Curbstone data processing method, curbstone data processing device and storage medium
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-13
- Publication Date
- 2026-03-13
AI Technical Summary
In existing technologies, curb detection is difficult, inefficient, and unstable. Manual drawing is costly, and edge segmentation and line fitting methods are affected by detection distance and interval, making it impossible to reliably detect curbs.
By performing elevation jump statistical processing on curb point cloud data, curb elevation feature data is obtained, and data fusion is performed to determine curb boundary and height data, thereby improving extraction efficiency, accuracy and stability.
It improves the mapping accuracy of curb maps, enhances the robustness of curb boundary extraction, and ensures the stability and accuracy of curb detection.
Smart Images
Figure CN121661603A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of autonomous driving technology, and in particular to a curb data processing method, apparatus and storage medium. Background Technology
[0002] Current autonomous driving technology has not yet completely broken free from its reliance on maps. Among these, curbs (shoulders) are elements that interact very frequently with autonomous driving technology. However, as low-lying obstacles, curbs present unique detection challenges. Existing technologies for detecting curbs mainly employ manual drawing, edge segmentation, or line fitting methods. However, manually drawing curbs is difficult, inefficient, and costly. Furthermore, using edge segmentation or line fitting methods to detect curbs is affected by the detection distance and detection interval, making it impossible to achieve stable curb detection. Summary of the Invention
[0003] To address the aforementioned technical problems, this application provides a technical solution for processing curb data, including a method, apparatus, and storage medium. Specifically, this application performs elevation jump statistical processing on first curb point cloud data to obtain curb elevation feature data. This data is then used to fuse curb elevation feature data, road surface point cloud data, and driving trajectory data to determine first data for the first curb boundary and curb height data. Furthermore, based on the curb height data and the first data for the first curb boundary, second data for the second curb boundary is determined. The second curb boundary and the first curb boundary are two opposing boundaries. Utilizing the technical solution provided by this application can improve the extraction efficiency, accuracy, and stability of the upper and lower curb boundaries, enhance the robustness of curb boundary extraction, and thereby improve the mapping accuracy of curb maps.
[0004] On one hand, embodiments of this application provide a curb data processing method, the method comprising: Acquire first curb point cloud data, road surface point cloud data, and driving trajectory data. The road surface point cloud data is the point cloud data of the road surface in the area near the curb, and the driving trajectory data is the trajectory data corresponding to the acquisition device used to collect the first curb point cloud data. The first curb point cloud data is subjected to elevation jump statistical processing to obtain curb elevation feature data. The curb elevation feature data, the road surface point cloud data, and the driving trajectory data are fused together to determine the first data of the first curb boundary and the curb height data. Based on the curb height data and the first data of the first curb boundary, the second data of the second curb boundary is determined, wherein the second curb boundary and the first curb boundary are two opposite boundaries.
[0005] Furthermore, the curb elevation feature data includes multiple curb elevation jump data and multiple curb elevation minimum value data; The elevation jump statistical processing of the first curb point cloud data to obtain curb elevation feature data includes: The first tooth point cloud data is interpolated to obtain the second tooth point cloud data; The first curb point cloud data is divided into regions and statistically analyzed to determine the first elevation maximum value corresponding to each of the multiple regions and the first quantity value of curb point cloud data contained in each of the multiple regions. The second curb point cloud data is divided into regions and statistically analyzed to determine the second elevation maximum value corresponding to each of the multiple regions and the second quantity value corresponding to the curb point cloud data contained in each of the multiple regions. Data processing is performed on the first elevation maximum and minimum values corresponding to each of the multiple division regions, the second elevation maximum and minimum values corresponding to each of the multiple division regions, the first quantity value contained in each of the multiple division regions, and the second quantity value contained in each of the multiple division regions to determine the multiple curb elevation jump data and the multiple curb elevation minimum value data.
[0006] Further, the interpolation processing of the first tooth point cloud data to obtain the second tooth point cloud data includes: Based on the positional relationship between multiple curb point cloud data in the first curb point cloud data, a curb point cloud index tree data structure is established. Based on the curb point cloud index tree data structure, query the first preset number of neighboring curb point cloud data that are closest to the target curb point cloud data, where the target curb point cloud data is any curb point cloud data in the first curb point cloud data; Interpolation processing is performed on the target curb point cloud data and the first preset number of neighboring curb point cloud data respectively to obtain interpolated curb point cloud data; The interpolated curb point cloud data is inserted into the corresponding position in the first curb point cloud data to obtain the second curb point cloud data.
[0007] Furthermore, the first elevation extreme value includes a first elevation maximum value and a first elevation minimum value, and the second elevation extreme value includes a second elevation maximum value and a second elevation minimum value; The step of processing the data for the first elevation maximum / minimum value, the second elevation maximum / minimum value, the first quantity value, and the second quantity value contained in each of the multiple partitioned regions to determine the multiple curb elevation jump data and the multiple curb elevation minimum value data includes: The difference between the maximum and minimum first elevation values corresponding to each of the multiple divided regions is calculated to obtain multiple first elevation jump values; The difference between the maximum and minimum second elevation values corresponding to each of the multiple divided regions is calculated to obtain multiple second elevation jump values. If the first quantity value is greater than the preset quantity value, the plurality of first elevation jump values are determined as the plurality of curb elevation jump data, and the plurality of first elevation minimum values are determined as the plurality of curb elevation minimum data; If the first quantity value is less than or equal to the preset quantity value, the plurality of second elevation jump values are determined as the plurality of curb elevation jump data, and the plurality of second elevation minimum values are determined as the plurality of curb elevation minimum data.
[0008] Further, the data fusion processing of the curb elevation feature data, the road surface point cloud data, and the driving trajectory data to determine the first data of the first curb boundary and the curb height data includes: Based on the multiple curb elevation jump data, the multiple curb elevation minimum data, and the driving trajectory data, the initial data of the first curb boundary are determined; The road surface point cloud data is subjected to elevation jump statistics processing to obtain road surface elevation jump data and road surface elevation minimum value data; Determine a second preset number of neighboring curb point cloud data that are closest to the target road point cloud data and a third preset number of neighboring road point cloud data that are closest to the target road point cloud data, wherein the target road point cloud data is any one of the road point cloud data; The curb height data is obtained based on the target road surface point cloud data, the second preset number of adjacent curb point cloud data, and the multiple curb elevation jump data. The initial data of the first curb boundary and the minimum road elevation data corresponding to the third preset number of neighboring road surface point cloud data are subjected to dimensional transformation to obtain the first data of the first curb boundary.
