Feature extraction method, device and equipment of lidar point cloud and storage medium

Abstract

The application discloses a feature extraction method and device of a laser radar point cloud, equipment and a storage medium, relates to the technical field of laser radar, and solves the problem that the current point cloud positioning technology is not high in applicability. The specific scheme comprises the following steps: acquiring horizontal angles and pitching angles of each detection point included in point cloud data currently detected by a laser radar, and acquiring attribute parameters of the laser radar; determining a two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar, the two-dimensional matrix being used for indicating a current detection range of the laser radar; determining positions of the detection points in the two-dimensional matrix according to the horizontal angles, the pitching angles and the attribute parameters of the detection points; determining curvatures of the detection points according to the positions of the detection points in the two-dimensional matrix, and determining whether the detection points are line feature points or surface feature points according to the curvatures of the detection points.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of laser radar, in particular to a feature extraction method and device of laser radar point cloud, equipment and storage medium. BACKGROUND

[0002] With the maturity of laser radar technology, solid-state laser radar has become a mainstream sensor for automatic driving, and the point cloud positioning technology involved in automatic driving is a technology of positioning by using laser radar detection results.

[0003] The LOAM technology is to extract feature point clouds by using the scan lines of a mechanical laser radar and then perform matching to realize point cloud positioning. This method has high positioning efficiency and is currently the mainstream positioning method. Therefore, the commonly used point cloud feature extraction method is the method of extracting feature point clouds by using the scan lines of a mechanical laser radar in the LOAM technology. However, this method uses the scan lines of the mechanical laser radar itself to extract feature point clouds, and thus cannot be applied to other solid-state laser radars, which has the problem of low applicability. SUMMARY

[0004] The present application provides a feature extraction method and device of laser radar point cloud, equipment and storage medium, which can solve the problem of low applicability of the current point cloud feature extraction method.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0006] The present application provides a feature extraction method of laser radar point cloud, which comprises:

[0007] Obtaining the horizontal angle and the pitch angle of each detection point included in the point cloud data currently detected by the laser radar, and obtaining the attribute parameters of the laser radar;

[0008] Determining a two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar, the two-dimensional matrix being used to indicate the current detection range of the laser radar;

[0009] Determining the position of each detection point in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameters of each detection point;

[0010] Determining the curvature of each detection point according to the position of each detection point in the two-dimensional matrix, and determining whether each detection point is a line feature point or a surface feature point according to the curvature of each detection point.

[0011] In a possible implementation manner, the attribute parameters of the laser radar include a horizontal resolution, a vertical resolution, a horizontal field of view angle range and a pitch field of view angle range; and determining the two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar comprises:

[0012] determining a first size of the two-dimensional matrix according to the range of the vertical field of view angle and the vertical resolution;

[0013] determining a second size of the two-dimensional matrix according to the range of the horizontal field of view angle and the horizontal resolution.

[0014] In a possible implementation, the position of each detection point in the two-dimensional matrix is determined according to the horizontal angle, the vertical angle and the attribute parameter of each detection point, including:

[0015] determining a first position of the detection point according to the horizontal angle of each detection point and the minimum horizontal field of view angle of the range of the horizontal field of view angle;

[0016] determining a second position of the detection point according to the vertical angle of each detection point and the minimum vertical field of view angle of the range of the vertical field of view angle;

[0017] determining the position of each detection point in the two-dimensional matrix according to the first position and the second position.

[0018] In a possible implementation, the curvature of each detection point is determined according to the position of each detection point in the two-dimensional matrix, including:

[0019] obtaining at least one reference detection point included in the preset range of the detection point based on the position of the detection point;

[0020] determining the curvature of each detection point according to each reference detection point and the preset range.

[0021] In a possible implementation, the curvature of each detection point is determined according to each reference detection point and the preset range, including:

[0022] determining a first distance of each reference feature point according to the coordinates of each reference detection point, and determining a distance sum of each first distance;

[0023] determining a second distance of the detection point according to the coordinates of the detection point;

[0024] determining the curvature of the detection point according to the distance sum, the preset range and the second distance.

