Laser point cloud segmentation method, device, electronic device and storage medium

Abstract

The present invention discloses a laser point cloud segmentation method, device, equipment and storage medium. The method comprises: obtaining a first segmentation result of a laser point cloud; wherein the first segmentation result comprises a first ground point cloud and a first non-ground point cloud; clustering the first non-ground point cloud in a pre-set target area to obtain at least one point cloud cluster in the target area; determining the point cloud cluster features that match each point cloud cluster based on the coordinates of the midpoints of each point cloud cluster, and determining the type of each point cloud cluster based on the point cloud cluster features that match each point cloud cluster; determining a second segmentation result of the laser point cloud based on the type of each point cloud cluster and the first segmentation result. This technical solution solves the problems of low accuracy and poor reliability of laser point cloud segmentation under complex working conditions, and can further enhance the reliability of ground and non-ground point cloud segmentation while improving the accuracy of laser point cloud segmentation.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a laser point cloud segmentation method, device, electronic device and storage medium. Background Art

[0002] In the field of autonomous driving, laser point clouds acquired by autonomous vehicles using LiDAR (LiDAR) are crucial for obstacle detection, path planning, and other tasks. The segmentation of these point clouds determines the accuracy of obstacle detection and the reliability of path planning.

[0003] Currently, existing technologies usually distinguish between ground point clouds and non-ground point clouds based on the geometric features of laser point clouds, such as height and curvature, through threshold comparison; or, based on segmentation models, the features of laser point cloud data are extracted to achieve the division of ground point clouds and non-ground point clouds.

[0004] However, in complex driving conditions, such as those with large slope variations and bumpy driving conditions, where the laser point cloud may contain errors in acquisition, existing technologies struggle to accurately segment ground and non-ground point clouds, and are prone to segmentation errors. Therefore, a method for performing segmentation verification after the initial segmentation of the laser point cloud is urgently needed to improve the reliability of the region segmentation. Summary of the Invention

[0005] The present invention provides a laser point cloud segmentation method, device, equipment and storage medium to solve the problems of low laser point cloud segmentation accuracy and poor reliability under complex working conditions. While improving the accuracy of laser point cloud segmentation, it can further enhance the reliability of ground and non-ground point cloud division.

[0006] According to one aspect of the present invention, a laser point cloud segmentation method is provided, the method comprising:

[0007] Obtaining a first segmentation result of the laser point cloud; wherein the first segmentation result includes a first ground point cloud and a first non-ground point cloud;

[0008] Performing clustering processing on a first non-ground point cloud within a preset target area to obtain at least one point cloud cluster within the target area;

[0009] According to the coordinates of the midpoints of each point cloud cluster, the point cloud cluster features of each point cloud cluster are determined, and according to the point cloud cluster features of each point cloud cluster, the type of each point cloud cluster is determined;

[0010] A second segmentation result of the laser point cloud is determined according to the type of each point cloud cluster and the first segmentation result.

[0011] According to another aspect of the present invention, a laser point cloud segmentation device is provided, the device comprising:

[0012] A first segmentation result acquisition module, configured to acquire a first segmentation result of the laser point cloud; wherein the first segmentation result includes a first ground point cloud and a first non-ground point cloud;

[0013] a point cloud cluster generation module, configured to perform clustering processing on a first non-ground point cloud within a preset target area to obtain at least one point cloud cluster within the target area;

[0014] The point cloud cluster type determination module is used to determine the point cloud cluster features matched by each point cloud cluster according to the coordinates of the midpoints of each point cloud cluster, and to determine the type of each point cloud cluster according to the point cloud cluster features matched by each point cloud cluster;

[0015] The second segmentation result determination module is used to determine the second segmentation result of the laser point cloud according to the type of each point cloud cluster and the first segmentation result.

[0016] According to another aspect of the present invention, an electronic device is provided, comprising:

[0017] at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor. The computer program is executed by the at least one processor so that the at least one processor can perform the laser point cloud segmentation method described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the laser point cloud segmentation method according to any embodiment of the present invention when executed.

