Lidar point cloud data enhancement method, device, equipment, medium and product

Abstract

The embodiments of the present application provide a method, device, equipment, medium and product for enhancing laser radar point cloud data. In this method, after generating a screening area according to the collection position of the laser radar point cloud data to be supplemented, the multiple laser radar point cloud data to be screened are screened in combination with the height values ​​of the collection positions of the multiple laser radar point cloud data to be screened to obtain the first point cloud data, which is then fused with the laser radar point cloud data to be supplemented to obtain the target point cloud data. This solution increases the number of points in the laser radar point cloud data by fusing the screened first point cloud data with the laser radar point cloud data to be supplemented, and because the collection position of the first point cloud data is close to the collection position of the laser radar point cloud data to be supplemented and is in the same height range, the first point cloud data and the laser radar point cloud data to be supplemented have point clouds obtained by scanning the same map elements, which can improve the accuracy of subsequent target detection.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a laser radar point cloud data enhancement method, device, equipment, medium and product. Background Art

[0002] In the field of intelligent driving technology, high-precision maps are an indispensable part. In the process of building high-precision maps, it is necessary to use laser radar point cloud data for target detection to obtain map elements such as lane lines, sidewalks, street lights, and signboards.

[0003] In the prior art, a vehicle is equipped with a laser radar. When the vehicle is traveling on the road, the laser radar is used to collect laser radar point cloud data, and then the laser radar point cloud data collected each time is used for target detection.

[0004] However, since the points in the collected LiDAR point cloud data are relatively sparse and there are also missing points, it is easy to cause inaccurate subsequent target detection. Summary of the invention

[0005] The embodiments of the present application provide a laser radar point cloud data enhancement method, device, equipment, medium and product to solve the problem that the laser radar point cloud data in the prior art is relatively sparse, resulting in inaccurate subsequent target detection.

[0006] In a first aspect, an embodiment of the present application provides a method for enhancing laser radar point cloud data, comprising:

[0007] Generate a screening area according to the collection location of the laser radar point cloud data to be supplemented;

[0008] According to the screening area and the height values ​​of the collection positions of the plurality of laser radar point cloud data to be screened, the plurality of laser radar point cloud data to be screened are screened to obtain first point cloud data;

[0009] The first point cloud data is fused with the laser radar point cloud data to be supplemented to obtain target point cloud data.

[0010] In a possible implementation, the filtering of the plurality of to-be-filtered laser radar point cloud data according to the filtering area and the height values ​​of the collection positions of the plurality of to-be-filtered laser radar point cloud data to obtain the first point cloud data includes:

[0011] Filter out point cloud data whose acquisition position belongs to the screening area from the plurality of laser radar point cloud data to be screened, and obtain second point cloud data;

[0012] The first point cloud data are screened out from each of the second point cloud data according to the height value of the acquisition position of each of the second point cloud data.

[0013] In a possible implementation, the collection position of the first point cloud data and the collection position of the to-be-supplemented laser radar point cloud data belong to the same altitude range.

[0014] In a possible implementation, the step of selecting the first point cloud data from each of the second point cloud data according to the height value of the acquisition position of each of the second point cloud data includes:

[0015] Clustering all the second point cloud data and the to-be-supplemented laser radar point cloud data according to the height value of the acquisition position to obtain a plurality of point cloud classes;

[0016] The second point cloud data in the point cloud class including the laser radar point cloud data to be supplemented is used as the first point cloud data.

[0017] In a possible implementation, the filtering of the plurality of to-be-filtered laser radar point cloud data according to the filtering area and the height values ​​of the collection positions of the plurality of to-be-filtered laser radar point cloud data to obtain the first point cloud data includes:

[0018] Clustering all the laser radar point cloud data to be screened and the laser radar point cloud data to be supplemented according to the height value of the collection position to obtain multiple point cloud classes;

[0019] Using the to-be-screened laser radar point cloud data in the point cloud class including the to-be-supplemented laser radar point cloud data as the third point cloud data;

[0020] The point cloud data whose acquisition position belongs to the screening area in the third point cloud data is screened out to obtain the first point cloud data.

[0021] In a possible implementation manner, before fusing the first point cloud data with the to-be-supplemented laser radar point cloud data to obtain target point cloud data, the method further includes:

[0022] Calculate the relative angle of collection of each first point cloud data according to the collection position of each first point cloud data and the collection position of the laser radar point cloud data to be supplemented;

[0023] According to the relative angle of collection of each first point cloud data, all the first point cloud data are screened to obtain fourth point cloud data;

[0024] The step of fusing the first point cloud data with the laser radar point cloud data to be supplemented to obtain target point cloud data includes:

[0025] The fourth point cloud data is fused with the laser radar point cloud data to be supplemented to obtain target point cloud data.

[0026] In a possible implementation, the relative angle of collection of each first point cloud data is the angle between a projection line and a preset direction, the projection line is the projection of a line connecting a collection position of the first point cloud data and a collection position of the laser radar point cloud data to be supplemented on a horizontal plane, and the preset direction is the direction of travel of the vehicle at the collection position of the laser radar point cloud data to be supplemented.