[0025] In a possible implementation, each detection point is determined to be a line feature point or a surface feature point according to the curvature of each detection point, including:

[0026] if the curvature is greater than a preset line feature threshold, the detection point is determined to be a line feature point;

[0027] if the curvature is less than a preset surface feature threshold, the detection point is determined to be a surface feature point.

[0028] In a possible implementation, after each detection point is determined to be a line feature point or a surface feature point according to the curvature of each detection point, the method further includes:

[0029] Line feature points and surface feature points in point cloud data are acquired, and point cloud positioning is performed based on the line feature points and the surface feature points.

[0030] In a second aspect, the application provides a feature extraction device for laser radar point cloud, which comprises:

[0031] The acquisition module is configured to acquire the horizontal angle and the pitch angle of each detection point included in the point cloud data currently detected by the laser radar, and acquire the attribute parameter of the laser radar.

[0032] The first determination module is configured to determine a two-dimensional matrix of the laser radar according to the attribute parameter of the laser radar, the two-dimensional matrix being used to indicate the current detection range of the laser radar.

[0033] The second determination module is configured to determine the position of each detection point in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameter of each detection point.

[0034] The third determination module is configured to determine the curvature of each detection point according to the position of each detection point in the two-dimensional matrix, and determine whether each detection point is a line feature point or a surface feature point according to the curvature of each detection point.

[0035] In a third aspect, the application provides an electronic device, which comprises a memory and a processor, the memory storing a computer program, and the computer program being executed by the processor to implement the feature extraction method for laser radar point cloud in the first aspect of the application.

[0036] In a fourth aspect, the application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the feature extraction method for laser radar point cloud in the first aspect of the application.

[0037] The technical scheme provided by the application has at least the following beneficial effects:

[0038] The feature extraction method of the laser radar point cloud provided in the embodiments of the present application comprises the following steps: acquiring the horizontal angle and the pitch angle of each detection point included in the point cloud data currently detected by the laser radar, and acquiring the attribute parameters of the laser radar; then determining a two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar, the two-dimensional matrix being the current detection range of the laser radar; determining the positions of the detection points in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameters of the detection points; finally determining the curvatures of the detection points according to the positions of the detection points in the two-dimensional matrix, and determining whether each detection point is a line feature point or a surface feature point according to the curvatures of the detection points. The feature extraction method of the laser radar point cloud provided in the embodiments of the present application has high applicability, because the positions of the points in the two-dimensional matrix are determined according to the horizontal angle and the pitch angle of each point in the point cloud data and the attribute parameters of the radar, which is equivalent to sorting the points according to the horizontal angle and the pitch angle, then calculating the curvatures of the points, and then extracting the feature points according to the curvatures of the points, so that the feature points can be extracted without using the scanning line of the mechanical laser radar in the extraction process, and the method can be applied to any laser radar. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 An internal structure schematic diagram of a computer device provided in the embodiments of the present application;

[0040] Figure 2 A flowchart of a feature extraction method of a laser radar point cloud provided in the embodiments of the present application;

[0041] Figure 3 A structure diagram of a feature extraction device of a laser radar point cloud provided in the embodiments of the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0043] Hereinafter, the terms “first” and “second” are only used for description purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with “first” and “second” can explicitly or implicitly include one or more features. In the description of the embodiments of the present disclosure, unless otherwise specified, the meaning of “a plurality of” is two or more.

[0044] In addition, the use of "based on" or "according to" means open and inclusive, as a process, step, calculation or other action that is "based on" or "according to" one or more conditions or values can be based on additional conditions or values in practice.

[0045] With the maturity of laser radar technology, solid-state laser radar has become the mainstream sensor of autonomous driving, and the point cloud positioning technology involved in autonomous driving is the technology of positioning by using the detection results of laser radar.