[0021] The technical solution of the embodiment of the present invention obtains a first segmentation result of the laser point cloud, clusters the first non-ground point cloud in the first segmentation result within a preset target area, and obtains at least one point cloud cluster within the target area. The solution then determines the matching point cloud cluster features of each point cloud cluster based on the coordinates of the midpoints of each point cloud cluster, and determines the type of each point cloud cluster based on the matching point cloud cluster features of each point cloud cluster. Finally, the solution determines a second segmentation result of the laser point cloud based on the point cloud cluster type and the first segmentation result. This technical solution solves the problems of low accuracy and poor reliability of laser point cloud segmentation under complex working conditions, improving the accuracy of laser point cloud segmentation while further enhancing the reliability of the classification of ground and non-ground point clouds.

[0022] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 This is a flow chart of a laser point cloud segmentation method provided according to the first embodiment of the present invention;

[0025] Figure 2 This is a flow chart of a laser point cloud segmentation method provided according to the second embodiment of the present invention;

[0026] Figure 3 This is a structural diagram of a laser point cloud segmentation device provided according to a third embodiment of the present invention;

[0027] Figure 4 It is a structural diagram of an electronic device for implementing the laser point cloud segmentation method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention 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 interchangeable where appropriate, so that the embodiments of the present invention 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. The acquisition, storage, use, processing, etc. of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0030] Example 1

[0031] Figure 1 A flowchart of a laser point cloud segmentation method is provided for the first embodiment of the present invention. This embodiment is applicable to the scene of segmenting laser point clouds in an autonomous driving environment. The method can be executed by a laser point cloud segmentation device, which can be implemented in the form of hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0032] S110. Obtain a first segmentation result of the laser point cloud; wherein the first segmentation result includes a first ground point cloud and a first non-ground point cloud.

[0033] This solution can be implemented by an autonomous vehicle, which uses a LiDAR to acquire laser point cloud data of the driving environment. Based on a ground segmentation algorithm, the autonomous vehicle can obtain a first segmentation result for the laser point cloud, dividing the laser point cloud of the driving environment into a first ground point cloud and a first non-ground point cloud. The ground segmentation algorithm can be implemented using a threshold comparison based on geometric features or a segmentation model constructed based on machine learning principles.

[0034] Due to the large changes in ground slope in the driving environment and vehicle bumps, it is easy to cause erroneous segmentation of point clouds in some areas of the laser point cloud. Therefore, after the ground segmentation algorithm outputs the first segmentation result, the autonomous driving vehicle can verify the first segmentation result to correct the erroneous segmentation of point clouds in some areas, improve the accuracy of laser point cloud segmentation, and thus enhance the driving safety and reliability of the autonomous driving vehicle.

[0035] S120 : Perform clustering processing on the first non-ground point cloud in the preset target area to obtain at least one point cloud cluster in the target area.

[0036] The autonomous vehicle can determine a target area based on information such as the lidar's sensing range, the vehicle's status, and the priority of areas in the driving environment. The target area can be an area prone to laser point cloud segmentation errors or an area of ​​user interest, such as the area in front of and behind the autonomous vehicle. The autonomous vehicle can cluster the first non-ground point cloud within the target area based on a clustering model to generate one or more point cloud clusters. The clustering model can be implemented using a density-based clustering algorithm.

[0037] S130 , determining the point cloud cluster features matched by each point cloud cluster according to the coordinates of the midpoints of each point cloud cluster, and determining the type of each point cloud cluster according to the point cloud cluster features matched by each point cloud cluster.

[0038] As will be appreciated, the autonomous vehicle can obtain the coordinates of the midpoints of each point cloud cluster and, based on these coordinates, determine the characteristics of the point cloud cluster. These characteristics may include one or more of average height, maximum height, minimum height, maximum height difference, and point cloud distribution. Based on these characteristics, the autonomous vehicle can determine the type of each point cloud cluster. These types may include ground point clouds and non-ground point clouds.

[0039] S140 : Determine a second segmentation result of the laser point cloud according to the type of each point cloud cluster and the first segmentation result.

[0040] It is easy to understand that after determining the type of each point cloud cluster, the autonomous driving vehicle can merge the ground point cloud into the first ground point cloud, and merge the non-ground point cloud into the first non-ground point cloud to obtain a second segmentation result of the laser point cloud.

[0041] The technical solution of the embodiment of the present invention obtains a first segmentation result of the laser point cloud, clusters the first non-ground point cloud in the first segmentation result within a preset target area, and obtains at least one point cloud cluster within the target area. The solution then determines the matching point cloud cluster features of each point cloud cluster based on the coordinates of the midpoints of each point cloud cluster, and determines the type of each point cloud cluster based on the matching point cloud cluster features of each point cloud cluster. Finally, the solution determines a second segmentation result of the laser point cloud based on the point cloud cluster type and the first segmentation result. This technical solution solves the problems of low accuracy and poor reliability of laser point cloud segmentation under complex working conditions, improving the accuracy of laser point cloud segmentation while further enhancing the reliability of the classification of ground and non-ground point clouds.

