Plane parameter extraction method and device of plane point cloud and electronic equipment

Abstract

The application provides a plane parameter extraction method and device of a plane point cloud and electronic equipment, and relates to the technical field of computers. The plane parameter extraction method of the plane point cloud comprises: obtaining a point cloud frame to be processed; dividing the point cloud frame to be processed into a plurality of sub-block point cloud sets by using a spherical coordinate system; identifying plane point clouds from the plurality of sub-block point cloud sets according to characteristic values of each sub-block point cloud set, to obtain a plurality of first plane point clouds; and extracting plane parameters of the plurality of first plane point clouds. The application can efficiently and accurately extract the plane parameters of the plane point clouds from the point cloud frame.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a plane parameter extraction method and device of plane point cloud and electronic equipment. BACKGROUND

[0002] In recent years, with the rapid development of science and technology, the robot industry has made remarkable progress, and mobile robots have been widely used in many fields and have become an important part of production and life. However, with the continuous expansion and deepening of application fields, mobile robots are facing more and more complex and diverse task requirements and working environments.

[0003] In mobile robots, laser SLAM (Simultaneous Localization and Mapping) system is an important perception and positioning method, which can effectively perceive the environmental geometric information around the robot and accurately estimate the pose state of the robot. In real environment, there are often a large number of geometric features, such as line segment features, plane features, curve features, and surface features. In addition to the commonly used point features, these geometric features can provide additional constraints for state estimation. Especially in indoor scenes where feature points are difficult to accurately track, these scenes often contain rich line and surface features. The extraction of geometric features can provide effective constraint information for the SLAM system and optimize the mapping results. However, in the current laser SLAM technology, there is still a lack of effective extraction method for the plane parameters of plane point cloud.

[0004] Therefore, how to efficiently and accurately extract the plane parameters of the plane point cloud from the point cloud frame is a technical problem to be solved. SUMMARY

[0005] The present application provides a plane parameter extraction method and device of plane point cloud and electronic equipment to solve the defect of lack of effective extraction method for the plane parameters of plane point cloud in the prior art, so as to efficiently and accurately extract the plane parameters of the plane point cloud from the point cloud frame.

[0006] The present application provides a plane parameter extraction method of plane point cloud, comprising the following steps.

[0007] Obtain a point cloud frame to be processed; divide the point cloud frame to be processed into a plurality of sub-block point cloud sets by using a spherical coordinate system; identify a plane point cloud from the plurality of sub-block point cloud sets according to a characteristic value of each sub-block point cloud set to obtain a plurality of first plane point clouds; and extract plane parameters of the plurality of first plane point clouds.

[0008] According to the plane parameter extraction method of the plane point cloud provided by the application, the first plane point cloud with both adjacent relationship and similar relationship is merged into a same adjacent similar plane point cloud set to obtain a plurality of second plane point clouds.

[0009] According to the plane parameter extraction method of the plane point cloud provided by the application, the first plane point cloud with both adjacent relationship and similar relationship is merged into a same adjacent similar plane point cloud set to obtain a plurality of second plane point clouds.

[0010] According to the plane parameter extraction method of the plane point cloud provided by the application, the first plane point cloud with both adjacent relationship and similar relationship is merged into a same adjacent similar plane point cloud set to obtain a plurality of second plane point clouds.

[0011] According to the plane parameter extraction method of the plane point cloud provided by the application, the first plane point cloud with both adjacent relationship and similar relationship is merged into a same adjacent similar plane point cloud set to obtain a plurality of second plane point clouds.

[0012] According to the plane parameter extraction method of the plane point cloud provided in the application, the method further comprises: determining the association relationship between the second plane point cloud of the to-be-processed point cloud frame and the second plane point cloud of the processed point cloud frame by using a global coordinate system; in response to the second plane point cloud of the to-be-processed point cloud frame not having the association relationship with the second plane point cloud of the processed point cloud frame, establishing a global plane according to the second plane point cloud and adding the global plane into a sliding window; optimizing the pose information of each point cloud frame in the sliding window in a manner of minimizing the minimum eigenvalue; and performing a mapping operation by using the to-be-processed point cloud frame after the pose optimization.