[0009] Further, determining the initial data of the first curb boundary based on the plurality of curb elevation jump data, the plurality of curb elevation minimum data, and the driving trajectory data includes: The multiple curb elevation jump data are binarized to obtain multiple connected components corresponding to the curbs; Based on preset filtering conditions, a target connected component is determined from the plurality of connected components; The curb endpoints are determined based on the target connected component and the trajectory direction corresponding to the driving trajectory data; Based on the curb endpoint, the target connected region, and the preset radius annulus, determine multiple curb point cloud data along the target direction corresponding to the driving trajectory data, wherein the target direction is the direction consistent with the trajectory direction; The initial data of the first curb boundary are determined based on the minimum curb elevation data and the curb elevation jump data corresponding to the divided regions where the multiple curb point cloud data along the target direction are located.
[0010] Further, the initial data of the first curb boundary includes initial data of multiple first curb boundaries in two-dimensional coordinates; the step of performing a dimensional transformation on the initial data of the first curb boundary and the minimum road elevation data corresponding to the third preset number of neighboring road surface point cloud data to obtain the first data of the first curb boundary includes: The minimum road surface elevation data corresponding to each of the third preset number of adjacent road surface point cloud data is determined as the curb elevation data in the third dimension; The initial data of multiple first curb boundaries in the two-dimensional coordinate system and the curb elevation data in the third dimension are subjected to dimensional transformation to obtain the first data of the first curb boundaries in the three-dimensional coordinate system.
[0011] Further, determining the second data of the second curb boundary based on the curb height data and the first data of the first curb boundary includes: The curb height data and the first data of the first curb boundary are subjected to dimensional transformation to obtain the second data of the second curb boundary.
[0012] On the other hand, embodiments of this application provide a curb data processing apparatus, the apparatus comprising: The data acquisition module is used to acquire first curb point cloud data, road surface point cloud data and driving trajectory data. The road surface point cloud data is the point cloud data of the road surface in the area near the curb, and the driving trajectory data is the trajectory data corresponding to the acquisition device used to collect the first curb point cloud data. The curb elevation feature data determination module is used to perform elevation jump statistical processing on the first curb point cloud data to obtain curb elevation feature data. The first curb boundary data determination module is used to perform data fusion processing on the curb elevation feature data, the road surface point cloud data and the driving trajectory data to determine the first data of the first curb boundary and the curb height data. The second curb boundary data determination module is used to determine the second data of the second curb boundary based on the curb height data and the first data of the first curb boundary, wherein the second curb boundary and the first curb boundary are two opposite boundaries.
[0013] On the other hand, a curb data processing device is provided, which includes a processor and a memory. The memory stores at least one instruction, at least one program, code set, or instruction set. The at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the curb data processing method described above.
[0014] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction, at least one program, code set, or instruction set is stored therein, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the curb data processing method described above.
[0015] Implementing this application will have the following beneficial effects: This application obtains curb elevation feature data by performing elevation jump statistical processing on the first curb point cloud data. This data is then used to fuse the curb elevation feature data, road surface point cloud data, and driving trajectory data to determine the first data of the first curb boundary and the curb height data. Furthermore, based on the curb height data and the first data of the first curb boundary, the second data of the second curb boundary is determined. The second curb boundary and the first curb boundary are two opposite boundaries. The technical solution provided in this application can improve the extraction efficiency, accuracy, and stability of the upper and lower curb boundaries, enhance the robustness of curb boundary extraction, and thus improve the mapping accuracy of curb maps. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 A schematic diagram of an implementation environment provided for an embodiment of this application; Figure 2 A flowchart illustrating a curb data processing method provided in an embodiment of this application; Figure 3 A flowchart illustrating a method for determining curb elevation feature data provided in an embodiment of this application; Figure 4 A flowchart illustrating the second-path tooth point cloud data determination method provided in this application embodiment; Figure 5 A flowchart illustrating the method for determining multiple curb elevation jump data and multiple curb elevation minimum data provided in the embodiments of this application; Figure 6 A flowchart illustrating the method for determining the first data and curb height data provided in the embodiments of this application; Figure 7 A flowchart illustrating the method for determining the initial data of the first tooth boundary provided in an embodiment of this application; Figure 8 This is a lookup diagram for curb nodes provided in an embodiment of this application; Figure 9 A curb merging lookup map provided in an embodiment of this application; Figure 10 This is a schematic diagram of the structure of a curb data processing device provided in an embodiment of this application; Figure 11 This is a schematic diagram of the curb elevation feature data determination module provided in an embodiment of this application; Figure 12 This is a schematic diagram of the structure of the second-path tooth point cloud data determination submodule provided in the embodiments of this application; Figure 13 This is a schematic diagram of the structure of the curb elevation feature data determination submodule provided in the embodiments of this application; Figure 14 This is a schematic diagram of the structure of a server provided in an embodiment of this application. Detailed Implementation
[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.
[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.
[0020] Please see Figure 1 This is a schematic diagram of an implementation environment provided in an embodiment of this application, such as... Figure 1The implementation environment shown may include terminal 01 and server 02. In practical applications, terminal 01 and server 02 can be connected wirelessly to realize interaction between terminal 01 and server 02.
[0021] In this application embodiment, terminal 01 can be a mobile device. For example, terminal 01 can include, but is not limited to, smartphones, tablets, PDA smart terminals, in-vehicle smart terminals, wearable devices, and smart home control terminals. Both terminal 01 and server 02 can serve as the execution subject of this application embodiment.
[0022] In one specific embodiment, when terminal 01 is the executing entity, terminal 01 is used to collect curb point cloud data and road surface point cloud data, and to analyze and process the curb point cloud data and road surface point cloud data to determine the data corresponding to the upper and lower boundaries of the curb. In one specific embodiment, terminal 01 can be a data processing module in the acquisition device, wherein the data processing module is communicatively connected to the acquisition module in the acquisition device. After the acquisition module in the acquisition device collects the curb point cloud data and road surface point cloud data, it can send the collected data to the data processing module so that the data processing module can analyze and process it. The results of the analysis and processing can be sent to server 02 for cloud storage by server 02.
[0023] In another specific embodiment, when the server 02 is the execution subject, the terminal 01 is used to collect curb point cloud data and road surface point cloud data, and transmit the collected curb point cloud data and road surface point cloud data to the server 02, so that the server 02 can analyze and process the curb point cloud data and road surface point cloud data to determine the data corresponding to the upper and lower boundaries of the curb, and analyze and store the data corresponding to the determined upper and lower boundaries of the curb.