[0042] Example 2

[0043] Figure 2 This is a flow chart of a laser point cloud segmentation method provided in the second embodiment of the present invention. This embodiment is based on the above embodiment and is refined. Figure 2 As shown, the method includes:

[0044] S210. Obtain a first segmentation result of the laser point cloud; wherein the first segmentation result includes a first ground point cloud and a first non-ground point cloud.

[0045] In this solution, the laser point cloud can be represented by a three-dimensional coordinate system. Taking the three-dimensional rectangular system as an example, each point in the laser point cloud can be represented as a three-dimensional coordinate.

[0046] S220 : Perform clustering processing on the first non-ground point cloud in the preset target area to obtain at least one point cloud cluster in the target area.

[0047] In a feasible solution, clustering the first non-ground point cloud in the preset target area to obtain at least one point cloud cluster in the target area includes:

[0048] Based on a predetermined clustering model, a first non-ground point cloud in a preset target area is clustered to obtain at least one point cloud cluster in the target area; wherein the clustering model is one of a Euclidean distance clustering model, a mean shift clustering model, and a spatial clustering model with noise.

[0049] Specifically, the Euclidean distance clustering model can be implemented based on the Euclidean clustering algorithm, the mean shift clustering model can be implemented based on the mean shift clustering algorithm, and the spatial clustering model with noise can be implemented based on the DBSCAN clustering algorithm (Density-Based Spatial Clustering of Applications with Noise).

[0050] In a specific example, the step of implementing clustering processing on the first non-ground point cloud using a spatial clustering model with noise may include:

[0051] (1) The autonomous vehicle can randomly select an unvisited point P in the first non-ground point cloud;

[0052] (2) If P is a core point, create a point cloud cluster and add P to the point cloud cluster;

[0053] (3) Determine all points that are density-reachable by P and add them to the point cloud cluster;

[0054] (4) For each newly added point, if it is a core point, continue to find its density-reachable points and add them to the cluster;

[0055] (5) Continue to use the density-reachable points of P as new points P, and return to execute steps (3)-(4) until there is no density-reachable point in P;

[0056] (6) After forming a point cloud cluster, return to step (1) to select a new unvisited point and continue iterating until all points in the first non-ground point cloud have been visited.

[0057] It should be noted that the DBSCAN model defines clusters by density, presetting a radius ε and a minimum density threshold MinPts. For a point P, if there are at least MinPts points (including P) within the radius ε, then P is considered a core point. If the number of points within the radius ε is less than MinPts, but P is within the ε neighborhood of other core points, then P is considered a boundary point. Except for core points and boundary points, other points are considered noise points. For two core points P and Q, there is a core point sequence P1, P2, ..., P n , if P1=P, P n =Q, and for any adjacent core point P i and P i+1 , P i+1 In P i If P and Q are in the ε neighborhood of , then P and Q are said to be density-reachable. For a core point P and a boundary point Q, there exists a core point sequence P1, P2, ..., P n , if P1=P, P n =Q, and for any adjacent core point P i and P i+1 , P i+1 In P i If P and Q are within the ε neighborhood of , then P and Q are said to be density connected.

[0058] After obtaining the point cloud clusters within the target area, the autonomous vehicle can extract the features of each point cloud cluster and determine the point cloud cluster type based on these features. In this solution, the point cloud cluster features may include average height, maximum height difference, and point cloud straight line distribution evaluation value. The autonomous vehicle can calculate the average height, maximum height difference, and point cloud straight line distribution evaluation value for each point cloud cluster and determine the point cloud cluster type based on these values.

[0059] S230 , calculating the mean value of the coordinates of the points in each point cloud cluster, determining the center point of each point cloud cluster based on the mean value of the coordinates of the points, and taking the height coordinate of the center point as the average height.

[0060] It can be understood that the autonomous driving vehicle can calculate the mean coordinate of each point in the point cloud cluster, use the mean coordinate as the coordinate of the center point, and use the height coordinate of the center point as the average height of the point cloud cluster.