[0027] In a possible implementation manner, the step of screening all first point cloud data according to the acquisition relative angle of each first point cloud data to obtain the third point cloud data includes:

[0028] Performing a point cloud screening operation on each preset angle range of at least two preset angle ranges to obtain fifth point cloud data corresponding to each preset angle range, and determining the fifth point cloud data corresponding to each preset angle range as the fourth point cloud data; wherein,

[0029] The point cloud screening operation includes: for the preset angle range, if the number of first point cloud data collected whose relative angles belong to the preset angle range is greater than a preset number, then selecting the preset number of point cloud data from the first point cloud data collected whose relative angles belong to the preset angle range as the fifth point cloud data corresponding to the preset angle range.

[0030] In a possible implementation, according to the acquired collection positions of the plurality of laser radar point cloud data to be screened and the collection positions of the laser radar point cloud data to be supplemented, the plurality of laser radar point cloud data to be screened are screened, and before obtaining the first point cloud data, the method further includes:

[0031] For each initial laser radar point cloud data, determining a preset collection range to which a collection position of the initial laser radar point cloud data belongs;

[0032] If the number of points in the initial laser radar point cloud data is less than the preset sparse threshold corresponding to the preset acquisition range, the initial laser radar point cloud data is used as the laser radar point cloud data to be supplemented.

[0033] In a second aspect, an embodiment of the present application provides a laser radar point cloud data enhancement device, comprising:

[0034] Processing modules for:

[0035] Generate a screening area according to the collection location of the laser radar point cloud data to be supplemented;

[0036] According to the screening area and the height values ​​of the collection positions of the plurality of laser radar point cloud data to be screened, the plurality of laser radar point cloud data to be screened are screened to obtain first point cloud data;

[0037] The data enhancement module is used to fuse the first point cloud data with the laser radar point cloud data to be supplemented to obtain target point cloud data.

[0038] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0039] Processor, memory, communication interface;

[0040] The memory is used to store executable instructions of the processor;

[0041] Wherein, the processor is configured to execute the lidar point cloud data enhancement method described in any one of the first aspects by executing the executable instructions.

[0042] In a fourth aspect, an embodiment of the present application provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the laser radar point cloud data enhancement method described in any one of the first aspects is implemented.

[0043] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, is used to implement the laser radar point cloud data enhancement method described in any one of the first aspects.

[0044] The laser radar point cloud data enhancement method, device, equipment, medium and product provided in the embodiments of the present application generate a screening area according to the collection position of the laser radar point cloud data to be supplemented; then, according to the screening area and the height values ​​of the collection positions of the multiple laser radar point cloud data to be supplemented, multiple laser radar point cloud data to be supplemented are screened to obtain the first point cloud data whose collection position belongs to the screening area and is in the same height range as the collection position of the laser radar point cloud data to be supplemented. Then, the first point cloud data is fused with the laser radar point cloud data to be supplemented to obtain the target point cloud data. This solution increases the number of points in the laser radar point cloud data by fusing the first point cloud data after screening with the laser radar point cloud data to be supplemented, and because the collection position of the first point cloud data is close to the collection position of the laser radar point cloud data to be supplemented and is in the same height range, the first point cloud data and the laser radar point cloud data to be supplemented have point clouds obtained by scanning the same map elements, which can improve the accuracy of subsequent target detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0046] Figure 1 Schematic diagram of the collected LiDAR point cloud data provided for this application;

[0047] Figure 2a A schematic diagram of the process of the first embodiment of the laser radar point cloud data enhancement method provided in this application;

[0048] Figure 2b Schematic diagram of the collection scenario provided for this application;

[0049] Figure 2c A point cloud diagram of the target point cloud data provided for this application;

[0050] Figure 3a A schematic diagram of the process of Embodiment 2 of the laser radar point cloud data enhancement method provided in this application;

[0051] Figure 3b A schematic diagram of the relative angles of acquisition provided for this application;

[0052] Figure 3c A schematic diagram of the fourth point cloud data provided by this application;

[0053] Figure 4 A schematic diagram of the process of Embodiment 3 of the laser radar point cloud data enhancement method provided in this application;

[0054] Figure 5 A schematic diagram of the structure of an embodiment of a laser radar point cloud data enhancement device provided in this application;

[0055] Figure 6 A schematic diagram of the structure of an electronic device provided in this application.

[0056] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0057] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0058] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0059] In the field of intelligent driving, lidar point cloud data is needed for target detection to obtain map elements such as lane lines, sidewalks, street lights, and signboards, so as to build high-precision maps based on the detected targets.

[0060] In the prior art, a vehicle is equipped with a laser radar. When the vehicle is driving on the road, the laser radar is used to scan the surrounding environment. The surrounding environment contains map elements, and the laser radar point cloud data is obtained. Then, the laser radar point cloud data collected each time is used for target detection. However, since the points in the collected laser radar point cloud data are relatively sparse, the number of points is small, and there are also missing points, it is easy to cause inaccurate subsequent target detection.

[0061] For example, Figure 1 A schematic diagram of the collected LiDAR point cloud data provided for this application, such as Figure 1 As shown, there are two lane lines and a straight ground sign in the external environment of the collection location. Some points of the right lane line and the straight ground sign are missing in the collected lidar point cloud data. It is a sparse lidar point cloud data with a small number of points, which affects the accuracy of subsequent target detection.