[0024] It should be noted that the following description of the curb data processing method is based on terminal 01 as the execution subject. Furthermore, it should be noted that... Figure 1 The diagram shown is merely a schematic representation of an implementation environment, which may include more or fewer nodes; this application makes no limitation on this.
[0025] Figure 2 This is a flowchart illustrating a curb data processing method provided in an embodiment of this application. The following is a summary of the process. Figure 2 The technical solution of this application is described in detail. It should be noted that this specification provides the method operation steps as shown in the embodiments or flowcharts, but based on conventional or non-inventive labor, more or fewer operation steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many steps and does not represent the only execution order. The method specifically includes the following steps: S101: Acquire the first curb point cloud data, road surface point cloud data, and driving trajectory data. The road surface point cloud data is the point cloud data of the road surface in the area near the curb, and the driving trajectory data is the trajectory data corresponding to the acquisition device used to collect the first curb point cloud data.
[0026] In this embodiment of the application, the curb is referred to as the curbstone or curb stone, which is located in the boundary area of the road and has a certain height. The first curb point cloud data is the curb point cloud data collected by the acquisition device before the curb is interpolated. The acquisition device can be a collection vehicle equipped with a laser scanner, a small UAV-borne laser radar device or a mobile scanning device, etc. The acquisition device is used to collect curb point cloud data and road surface point cloud data along the extension direction of the road.
[0027] It should be noted that a point cloud segmentation model can be trained in advance, and then the point cloud data of the first curb and the road surface can be segmented by the point cloud segmentation model to improve the efficiency of distinguishing point cloud data.
[0028] S102: Perform elevation jump statistical processing on the first curb point cloud data to obtain curb elevation feature data.
[0029] In this embodiment, the elevation jump statistical processing involves dividing the first curb point cloud data into regions to obtain characteristic data related to curb elevation for each region. The curb elevation characteristic data is used to characterize the feature data related to curb elevation. Specifically, the curb elevation characteristic data may include multiple curb elevation jump data and multiple curb elevation minimum value data. The curb elevation jump data is the maximum and minimum value change of the curb point cloud data in the divided region, and the curb elevation minimum value data is the minimum value of the curb point cloud data in the divided region. Thus, a curb elevation jump map can be plotted in two-dimensional coordinates using multiple curb elevation jump data, and a curb elevation minimum value map can be plotted in two-dimensional coordinates using multiple curb elevation minimum value data.
[0030] In one alternative implementation, such as Figure 3 As shown, this is a flowchart illustrating a method for determining curb elevation feature data according to an embodiment of this application. Specifically, the curb elevation feature data includes multiple curb elevation jump data and multiple curb elevation minimum value data. Step S102 includes: S1021: Perform interpolation processing on the first channel tooth point cloud data to obtain the second channel tooth point cloud data; S1022: Perform regional division statistics on the first curb point cloud data, and determine the first elevation maximum value corresponding to each of the multiple division regions and the first quantity value of curb point cloud data contained in each of the multiple division regions. S1023: Perform regional division statistics on the second curb point cloud data, and determine the second elevation maximum value corresponding to each of the multiple division regions and the second quantity value corresponding to the curb point cloud data contained in each of the multiple division regions. S1024: Perform data processing on the first elevation maximum and minimum values, the second elevation maximum and minimum values, the first quantity values, and the second quantity values contained in each of the multiple division regions to determine multiple curb elevation jump data and multiple curb elevation minimum value data.
[0031] In this embodiment, interpolation processing is used to interpolate and encrypt the first curb point cloud data, thereby solving the technical problem that the accuracy of the obtained curb elevation feature data is low due to the excessive sparseness of the collected first curb point cloud data. The second curb point cloud data is the curb point cloud data after interpolation processing. The region division statistics are used to divide the curb point cloud data into regions and to statistically analyze the elevation feature data of the curb point cloud data in the divided regions. Specifically, the maximum and minimum values of the vertical and horizontal coordinates corresponding to the first curb point cloud data can be determined first, and then the size of the two-dimensional area corresponding to the first curb point cloud data can be determined according to the maximum and minimum values of the vertical and horizontal coordinates, so as to divide the first curb point cloud data into regions. For example, the region can be divided according to a division area of 0.03m × 0.03m, or according to a division area of 0.06m × 0.06m.
[0032] Furthermore, in this embodiment of the application, both the first curb point cloud data and the second curb point cloud data before and after interpolation are divided into regions. In cases where some of the first curb point cloud data is too sparse, the curb elevation jump data and curb elevation minimum value data corresponding to the second curb point cloud data can be taken, thereby improving the accuracy of the curb elevation jump data and curb elevation minimum value data. Specifically, it can be determined whether to take the curb elevation jump data and curb elevation minimum value data corresponding to the second curb point cloud data based on the first quantity value corresponding to the curb point cloud data contained in each of the multiple divided regions.
[0033] In one specific implementation, such as Figure 4 As shown, this is a flowchart illustrating the second-way tooth point cloud data determination method provided in this application embodiment. Specifically, step S1021 includes: S10211: Based on the positional relationship between multiple curb point cloud data in the first curb point cloud data, establish a curb point cloud index tree data structure; S10212: Based on the curb point cloud index tree data structure, query the first preset number of neighboring curb point cloud data that are closest to the target curb point cloud data. The target curb point cloud data is any curb point cloud data in the first curb point cloud data. S10213: Interpolate the target curb point cloud data and the first preset number of neighboring curb point cloud data respectively to obtain interpolated curb point cloud data; S10214: Insert the interpolated tooth point cloud data into the corresponding position in the first tooth point cloud data to obtain the second tooth point cloud data.
[0034] In this embodiment, the curb point cloud index tree data structure is used to characterize the feature dependency relationship between multiple curb point cloud data in the first curb point cloud data. Specifically, the feature dependency relationship can be the positional relationship between multiple curb point cloud data in the first curb point cloud data. Then, through the curb point cloud index tree data structure, the first preset number of neighboring curb point cloud data closest to the target curb point cloud data can be found. The first preset number can be 10, 15, or 20, etc. Preferably, the first preset number is 10. Specifically, the interpolation process can be to calculate the midpoint between the target curb point cloud data and each neighboring curb point cloud data, thereby obtaining multiple interpolated curb point cloud data. The multiple interpolated curb point cloud data are then inserted into the corresponding positions in the first curb point cloud data to obtain the second curb point cloud data. It should be noted that the interpolation process can also be to average the target curb point cloud data and each neighboring curb point cloud data.