[0061] S240 , taking the difference between the maximum height coordinate and the minimum height coordinate in each point cloud cluster as the maximum height difference.

[0062] The autonomous driving vehicle can compare the height coordinates of each point in the point cloud cluster, determine the maximum height coordinate and the minimum height coordinate, and calculate the difference between the maximum height coordinate and the minimum height coordinate, and use the difference between the maximum height coordinate and the minimum height coordinate as the maximum height difference of the point cloud cluster.

[0063] S250 , determining the covariance matrix of each point cloud cluster match based on the coordinates of the midpoints of each point cloud cluster.

[0064] It is easy to understand that based on the point coordinates in the point cloud cluster, the autonomous driving vehicle can calculate the covariance matrix corresponding to the point cloud cluster.

[0065] S260 , performing singular value decomposition on the covariance matrix of each point cloud cluster match to obtain eigenvalues ​​of each coordinate direction match.

[0066] By performing singular value decomposition on the covariance matrix corresponding to the point cloud cluster, the autonomous driving vehicle can obtain the eigenvalues ​​corresponding to each coordinate direction of the point cloud cluster.

[0067] S270 , sorting the eigenvalues ​​matched in each coordinate direction, and determining the point cloud straight line distribution evaluation value based on the eigenvalue sorting result.

[0068] The autonomous vehicle can sort the eigenvalues ​​corresponding to each coordinate direction and determine a first eigenvalue and a second eigenvalue based on the eigenvalue sorting results. The first eigenvalue can be the maximum eigenvalue, and the second eigenvalue can be the largest eigenvalue smaller than the maximum eigenvalue, i.e., the second largest eigenvalue. The autonomous vehicle can use the ratio of the first eigenvalue to the second eigenvalue as the point cloud straight line distribution evaluation value.

[0069] S280 : Determine the type of each point cloud cluster according to the matched point cloud cluster features of each point cloud cluster.

[0070] It is easy to understand that the autonomous driving vehicle can determine the type of each point cloud cluster based on the average height, maximum height difference and point cloud straight line distribution evaluation value of each point cloud cluster.

[0071] In this solution, optionally, determining the type of each point cloud cluster according to the point cloud cluster features matched by each point cloud cluster includes:

[0072] If the average height of the point cloud cluster is less than the preset average height threshold, the maximum height difference is less than the preset maximum height difference threshold, and the point cloud straight line distribution evaluation value is greater than the preset evaluation value threshold, then the point cloud cluster type is determined to be a ground point cloud;

[0073] If the average height of the point cloud cluster is greater than or equal to the preset average height threshold, the maximum height difference is greater than or equal to the preset maximum height difference threshold, and the point cloud straight line distribution evaluation value is less than or equal to the preset evaluation value threshold, then the point cloud cluster type is determined to be a non-ground point cloud;

[0074] The average height threshold is determined based on the average height of the ground point cloud within a preset range of the center point of the point cloud cluster in the target area.

[0075] It should be noted that within the target area, the autonomous vehicle can calculate the average height of the ground point clouds within a preset range, centered on the center point of each point cloud cluster, and set the average height threshold of the point cloud cluster based on the average height of the ground point clouds. For example, the average height H of the ground point clouds within a circular area with a radius R and a point cloud cluster center point as the center can be calculated. A tolerance parameter h can be added to H, and H + h can be used as the average height threshold of the point cloud cluster.

[0076] By setting the average height threshold according to the average height of the ground point cloud within the preset range of the point cloud cluster center point in the target area, it is beneficial to accurately judge the ground point cloud with large slope changes and improve the environmental adaptability of the threshold.

[0077] It is understandable that the distribution characteristic of laser point clouds is that the height difference between each point in the ground point cloud within a small range is small. The point cloud cluster after clustering can ensure the small range of the spatial distribution of the first non-ground point cloud. If the height difference between each point in the point cloud cluster is too large, then the points in the point cloud cluster do not meet the characteristics of the ground point cloud. The point cloud straight line distribution evaluation value can indicate the possibility of the point cloud straight line distribution. The smaller the point cloud straight line distribution evaluation value, the point cloud distribution is a curve with less curvature. The larger the point cloud straight line distribution evaluation value, the closer the point cloud distribution is to a straight line.