[0062] In view of the problems existing in the prior art, the inventor discovered in the process of studying the laser radar point cloud data enhancement method that the vehicle collects laser radar point cloud data during driving. For each road, the vehicle will pass through the road many times, and because the frequency of collection is large, there are usually multiple other collection locations near a collection location of the laser radar point cloud data to be supplemented, so the laser radar point cloud data of these collection locations can be screened out. The laser radar point cloud data of these collection locations and the laser radar point cloud data to be supplemented have point clouds obtained by scanning the same map elements, and then merged with the laser radar point cloud data to be supplemented to increase the number of points in the laser radar point cloud data. Based on the above invention concept, the laser radar point cloud data enhancement scheme in this application is designed.

[0063] In the present application, the executor of the laser radar point cloud data enhancement method can be a computer, or a server, a terminal device, a vehicle, etc. This application does not limit it, and the following description will take a computer as an example.

[0064] The following is an example of the application scenario of the lidar point cloud data enhancement method provided in this application.

[0065] For example, in this application scenario, a vehicle is equipped with lidar point cloud data and drives multiple times in an area where a high-precision map needs to be built. The vehicle collects data through the lidar, and then sends the collected lidar point cloud data and the collection location of each lidar point cloud data to a computer.

[0066] The computer sequentially treats each laser radar point cloud data as the laser radar point cloud data to be supplemented, and treats other laser radar point cloud data other than the laser radar point cloud data as the laser radar point cloud data to be screened.

[0067] Generate a screening area according to the collection position of the laser radar point cloud data to be supplemented; then screen out the point cloud data whose collection position belongs to the screening area from the plurality of laser radar point cloud data to be screened, and obtain the second point cloud data;

[0068] It should be noted that the screening area includes the collection position of the laser radar point cloud data to be supplemented, which can be a spherical area with the collection position of the laser radar point cloud data to be supplemented as the center and the preset radius length as the radius; it can also be a regular polyhedron area with the collection position of the laser radar point cloud data to be supplemented as the center and the preset side length as the side length; it can also be a cylindrical area with the collection position of the laser radar point cloud data to be supplemented as the center, the preset radius length as the radius, and the preset height as the height. The preset radius length, preset side length, and preset height can be 30 meters, 50 meters, 70 meters, etc. The embodiment of the present application does not limit the screening area, preset radius length, preset side length, and preset height, and can be determined according to actual conditions.

[0069] Then, according to the acquisition position of each second point cloud data and the acquisition position of the laser radar point cloud data to be supplemented, the second point cloud data is screened, and the point cloud data whose acquisition position is in the same height range as the acquisition position of the laser radar point cloud data to be supplemented is screened out to obtain the first point cloud data. All the second point cloud data and the laser radar point cloud data to be supplemented can be clustered according to the height value of the acquisition position to obtain multiple point cloud classes. Since the acquisition position of the point cloud data in each point cloud class belongs to the same height range, the second point cloud data in the point cloud class including the laser radar point cloud data to be supplemented is used as the first point cloud data.

[0070] The first point cloud data is then fused with the laser radar point cloud data to be supplemented to obtain the target point cloud data, thus completing the laser radar point cloud data enhancement.

[0071] The target point cloud data can then be used for target detection to obtain map elements such as lane lines, sidewalks, street lights, and signboards to build a high-precision map.

[0072] It should be noted that the above scenario is only an example of an application scenario provided by an embodiment of the present application. The embodiment of the present application does not limit the actual form of the various devices included in the scenario, nor does it limit the interaction method between the devices. In the specific application of the solution, it can be set according to actual needs.

[0073] The technical solution of the present application is described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0074] Figure 2a This is a flow chart of the first embodiment of the laser radar point cloud data enhancement method provided by the present application. The present embodiment of the present application describes the situation in which a computer screens a plurality of laser radar point cloud data to be screened, obtains the first point cloud data, and then fuses the first point cloud data with the laser radar point cloud data to be supplemented to obtain the target point cloud data. Figure 2aAs shown, the laser radar point cloud data enhancement method specifically includes the following steps:

[0075] S201: Generate a screening area according to the collection location of the laser radar point cloud data to be supplemented.

[0076] After the computer obtains multiple laser radar point cloud data to be screened and their collection positions, as well as laser radar point cloud data to be supplemented and their collection positions, in order to increase the number of points in the laser radar point cloud data to be supplemented, it is necessary to select some point cloud data from the multiple laser radar point cloud data to be screened and add them to the laser radar point cloud data to be supplemented. The collection positions of these point cloud data are close to the collection positions of the laser radar point cloud data to be supplemented, and are in the same height range.

[0077] In this step, in order to complete the screening, it is necessary to generate a screening area based on the collection location of the lidar point cloud data to be supplemented.

[0078] S202: Filter the plurality of laser radar point cloud data to be filtered according to the filtering area and the height values ​​of the collection positions of the plurality of laser radar point cloud data to be filtered, and obtain the first point cloud data.