[0013] The application further provides a plane parameter extraction device of a plane point cloud, comprising the following modules: a first acquisition module, configured to acquire a to-be-processed point cloud frame; a division module, configured to divide the to-be-processed point cloud frame into a plurality of block point cloud sets by using a spherical coordinate system; a second acquisition module, configured to identify a plane point cloud from the plurality of block point cloud sets according to the eigenvalue of each block point cloud set, and obtain a plurality of first plane point clouds; and an extraction module, configured to extract the plane parameters of the plurality of first plane point clouds.

[0014] The application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the plane parameter extraction method of the plane point cloud according to any one of the above when executing the computer program.

[0015] The application further provides a non-transitory computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the plane parameter extraction method of the plane point cloud according to any one of the above.

[0016] The application further provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement the plane parameter extraction method of the plane point cloud according to any one of the above.

[0017] The plane parameter extraction method, device and electronic device provided by the application can effectively improve the identification accuracy and processing efficiency of the plane point cloud in the to-be-processed point cloud frame by dividing the to-be-processed point cloud frame into a plurality of block point cloud sets by using a spherical coordinate system. Furthermore, the first plane point cloud can be accurately identified from the to-be-processed point cloud frame according to the eigenvalue of each block point cloud set, and the plane parameters can be efficiently and accurately extracted from the first plane point cloud. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to make the technical solutions in the present application or prior art clearer, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings described below are some embodiments of the present application, and all other drawings obtained by those skilled in the art without creative effort based on these drawings also belong to the protection scope of the present application.

[0019] Figure 1 is one of the flowcharts of the plane parameter extraction method of the plane point cloud provided by the present application.

[0020] Figure 2 is another flowchart of the plane parameter extraction method of the plane point cloud provided by the present application.

[0021] Figure 3 is the flowchart of the method for obtaining a plurality of second plane point clouds provided by the present application.

[0022] Figure 4 is the flowchart of the method for performing mapping operation based on the second plane point cloud provided by the present application.

[0023] Figure 5 is an exemplary schematic diagram of the plane feature extraction result of the second plane point cloud provided by the present application.

[0024] Figure 6 is an exemplary schematic diagram of the plane feature extraction result before global optimization provided by the present application.

[0025] Figure 7 is an exemplary schematic diagram of the plane feature extraction result after global optimization provided by the present application.

[0026] Figure 8 is the structural schematic diagram of the plane parameter extraction device of the plane point cloud provided by the present application.

[0027] Figure 9 is the structural schematic diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0028] In order to make the technical solutions in the present application or prior art clearer, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings described below are some embodiments of the present application, and all other drawings obtained by those skilled in the art without creative effort based on these drawings also belong to the protection scope of the present application.

[0029] The plane parameter extraction method of the plane point cloud provided by the present application will be described below. Figures 1-6

[0030] ​Figure 1 This is one of the flowcharts illustrating the planar parameter extraction method for planar point clouds provided by the present invention, such as... Figure 1 As shown, the method includes the following:

[0031] Step 101: Obtain the point cloud frame to be processed.

[0032] In practice, the planar parameter extraction method for planar point clouds provided by this invention can be executed by a self-moving device (e.g., an inspection robot).

[0033] The point cloud frame to be processed is a point cloud data frame acquired by the self-moving device using its own configured sensors (e.g., radar or laser sensors). As an example only, the point cloud frame to be processed could be a point cloud frame of the current location acquired by an inspection robot using radar.

[0034] Step 102: Using a spherical coordinate system, divide the point cloud frame to be processed into multiple block point cloud sets.

[0035] In practice, the point cloud frame to be processed can be projected onto a spherical coordinate system. For example, if the point cloud frame to be processed is a point cloud frame acquired using a radar sensor, the point cloud data can be projected onto a spherical coordinate system centered at the radar position.

[0036] In the specific implementation process, the spherical coordinates of each point cloud data in the point cloud frame to be processed can be obtained, and point cloud data with the same spherical coordinates can be divided into a block point cloud set to obtain multiple block point cloud sets.

[0037] As an example only, for those with N Radar point cloud frames of individual points For the i-th point cloud of this radar point cloud frame First, calculate the pitch angle based on the three-dimensional coordinates of the point cloud. and yaw angle The calculation formula is as follows:

[0038] (1)

[0039] (2)

[0040] in:

[0041] (3)

[0042] in, , as well as Point clouds Coordinates on the x-axis, y-axis, and z-axis.

[0043] The pitch angle resolution and the yaw angle resolution of the spherical surface grid are respectively The spherical coordinates of the point cloud are:

[0044] (4)

[0045] wherein, is the spherical coordinate of the point cloud along the X-axis direction, is the spherical coordinate of the point cloud along the Y-axis direction.