[0078] If the average height of a point cloud cluster is less than the preset average height threshold, the maximum height difference is less than the preset maximum height difference threshold, and the point cloud straight line distribution evaluation value is greater than the preset evaluation value threshold, then the point cloud cluster meets the ground point cloud characteristics and the point cloud cluster type is determined to be a ground point cloud. If the average height of a point cloud cluster is greater than or equal to the preset average height threshold, the maximum height difference is greater than or equal to the preset maximum height difference threshold, and the point cloud straight line distribution evaluation value is less than or equal to the preset evaluation value threshold, then the point cloud cluster does not meet the ground point cloud characteristics and the point cloud cluster type is determined to be a non-ground point cloud.

[0079] S290 : Determine a second segmentation result of the laser point cloud according to the type of each point cloud cluster and the first segmentation result.

[0080] Based on the above solution, the second segmentation result includes a second ground point cloud and a second non-ground point cloud;

[0081] Determining a second segmentation result of the laser point cloud according to the type of each point cloud cluster and the first segmentation result includes:

[0082] The point cloud clusters whose point cloud cluster type is ground point cloud are merged into the first ground point cloud to obtain a second ground point cloud, and the point cloud clusters whose ground type is non-ground point cloud are merged with the first non-ground point cloud outside the target area to obtain a second non-ground point cloud.

[0083] The above scheme can verify the first non-ground point cloud, determine whether there is an incorrectly segmented laser point cloud in the first non-ground point cloud based on the verification result, and re-plan the incorrectly segmented laser point cloud, which is conducive to achieving accurate segmentation of the laser point cloud and improving the safety and reliability of autonomous driving.

[0084] The technical solution of the embodiment of the present invention obtains a first segmentation result of the laser point cloud, clusters the first non-ground point cloud in the first segmentation result within a preset target area, and obtains at least one point cloud cluster within the target area. The solution then determines the matching point cloud cluster features of each point cloud cluster based on the coordinates of the midpoints of each point cloud cluster, and determines the type of each point cloud cluster based on the matching point cloud cluster features of each point cloud cluster. Finally, the solution determines a second segmentation result of the laser point cloud based on the point cloud cluster type and the first segmentation result. This technical solution solves the problems of low accuracy and poor reliability of laser point cloud segmentation under complex working conditions, improving the accuracy of laser point cloud segmentation while further enhancing the reliability of the classification of ground and non-ground point clouds.

[0085] Example 3

[0086] Figure 3 This is a schematic diagram of the structure of a laser point cloud segmentation device provided in the third embodiment of the present invention. Figure 3 As shown, the device includes:

[0087] A first segmentation result acquisition module 310 is configured to acquire a first segmentation result of the laser point cloud; wherein the first segmentation result includes a first ground point cloud and a first non-ground point cloud;

[0088] The point cloud cluster generation module 320 is configured to perform clustering processing on the first non-ground point cloud in the preset target area to obtain at least one point cloud cluster in the target area;

[0089] The point cloud cluster type determination module 330 is used to determine the point cloud cluster features matched by each point cloud cluster according to the coordinates of the midpoints of each point cloud cluster, and determine the type of each point cloud cluster according to the point cloud cluster features matched by each point cloud cluster;

[0090] The second segmentation result determination module 340 is configured to determine a second segmentation result of the laser point cloud according to the point cloud cluster types and the first segmentation result.

[0091] In this solution, optionally, the point cloud cluster feature includes an average height;

[0092] The point cloud cluster type determination module 330 includes a point cloud cluster feature determination unit, which is specifically configured to:

[0093] Calculate the mean coordinates of the points in each point cloud cluster, determine the center point of each point cloud cluster based on the mean coordinates, and use the height coordinate of the center point as the average height.

[0094] In this embodiment, optionally, the point cloud cluster feature further includes a maximum height difference;

[0095] The point cloud cluster feature determination unit is further configured to:

[0096] The difference between the maximum height coordinate and the minimum height coordinate in each point cloud cluster is taken as the maximum height difference.

[0097] On the basis of the above solution, the point cloud cluster feature further includes a point cloud straight line distribution evaluation value;

[0098] The point cloud cluster feature determination unit is further configured to:

[0099] According to the coordinates of the midpoints of each point cloud cluster, the covariance matrix of each point cloud cluster matching is determined;

[0100] Perform singular value decomposition on the covariance matrix of each point cloud cluster matching to obtain the eigenvalues ​​of each coordinate direction matching;

[0101] The eigenvalues ​​matching each coordinate direction are sorted, and the point cloud straight line distribution evaluation value is determined based on the eigenvalue sorting results.