[0046] The LOAM technology is to extract feature point clouds by using the scanning lines of mechanical laser radar and then match to realize point cloud positioning. The positioning efficiency of this method is high, and it is currently the mainstream positioning method. Therefore, the commonly used point cloud feature extraction method is the method of extracting feature point clouds by using the scanning lines of mechanical laser radar in the LOAM technology. However, this method uses the scanning lines of mechanical laser radar to extract feature point clouds, and therefore cannot be applied to other solid-state laser radars, which has the problem of low applicability.

[0047] To solve the above problems, the embodiment of the present application provides a feature extraction method for laser radar point cloud. The horizontal angle and the pitch angle of each detection point included in the point cloud data currently detected by the laser radar are obtained, and the attribute parameters of the laser radar are obtained. Then, the two-dimensional matrix of the laser radar is determined according to the attribute parameters of the laser radar, the two-dimensional matrix refers to the current detection range of the laser radar, and the position of each detection point in the two-dimensional matrix is determined according to the horizontal angle, the pitch angle and the attribute parameters of each detection point. Finally, the curvature of each detection point is determined according to the position of each detection point in the two-dimensional matrix, and each detection point is determined to be a line feature point or a surface feature point according to the curvature of each detection point. The feature extraction method for laser radar point cloud provided by the embodiment of the present application is to determine the position of each point in the two-dimensional matrix according to the horizontal angle and the pitch angle of each point in the point cloud data and the attribute parameters of the radar, which is equivalent to sorting each point according to the horizontal angle and the pitch angle of each point, then calculating the curvature of each point, and then extracting the feature point according to the curvature of each point. In the extraction process, the scanning line of the mechanical laser radar itself is not used, so it can be applied to any kind of laser radar, and has high applicability.

[0048] The execution subject of the feature extraction method for laser radar point cloud provided by the embodiment of the present application can be a computer device, a terminal device, or a server. The terminal device can be various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, etc. The present application does not make specific limitation.

[0049] Figure 1 An internal structure schematic diagram of a computer device provided by the embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the computer device includes a central processing unit (CPU), a memory, a storage device, a keyboard, a mouse, a display screen, a communication interface and a bus. Figure 1As shown, the computer device includes a processor and a memory connected through a system bus. Among them, the processor is used to provide computing and control capabilities. The memory can include a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The computer program can be executed by the processor to implement the steps of the feature extraction method of the laser radar point cloud provided by each of the above embodiments. The internal memory provides a cache running environment for the operating system and the computer program in the non-volatile storage medium.

[0050] Those skilled in the art can understand that, Figure 1 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0051] Based on the above execution subject, the embodiment of the present application provides a feature extraction method of laser radar point cloud. As Figure 2 As shown, the method comprises the following steps:

[0052] Step 201, acquiring the horizontal angle and the pitch angle of each detection point included in the point cloud data currently detected by the laser radar, and acquiring the attribute parameters of the laser radar.

[0053] Among them, each detection point is each point in the point cloud data, and a plurality of detection points constitute the point cloud data. The attribute parameters of the laser radar include: the horizontal resolution of the laser radar, the vertical resolution of the laser radar, the horizontal field of view angle range of the laser radar and the pitch field of view angle range of the laser radar.

[0054] Optionally, the process of acquiring the horizontal angle and the pitch angle of each detection point can be: acquiring the coordinates of each detection point, then calculating the pitch angle of the detection point by using formula 1, and obtaining the horizontal angle of the detection point by using formula 2.

[0055]

[0056] azmuth=atan(y / x) Formula 2

[0057] Among them, pitch in formula 1 is the pitch angle of the detection point, azmuth in formula 2 is the horizontal angle of the detection point, x, y, z in formula 1 and formula 2 are the coordinates of the detection point, and atan in formula 1 and formula 2 is the inverse tangent function.

[0058] Step 202, determining a two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar.

[0059] Among them, the two-dimensional matrix is used to indicate the current detection range of the laser radar.

[0060] Optionally, the current detection range of the laser radar, that is, the size of the two-dimensional matrix, can be determined according to the horizontal resolution of the laser radar, the vertical resolution of the laser radar, the horizontal field of view range of the laser radar, and the pitch field of view range of the laser radar.