[0079] In this step, after the computer generates the screening area, multiple lidar point cloud data to be screened can be screened according to the screening area and the height values ​​of the collection positions of the multiple lidar point cloud data to be screened, so as to obtain the first point cloud data whose collection position is within the screening area and belongs to the same height range as the collection position of the lidar point cloud data to be supplemented.

[0080] In one implementation, point cloud data whose acquisition position belongs to the screening area among a plurality of to-be-screened laser radar point cloud data are first screened out to obtain the second point cloud data. In this way, the to-be-screened laser radar point cloud data whose acquisition position is close to the acquisition position of the to-be-supplemented laser radar point cloud data can be screened out.

[0081] Then, the first point cloud data are screened out from each of the second point cloud data according to the height value of the acquisition position of each of the second point cloud data.

[0082] In some scenarios, the collection positions of some second point cloud data are not in the same height range as the collection positions of the lidar point cloud data to be supplemented. If the second point cloud data are fused with the lidar point cloud data to be supplemented, not only will the lidar point cloud data to be supplemented fail to be supplemented to improve the accuracy of subsequent target detection, but it may also reduce the accuracy of target detection.

[0083] For example, Figure 2b The collection scene diagram provided for this application is as follows: Figure 2bAs shown, the collection position A of the laser radar point cloud data to be supplemented is located under the viaduct, and the collection position B of the second point cloud data is located on the viaduct. There is a sign on the viaduct, so the second point cloud data includes the point cloud data of the sign. If the second point cloud data is fused with the laser radar point cloud data to be supplemented, target detection will detect the presence of the sign, but in fact there is no sign, which will reduce the accuracy of target detection.

[0084] Therefore, it is necessary to determine the second point cloud data whose collection position belongs to the same height range as the collection position of the laser radar point cloud data to be supplemented, and obtain the first point cloud data to improve the accuracy of subsequent target detection.

[0085] According to the height value of the acquisition position, all the second point cloud data and the laser radar point cloud data to be supplemented are clustered to obtain multiple point cloud classes; since the acquisition position can be represented by three-dimensional coordinates, the height value in the acquisition position can be determined.

[0086] Since the collection positions of the point cloud data in the same point cloud class are in the same height range, the second point cloud data in the point cloud class including the laser radar point cloud data to be supplemented is used as the first point cloud data.

[0087] Exemplarily, there are ten second point cloud data, namely data A-data J. According to the height value of the acquisition position, all the second point cloud data and the laser radar point cloud data to be supplemented are clustered to obtain three point cloud classes. The first point cloud class includes data A-data D, the second point cloud class includes data E-data G, and the third point cloud class includes data H-data J and the laser radar point cloud data to be supplemented. Therefore, data H-data J is the first point cloud data.

[0088] It should be noted that, according to the height value H of the acquisition position of the laser radar point cloud data to be supplemented, a height screening range [HX, H+X] can be generated, where X is a preset height difference. The second point cloud data whose height value at the acquisition position belongs to the height screening range is used as the first point cloud data.

[0089] It should be noted that the preset height difference may be 5 meters, 10 meters, 15 meters, etc. The embodiment of the present application does not limit the preset height difference and may be set according to actual conditions.

[0090] In another implementation, all the laser radar point cloud data to be screened and the laser radar point cloud data to be supplemented are clustered according to the height value of the collection position to obtain multiple point cloud classes.

[0091] The laser radar point cloud data to be screened in the point cloud class including the laser radar point cloud data to be supplemented is used as the third point cloud data. That is, the acquisition position of the third acquisition data is in the same height range as the acquisition position of the laser radar point cloud data to be supplemented.

[0092] The point cloud data whose acquisition position belongs to the screening area in the third point cloud data is screened out to obtain the first point cloud data. That is, the third point cloud data whose acquisition position is close to the acquisition position of the supplementary laser radar point cloud data is determined to obtain the first point cloud data.

[0093] S203: Fusing the first point cloud data with the laser radar point cloud data to be supplemented to obtain target point cloud data.

[0094] In this step, after the computer screens and obtains the first point cloud data, that is, after obtaining the point cloud data whose collection position is close to the collection position of the laser radar point cloud data to be supplemented and whose collection position is in the same height range as the collection position of the laser radar point cloud data to be supplemented, the first point cloud data is fused with the laser radar point cloud data to be supplemented, that is, the first point cloud data is added to the laser radar point cloud data to be supplemented to obtain the target point cloud data.

[0095] For example, Figure 2c The point cloud diagram of the target point cloud data provided for this application is as follows: Figure 2c As shown, Figure 1 In comparison, not only some points of the right lane line and some points of the straight ground markings are completed, but the number of points is also greater.

[0096] The laser radar point cloud data enhancement method provided in this embodiment generates a screening area according to the collection position of the laser radar point cloud data to be supplemented; then, according to the screening area and the height values ​​of the collection positions of the multiple laser radar point cloud data to be supplemented, multiple laser radar point cloud data to be supplemented are screened to obtain the first point cloud data whose collection position belongs to the screening area and is in the same height range as the collection position of the laser radar point cloud data to be supplemented. Then, the first point cloud data is fused with the laser radar point cloud data to be supplemented to obtain the target point cloud data. Compared with the prior art that does not process the collected laser radar point cloud data, this solution increases the number of points in the laser radar point cloud data by fusing the first point cloud data after screening with the laser radar point cloud data to be supplemented, and because the collection position of the first point cloud data is close to the collection position of the laser radar point cloud data to be supplemented and is in the same height range, the first point cloud data and the laser radar point cloud data to be supplemented have point clouds obtained by scanning the same map elements, which can improve the accuracy of subsequent target detection.