[0046] Traverse all point clouds of the radar point cloud frame , and list the point clouds with the same spherical coordinates into a set to obtain M piecewise point cloud sets .

[0047] The embodiment provided by the application is suitable for sensors with different scanning modes for obtaining a point cloud frame to be processed, and overcomes the defect that the common feature extraction mode of point cloud data can only be used for sensors with a rotating scanning mode (for example, a rotating laser radar).

[0048] In step 103, a planar point cloud is identified from the multiple piecewise point cloud sets according to the characteristic values of each piecewise point cloud set, and multiple first planar point clouds are obtained.

[0049] In the specific implementation process, the covariance of all point cloud data in the piecewise point cloud set can be determined, and the characteristic values of the piecewise point cloud set are obtained by performing eigenvalue decomposition on the covariance; wherein the characteristic values include a first characteristic value, a second characteristic value and a third characteristic value; in the case that the ratio between the first characteristic value and the second characteristic value is greater than a preset threshold value, the piecewise point cloud set is identified as a planar point cloud; in the case that the ratio between the first characteristic value and the second characteristic value is not greater than the preset threshold value, the piecewise point cloud set is identified as a non-planar point cloud set.

[0050] For example, in order to obtain a high-precision calculation result, the piecewise point cloud set with a point cloud quantity greater than a preset quantity threshold value (for example, 50) is selected from the point cloud set of step 102, and covariance construction and eigenvalue decomposition are performed thereon. Denote the piecewise point cloud set as The covariance of the piecewise point cloud set can be calculated by using the following formula is:

[0051] (5)

[0052] (6)

[0053] wherein, ​a center point of the block point cloud set.

[0054] perform eigenvalue decomposition on the covariance to obtain a first eigenvalue, a second eigenvalue and a third eigenvalue: (in ascending order), and the corresponding eigenroots if , is a preset threshold, marking the block point cloud set as a planar point cloud, and taking it as a first planar point cloud.

[0055] Step 104, extracting the plane parameters of the plurality of first planar point clouds.

[0056] After extracting the first planar point cloud from the to-be-processed point cloud frame, the plane parameters of the first planar point cloud can be extracted ( j is the number of the first planar point cloud). The plane parameters π are usually represented as a vector or a set for uniquely determining a plane in a three-dimensional space. Usually, a combination of a normal vector n and a distance d (a signed distance from the origin to the plane) can be used for representation, that is, .

[0057] Subsequently, SLAM mapping can be performed based on the extracted plane parameters of the first planar point cloud.

[0058] Figure 2 is a second flowchart of a plane parameter extraction method of a planar point cloud provided by the present application, as shown in Figure 2 , the method comprises the following:

[0059] Step 201, obtaining a to-be-processed point cloud frame.

[0060] For detailed description of this step, see the related content in step 101, which will not be repeated here.

[0061] Step 202, dividing the to-be-processed point cloud frame into a plurality of block point cloud sets by using a spherical coordinate system.

[0062] For detailed description of this step, see the related content in step 102, which will not be repeated here.

[0063] Step 203, identifying planar point clouds from the plurality of block point cloud sets according to the eigenvalues of each block point cloud set, to obtain a plurality of first planar point clouds.

[0064] For detailed description of this step, see the related content in step 103, which will not be repeated here.

[0065] Step 204, merging the first planar point clouds that simultaneously exist in adjacent relationships and similar relationships into the same adjacent and similar planar point cloud set, to obtain a plurality of second planar point clouds.

[0066] In some embodiments, in order to obtain a larger planar point cloud than the first planar point cloud range, and to obtain more accurate planar parameters based on the larger range of planar point cloud, the first planar point cloud with adjacent similar relationship can be merged in a planar region expansion manner to obtain a plurality of second planar point clouds.

[0067] In the specific implementation process, the first planar point cloud with both adjacent relationship and similar relationship can be merged into the same adjacent similar planar point cloud set in various ways, which is not limited by the description in the specification.

[0068] For an embodiment of merging the first planar point cloud with both adjacent relationship and similar relationship into the same adjacent similar planar point cloud set to obtain a plurality of second planar point clouds, see the related content in Figure 3 , which will not be described here.

[0069] Step 205, extracting the planar parameters of the plurality of second planar point clouds.