[0061] In step 203, the positions of the detection points in the two-dimensional matrix are determined according to the horizontal angle, the pitch angle, and the attribute parameter of each detection point.

[0062] It can be understood that determining the positions of the detection points in the two-dimensional matrix according to the horizontal angle and the pitch angle of the detection points is equivalent to sorting the detection points according to the pitch angle and the horizontal angle, which facilitates subsequent extraction of feature points.

[0063] In step 204, the curvatures of the detection points are determined according to the positions of the detection points in the two-dimensional matrix, and the detection points are determined to be line feature points or surface feature points according to the curvatures of the detection points.

[0064] Optionally, if the curvature is greater than a preset line feature threshold, the detection point is determined to be a line feature point; and if the curvature is less than a preset surface feature threshold, the detection point is determined to be a surface feature point.

[0065] For example, the line feature threshold can be 10, and the surface feature threshold can be 0.1, or the line feature threshold or the surface feature threshold can be set according to different feature extraction accuracies or application scenarios of the feature point cloud, and the embodiments of the present application do not make specific limitations thereto.

[0066] In actual execution, after the detection points are determined to be line feature points or surface feature points, the line feature point set and the surface feature point set included in the point cloud data are obtained, and the point cloud points are determined based on the line feature point set and the surface feature point set.

[0067] The feature extraction method of the laser radar point cloud provided in the embodiments of the present application comprises the following steps: acquiring the horizontal angle and the pitch angle of each detection point included in the point cloud data currently detected by the laser radar, and acquiring the attribute parameters of the laser radar; then determining a two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar, wherein the two-dimensional matrix refers to the current detection range of the laser radar; determining the position of each detection point in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameters of each detection point; and finally determining the curvature of each detection point according to the position of each detection point in the two-dimensional matrix, and determining whether each detection point is a line feature point or a surface feature point according to the curvature of each detection point. The feature extraction method of the laser radar point cloud provided in the embodiments of the present application is characterized in that the position of each point in the two-dimensional matrix is determined according to the horizontal angle and the pitch angle of each point in the point cloud data and the attribute parameters of the radar, which is equivalent to sorting each point according to the horizontal angle and the pitch angle of each point, then calculating the curvature of each point, and then extracting the feature points according to the curvature of each point. In the extraction process, the scanning line of the mechanical laser radar itself is not used, so the method can be applied to any kind of laser radar and has high applicability.

[0068] Optionally, the process of determining the two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar in the above step 202 can be as follows:

[0069] The first size of the two-dimensional matrix is determined according to the pitch field of view angle range and the vertical resolution, and the second size of the two-dimensional matrix is determined according to the horizontal field of view angle range and the horizontal resolution.

[0070] For example, the size of the two-dimensional matrix (Matrix) can be set as M*N, wherein M is the first size and N is the second size. Then, M=pitch field of view angle range / vertical resolution, and N=horizontal FOV range / horizontal resolution, so as to obtain the first size and the second size of the two-dimensional matrix.

[0071] Optionally, the process of determining the position of each detection point in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameters of each detection point can be as follows:

[0072] The first position of the detection point is determined according to the horizontal angle of each detection point and the minimum horizontal field of view angle of the horizontal field of view angle range, then the second position of the detection point is determined according to the pitch angle of each detection point and the minimum pitch field of view angle of the pitch field of view angle range, and finally the position of each detection point in the two-dimensional matrix is determined according to the first position and the second position.

[0073] In actual execution, the first position can be n, and the second position can be m, so that m=(pitch-pitch0) / vertical resolution, and n=(azmuth-azmuth0) / horizontal resolution. In the formula, pitch is the pitch angle of the detection point, pitch0 is the minimum pitch angle in the pitch field of view, azmuth is the horizontal angle of the detection point, and azmuth0 is the minimum horizontal angle in the horizontal field of view.