[0097] Figure 3aThis is a flow chart of the second embodiment of the laser radar point cloud data enhancement method provided by the present application. Based on the above embodiment, the present embodiment of the present application describes the situation where after the computer obtains the first point cloud data, the first point cloud data is screened to obtain the third point cloud data, and then the third point cloud data is merged with the laser radar point cloud data to be supplemented to obtain the target point cloud data. Figure 3a As shown, the laser radar point cloud data enhancement method specifically includes the following steps:

[0098] S301: Calculate the relative angle of collection of each first point cloud data according to the collection position of each first point cloud data and the collection position of the laser radar point cloud data to be supplemented.

[0099] After the computer obtains the first point cloud data, the collection positions of the first point cloud data may be unevenly distributed. If the number of first point cloud data is large, it will affect the accuracy and efficiency of subsequent target detection, so the first point cloud data needs to be screened.

[0100] In this step, the computer calculates the relative angle of collection of each first point cloud data according to the collection position of each first point cloud data and the collection position of the laser radar point cloud data to be supplemented.

[0101] The relative angle of collection of each first point cloud data is the angle between the projection line and the preset direction. The projection line is the projection of the line connecting the collection position of the first point cloud data and the collection position of the laser radar point cloud data to be supplemented on the horizontal plane. The preset direction is the driving direction of the vehicle at the collection position of the laser radar point cloud data to be supplemented.

[0102] For example, Figure 3b A schematic diagram of the relative angle of acquisition provided for this application, such as Figure 3b As shown, point A is the projection point of the acquisition position of the laser radar point cloud data to be supplemented on the horizontal plane, point B is the projection point of the acquisition position of the first point cloud data on the horizontal plane, and the line connecting point A and point B is the projection of the line connecting the acquisition position of the first point cloud data and the acquisition position of the laser radar point cloud data to be supplemented on the horizontal plane, that is, the projection line. The direction of the arrow is the driving direction of the vehicle at the acquisition position of the laser radar point cloud data to be supplemented, that is, the preset direction. Angle a is the relative angle of acquisition.

[0103] Specifically, the coordinates of the collection position can be converted into the Front-Left-Up (FLU) coordinate system with the collection position of the lidar point cloud data to be supplemented as the origin to obtain the target coordinates, and then the collection relative angle of each first point cloud data can be calculated based on the target coordinates of each first point cloud data and the target coordinates of the lidar point cloud data to be supplemented.

[0104] S302: Filtering all first point cloud data according to the acquisition relative angle of each first point cloud data to obtain fourth point cloud data.

[0105] In this step, after the computer obtains the acquisition relative angle of each first point cloud data, it can combine at least two preset angle ranges to filter all the first point cloud data to obtain the third point cloud data.

[0106] Specifically, a point cloud screening operation is performed on each preset angle range of at least two preset angle ranges to obtain fifth point cloud data corresponding to each preset angle range, and the fourth point cloud data corresponding to each preset angle range is determined as the fourth point cloud data.

[0107] Among them, the point cloud screening operation includes: for the preset angle range, if the number of first point cloud data collected whose relative angle belongs to the preset angle range is greater than the preset number, then a preset number of point cloud data are selected from the first point cloud data collected whose relative angle belongs to the preset angle range as the fifth point cloud data corresponding to the preset angle range.

[0108] In one implementation, a preset number of point cloud data may be selected by sorting the first point cloud data whose relative angles belong to the preset angle range in ascending order of the distance from the collection position of the laser radar point cloud data to be supplemented to obtain a point cloud data sequence, and selecting the first preset number of point cloud data from the point cloud data sequence as the fifth point cloud data.

[0109] In another implementation, the method of selecting the preset number of point cloud data may be: randomly selecting a preset number of point cloud data from the first point cloud data whose relative angles belong to the preset angle range as the fifth point cloud data.

[0110] If the number of first point cloud data whose relative angles belong to the preset angle range is less than or equal to the preset number, all the first point cloud data whose relative angles belong to the preset angle range are fifth point cloud data corresponding to the preset angle range.

[0111] For example, Figure 3c A schematic diagram of the fourth point cloud data provided in this application, such as Figure 3cAs shown, the dots in the figure represent the projection points of the first point cloud data on the horizontal plane, and the gray dots represent the projection points of the fourth point cloud data on the horizontal plane. There are four preset angle ranges, namely (0, 90], (90, 180], (180, 270], and (270, 360], in degrees. The preset number is 3, and the number of first point cloud data collected with a relative angle belonging to (0, 90] is 5, from which 3 first point cloud data are selected as the fifth point cloud data. The number of first point cloud data collected with a relative angle belonging to (90, 180] is 4, from which 3 first point cloud data are selected as the fifth point cloud data. The number of first point cloud data collected with a relative angle belonging to (180, 270] is 2, and these first point cloud data are all fifth point cloud data. The number of first point cloud data collected with a relative angle belonging to (270, 360] is 3, and these first point cloud data are all fifth point cloud data. Then the fifth point cloud data is used as the fourth point cloud data.