[0070] In the specific implementation process, the planar parameters of the second planar point cloud can be extracted ( i , the number of the second planar point cloud). The planar parameter π is usually represented as a vector or a set for uniquely determining a plane in a three-dimensional space. Usually, a combination of normal vector n and distance d can be used for representation, that is, .

[0071] For each second planar point cloud, according to all the point cloud data contained therein, the covariance is constructed and the eigenvalue decomposition is performed to obtain the planar parameters of the second planar point cloud; the distance of all point clouds to the center of the second planar point cloud is solved according to the planar parameters, and the point clouds with large distance (for example, greater than 0.01 meters) are removed; the covariance of the remaining point clouds is calculated again and the eigenvalue decomposition is performed to obtain the planar parameters of the second planar point cloud, so as to filter out some slightly distant points and obtain more optimal planar parameters.

[0072] In the embodiments provided by the present application, the first planar point cloud with both adjacent relationship and similar relationship is merged into the same adjacent similar planar point cloud set, so that a plurality of second planar point clouds with larger range can be accurately obtained. The planar feature extraction result of the second planar point cloud is shown in Figure 5 , in which the larger plane is the result after processing the point cloud of the point cloud frame to be processed by using the embodiments provided by the present application. Obviously, the planar feature has clear and rich surface features, which can provide effective constraint information for the SLAM system.

[0073] Figure 3 is a flowchart of the method for obtaining a plurality of second planar point clouds provided by the present application.

[0074] In the embodiment, the current first plane point cloud is determined from all first plane point clouds, and operations as shown in the following are performed for the current first plane point cloud and each adjacent first plane point cloud corresponding to the current first plane point cloud. Figure 3

[0075] In step 301, whether the current first plane point cloud and the adjacent first plane point cloud have a similar relationship is determined by using the plane angle cosine between the two and the distance between the center points of the two.

[0076] In the specific implementation process, the adjacent first plane point cloud corresponding to the current first plane point cloud is determined according to the spherical coordinates of the current first plane point cloud in the spherical coordinate system, and the nearest neighbor plane of each first plane point cloud is at most 4.

[0077] For example, for the current first plane point cloud and the adjacent first plane point cloud corresponding thereto, if and have been in the same adjacent similar plane set, the next adjacent first plane point cloud corresponding to the current first plane point cloud is traversed, otherwise whether and have a similar relationship is determined by using the following method:

[0078] The cosine value of the angle between and is calculated by using the following formula :

[0079] (7)

[0080] wherein, is the normal vector of ; is the normal vector of .

[0081] The center point distance between is calculated by using the following formula :

[0082] (8)

[0083] (9)

[0084] wherein, is the center point of ; is the center point of .​​ is a signed distance to the coordinate origin; is a signed distance to the coordinate origin.

[0085] If the above calculation result satisfies the following conditions: γ is not greater than a first preset threshold (for example, γ ), is not greater than a second preset threshold (for example, γ ), is not greater than a third preset threshold (for example, γ ), then it is determined that there is a similarity relationship between and .

[0086] If it is determined that there is no similarity relationship between and , then the next adjacent plane point cloud of the current first plane point cloud is traversed, and step 301 is continued to be executed until all adjacent plane point clouds of the current first plane point cloud are traversed.

[0087] If it is determined that there is a similarity relationship between and , the following steps are executed.

[0088] Step 302, in response to the existence of a similarity relationship between the current first plane point cloud and the adjacent first plane point cloud, the current first plane point cloud and the adjacent first plane point cloud are merged into the same adjacent similar plane point cloud set.

[0089] For example only, the current first plane point cloud and the adjacent first plane point cloud may be merged into the same adjacent similar plane point cloud set in the following manner.

[0090] If neither nor is in the adjacent similar plane set, a new adjacent similar plane set is constructed to contain and ; if only is in the adjacent similar plane set, then is added to the adjacent similar plane set in which is located; if both and are in the adjacent similar plane set, then the one with fewer planes in the adjacent similar plane set is merged into the one with more planes; if only is in the adjacent similar plane set, then is added to the adjacent similar plane set in which is located.

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] ​​​​​​​​​​​​​​​​​​​​​​​​The set of non-planar point cloud blocks is used as the set of non-planar point cloud blocks corresponding to the new adjacent similar planar sets; if only In adjacent similar plane sets, then join in The set of adjacent similar planes will contain Adding nonplanar point cloud block sets The set of non-planar point cloud blocks corresponding to the adjacent similar plane set; if and If all planes are in adjacent similar plane sets, then merge the set with the smaller number of planes into the set with the larger number, and simultaneously merge the corresponding non-planar point cloud patch sets; if only In adjacent similar plane sets, then join in Set up and update the adjacent similar planes. A nonplanar point cloud block set.