[0074] It should be noted that if the positions of multiple detection points in the two-dimensional matrix are the same, the distance of the point needs to be calculated, the detection point with the minimum distance is determined as the detection point at the position, and other detection points at the position are deleted, so as to facilitate subsequent feature extraction.

[0075] Optionally, the distance of the detection point can be calculated by formula 3. In the formula, dis represents the distance of the detection point, and x, y and z are the coordinates of the detection point.

[0076]

[0077] Optionally, the process of determining the curvature of each detection point according to the position of each detection point in the two-dimensional matrix in step 204 can be: obtaining at least one reference detection point included in the preset range of the detection point based on the position of the detection point, and then determining the curvature of each detection point according to each reference detection point and the preset range.

[0078] In the formula, the preset range of each detection point is a preset range with the detection point as the center, and the size of the preset range can be a 5*5 range in the two-dimensional matrix, or can be set according to the size of the two-dimensional matrix, which is not limited in the present application. Correspondingly, obtaining at least one reference detection point included in the preset range of the detection point can be understood as obtaining the detection points included in the 5*5 range around the detection point.

[0079] In one possible implementation, the process of determining the curvature of each detection point according to each reference detection point and the preset range can be:

[0080] determining the first distance of each reference feature point according to the coordinates of each reference detection point, and determining the sum of the distances of each first distance, then determining the second distance of the detection point according to the coordinates of the detection point, and finally determining the curvature of the detection point according to the sum of the distances, the preset range and the second distance.

[0081] In actual execution, the curvature of each detection point can be calculated by formula 4.

[0082]

[0083] In the formula, curvature(m, n) represents the curvature of the detection point, dis represents the distance of the detection point, and x, y and z are the coordinates of the detection point.ij dis represents the distance of each reference detection point, that is, the first distance mn w*w represents the size of the preset range.

[0084] It can be understood that, in the calculation of the curvature of the detection point, the detection points around the point are used, which more fully represents the local features of the detection point than the traditional LOAM technology which only uses the detection points on a scanning line to calculate the curvature, so that the extracted feature points are more accurate.

[0085] The feature extraction method of the laser radar point cloud provided in the embodiment of the application comprises the following steps: acquiring the horizontal angle and the pitch angle of each detection point included in the point cloud data currently detected by the laser radar, and acquiring the attribute parameters of the laser radar; then determining a two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar, the two-dimensional matrix being used to indicate the current detection range of the laser radar; determining the positions of the detection points in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameters of each detection point; finally, determining the curvature of each detection point according to the positions of the detection points in the two-dimensional matrix, and determining whether each detection point is a line feature point or a surface feature point according to the curvature of each detection point. The feature extraction method of the laser radar point cloud provided in the embodiment of the application is equivalent to sorting each point according to the horizontal angle and the pitch angle of each point and then calculating the curvature of each point, and then extracting the feature points according to the curvature of each point, so that the feature extraction method is applicable to any laser radar and has high applicability.

[0086] Further, in the calculation of the curvature of the detection point, the detection points around the point are used, which more fully represents the local features of the detection point than the traditional LOAM technology which only uses the detection points on a scanning line to calculate the curvature, so that the extracted feature points are more accurate.

[0087] As shown in Figure 3 , the embodiment of the application provides a feature extraction device for laser radar point cloud, which comprises:

[0088] The acquisition module 11 is configured to acquire the horizontal angle and the pitch angle of each detection point included in the point cloud data currently detected by the laser radar, and acquire the attribute parameters of the laser radar.

[0089] The first determination module 12 is configured to determine a two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar, the two-dimensional matrix being used to indicate the current detection range of the laser radar.

[0090] The second determining module 13 is configured to determine the position of each detection point in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameter of each detection point.

[0091] The third determining module 14 is configured to determine the curvature of each detection point according to the position of each detection point in the two-dimensional matrix, and determine whether each detection point is a line feature point or a surface feature point according to the curvature of each detection point.