[0112] It should be noted that the preset number can be 3, 5, 10, 15, 20, etc. The embodiment of the present application does not limit the preset number and can be determined according to actual conditions.

[0113] It should be noted that the multiple preset angle ranges may be: (0, 45], (45, 135], (135, 225], (225, 315], (315, 360], in degrees. It may also be: (0, 45], (45, 90], (90, 135], (135, 180], (180, 225], (225, 270], (270, 315], (315, 360], in degrees. It may also be: (0, 90], (90, 180], (180, 270], (270, 360], in degrees. The embodiment of the present application does not limit the multiple preset angle ranges, and may be set according to actual conditions.

[0114] S303: Fusing the fourth point cloud data with the laser radar point cloud data to be supplemented to obtain target point cloud data.

[0115] In this step, after the computer screens and obtains the fourth point cloud data, the fourth point cloud data is merged with the laser radar point cloud data to be supplemented, that is, the fourth point cloud data is added to the laser radar point cloud data to be supplemented to obtain the target point cloud data.

[0116] The laser radar point cloud data enhancement method provided in this embodiment obtains the third data by collecting relative angles and filtering the first point cloud data. This not only makes the collection position of the fourth point cloud data evenly distributed, but also reduces the amount of point cloud data, thereby improving the accuracy and efficiency of subsequent target detection.

[0117] Figure 4This is a flow chart of the third embodiment of the laser radar point cloud data enhancement method provided by this application. Based on the above embodiment, this embodiment of the application describes the situation of determining the laser radar point cloud data to be supplemented before supplementing the laser radar point cloud data. Figure 4 As shown, the laser radar point cloud data enhancement method specifically includes the following steps:

[0118] S401: For each initial laser radar point cloud data, determine a preset collection range to which the collection position of the initial laser radar point cloud data belongs.

[0119] Since some of the collected lidar point cloud data are sparse and need to be enhanced, and some point cloud data are dense and do not need to be enhanced, it is necessary to determine the sparse point cloud data, that is, to determine the lidar point cloud data to be supplemented.

[0120] In this step, for each initial laser radar point cloud data, the preset collection range to which the collection position of the initial laser radar point cloud data belongs is determined.

[0121] S402: If the number of points in the initial laser radar point cloud data is less than the preset sparse threshold corresponding to the preset acquisition range, the initial laser radar point cloud data is used as the laser radar point cloud data to be supplemented.

[0122] In this step, after the computer determines the preset collection range to which the collection position of the initial laser radar point cloud data belongs, it determines whether the number of points in the initial laser radar point cloud data is less than the preset sparse threshold corresponding to the preset collection range.

[0123] If the number of points in the initial laser radar point cloud data is less than the preset sparse threshold corresponding to the preset acquisition range, it means that the initial laser radar point cloud data is sparse, and the initial laser radar point cloud data is used as the laser radar point cloud data to be supplemented.

[0124] If the number of points in the initial lidar point cloud data is greater than or equal to the preset sparse threshold corresponding to the preset acquisition range, it means that the initial lidar point cloud data is dense and there is no need to perform lidar point cloud data enhancement subsequently.

[0125] The preset sparse thresholds corresponding to different preset acquisition ranges can be the same or different. Some preset acquisition ranges have larger preset sparse thresholds. This is because there are more map elements within these preset acquisition ranges, and accordingly, there are more points in the lidar point cloud data. Map elements can be lane lines, sidewalks, street lights, signboards, etc.

[0126] It should be noted that the preset sparse threshold may be 120, 200, 300, etc. The embodiment of the present application does not limit the preset sparse threshold, which may be determined according to actual conditions.

[0127] The laser radar point cloud data enhancement method provided in this embodiment determines the laser radar point cloud data to be supplemented according to the number of points in the initial laser radar point cloud data and the preset sparse threshold corresponding to the preset acquisition range to which the acquisition position belongs, thereby improving the accuracy of the laser radar point cloud data to be supplemented. In addition, laser radar point cloud data enhancement is performed only on the laser radar point cloud data to be supplemented, thereby improving processing efficiency.

[0128] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0129] Figure 5 This is a schematic diagram of the structure of an embodiment of the laser radar point cloud data enhancement device provided in this application. Figure 5 As shown, the laser radar point cloud data enhancement device 50 includes:

[0130] The processing module 51 is used for:

[0131] Generate a screening area according to the collection location of the laser radar point cloud data to be supplemented;

[0132] According to the screening area and the height values ​​of the collection positions of the plurality of laser radar point cloud data to be screened, the plurality of laser radar point cloud data to be screened are screened to obtain first point cloud data;

[0133] The data enhancement module 52 is used to fuse the first point cloud data with the laser radar point cloud data to be supplemented to obtain target point cloud data.

