Method, device, equipment, medium and program product for determining background point cloud

Abstract

The application relates to a background point cloud determination method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a set of point clouds of interest in a current scene, acquiring a height value of each point cloud of interest in the set of point clouds of interest, and acquiring a corresponding background height threshold of each point cloud of interest in different background query tables, and then determining a background point cloud in the set of point clouds of interest according to the height value of each point cloud of interest and the corresponding background height threshold of each point cloud of interest in the background query tables. Different background query tables correspond to different specifications of grid maps, each specification of grid map is a planar map obtained by performing two-dimensional gridding on the current scene by using the same specification of grid, and each background query table comprises a background height threshold of each grid in the corresponding grid map. The above method can improve the accuracy of background point cloud determination and improve the accuracy of subsequent target identification.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, device, medium, and program product for determining background point clouds. Background Technology

[0002] Point clouds are collections of massive points representing the surface characteristics of an object within the same spatial reference frame, and can be acquired using lidar. LiDAR emits probe light in multiple directions, and calculates the distance between the object and itself in the current direction by the time difference between the emitted and returned probe light, thereby achieving a certain degree of environmental detection.

[0003] For example, point cloud data of a certain scene can be collected by roadside radar, and then background point clouds (such as ground point clouds) in the point cloud data of that scene can be filtered out to identify targets.

[0004] However, the relevant technologies cannot accurately determine the background point cloud, resulting in poor accuracy in target recognition. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer device, computer-readable storage medium, and computer program product for determining background point clouds to address the aforementioned technical problems.

[0006] Firstly, this application provides a method for determining background point clouds, including:

[0007] Get the set of points of interest (POIs) in the current scene, and get the height value of each POI in the set of POIs;

[0008] Obtain the background height threshold corresponding to each point cloud of interest in different background lookup tables; different background lookup tables correspond to raster images of different specifications. Each raster image of a specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specifications of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image.

[0009] The background point cloud in the set of point clouds of interest is determined based on the height value of each point cloud of interest and the background height threshold corresponding to each point cloud of interest in each background lookup table.

[0010] In one embodiment, obtaining the background height threshold corresponding to each point cloud of interest in different background lookup tables includes:

[0011] Determine the target raster in the raster map corresponding to each point cloud of interest in different background lookup tables;

[0012] The background height threshold corresponding to the target raster of each point cloud of interest is determined as the background height threshold corresponding to each point cloud of interest in different background lookup tables.

[0013] In one embodiment, determining the target raster in the raster map corresponding to each point cloud of interest in different background lookup tables includes:

[0014] Each point cloud of interest is mapped to a raster map of different sizes to obtain the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes;

[0015] Obtain the raster region range of each raster in the raster image corresponding to each background lookup table;

[0016] The grid cells within the grid regions to which the two-dimensional coordinates of each point cloud of interest belong in different grid maps are determined as the target grid cells of each point cloud of interest in the corresponding grid map.

[0017] In one embodiment, mapping each point cloud of interest to a raster map of different sizes, and obtaining the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes, includes:

[0018] Obtain the two-dimensional coordinates of a preset reference point; where the preset reference point is the origin of a raster map of different sizes;

[0019] Based on the two-dimensional coordinates of the preset reference point and the size of the raster images of different specifications, the point clouds of interest are mapped to obtain the two-dimensional coordinates of each point cloud of interest in the raster images of different specifications.

[0020] In one embodiment, each point cloud of interest is mapped according to the two-dimensional coordinates of a preset reference point and the size of the raster images of different specifications, to obtain the two-dimensional coordinates of each point cloud of interest in the raster images of different specifications, including:

[0021] Based on the X-axis coordinates of the preset reference point in the two-dimensional coordinate system, the X-axis coordinates of each point of interest cloud, and the length of the grid in the X-axis direction of each grid in the grid map of different specifications, determine the X-axis coordinate values ​​of each point of interest cloud in the corresponding grid map.