[0098] In the embodiments provided by this invention, all first planar point clouds are traversed, and adjacent similar planar point cloud sets are used to merge first planar point clouds with adjacent similarity relationships into second planar point clouds. This allows for the accurate acquisition of planar point clouds with a larger range.

[0099] Figure 4 This is a flowchart illustrating the mapping operation method based on a second-plane point cloud provided by the present invention, as shown below. Figure 4 As shown, the method includes the following:

[0100] Step 401: Using the global coordinate system, determine the association between the second plane point cloud of the point cloud frame to be processed and the second plane point cloud of the processed point cloud frame.

[0101] In specific implementation, the embodiments provided by the present invention can be used to process each point cloud frame acquired by the mobile device to obtain the second planar point cloud of each point cloud frame.

[0102] For the initial point cloud frame, a global coordinate system can be established based on this point cloud frame. Each second-plane point cloud of the point cloud frame is transformed into a global plane, and the center point, parameters, and index of the global plane among all generated global planes are recorded. Then, the information of these global planes is added to an octree, and the initial point cloud frame is added to a sliding window.

[0103] When a point cloud frame to be processed arrives, each second plane point cloud of the point cloud frame to be processed is traversed, the second plane point cloud of the point cloud frame to be processed is projected onto the established global coordinate system, and the global plane of the processed point cloud frame closest to the center point is found by means of the octree, and whether there is an association relationship between the second plane point cloud of the point cloud frame to be processed and the global plane of the processed point cloud frame is determined according to the similarity between the parameters of the second plane point cloud of the point cloud frame to be processed and the parameters of the global plane of the processed point cloud frame.

[0104] For example, the parameters of the second plane point cloud of the point cloud frame to be processed are , the parameters of the global plane of the processed point cloud frame are , if is satisfied, it is determined that there is an association relationship between them, that is, the second plane point cloud of the point cloud frame to be processed belongs to the global plane of the processed point cloud frame, and the association relationship is recorded for subsequent processing. Otherwise, it is determined that there is no association relationship between them.

[0105] Step 402, in response to the second plane point cloud of the point cloud frame to be processed not having an association relationship with the second plane point cloud of the processed point cloud frame, a global plane is established according to the second plane point cloud, and the global plane is added to the sliding window.

[0106] In the case where it is determined that the second plane point cloud of the point cloud frame to be processed does not have an association relationship with the second plane point cloud of the processed point cloud frame, a global plane is established according to the second plane point cloud, and then the information of the global plane is added to the octree and the global plane is added to the sliding window.

[0107] Step 403, the pose information of each point cloud frame in the sliding window is optimized by minimizing the minimum eigenvalue.

[0108] In the case where the number of global planes contained in the sliding window is greater than 1, a covariance matrix is constructed according to the point cloud data contained in all global planes in the sliding window and the corresponding pose information. Then, the covariance matrix is subjected to eigenvalue decomposition to obtain the minimum eigenvalue. By minimizing the minimum eigenvalue, the pose information of each point cloud frame in the sliding window is optimized. After optimization, if the length of the sliding window is greater than a preset threshold, the earliest added point cloud frame in the sliding window is marginalized.

[0109] Step 404, using the point cloud frame to be processed after optimization of the pose, a mapping operation is performed.

[0110] In the process of sliding window optimization, key frames can be retained according to the change of the pose, and the first point cloud frame is automatically regarded as a key frame. Whenever a new key frame is added, the pose information of the key frame and the parameters of the global plane can be adjusted through global optimization to ensure accuracy. Then, the pose information of all point cloud frames is updated, and a map is constructed based on the information.

[0111] In the SLAM initialization process, when the second frame of point cloud data is received, a pose optimization operation is performed to obtain accurate pose information of the new frame. Thereafter, for each frame of newly arrived data, the relative pose information between the previous two frames is used to preliminarily estimate the pose of the new frame, so as to perform subsequent mapping processing and analysis.