[0102] In a feasible solution, the point cloud cluster type determination module 330 includes a point cloud cluster type determination unit, and the point cloud cluster type determination unit is specifically configured to:

[0103] If the average height of the point cloud cluster is less than the preset average height threshold, the maximum height difference is less than the preset maximum height difference threshold, and the point cloud straight line distribution evaluation value is greater than the preset evaluation value threshold, then the point cloud cluster type is determined to be a ground point cloud;

[0104] If the average height of the point cloud cluster is greater than or equal to the preset average height threshold, the maximum height difference is greater than or equal to the preset maximum height difference threshold, and the point cloud straight line distribution evaluation value is less than or equal to the preset evaluation value threshold, then the point cloud cluster type is determined to be a non-ground point cloud;

[0105] The average height threshold is determined based on the average height of the ground point cloud within a preset range of the center point of the point cloud cluster in the target area.

[0106] Based on the above solution, optionally, the second segmentation result includes a second ground point cloud and a second non-ground point cloud;

[0107] The second segmentation result determination module 340 is specifically configured to:

[0108] The point cloud clusters whose point cloud cluster type is ground point cloud are merged into the first ground point cloud to obtain a second ground point cloud, and the point cloud clusters whose ground type is non-ground point cloud are merged with the first non-ground point cloud outside the target area to obtain a second non-ground point cloud.

[0109] In a preferred solution, the point cloud cluster generation module 320 is specifically configured to:

[0110] Based on a predetermined clustering model, a first non-ground point cloud in a preset target area is clustered to obtain at least one point cloud cluster in the target area; wherein the clustering model is one of a Euclidean distance clustering model, a mean shift clustering model, and a spatial clustering model with noise.

[0111] The laser point cloud segmentation device provided in the embodiment of the present invention can execute the laser point cloud segmentation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0112] Example 4

[0113] Figure 4 A schematic diagram of the structure of an electronic device 410 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0114] like Figure 4As shown, the electronic device 410 includes at least one processor 411, and a memory connected to the at least one processor 411, such as a read-only memory (ROM) 412, a random access memory (RAM) 413, etc., wherein the memory stores a computer program that can be executed by the at least one processor, and the processor 411 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 412 or the computer program loaded from the storage unit 418 to the random access memory (RAM) 413. Various programs and data required for the operation of the electronic device 410 can also be stored in the RAM 413. The processor 411, ROM 412 and RAM 413 are connected to each other via a bus 414. An input / output (I / O) interface 415 is also connected to the bus 414.

[0115] Multiple components in electronic device 410 are connected to I / O interface 415, including an input unit 416, such as a keyboard, mouse, etc.; an output unit 417, such as various types of displays, speakers, etc.; a storage unit 418, such as a magnetic disk, optical disk, etc.; and a communication unit 419, such as a network card, modem, wireless communication transceiver, etc. The communication unit 419 allows electronic device 410 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0116] Processor 411 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of processor 411 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processor, controller, microcontroller, etc. Processor 411 executes the various methods and processes described above, such as the laser point cloud segmentation method.

[0117] In some embodiments, the laser point cloud segmentation method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 418. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 410 via the ROM 412 and / or the communication unit 419. When the computer program is loaded into the RAM 413 and executed by the processor 411, one or more steps of the laser point cloud segmentation method described above can be performed. Alternatively, in other embodiments, the processor 411 can be configured to perform the laser point cloud segmentation method in any other appropriate manner (for example, by means of firmware).