[0035] In one specific embodiment, before performing interpolation processing on the target curb point cloud data and a first preset number of neighboring curb point cloud data to obtain interpolated curb point cloud data, the first preset number of neighboring curb point cloud data is filtered to remove curb point cloud data that meet the conditions for interpolation processing. Specifically, neighboring curb point cloud data with a distance of less than 0.06m between the target curb point cloud data and neighboring curb point cloud data can be filtered out. No interpolation processing is performed on the filtered neighboring curb point cloud data, thereby ensuring that interpolation processing is only performed on sparse curb point cloud data to avoid the second curb point cloud data after interpolation processing being too dense. It should be noted that the interpolated curb point cloud data does not need to participate in interpolation processing again, nor does it need to be the neighboring curb point cloud data closest to the target curb point cloud data.
[0036] In one specific implementation, such as Figure 5 As shown, this is a flowchart illustrating the method for determining multiple curb elevation jump data and multiple curb elevation minimum data provided in this application embodiment. Specifically, the first elevation extreme value includes a first elevation maximum value and a first elevation minimum value, and the second elevation extreme value includes a second elevation maximum value and a second elevation minimum value. Then, step S1024 includes: S10241: Subtract the maximum and minimum values of the first elevation corresponding to each of the multiple divided regions to obtain multiple first elevation jump values; S10242: Subtract the maximum and minimum values of the second elevation corresponding to each of the multiple divided regions to obtain multiple second elevation jump values; S10243: If the first quantity value is greater than the preset quantity value, the multiple first elevation jump values are determined as multiple curb elevation jump data, and the multiple first elevation minimum values are determined as multiple curb elevation minimum data; S10244: If the first quantity value is less than or equal to the preset quantity value, the multiple second elevation jump values are determined as multiple curb elevation jump data, and the multiple second elevation minimum values are determined as multiple curb elevation minimum data.
[0037] In this embodiment, the region can be divided into 0.03m × 0.03m areas. The first maximum elevation value is the maximum elevation value corresponding to each area after dividing the first tooth point cloud data into 0.03m × 0.03m areas. The first minimum elevation value is the minimum elevation value corresponding to each area after dividing the first tooth point cloud data into 0.03m × 0.03m areas. The first elevation jump value is obtained by subtracting the first minimum elevation value from the first maximum elevation value. The second maximum elevation value is the maximum elevation value corresponding to each area after dividing the second tooth point cloud data into 0.03m × 0.03m areas. The second minimum elevation value is the minimum elevation value corresponding to each area after dividing the second tooth point cloud data into 0.03m × 0.03m areas. The second elevation jump value is obtained by subtracting the second minimum elevation value from the second maximum elevation value.
[0038] In one specific embodiment, the preset quantity value can be 2, 3, 4, 5, 6, or 7, etc. Preferably, the preset quantity value can be 2. When the first quantity in the divided area is greater than the preset quantity value, it indicates that the collected first curb point cloud data is not too sparse. Therefore, multiple first elevation jump values determined based on the first curb point cloud data can be determined as multiple curb elevation jump data, and multiple first elevation minimum values determined based on the first curb point cloud data can be determined as multiple curb elevation minimum value data. When the first quantity value is less than or equal to the preset quantity value, it indicates that the collected first curb point cloud data is too sparse. Therefore, multiple second elevation jump values determined based on the second curb point cloud data can be determined as multiple curb elevation jump data, and multiple second elevation minimum values determined based on the second curb point cloud data can be determined as multiple curb elevation minimum value data, so as to improve the accuracy of curb elevation jump data and curb elevation minimum value data.
[0039] In another specific embodiment, since the curb point cloud data in more than one-third of the divided areas are too sparse when the area is divided according to 0.03m×0.03m, the area can be divided according to 0.06m×0.06m. It should be noted that the methods for obtaining multiple curb elevation jump data and multiple curb elevation minimum data are the same in different area divisions. The only difference is the selection of the first preset number. In this embodiment, the first preset number is 20.
[0040] S103: Perform data fusion processing on curb elevation feature data, road surface point cloud data, and driving trajectory data to determine the first data of the first curb boundary and curb height data.
[0041] In this embodiment, the first curb boundary is the lower curb boundary, the first data of the first curb boundary is the data corresponding to the first curb boundary in the three-dimensional coordinate system, and the curb height data is the data used to characterize the curb height. Then, by performing data fusion processing on the curb elevation feature data, road surface point cloud data, and driving trajectory data, the first data of the first curb boundary and the curb height data can be obtained, thereby improving the extraction efficiency and accuracy of the lower curb boundary, making the vectorization result of the lower curb boundary robust, and thus improving the accuracy and efficiency of curb mapping.
[0042] In one alternative implementation, such as Figure 6 As shown, it is a flowchart illustrating the method for determining the first data and curb height data provided in this application embodiment. Specifically, step S103 includes: S1031: Determine the initial data of the first curb boundary based on multiple curb elevation jump data, multiple curb elevation minimum data and driving trajectory data; S1032: Perform elevation jump statistics on road surface point cloud data to obtain road surface elevation jump data and road surface elevation minimum value data; S1033: Determine the second preset number of neighboring curb point cloud data and the third preset number of neighboring road point cloud data that are closest to the target road point cloud data. The target road point cloud data is any road point cloud data in the road point cloud data. S1034: Obtain curb height data based on the target road surface point cloud data, the second preset number of adjacent curb point cloud data, and multiple curb elevation jump data; S1035: Perform dimensional transformation on the initial data of the first curb boundary and the minimum road elevation data corresponding to the third preset number of neighboring road point cloud data to obtain the first data of the first curb boundary.
[0043] In this embodiment, the initial data of the first curb boundary represents the initial data of the first curb boundary in two-dimensional coordinates. The method for determining the road surface elevation jump data is the same as the method for determining the curb elevation jump data. The method for determining the minimum road surface elevation data is the same as the method for determining the minimum curb elevation data, and will not be elaborated here. The second preset number can be 20. Then, the curb height data can be obtained based on the target road surface point cloud data, the second preset number of adjacent curb point cloud data and multiple curb elevation jump data.
[0044] Specifically, the curb elevation jump data corresponding to the 20 nearest neighbor curb point cloud data to the target road surface is obtained. The curb point cloud data with curb elevation jump data greater than 0.018m are filtered out, and the curb elevation jump data corresponding to each curb point cloud data with a jump data greater than 0.018m is determined. The median of the above-mentioned curb elevation jump data that meet the above conditions is calculated, and then the curb height data of the corresponding area can be obtained. By analogy, the curb height data corresponding to the entire curb can be obtained.