[0022] Based on the Y-axis coordinates of the preset reference point in the two-dimensional coordinate system, the Y-axis coordinates of each point of interest cloud, and the length of the grid in the Y-axis direction of each grid in the different sizes of the grid, determine the Y-axis coordinate value of each point of interest cloud in the corresponding grid.

[0023] Based on the X-axis and Y-axis coordinates of each point cloud of interest within the corresponding raster image, the two-dimensional coordinates of each point cloud of interest in raster images of different sizes are obtained.

[0024] In one embodiment, the different background lookup tables include a first background lookup table corresponding to a first raster image and a second background lookup table corresponding to a second raster image; wherein the size of the raster in the first raster image is smaller than the size of the raster in the second raster image;

[0025] Then, based on the height value of each point cloud of interest and the corresponding background height threshold in each background lookup table, the background point clouds in the set of point clouds of interest are determined, including:

[0026] Candidate background point clouds are determined from the set of point clouds of interest based on the height value of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table.

[0027] The background point cloud is determined from the candidate background point clouds based on the height value of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table.

[0028] In one embodiment, candidate background point clouds are determined from the set of point clouds of interest based on the height values ​​of each point cloud of interest and a first background height threshold corresponding to each point cloud of interest in a first background lookup table, including:

[0029] Obtain the first height difference between the height value of each point cloud of interest and the first background height threshold;

[0030] Point clouds of interest whose first height difference is less than or equal to the first threshold are identified as candidate background point clouds.

[0031] In one embodiment, determining the background point cloud from the candidate background point cloud based on the height values ​​of the candidate background point cloud and the second background height threshold corresponding to the point cloud of interest in the second background lookup table includes:

[0032] Obtain the second height difference between the height value of each candidate background point cloud and the second background height threshold;

[0033] Candidate background point clouds with a second height difference less than or equal to a second threshold are identified as background point clouds.

[0034] In one embodiment, the method further includes:

[0035] Obtain a set of reference point clouds for the current scene, and convert the reference point clouds in the set to raster maps of different sizes to obtain the two-dimensional coordinates of each reference point cloud in the corresponding raster map; the target object does not exist in the set of reference point clouds.

[0036] Based on the two-dimensional coordinates of each reference point cloud in the corresponding raster image, obtain the height value of the reference point cloud in the region of interest in each raster image of different sizes.

[0037] Different background lookup tables are determined based on the height values ​​of the reference point clouds within the regions of interest in raster maps of different sizes and the raster cells within the regions of interest in raster maps of different sizes.

[0038] In one embodiment, different background lookup tables are determined based on the height values ​​of reference point clouds within regions of interest in raster images of different sizes and the rasters within those regions of interest in raster images of different sizes, including:

[0039] Obtain the minimum height value of the reference point cloud in each grid cell within the region of interest in raster maps of different specifications;

[0040] The minimum height value of the reference point cloud in each grid cell within the region of interest in each grid image of different specifications is determined as the background height threshold for the corresponding grid cell, thus obtaining the background lookup table corresponding to each grid image of different specifications.

[0041] In one embodiment, obtaining a reference point cloud set for the current scene includes:

[0042] Obtain the initial point cloud set for the current scene, and extract the ground point cloud from the initial point cloud set;

[0043] The ground point cloud is expanded using the three-dimensional coordinates and interpolation algorithm to obtain a reference point cloud set.

[0044] Secondly, this application also provides a background point cloud determination device, comprising:

[0045] The height determination module is used to obtain the set of points of interest (POIs) in the current scene and to obtain the height value of each POI in the set of POIs;

[0046] The threshold determination module is used to obtain the background height threshold corresponding to each point cloud of interest in different background lookup tables. Different background lookup tables correspond to raster images of different specifications. Each raster image of a specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specification of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image.

[0047] The background determination module is used to determine the background point cloud in the set of point clouds of interest based on the height value of each point cloud of interest and the background height threshold corresponding to each point cloud of interest in each background lookup table.

[0048] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0049] Obtain the set of points of interest (POIs) in the current scene, and obtain the height value of each POI in the set of POIs;

[0050] Obtain the background height threshold corresponding to each point cloud of interest in different background lookup tables; different background lookup tables correspond to raster images of different specifications. Each raster image of a specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specifications of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image.