[0092] In an embodiment, the attribute parameter of the laser radar comprises a horizontal resolution, a vertical resolution, a horizontal field of view range and a pitch field of view range; and the first determining module 12 is specifically configured to:

[0093] determine the first size of the two-dimensional matrix according to the pitch field of view range and the vertical resolution;

[0094] determine the second size of the two-dimensional matrix according to the horizontal field of view range and the horizontal resolution.

[0095] In an embodiment, the second determining module 13 is specifically configured to:

[0096] determine the first position of each detection point according to the horizontal angle of each detection point and the minimum horizontal field of view angle of the horizontal field of view range;

[0097] determine the second position of each detection point according to the pitch angle of each detection point and the minimum pitch field of view angle of the pitch field of view range;

[0098] determine the position of each detection point in the two-dimensional matrix according to the first position and the second position.

[0099] In an embodiment, the third determining module 14 is specifically configured to:

[0100] obtain at least one reference detection point included in the preset range of the detection point based on the position of the detection point;

[0101] determine the curvature of each detection point according to each reference detection point and the preset range.

[0102] In an embodiment, the third determining module 14 is specifically configured to:

[0103] determine the first distance of each reference feature point according to the coordinates of each reference detection point, and determine the sum of distances of each first distance;

[0104] determine the second distance of the detection point according to the coordinates of the detection point;

[0105] determine the curvature of the detection point according to the sum of distances, the preset range and the second distance.

[0106] In an embodiment, the third determining module 14 is specifically configured to:

[0107] If the curvature is greater than the preset line feature threshold, it is determined that the detection point is a line feature point.

[0108] If the curvature is less than the preset surface feature threshold, it is determined that the detection point is a surface feature point.

[0109] In one embodiment, the device further comprises a positioning module 15, which is configured to

[0110] Obtaining the line feature points and the surface feature points in the point cloud data, and performing point cloud positioning based on the line feature points and the surface feature points.

[0111] The laser radar point cloud feature extraction device provided in the embodiment can execute the method embodiments described above, and has similar implementation principles and technical effects, which will not be described in detail here.

[0112] For specific limitations of the laser radar point cloud feature extraction device, refer to the limitations of the laser radar point cloud feature extraction method described above, which will not be described in detail here. Each module in the laser radar point cloud feature extraction device described above can be realized by software, hardware, and a combination thereof, in whole or in part. The above-mentioned modules can be embedded in or independent of the processor in the server in hardware form, or can be stored in the memory in the server in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0113] In another embodiment of the present application, a computer device is also provided, which includes a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to realize the steps of the laser radar point cloud feature extraction method according to the embodiments of the present application.

[0114] In another embodiment of the present application, a computer readable storage medium is also provided, which stores a computer program, and the computer program is executed by the processor to realize the steps of the laser radar point cloud feature extraction method according to the embodiments of the present application.

[0115] In another embodiment of the present application, a computer program product is also provided, which includes computer instructions, when the computer instructions run on the laser radar point cloud feature extraction device, the laser radar point cloud feature extraction device executes each step of the laser radar point cloud feature extraction method in the method flow shown in the method embodiments described above.

[0116] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer executes the computer instructions, all or part of the processes or functions according to the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (for example, coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (for example, infrared, wireless, microwave, etc.). The computer readable storage medium can be any available medium that can be accessed by a computer or data storage device such as one or more servers, data centers, etc. integrated with one or more media. The available media can be magnetic media (for example, floppy disk, hard disk, magnetic tape), optical media (for example, DVD), or semiconductor media (for example, solid state disk (SSD)) and the like.

[0117] The technical features of the above embodiments can be combined in any way. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, but as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0118] The above embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it should not be understood as limiting the scope of the patent. It should be noted that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application. Therefore, the scope of the patent of the present application should be subject to the appended claims.