[0045] In one specific embodiment, the third preset number can be 20. Based on the minimum road surface elevation data, the minimum road surface elevation data corresponding to each of the 20 adjacent road surface point cloud data can be determined. Then, the minimum road surface elevation data can be determined as the minimum curb elevation information corresponding to the corresponding area. By analogy, the minimum curb elevation information corresponding to the entire curb can be obtained. The minimum curb elevation information is the height information of the lower boundary of the curb. Thus, the initial data of the first curb boundary and the minimum curb elevation information corresponding to the entire curb can be dimensionally transformed to obtain the first data of the first curb boundary in three-dimensional coordinates. It should be noted that since the curb is often located above the ground and there are obstacles on the road surface when acquiring road surface point cloud data, the minimum road surface elevation data can be selected as the minimum curb elevation information corresponding to the corresponding area to improve the accuracy of detecting the lower boundary of the curb.
[0046] In one specific implementation, such as Figure 7 As shown, it is a flowchart illustrating the method for determining the initial data of the first tooth boundary provided in an embodiment of this application. Specifically, step S1031 includes: S10311: Binarize multiple curb elevation jump data to obtain multiple connected components corresponding to the curbs; S10312: Determine the target connected component from multiple connected components based on preset filtering conditions; S10313: Determine the curb endpoints based on the target connected components and the trajectory direction corresponding to the driving trajectory data; S10314: Based on the curb endpoints, target connected components, and a preset radius ring, determine multiple curb point cloud data along the target direction corresponding to the driving trajectory data. The target direction is the direction consistent with the trajectory direction. S10315: Determine the initial data of the first curb boundary based on the minimum curb elevation data and the curb elevation jump data corresponding to the divided area where multiple curb point cloud data along the target direction are located.
[0047] In this embodiment, the curb elevation jump data with a value greater than zero is assigned a value of 1, and the curb elevation jump data with a value less than zero is assigned a value of 0. Since the collected curb point cloud data has low precision, multiple connected components can be formed when binarizing the multiple curb elevation jump data. Furthermore, after binarization, 3*3 image dilation is performed on the multiple connected components to facilitate subsequent task processing and improve the accuracy of data processing.
[0048] In one specific embodiment, the preset filtering condition is a connected region with a length greater than 60 pixels, a width greater than 3 pixels, and an area greater than 100 pixels. Based on this preset filtering condition, a target connected region can be determined from multiple connected regions. Further, curb point cloud data with only one neighboring curb point cloud data is filtered from the target connected region as candidate points. The curb endpoint can then be determined based on the candidate points and the trajectory direction corresponding to the driving trajectory data. The trajectory direction corresponding to the driving trajectory data is the direction extending from the data acquisition endpoint corresponding to the curb towards the curb. Based on the curb endpoint, the target connected region, and a preset radius annulus, multiple curb point cloud data along the target direction corresponding to the driving trajectory data can be determined. This allows for the determination of the initial data for the first curb boundary based on the minimum curb elevation data and the curb elevation jump data corresponding to the divided area where the multiple curb point cloud data along the target direction are located, thereby improving the extraction stability and accuracy of the lower curb boundary.
[0049] In practical applications, based on the curb endpoints, target connected components, and a pre-defined radius annulus, multiple curb point cloud data points along the target direction corresponding to the driving trajectory data can be determined. The target direction is the direction consistent with the trajectory direction. Specifically, for example... Figure 8 As shown, it is a curb node lookup diagram provided in the embodiment of this application. The preset radius ring is a ring with P1 as the center, an inner radius of 15 pixels and an outer radius of 30 pixels. The intersection point of the ring and the target connected region can be determined. It should be noted that the selection of the preset radius ring must ensure that it has an intersection point with the target connected region.
[0050] In one embodiment, P2 is the centroid of the set of intersection points of the annulus and the target connected domain. Therefore, the centroid P2 of the set of intersection points of the annulus and the target connected domain can be determined based on this set. Then, for any curb point cloud data P3, the angle between the first straight line formed by points P1 and P2 and the second straight line formed by points P1 and P3 can be determined. Multiple arbitrary curb point cloud data points whose angle between the first and second straight lines is less than a right angle are considered as multiple curb point cloud data points along the target direction corresponding to the driving trajectory data. Furthermore, from the curb elevation jump data corresponding to multiple curb point cloud data along the target direction, curb point cloud data with curb elevation jump data greater than 0.018m can be identified. Thus, the minimum curb elevation data corresponding to the curb point cloud data with curb elevation jump data greater than 0.018m can be used as the initial data of the first curb boundary corresponding to a part of the region. Further, with centroid P2 as the center, the relationship between the next target connected domain and the annulus with centroid P2 as the center is determined. By analogy, the initial data of the entire first curb boundary can be determined.
[0051] In one specific embodiment, the initial data of the first curb boundary is merged to obtain complete initial data of the first curb boundary. It should be noted that the determined initial data of the first curb boundary may have gaps, therefore, the gaps need to be merged. In practical applications, such as... Figure 9 As shown, this is a curb merging lookup map provided in an embodiment of this application. The curb endpoint is used as the starting point for the search, and the tail point of the initial data of the first curb boundary is used as the final tail point. The tail point Q2 of the vector line corresponding to curb endpoint Q1 is determined. Other curb vectorization result starting points within the range of the tail point Q2 are searched. Any starting point Q3 is taken, along with the nearest node Q4 in the vectorized line L2 containing Q3. Then, the angle α between the third straight line formed by points Q1 and Q2 and the fourth straight line formed by points Q2 and Q3, and the angle β between the fifth straight line formed by points Q2 and Q3 and the sixth straight line formed by points Q3 and Q4, can be calculated. If both angles are less than 10 degrees, the curb endpoint Q1 and point Q4 are merged and connected to ensure the smoothness of some initial data of the first curb boundary. Then, using the tail point Q4 of the vectorized line L2 as the starting point, the above steps are repeated to continue searching for mergeable curb vector lines until all curb vector lines have been traversed.
[0052] In one specific implementation, the initial data of the first tooth boundary includes initial data of multiple first tooth boundaries in two-dimensional coordinates, and step S1035 includes: S10351: Determine the minimum road surface elevation data corresponding to each of the third preset number of adjacent road surface point cloud data as the curb elevation data in the third dimension; S10352: Perform dimensional transformation processing on the initial data of multiple first curb boundaries in two-dimensional coordinates and the curb elevation data in the third dimension to obtain the first data of the first curb boundaries in three-dimensional coordinates.