[0112] In the embodiments provided by the present application, through the global coordinate system, the point cloud data between different point cloud frames can be processed under a unified spatial reference, thereby improving the accuracy of point cloud registration. Global optimization in the sliding window can adjust the pose information of the key frame, so that the entire point cloud map remains consistent in the global range. At the same time, the number of point cloud frames participating in optimization is limited by the sliding window, reducing the computational complexity. The plane feature extraction results before and after global optimization are shown in Figure 6 and Figure 7 It is obvious that the plane features after global optimization are clearer and larger in range, which can provide more effective constraint information for the SLAM system.

[0113] The plane parameter extraction device for the plane point cloud provided by the present application will be described below. The plane parameter extraction device for the plane point cloud described below can be mutually corresponding to the plane parameter extraction method for the plane point cloud described above.

[0114] Figure 8 is a structural schematic diagram of the plane parameter extraction device for the plane point cloud provided by the present application. As shown in Figure 8 The device 800 includes the following modules.

[0115] The first acquisition module 810 is configured to acquire a point cloud frame to be processed.

[0116] The division module 820 is configured to divide the point cloud frame to be processed into a plurality of sub-block point cloud sets by using a spherical coordinate system.

[0117] The second acquisition module 830 is configured to identify plane point clouds from the plurality of sub-block point cloud sets according to feature values of each of the sub-block point cloud sets, to obtain a plurality of first plane point clouds.

[0118] The extraction module 840 is configured to extract plane parameters of the plurality of first plane point clouds.

[0119] Figure 9 An example of an entity structure schematic diagram of an electronic device is shown in Figure 9As shown, the electronic device can include a processor 910, a communications interface 920, a memory 930, and a communications bus 940, wherein the processor 910, the communications interface 920, and the memory 930 complete mutual communication through the communications bus 940. The processor 910 can invoke a logic instruction in the memory 930 to execute a plane parameter extraction method of a plane point cloud, which includes: obtaining a point cloud frame to be processed; dividing the point cloud frame to be processed into a plurality of block point cloud sets using a spherical coordinate system; identifying a plane point cloud from the plurality of block point cloud sets according to a characteristic value of each block point cloud set, to obtain a plurality of first plane point clouds; and extracting plane parameters of the plurality of first plane point clouds.

[0120] In addition, the logic instruction in the memory 930 described above can be implemented in the form of a software functional unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0121] On the other hand, the present application also provides a computer program product, which includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program is executed by a processor, so that the computer can execute the plane parameter extraction method of a plane point cloud provided by the above-mentioned methods, which includes: obtaining a point cloud frame to be processed; dividing the point cloud frame to be processed into a plurality of block point cloud sets using a spherical coordinate system; identifying a plane point cloud from the plurality of block point cloud sets according to a characteristic value of each block point cloud set, to obtain a plurality of first plane point clouds; and extracting plane parameters of the plurality of first plane point clouds.

[0122] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the method for extracting plane parameters of a plane point cloud provided by each of the above methods, and the method comprises: obtaining a point cloud frame to be processed; dividing the point cloud frame to be processed into a plurality of block point cloud sets using a spherical coordinate system; identifying plane point clouds from the plurality of block point cloud sets according to characteristic values of each of the block point cloud sets, to obtain a plurality of first plane point clouds; and extracting plane parameters of the plurality of first plane point clouds.

[0123] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0124] From the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software plus a necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions, essentially or in terms of contribution to the prior art, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0125] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features thereof; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for plane parameter extraction of a planar point cloud, characterized in that, The method comprises: obtaining a to-be-processed point cloud frame; dividing the to-be-processed point cloud frame into a plurality of sub-block point cloud sets by using a spherical coordinate system, comprising: projecting the to-be-processed point cloud frame onto the spherical coordinate system to obtain the spherical coordinates of each point cloud data in the to-be-processed point cloud frame; dividing the point cloud data with the same spherical coordinates into a sub-block point cloud set to obtain the plurality of sub-block point cloud sets, comprising: For a radar point cloud frame with N one point , for the i-th point cloud of the radar point cloud frame , first calculate the pitch angle and the yaw angle according to the three-dimensional coordinates of the point cloud, and the calculation formula is as follows: ； wherein: ； wherein, , and , are the coordinates of the point cloud in the x-axis, y-axis, and z-axis, respectively. The pitch angle resolution and the yaw angle resolution of the spherical surface grid are respectively The spherical coordinates of the point cloud are: ​ ； wherein, is a point cloud spherical coordinates along the X-axis direction, is a point cloud spherical coordinates along the Y-axis direction; Traversing a radar point cloud frame of all point clouds, put the point clouds with the same spherical coordinates into a set, and obtain M piecewise point cloud sets ; identifying planar point clouds from the plurality of sub-block point cloud sets according to the characteristic values of each sub-block point cloud set to obtain a plurality of first planar point clouds; extracting the plane parameters of the plurality of first planar point clouds; wherein the plane parameters are represented by a combination of a normal vector and a distance.