Claims

1. A feature extraction method of a laser radar point cloud, characterized by, The method comprises: acquiring horizontal angles and pitch angles of each detection point included in point cloud data currently detected by a laser radar, and acquiring attribute parameters of the laser radar; determining a two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar, the two-dimensional matrix being used to indicate a current detection range of the laser radar; determining positions of each detection point in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameters of each detection point; determining curvatures of each detection point according to the positions of each detection point in the two-dimensional matrix, and determining each detection point as a line feature point or a surface feature point according to the curvatures of each detection point; the attribute parameters of the laser radar comprise a horizontal resolution, a vertical resolution, a horizontal field of view angle range and a pitch field of view angle range; and the determining of the two-dimensional matrix of the laser radar according to the attribute parameters of the laser radar comprises: determining a first size of the two-dimensional matrix according to the pitch field of view angle range and the vertical resolution; determining a second size of the two-dimensional matrix according to the horizontal field of view angle range and the horizontal resolution; the determining of the positions of each detection point in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameters of each detection point comprises: determining a first position of each detection point according to the horizontal angle of each detection point and a minimum horizontal field of view angle of the horizontal field of view angle range; determining a second position of each detection point according to the pitch angle of each detection point and a minimum pitch field of view angle of the pitch field of view angle range; determining the position of each detection point in the two-dimensional matrix according to the first position and the second position.

2. The method of claim 1, wherein, the determining of the curvatures of each detection point according to the positions of each detection point in the two-dimensional matrix comprises: acquiring at least one reference detection point included in a preset range of the detection point based on the position of the detection point; determining the curvature of each detection point according to each reference detection point and the preset range.

3. The method of claim 2, wherein, the determining of the curvature of each detection point according to each reference detection point and the preset range comprises: determining a first distance of each reference feature point according to the coordinates of each reference detection point, and determining a distance sum of each first distance; determining a second distance of the detection point according to the coordinates of the detection point; determining the curvature of the detection point according to the distance sum, the preset range and the second distance.

4. The method of claim 1, wherein, the determining of each detection point as a line feature point or a surface feature point according to the curvature of each detection point comprises: if the curvature is greater than a preset line feature threshold value, determining the detection point as a line feature point; if the curvature is less than a preset surface feature threshold value, determining the detection point as a surface feature point.

5. The method of claim 1, wherein, after the determining of each detection point as a line feature point or a surface feature point according to the curvature of each detection point, the method further comprises: acquiring line feature points and surface feature points in the point cloud data, and performing point cloud positioning based on the line feature points and the surface feature points.

6. An apparatus for feature extraction of a lidar point cloud, the apparatus comprising: The device comprises: an acquisition module, configured to acquire horizontal angles and pitch angles of each detection point included in point cloud data currently detected by a laser radar, and acquire attribute parameters of the laser radar; a first determining module configured to determine a two-dimensional matrix of the laser radar according to attribute parameters of the laser radar, the two-dimensional matrix being used to indicate a current detection range of the laser radar; a second determining module configured to determine positions of the detection points in the two-dimensional matrix according to the horizontal angle, the pitch angle and the attribute parameters of each detection point; a third determining module configured to determine curvatures of the detection points according to the positions of the detection points in the two-dimensional matrix, and determine whether each detection point is a line feature point or a surface feature point according to the curvatures of the detection points; the attribute parameters of the laser radar include a horizontal resolution, a vertical resolution, a horizontal field of view range and a pitch field of view range; the first determining module is specifically configured to determine a first size of the two-dimensional matrix according to the pitch field of view range and the vertical resolution, and determine a second size of the two-dimensional matrix according to the horizontal field of view range and the horizontal resolution; the second determining module is specifically configured to determine a first position of each detection point according to the horizontal angle of the detection point and a minimum horizontal field of view angle of the horizontal field of view range, and determine a second position of the detection point according to the pitch angle of the detection point and a minimum pitch field of view angle of the pitch field of view range; the positions of the detection points in the two-dimensional matrix are determined according to the first position and the second position.

7. An electronic device, comprising: a memory and a processor, the memory storing a computer program, the computer program being executed by the processor to implement the feature extraction method of the laser radar point cloud according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, a computer program stored thereon, the computer program being executed by a processor to implement the feature extraction method of the laser radar point cloud according to any one of claims 1 to 5.

Images