[0053] In this embodiment of the application, the initial data of the first curb boundary includes the initial data of multiple first curb boundaries in two-dimensional coordinates. That is, the initial data of the first curb boundary is the first data of the first curb boundary in two-dimensional coordinates. The third preset number can be 20. By determining the minimum road surface elevation data corresponding to each of the third preset number of adjacent road surface point cloud data as the curb elevation data in the third dimension, the accuracy of the first data of the first curb boundary in three-dimensional coordinates can be improved.
[0054] It should be noted that, since there may be obstacles on the road surface, the minimum road surface elevation data corresponding to each of the adjacent road surface point cloud data is used to determine the curb elevation data in the third dimension. This can eliminate the influence of obstacles on the lower boundary height of the curb, thereby improving the accuracy of the first data of the first curb boundary in three-dimensional coordinates. Furthermore, the curb elevation data is the data corresponding to the curb height in the third dimension. Based on the initial data of the first curb boundary which only has two dimensions, the curb elevation data in the third dimension is added, thereby obtaining the curb elevation data of the corresponding area. Then, by performing dimensional transformation processing on the initial data of multiple first curb boundaries in two-dimensional coordinates and the curb elevation data in the third dimension, the first data of the first curb boundary in three-dimensional coordinates is obtained, so as to realize the representation of the lower boundary of the curb in three-dimensional coordinates, so as to improve the accuracy of the subsequent drawing of the curb on the map.
[0055] S104: Based on the curb height data and the first data of the first curb boundary, determine the second data of the second curb boundary. The second curb boundary and the first curb boundary are two opposite boundaries.
[0056] In this embodiment, the second curb boundary is the upper boundary of the curb, and the second data of the second curb boundary is the data corresponding to the second curb boundary in the three-dimensional coordinate system. In an optional implementation, step S104 includes: S1041: Perform dimensional transformation on the curb height data and the first data of the first curb boundary to obtain the second data of the second curb boundary.
[0057] Specifically, given the curb height data and the first data of the first curb boundary, a dimensional transformation can be performed on the curb height data and the first data of the first curb boundary. Based on the first data of the first curb boundary in the obtained dimensional coordinates, the curb height data can be added to obtain the second data of the second curb boundary. Then, based on the first data of the first curb boundary and the second data of the second curb boundary, the curbs on the map can be drawn, thereby improving the accuracy and efficiency of curb mapping.
[0058] As can be seen from the above technical solutions of the embodiments of this application, the following technical effects are achieved: This application obtains curb elevation feature data by performing elevation jump statistical processing on the first curb point cloud data. This data is then used to fuse the curb elevation feature data, road surface point cloud data, and driving trajectory data to determine the first data of the first curb boundary and the curb height data. Furthermore, based on the curb height data and the first data of the first curb boundary, the second data of the second curb boundary is determined. The second curb boundary and the first curb boundary are two opposite boundaries. The technical solution provided in this application can improve the extraction efficiency, accuracy, and stability of the upper and lower curb boundaries, enhance the robustness of curb boundary extraction, and thus improve the mapping accuracy of curb maps.
[0059] This application also provides a curb data processing device, such as... Figure 10 The diagram shown is a structural schematic of a curb data processing device provided in an embodiment of this application. The curb data processing device includes: The data acquisition module 10 is used to acquire first curb point cloud data, road surface point cloud data and driving trajectory data. The road surface point cloud data is the point cloud data of the road surface in the area near the curb, and the driving trajectory data is the trajectory data corresponding to the acquisition device used to collect the first curb point cloud data.
[0060] The curb elevation feature data determination module 20 is used to perform elevation jump statistical processing on the first curb point cloud data to obtain curb elevation feature data.
[0061] The first curb boundary data determination module 30 is used to perform data fusion processing on curb elevation feature data, road surface point cloud data and driving trajectory data to determine the first data of the first curb boundary and curb height data.
[0062] The second curb boundary data determination module 40 is used to determine the second data of the second curb boundary based on the curb height data and the first data of the first curb boundary. The second curb boundary and the first curb boundary are two opposite boundaries.
[0063] Furthermore, such as Figure 11As shown, this is a structural schematic diagram of the curb elevation feature data determination module provided in this application embodiment. The curb elevation feature data includes multiple curb elevation jump data and multiple curb elevation minimum value data; the curb elevation feature data determination module 20 includes: The second-path tooth point cloud data determination submodule 201 is used to perform interpolation processing on the first-path tooth point cloud data to obtain the second-path tooth point cloud data.
[0064] The first quantity value determination submodule 202 is used to perform regional division statistics on the first curb point cloud data, and determine the first elevation maximum value corresponding to each of the multiple division regions and the first quantity value of the curb point cloud data contained in each of the multiple division regions.
[0065] The second quantity value determination submodule 203 is used to perform regional division statistics on the second curb point cloud data, and determine the second elevation maximum value corresponding to each of the multiple division regions and the second quantity value corresponding to the curb point cloud data contained in each of the multiple division regions. The curb elevation feature data determination submodule 204 is used to process the data of the first elevation maximum value, the second elevation maximum value, the first quantity value and the second quantity value contained in each of the multiple division regions, and to determine multiple curb elevation jump data and multiple curb elevation minimum value data.
[0066] Furthermore, such as Figure 12 As shown, this is a schematic diagram of the structure of the second tooth point cloud data determination submodule provided in an embodiment of this application. The second tooth point cloud data determination submodule 201 includes: The index tree data structure establishment unit 2011 is used to establish an index tree data structure for curb points based on the positional relationship between multiple curb point cloud data in the first curb point cloud data.
[0067] The neighboring curb point cloud data query unit 2012 is used to query a first preset number of neighboring curb point cloud data that are closest to the target curb point cloud data based on the curb point cloud index tree data structure. The target curb point cloud data is any curb point cloud data in the first curb point cloud data.
[0068] The interpolation curb point cloud data determination unit 2013 is used to perform interpolation processing on the target curb point cloud data and the first preset number of neighboring curb point cloud data respectively to obtain interpolated curb point cloud data.
[0069] The second-path tooth point cloud data determination unit 2014 is used to insert the interpolated tooth point cloud data into the corresponding position in the first-path tooth point cloud data to obtain the second-path tooth point cloud data.