[0134] Furthermore, the processing module 51 is specifically used for:

[0135] Filter out point cloud data whose acquisition position belongs to the screening area from the plurality of laser radar point cloud data to be screened, and obtain second point cloud data;

[0136] The first point cloud data are screened out from each of the second point cloud data according to the height value of the acquisition position of each of the second point cloud data.

[0137] Furthermore, the collection position of the first point cloud data and the collection position of the to-be-supplemented laser radar point cloud data belong to the same height range.

[0138] Furthermore, the processing module 51 is further configured to:

[0139] Clustering all the second point cloud data and the to-be-supplemented laser radar point cloud data according to the height value of the acquisition position to obtain a plurality of point cloud classes;

[0140] The second point cloud data in the point cloud class including the laser radar point cloud data to be supplemented is used as the first point cloud data.

[0141] Furthermore, the processing module 51 is further configured to:

[0142] Clustering all the laser radar point cloud data to be screened and the laser radar point cloud data to be supplemented according to the height value of the collection position to obtain multiple point cloud classes;

[0143] Using the to-be-screened laser radar point cloud data in the point cloud class including the to-be-supplemented laser radar point cloud data as the third point cloud data;

[0144] The point cloud data whose acquisition position belongs to the screening area in the third point cloud data is screened out to obtain the first point cloud data.

[0145] Furthermore, the processing module 51 is also used for:

[0146] Calculate the relative angle of collection of each first point cloud data according to the collection position of each first point cloud data and the collection position of the laser radar point cloud data to be supplemented;

[0147] According to the relative angle of collection of each first point cloud data, all the first point cloud data are screened to obtain fourth point cloud data;

[0148] Furthermore, the data enhancement module 52 is specifically configured to:

[0149] The fourth point cloud data is fused with the laser radar point cloud data to be supplemented to obtain target point cloud data.

[0150] Furthermore, the relative angle of collection of each first point cloud data is the angle between the projection line and the preset direction, the projection line is the projection of the line connecting the collection position of the first point cloud data and the collection position of the laser radar point cloud data to be supplemented on the horizontal plane, and the preset direction is the driving direction of the vehicle at the collection position of the laser radar point cloud data to be supplemented.

[0151] Furthermore, the processing module 51 is further configured to:

[0152] Performing a point cloud screening operation on each preset angle range of at least two preset angle ranges to obtain fifth point cloud data corresponding to each preset angle range, and determining the fifth point cloud data corresponding to each preset angle range as the fourth point cloud data; wherein,

[0153] The point cloud screening operation includes: for the preset angle range, if the number of first point cloud data collected whose relative angles belong to the preset angle range is greater than a preset number, then selecting the preset number of point cloud data from the first point cloud data collected whose relative angles belong to the preset angle range as the fifth point cloud data corresponding to the preset angle range.

[0154] Furthermore, the processing module 51 is also used for:

[0155] For each initial laser radar point cloud data, determining a preset collection range to which a collection position of the initial laser radar point cloud data belongs;

[0156] If the number of points in the initial laser radar point cloud data is less than the preset sparse threshold corresponding to the preset acquisition range, the initial laser radar point cloud data is used as the laser radar point cloud data to be supplemented.

[0157] The laser radar point cloud data enhancement device provided in this embodiment is used to execute the technical solution in any of the aforementioned method embodiments. Its implementation principle and technical effects are similar and will not be repeated here.

[0158] Figure 6 This is a schematic diagram of the structure of an electronic device provided in this application. Figure 6 As shown, the electronic device 60 includes:

[0159] Processor 61, memory 62, and communication interface 63;

[0160] The memory 62 is used to store executable instructions of the processor 61;

[0161] The processor 61 is configured to execute the technical solution in any of the aforementioned method embodiments by executing the executable instructions.

[0162] Optionally, the memory 62 may be independent or integrated with the processor 61 .

[0163] Optionally, when the memory 62 is a device independent of the processor 61, the electronic device 60 may further include:

[0164] The bus 64 , the memory 62 and the communication interface 63 are connected to the processor 61 via the bus 64 and communicate with each other. The communication interface 63 is used to communicate with other devices.

[0165] Optionally, the communication interface 63 may be implemented by a transceiver. The communication interface is used to implement communication between the database access device and other devices (such as a client, a read-write library, and a read-only library). The memory may include a random access memory (RAM) and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage.

[0166] The bus 64 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0167] The above-mentioned processor can be a general-purpose processor, including a central processing unit CPU, a network processor (NP), etc.; it can 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 gate or transistor logic devices, discrete hardware components.

[0168] The electronic device is used to execute the technical solution in any of the aforementioned method embodiments, and its implementation principle and technical effect are similar and will not be repeated here.

[0169] An embodiment of the present application also provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the technical solution provided by any of the aforementioned method embodiments is implemented.

[0170] An embodiment of the present application also provides a computer program product, including a computer program, which is used to implement the technical solution provided by any of the aforementioned method embodiments when executed by a processor.