2. The method of claim 1, wherein, The method further comprises: merging the first planar point clouds that simultaneously have a neighboring relationship and a similar relationship into the same adjacent similar planar point cloud set to obtain a plurality of second planar point clouds; extracting the plane parameters of the plurality of second planar point clouds.

3. The plane parameter extraction method of a planar point cloud according to claim 2, characterized in that, The method further comprises: determining a current first planar point cloud from all the first planar point clouds; for the current first planar point cloud and each adjacent first planar point cloud corresponding to the current first planar point cloud, performing the following operations: determining whether there is a similar relationship between the current first planar point cloud and the adjacent first planar point cloud by using the plane angle cosine between the two and the distance between the center points of the two; in response to the existence of a similar relationship between the current first planar point cloud and the adjacent first planar point cloud, merging the current first planar point cloud and the adjacent first planar point cloud into the same adjacent similar planar point cloud set; returning to the step of determining the current first planar point cloud from all the first planar point clouds until all the first planar point clouds are traversed; taking each adjacent similar planar point cloud set generated in the traversal process as a second planar point cloud; wherein the adjacent first planar point cloud corresponding to the current first planar point cloud is determined according to the spherical coordinates of the current first planar point cloud in the spherical coordinate system.

4. The plane parameter extraction method of a planar point cloud according to claim 3, characterized in that, The method further comprises: adding a non-planar point cloud block set with a distance from the center point of the adjacent similar planar point cloud set less than a preset threshold to the adjacent similar planar point cloud set; wherein the non-planar point cloud block set is the sub-block point cloud set identified as a non-planar point cloud.

5. The method of claim 2, wherein, The method further comprises: determining the association relationship between the second planar point cloud of the to-be-processed point cloud frame and the second planar point cloud of the processed point cloud frame by using a global coordinate system; in response to the second planar point cloud of the to-be-processed point cloud frame not having an association relationship with the second planar point cloud of the processed point cloud frame, establishing a global plane according to the second planar point cloud and adding the global plane to a sliding window; optimizing the pose information of each point cloud frame in the sliding window by minimizing the minimum characteristic value; performing a mapping operation by using the to-be-processed point cloud frame after the pose is optimized.

6. A plane parameter extraction apparatus of a plane point cloud, characterized by comprising: The method comprises: a first obtaining module for obtaining a to-be-processed point cloud frame; The division module is configured to divide the to-be-processed point cloud frame into a plurality of sub-block point cloud sets by using a spherical coordinate system, including: projecting the to-be-processed point cloud frame onto the spherical coordinate system to obtain spherical coordinates of each point cloud data in the to-be-processed point cloud frame; dividing the point cloud data with the same spherical coordinates into a sub-block point cloud set to obtain the plurality of sub-block point cloud sets, including: For radar point cloud frames with N one point , for the i-th point cloud of the radar point cloud frame , first calculate the pitch angle and the yaw angle according to the three-dimensional coordinates of the point cloud, and the calculation formula is as follows: ； wherein: ； wherein, , and , are the coordinates of the point cloud in the x-axis, y-axis and z-axis, respectively; The pitch angle resolution and the yaw angle resolution of the spherical surface grid are respectively The spherical coordinates of the point cloud are: ​ ； wherein, is a point cloud spherical coordinates along the X-axis direction, is a point cloud spherical coordinates along the Y-axis direction; Traversing a radar point cloud frame of all point clouds, put the point clouds with the same spherical coordinates into a set, and obtain M piecewise point cloud sets ; The second acquisition module is configured to identify planar point clouds from the plurality of sub-block point cloud sets according to eigenvalues of each of the sub-block point cloud sets to obtain a plurality of first planar point clouds. The extraction module is configured to extract planar parameters of the plurality of first planar point clouds, wherein the planar parameters are represented by a combination of a normal vector and a distance.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the planar parameter extraction method of the planar point cloud according to any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the planar parameter extraction method of the planar point cloud according to any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the planar parameter extraction method of the planar point cloud according to any one of claims 1 to 5.

Images