[0070] Furthermore, such as Figure 13As shown, this is a structural diagram of the curb elevation feature data determination submodule provided in an embodiment of this application. The first elevation extreme value includes a first elevation maximum value and a first elevation minimum value, and the second elevation extreme value includes a second elevation maximum value and a second elevation minimum value. The curb elevation feature data determination submodule 204 includes: The first elevation jump value determination unit 2041 is used to perform subtraction processing on the first elevation maximum value and the first elevation minimum value corresponding to each of the multiple divided regions to obtain multiple first elevation jump values.
[0071] The second elevation jump value determination unit 2042 is used to perform subtraction on the second elevation maximum value and second elevation minimum value corresponding to each of the multiple divided regions to obtain multiple second elevation jump values.
[0072] The first assignment unit 2043 is used to determine multiple first elevation jump values as multiple curb elevation jump data and multiple first elevation minimum values as multiple curb elevation minimum data if the first quantity value is greater than the preset quantity value.
[0073] The second assignment unit 2044 is used to determine multiple second elevation jump values as multiple curb elevation jump data and multiple second elevation minimum values as multiple curb elevation minimum data if the first quantity value is less than or equal to the preset quantity value.
[0074] Furthermore, the first curb boundary data determination module 30 includes: The first curb boundary initial data determination submodule 301 is used to determine the first curb boundary initial data based on multiple curb elevation jump data, multiple curb elevation minimum value data and driving trajectory data.
[0075] The road surface elevation jump data determination submodule 302 is used to perform elevation jump statistical processing on the road surface point cloud data to obtain road surface elevation jump data and road surface elevation minimum value data.
[0076] The point cloud data determination submodule 303 is used to determine the second preset number of adjacent curb point cloud data and the third preset number of adjacent road point cloud data that are closest to the target road point cloud data. The target road point cloud data is any road point cloud data in the road point cloud data. The curb height data submodule 304 is used to obtain curb height data based on the target road surface point cloud data, a second preset number of adjacent curb point cloud data, and multiple curb elevation jump data.
[0077] The first data determination submodule 305 is used to perform dimensional transformation processing on the initial data of the first curb boundary and the minimum road elevation data corresponding to the third preset number of adjacent road surface point cloud data to obtain the first data of the first curb boundary.
[0078] Furthermore, the first curb boundary initial data determination submodule 301 includes: The connected component determination unit 3011 is used to perform binarization processing on multiple curb elevation jump data to obtain multiple connected components corresponding to the curb.
[0079] The target connected component determination unit 3012 is used to determine the target connected component from multiple connected components based on preset filtering conditions; The curb endpoint determination unit 3013 is used to determine the curb endpoint based on the target connected domain and the trajectory direction corresponding to the driving trajectory data.
[0080] The target direction point determination unit 3014 is used to determine multiple curb point cloud data along the target direction corresponding to the driving trajectory data based on the curb endpoint, the target connected region and the preset radius circle. The target direction is the direction consistent with the trajectory direction.
[0081] The first curb boundary initial data determination unit 3015 is used to determine the initial data of the first curb boundary based on the minimum curb elevation data and the curb elevation jump data corresponding to the divided area where multiple curb point cloud data along the target direction are located.
[0082] Furthermore, the initial data for the first tooth boundary includes initial data for multiple first tooth boundaries in two-dimensional coordinates, and the first data determination submodule 305 includes: The curb elevation data determination unit 3051 is used to determine the minimum road surface elevation data corresponding to each of the third preset number of adjacent road surface point cloud data as the curb elevation data in the third dimension.
[0083] The three-dimensional first data determination unit 3052 is used to perform dimensional transformation processing on the initial data of multiple first curb boundaries in two-dimensional coordinates and the curb elevation data in the third dimension to obtain the first data of the first curb boundaries in three-dimensional coordinates.
[0084] Furthermore, the second tooth boundary data determination module 40 includes: The second data determination submodule 401 for the second curb boundary is used to perform dimensional transformation processing on the curb height data and the first data of the first curb boundary to obtain the second data of the second curb boundary.
[0085] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.
[0086] This application provides a curb data processing device, which includes a processor and a memory. The memory stores at least one instruction, at least one program, code set, or instruction set. The at least one instruction, at least one program, code set, or instruction set is loaded and executed by the processor to implement the curb data processing method provided in the above method embodiments.
[0087] Memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. Memory can primarily include a program storage area and a data storage area. The program storage area can store the operating system, application programs required for the functions, etc.; the data storage area can store data created based on the use of the device, etc. Furthermore, memory can include high-speed random access memory, and can also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, memory can also include a memory controller to provide the processor with access to the memory.
[0088] The curb data processing device can be a server. This application embodiment also provides a schematic diagram of a server structure. Please refer to [link / reference]. Figure 14 The server 1400 is used to implement the data processing method provided in the above embodiments. The server 1400 can vary significantly due to different configurations or performance, and may include one or more processors 1410 (e.g., one or more processors) and storage 1430, and one or more storage media 1420 (e.g., one or more mass storage devices) for storing application programs 1423 or data 1422. The memory 1430 and storage media 1420 can be temporary or persistent storage. The program stored in the storage media 1420 may include one or more modules, each module including a series of instruction operations on the server. Furthermore, the processor 1410 may be configured to communicate with the storage media 1420 and execute the series of instruction operations in the storage media 1420 on the server 1400. Server 1400 may also include one or more power supplies 1460, one or more wired or wireless network interfaces 1450, one or more input / output interfaces 1440, and / or one or more operating systems 1421, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.
[0089] Embodiments of this application also provide a computer-readable storage medium, which can be disposed in a server to store at least one instruction, at least one program, code set, or instruction set related to implementing a curb data processing method in the method embodiments. The at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the curb data processing method provided in the above method embodiments.
[0090] Optionally, in this embodiment, the storage medium may be located at at least one of the multiple network servers in a computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.
[0091] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0092] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system and server embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.
[0093] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A method for processing curb data, characterized in that, The method includes: Acquire first curb point cloud data, road surface point cloud data, and driving trajectory data. The road surface point cloud data is the point cloud data of the road surface in the area near the curb, and the driving trajectory data is the trajectory data corresponding to the acquisition device used to collect the first curb point cloud data. The first curb point cloud data is subjected to elevation jump statistical processing to obtain curb elevation feature data. The curb elevation feature data, the road surface point cloud data, and the driving trajectory data are fused together to determine the first data of the first curb boundary and the curb height data. Based on the curb height data and the first data of the first curb boundary, the second data of the second curb boundary is determined, wherein the second curb boundary and the first curb boundary are two opposite boundaries.