[0118] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0119] Computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable laser point cloud segmentation device, so that when executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer programs can be executed entirely on the machine, partially on the machine, as a standalone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0120] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0121] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0122] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0123] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0124] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0125] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A laser point cloud segmentation method, characterized in that: The method comprises: Obtaining a first segmentation result of the laser point cloud; wherein the first segmentation result includes a first ground point cloud and a first non-ground point cloud; Performing clustering processing on a first non-ground point cloud within a preset target area to obtain at least one point cloud cluster within the target area; According to the coordinates of the midpoints of each point cloud cluster, the point cloud cluster features of each point cloud cluster are determined, and according to the point cloud cluster features of each point cloud cluster, the type of each point cloud cluster is determined; Determining a second segmentation result of the laser point cloud according to the type of each point cloud cluster and the first segmentation result; The point cloud cluster features include average height, maximum height difference and point cloud straight line distribution evaluation value; Determining the point cloud cluster features of each point cloud cluster according to the coordinates of the midpoints of each point cloud cluster includes: Calculate the mean coordinates of the points in each point cloud cluster, determine the center point of each point cloud cluster based on the mean coordinates, and use the height coordinate of the center point as the average height; The difference between the maximum height coordinate and the minimum height coordinate in each point cloud cluster is taken as the maximum height difference; According to the coordinates of the midpoints of each point cloud cluster, the covariance matrix of each point cloud cluster matching is determined; Perform singular value decomposition on the covariance matrix of each point cloud cluster matching to obtain the eigenvalues ​​of each coordinate direction matching; The eigenvalues ​​matching each coordinate direction are sorted, and the point cloud straight line distribution evaluation value is determined based on the eigenvalue sorting results.

2. The method according to claim 1, characterized in that The step of determining the type of each point cloud cluster according to the point cloud cluster features matched by each point cloud cluster includes: If the average height of the point cloud cluster is less than the preset average height threshold, the maximum height difference is less than the preset maximum height difference threshold, and the point cloud straight line distribution evaluation value is greater than the preset evaluation value threshold, then the point cloud cluster type is determined to be a ground point cloud; If the average height of the point cloud cluster is greater than or equal to the preset average height threshold, the maximum height difference is greater than or equal to the preset maximum height difference threshold, and the point cloud straight line distribution evaluation value is less than or equal to the preset evaluation value threshold, then the point cloud cluster type is determined to be a non-ground point cloud; The average height threshold is determined based on the average height of the ground point cloud within a preset range of the center point of the point cloud cluster in the target area.

3. The method according to claim 2, characterized in that The second segmentation result includes a second ground point cloud and a second non-ground point cloud; Determining a second segmentation result of the laser point cloud according to the type of each point cloud cluster and the first segmentation result includes: The point cloud clusters whose point cloud cluster type is ground point cloud are merged into the first ground point cloud to obtain a second ground point cloud, and the point cloud clusters whose ground type is non-ground point cloud are merged with the first non-ground point cloud outside the target area to obtain a second non-ground point cloud.

4. The method according to claim 1, wherein The clustering process is performed on the first non-ground point cloud in the preset target area to obtain at least one point cloud cluster in the target area, including: Based on a predetermined clustering model, a first non-ground point cloud in a preset target area is clustered to obtain at least one point cloud cluster in the target area; wherein the clustering model is one of a Euclidean distance clustering model, a mean shift clustering model, and a spatial clustering model with noise.

5. A laser point cloud segmentation device, characterized in that: include: A first segmentation result acquisition module, configured to acquire a first segmentation result of the laser point cloud; wherein the first segmentation result includes a first ground point cloud and a first non-ground point cloud; a point cloud cluster generation module, configured to perform clustering processing on a first non-ground point cloud within a preset target area to obtain at least one point cloud cluster within the target area; The point cloud cluster type determination module is used to determine the point cloud cluster features matched by each point cloud cluster according to the coordinates of the midpoints of each point cloud cluster, and to determine the type of each point cloud cluster according to the point cloud cluster features matched by each point cloud cluster; A second segmentation result determination module is used to determine a second segmentation result of the laser point cloud according to the type of each point cloud cluster and the first segmentation result; The point cloud cluster features include average height, maximum height difference and point cloud straight line distribution evaluation value; Determining the point cloud cluster features of each point cloud cluster according to the coordinates of the midpoints of each point cloud cluster includes: Calculate the mean coordinates of the points in each point cloud cluster, determine the center point of each point cloud cluster based on the mean coordinates, and use the height coordinate of the center point as the average height; The difference between the maximum height coordinate and the minimum height coordinate in each point cloud cluster is taken as the maximum height difference; According to the coordinates of the midpoints of each point cloud cluster, the covariance matrix of each point cloud cluster matching is determined; Perform singular value decomposition on the covariance matrix of each point cloud cluster matching to obtain the eigenvalues ​​of each coordinate direction matching; The eigenvalues ​​matching each coordinate direction are sorted, and the point cloud straight line distribution evaluation value is determined based on the eigenvalue sorting results.

6. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the laser point cloud segmentation method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the laser point cloud segmentation method according to any one of claims 1 to 4 when executed.

Images