[0171] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0172] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A laser radar point cloud data enhancement method, characterized in that: include: Generate a screening area according to the collection location of the laser radar point cloud data to be supplemented; According to the screening area and the height values ​​of the collection positions of the plurality of laser radar point cloud data to be screened, the plurality of laser radar point cloud data to be screened are screened to obtain first point cloud data; Calculate the relative angle of collection of each first point cloud data according to the collection position of each first point cloud data and the collection position of the laser radar point cloud data to be supplemented; According to the relative angle of collection of each first point cloud data, all the first point cloud data are screened to obtain fourth point cloud data; The fourth point cloud data is merged with the laser radar point cloud data to be supplemented to obtain target point cloud data; Among them, the relative angle of collection of each first point cloud data is the angle between the projection line and the preset direction, the projection line is the projection of the line connecting the collection position of the first point cloud data and the collection position of the laser radar point cloud data to be supplemented on the horizontal plane, and the preset direction is the driving direction of the vehicle at the collection position of the laser radar point cloud data to be supplemented.

2. The method according to claim 1, characterized in that The method of screening the plurality of laser radar point cloud data to be screened according to the screening area and the height values ​​of the collection positions of the plurality of laser radar point cloud data to be screened to obtain the first point cloud data includes: Filter out point cloud data whose acquisition position belongs to the screening area from the plurality of laser radar point cloud data to be screened, and obtain second point cloud data; The first point cloud data are screened out from the second point cloud data according to the height value of the acquisition position of each second point cloud data.

3. The method according to claim 2, characterized in that The collection position of the first point cloud data and the collection position of the to-be-supplemented laser radar point cloud data belong to the same height range.

4. The method according to claim 2, characterized in that: The step of selecting the first point cloud data from each of the second point cloud data according to the height value of the acquisition position of each of the second point cloud data comprises: Clustering all the second point cloud data and the to-be-supplemented laser radar point cloud data according to the height value of the acquisition position to obtain a plurality of point cloud classes; The second point cloud data in the point cloud class including the laser radar point cloud data to be supplemented is used as the first point cloud data.

5. The method according to claim 1, characterized in that The method of screening the plurality of laser radar point cloud data to be screened according to the screening area and the height values ​​of the collection positions of the plurality of laser radar point cloud data to be screened to obtain the first point cloud data includes: Clustering all the laser radar point cloud data to be screened and the laser radar point cloud data to be supplemented according to the height value of the collection position to obtain multiple point cloud classes; Using the to-be-screened laser radar point cloud data in the point cloud class including the to-be-supplemented laser radar point cloud data as the third point cloud data; The point cloud data whose acquisition position belongs to the screening area in the third point cloud data is screened out to obtain the first point cloud data.

6. The method according to claim 1, characterized in that The method of screening all first point cloud data according to the relative angle of collection of each first point cloud data to obtain fourth point cloud data includes: Performing a point cloud screening operation on each preset angle range of at least two preset angle ranges to obtain fifth point cloud data corresponding to each preset angle range, and determining the fifth point cloud data corresponding to each preset angle range as the fourth point cloud data; wherein, The point cloud screening operation includes: for the preset angle range, if the number of first point cloud data collected whose relative angles belong to the preset angle range is greater than a preset number, then selecting the preset number of point cloud data from the first point cloud data collected whose relative angles belong to the preset angle range as the fifth point cloud data corresponding to the preset angle range.

7. The method according to claim 1, characterized in that According to the acquired collection positions of the plurality of laser radar point cloud data to be screened and the collection positions of the laser radar point cloud data to be supplemented, the plurality of laser radar point cloud data to be screened are screened, and before obtaining the first point cloud data, the method further includes: For each initial laser radar point cloud data, determining a preset collection range to which a collection position of the initial laser radar point cloud data belongs; If the number of points in the initial laser radar point cloud data is less than the preset sparse threshold corresponding to the preset acquisition range, the initial laser radar point cloud data is used as the laser radar point cloud data to be supplemented.

8. A laser radar point cloud data enhancement device, characterized in that: include: Processing modules for: Generate a screening area according to the collection location of the laser radar point cloud data to be supplemented; According to the screening area and the height values ​​of the collection positions of the plurality of laser radar point cloud data to be screened, the plurality of laser radar point cloud data to be screened are screened to obtain first point cloud data; Calculate the relative angle of collection of each first point cloud data according to the collection position of each first point cloud data and the collection position of the laser radar point cloud data to be supplemented; According to the relative angle of collection of each first point cloud data, all the first point cloud data are screened to obtain fourth point cloud data; A data enhancement module, used for fusing the fourth point cloud data with the laser radar point cloud data to be supplemented to obtain target point cloud data; Among them, the relative angle of collection of each first point cloud data is the angle between the projection line and the preset direction, the projection line is the projection of the line connecting the collection position of the first point cloud data and the collection position of the laser radar point cloud data to be supplemented on the horizontal plane, and the preset direction is the driving direction of the vehicle at the collection position of the laser radar point cloud data to be supplemented.

9. An electronic device, characterized in that: include: Processor, memory, communication interface; The memory is used to store executable instructions of the processor; Wherein, the processor is configured to execute the lidar point cloud data enhancement method described in any one of claims 1 to 7 by executing the executable instructions.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the laser radar point cloud data enhancement method described in any one of claims 1 to 7 is implemented.

11. A computer program product, characterized in that It includes a computer program, which, when executed by a processor, is used to implement the laser radar point cloud data enhancement method described in any one of claims 1 to 7.

Images