2. The method according to claim 1, characterized in that, The curb elevation feature data includes multiple curb elevation jump data and multiple curb elevation minimum value data; The elevation jump statistical processing of the first curb point cloud data to obtain curb elevation feature data includes: The first tooth point cloud data is interpolated to obtain the second tooth point cloud data; The first curb point cloud data is divided into regions and statistically analyzed to determine the first elevation maximum value corresponding to each of the multiple regions and the first quantity value of curb point cloud data contained in each of the multiple regions. The second curb point cloud data is divided into regions and statistically analyzed to determine the second elevation maximum value corresponding to each of the multiple regions and the second quantity value corresponding to the curb point cloud data contained in each of the multiple regions. Data processing is performed on the first elevation maximum and minimum values corresponding to each of the multiple division regions, the second elevation maximum and minimum values corresponding to each of the multiple division regions, the first quantity value contained in each of the multiple division regions, and the second quantity value contained in each of the multiple division regions to determine the multiple curb elevation jump data and the multiple curb elevation minimum value data.
3. The method according to claim 2, characterized in that, The step of interpolating the first tooth point cloud data to obtain the second tooth point cloud data includes: Based on the positional relationship between multiple curb point cloud data in the first curb point cloud data, a curb point cloud index tree data structure is established. Based on the curb point cloud index tree data structure, query the first preset number of neighboring curb point cloud data that are closest to the target curb point cloud data, where the target curb point cloud data is any curb point cloud data in the first curb point cloud data; Interpolation processing is performed on the target curb point cloud data and the first preset number of neighboring curb point cloud data respectively to obtain interpolated curb point cloud data; The interpolated curb point cloud data is inserted into the corresponding position in the first curb point cloud data to obtain the second curb point cloud data.
4. The method according to claim 2, characterized in that, The first elevation extreme value includes the first elevation maximum value and the first elevation minimum value, and the second elevation extreme value includes the second elevation maximum value and the second elevation minimum value; The step of processing the data for the first elevation maximum / minimum value, the second elevation maximum / minimum value, the first quantity value, and the second quantity value contained in each of the multiple partitioned regions to determine the multiple curb elevation jump data and the multiple curb elevation minimum value data includes: The difference between the maximum and minimum first elevation values corresponding to each of the multiple divided regions is calculated to obtain multiple first elevation jump values; The difference between the maximum and minimum second elevation values corresponding to each of the multiple divided regions is calculated to obtain multiple second elevation jump values. If the first quantity value is greater than the preset quantity value, the plurality of first elevation jump values are determined as the plurality of curb elevation jump data, and the plurality of first elevation minimum values are determined as the plurality of curb elevation minimum data; If the first quantity value is less than or equal to the preset quantity value, the plurality of second elevation jump values are determined as the plurality of curb elevation jump data, and the plurality of second elevation minimum values are determined as the plurality of curb elevation minimum data.
5. The method according to claim 2, characterized in that, The process of fusing the curb elevation feature data, the road surface point cloud data, and the driving trajectory data to determine the first data of the first curb boundary and the curb height data includes: Based on the multiple curb elevation jump data, the multiple curb elevation minimum data, and the driving trajectory data, the initial data of the first curb boundary are determined; The road surface point cloud data is subjected to elevation jump statistics processing to obtain road surface elevation jump data and road surface elevation minimum value data; Determine a second preset number of neighboring curb point cloud data that are closest to the target road point cloud data and a third preset number of neighboring road point cloud data that are closest to the target road point cloud data, wherein the target road point cloud data is any one of the road point cloud data; The curb height data is obtained based on the target road surface point cloud data, the second preset number of adjacent curb point cloud data, and the multiple curb elevation jump data. The initial data of the first curb boundary and the minimum road elevation data corresponding to the third preset number of neighboring road surface point cloud data are subjected to dimensional transformation to obtain the first data of the first curb boundary.
6. The method according to claim 5, characterized in that, The step of determining the initial data of the first curb boundary based on the multiple curb elevation jump data, the multiple curb elevation minimum data, and the driving trajectory data includes: The multiple curb elevation jump data are binarized to obtain multiple connected components corresponding to the curbs; Based on preset filtering conditions, a target connected component is determined from the plurality of connected components; The curb endpoints are determined based on the target connected component and the trajectory direction corresponding to the driving trajectory data; Based on the curb endpoint, the target connected region, and the preset radius annulus, determine multiple curb point cloud data along the target direction corresponding to the driving trajectory data, wherein the target direction is the direction consistent with the trajectory direction; The initial data of the first curb boundary are determined based on the minimum curb elevation data and the curb elevation jump data corresponding to the divided regions where the multiple curb point cloud data along the target direction are located.
7. The method according to claim 5, characterized in that, The initial data of the first curb boundary includes initial data of multiple first curb boundaries in two-dimensional coordinates; the step of performing a dimensional transformation on the initial data of the first curb boundary and the minimum road elevation data corresponding to the third preset number of neighboring road surface point cloud data to obtain the first data of the first curb boundary includes: The minimum road surface elevation data corresponding to each of the third preset number of adjacent road surface point cloud data is determined as the curb elevation data in the third dimension; The initial data of multiple first curb boundaries in the two-dimensional coordinate system and the curb elevation data in the third dimension are subjected to dimensional transformation to obtain the first data of the first curb boundaries in the three-dimensional coordinate system.
8. The method according to claim 1, characterized in that, The step of determining the second data of the second curb boundary based on the curb height data and the first data of the first curb boundary includes: The curb height data and the first data of the first curb boundary are subjected to dimensional transformation to obtain the second data of the second curb boundary.
9. A curb data processing device, characterized in that, The device includes: The data acquisition module is used to acquire first curb point cloud data, road surface point cloud data and driving trajectory data. The road surface point cloud data is the point cloud data of the road surface in the area near the curb, and the driving trajectory data is the trajectory data corresponding to the acquisition device used to collect the first curb point cloud data. The curb elevation feature data determination module is used to perform elevation jump statistical processing on the first curb point cloud data to obtain curb elevation feature data. The first curb boundary data determination module is used to perform data fusion processing on the curb elevation feature data, the road surface point cloud data and the driving trajectory data to determine the first data of the first curb boundary and the curb height data. The second curb boundary data determination module is used to determine the second data of the second curb boundary based on the curb height data and the first data of the first curb boundary, wherein the second curb boundary and the first curb boundary are two opposite boundaries.
10. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction or at least one program segment, which is loaded and executed by a processor to implement the curb data processing method as described in any one of claims 1 to 8.