; the internal memory 1021 is also called internal memory, which is used to temporarily store the operation data in the processor 1010 and the data exchanged with the external memory 1022 such as a hard disk. The processor 1010 exchanges data with the external memory 1022 through the internal memory 1021.
[0136] In the embodiment of the present application, the memory 1020 is specifically used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 1010. That is, when the electronic device 1000 is running, the processor 1010 communicates with the memory 1020 through the bus 1030, or the processor 1010 communicates with the memory 1020 in other ways, so that the processor 1010 executes the application code stored in the memory 1020, and then executes the steps of the road marking information determination method described in any of the above embodiments.
[0137] Among them, the memory 1020 can be, but is not limited to, random access memory (Random Access Memory, RAM), read only memory (Read Only Memory, ROM), programmable read-only memory (Programmable Read-Only Memory, PROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable read-only memory (Electric Erasable Programmable Read-Only Memory, EEPROM), etc.
[0138] The processor 1010 may be an integrated circuit chip with signal processing capabilities. The above-mentioned processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present invention may be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0139] It is understood that the structure illustrated in the embodiment of the present application does not constitute a specific limitation on the electronic device 1000. In other embodiments of the present application, the electronic device 1000 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.
[0140] The embodiment of the present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the road marking information determination method described in the above method embodiment are executed. The storage medium can be a volatile or non-volatile computer-readable storage medium.
[0141] In addition, an embodiment of the present disclosure also provides a computer program product, which stores a computer program. When the computer program is executed by a processor, the steps of the road marking information determination method provided in any of the above embodiments of the present disclosure are executed. For details, please refer to the above method embodiments, which will not be repeated here.
[0142] The computer program product may be implemented in hardware, software or a combination thereof. In an optional embodiment, the computer program product is embodied as a computer storage medium, which may be a volatile or non-volatile computer-readable storage medium. In another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (SDK) and the like.
[0143] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the above-described equipment and devices can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed equipment, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0144] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0145] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0146] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0147] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.
Claims
1. A method for determining road marking information, characterized in that: The method comprises: Obtain a road area image containing target type markings; Performing binary classification segmentation processing on the road area image to obtain a segmented image; the segmented image includes a first pixel point having a first pixel value and a second pixel point having a second pixel value, the first pixel value indicating that the corresponding first pixel point belongs to the target type marking, and the second pixel value indicating that the corresponding second pixel point does not belong to the target type marking; Performing contour detection on the segmented image to determine the minimum circumscribed figure of each closed contour; the minimum circumscribed figure is a figure corresponding to the shape of the target type marking line; Based on the minimum circumscribed graph, target coordinate information of each target type marking line is determined; the target coordinate information is used to construct the target type marking line in the road map.
2. The method according to claim 1, characterized in that The step of acquiring a road area image containing target type markings comprises: Acquire a road image of a target road section captured by a vehicle camera; The target type marking detection model is used to detect and crop a road area image including the target type marking from the road image.
3. The method according to claim 2, characterized in that The adopting the target type marking detection model to detect the road area image from the road image includes: Determine whether the vehicle is currently located within a road range corresponding to a target scene according to the vehicle positioning position information and pre-acquired road map information; the target scene is a scene where target type markings can be deployed; When it is determined that the target scene is located within the road range corresponding to the target scene, the target type marking detection model is used to detect the road area image from the road image.
4. The method according to any one of claims 1 to 3, characterized in that The step of performing contour detection on the segmented image to determine the minimum circumscribed graph of each closed contour includes: Performing contour detection on the segmented image to obtain each initial closed contour; From the initial closed contours, filter out contours whose corresponding contour areas are smaller than a contour area threshold; Calculate the minimum circumscribed figure of the remaining closed contour.
5. The method according to any one of claims 1 to 3, characterized in that After determining the minimum circumscribed figure of each closed contour and before determining the target coordinate information of each target type marking line based on the minimum circumscribed figure, the method further includes: According to the reference attribute characteristics of the target type marking, from each determined minimum circumscribed graph, the minimum circumscribed graph whose corresponding attribute characteristics do not conform to the reference attribute characteristics is filtered out to obtain a filtered minimum circumscribed graph; the reference attribute characteristics include shape and position characteristics; The step of determining target coordinate information of each target type marking line based on the minimum circumscribed graphic includes: Based on the filtered minimum circumscribed graph, target coordinate information of each target type marking line is determined.
6. The method according to any one of claims 1 to 3, characterized in that After determining the minimum circumscribed figure of each closed contour and before determining the coordinate information of each target type marking line based on the minimum circumscribed figure, the method further includes: Calculating the ratio of pixels corresponding to the first pixel in each minimum circumscribed graph; the pixel ratio is the ratio of the number of the first pixel to the total number of pixels in the segmented image, or the ratio between the number of the first pixel and the number of the second pixel; If the pixel ratio is less than the corresponding ratio threshold, the corresponding minimum circumscribed graph is filtered out to obtain a filtered minimum circumscribed graph; The step of determining target coordinate information of each target type marking line based on the minimum circumscribed graphic includes: Based on the filtered minimum circumscribed graph, target coordinate information of each target type marking line is determined.
7. The method according to claim 2, characterized in that The method further comprises: Acquire road point cloud data scanned by a vehicle radar, and convert the point cloud data into a point cloud image; Using the target type marking detection model, detect the image area where the target type marking is located from the point cloud image, and determine a first coordinate range corresponding to the image area in the world coordinate system; Whether to adopt the determined target coordinate information of each target type marking line is determined based on the overlap between a second coordinate range of the road area image detected from the road image in the world coordinate system and the first coordinate range.
8. The method according to any one of claims 1 to 3, characterized in that After determining the target coordinate information of each target type marking line, the method further includes: Based on the target coordinate information of each target type marking line, as well as the collected road image and vehicle positioning data of the target road section, a target map for simulation testing is constructed; Based on the target map, a simulation test of autonomous driving is performed on the vehicle.
9. A vehicle, characterized in that: A controller is included, the controller comprising: a memory for storing machine-readable instructions; and A processor is used to call and execute the machine-readable instructions stored in the memory to implement the steps of the road marking information determination method according to any one of claims 1 to 8.
10. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the machine-readable instructions are executed by the processor to perform the steps of the road marking information determination method as described in any one of claims 1 to 8.
11. A computer program product, characterized in that The computer program product stores a computer program, and when the computer program is executed by a processor, the steps of the method for determining road marking line information as claimed in any one of claims 1 to 8 are executed.
Citation Information
Patent Citations
Method, device, medium and equipment for extracting road marking
CN115273033A
Road marking point cloud semi-automatic extraction and semantic recognition method
CN116246239A
Marking line damage detection method based on vehicle-mounted CCD (Charge Coupled Device) camera
CN116824532A
Point cloud road marking classification method and device based on improved U-Net network
CN118097601A
Expressway marking extraction method, device, equipment, medium and product
CN118658065A
Cited By
Complete marking line damage detection method and device based on geometric position of marking line and medium
CN119438226A