[0051] The background point cloud in the set of point clouds of interest is determined based on the height value of each point cloud of interest and the background height threshold corresponding to each point cloud of interest in each background lookup table.

[0052] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0053] Obtain the set of points of interest (POIs) in the current scene, and obtain the height value of each POI in the set of POIs;

[0054] Obtain the background height threshold corresponding to each point cloud of interest in different background lookup tables; different background lookup tables correspond to raster images of different specifications. Each raster image of a specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specifications of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image.

[0055] The background point cloud in the set of point clouds of interest is determined based on the height value of each point cloud of interest and the background height threshold corresponding to each point cloud of interest in each background lookup table.

[0056] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0057] Obtain the set of points of interest (POIs) in the current scene, and obtain the height value of each POI in the set of POIs;

[0058] Obtain the background height threshold corresponding to each point cloud of interest in different background lookup tables; different background lookup tables correspond to raster images of different specifications. Each raster image of a specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specifications of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image.

[0059] The background point cloud in the set of point clouds of interest is determined based on the height value of each point cloud of interest and the background height threshold corresponding to each point cloud of interest in each background lookup table.

[0060] The aforementioned method, apparatus, computer equipment, storage medium, and computer program product for determining background point clouds obtain a set of point clouds of interest (POIs) in the current scene, acquire the height value of each POI in the POI set, and obtain the background height threshold corresponding to each POI in different background lookup tables. Then, based on the height values ​​of each POI and the corresponding background height thresholds in each background lookup table, the background point clouds in the POI set are determined. Different background lookup tables correspond to raster images of different specifications. Each raster image of a specific specification is a planar image obtained by rasterizing the current scene using the same specification of raster, and each background lookup table includes the background height threshold for each raster in the corresponding raster image. Since different grid sizes define different scene areas, some may cover ground point clouds while others may not. Therefore, using background lookup tables corresponding to grid sizes of various types to determine the background point cloud in the set of points of interest can improve the accuracy of the judgment and avoid misjudging the background point cloud based on a background lookup table corresponding to a single grid size. This improves the accuracy of the background point cloud judgment and, in turn, accurately identifies the background point cloud, thereby improving the accuracy of subsequent target recognition. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating a method for determining background point clouds in one embodiment;

[0062] Figure 2 This is a schematic diagram of the structure of a grid map of different specifications in one embodiment;

[0063] Figure 3 This is a flowchart illustrating the process of determining a background height threshold in one embodiment;

[0064] Figure 4 This is a flowchart illustrating the process of determining the target grid in one embodiment;

[0065] Figure 5 This is a flowchart illustrating the process of determining the two-dimensional coordinates of a point cloud of interest in one embodiment.

[0066] Figure 6 This is a flowchart illustrating the process of determining the two-dimensional coordinates of a point cloud of interest in another embodiment;

[0067] Figure 7 This is a flowchart illustrating the process of determining a background point cloud in one embodiment;

[0068] Figure 8 This is a flowchart illustrating the process of determining candidate background point clouds from a set of point clouds of interest in one embodiment.

[0069] Figure 9This is a schematic diagram of the process of determining a candidate background point cloud from candidate background points in one embodiment;

[0070] Figure 10 This is a flowchart illustrating the process of constructing different background query tables in one embodiment;

[0071] Figure 11 A flowchart illustrating the process of constructing different background query tables in another embodiment;

[0072] Figure 12 This is a schematic diagram of the process for obtaining a reference point cloud set in one embodiment;

[0073] Figure 13 This is a structural block diagram of a background point cloud determination device in one embodiment;

[0074] Figure 14 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0075] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0076] In one embodiment, such as Figure 1 As shown, a method for determining background point clouds is provided. This embodiment illustrates the application of this method to a terminal. It can be understood that the terminal can be a roadside device, vehicle, drone, etc., that is, a terminal device used to achieve environmental perception, target localization, or tracking. In this embodiment, the method includes the following steps:

[0077] S110. Obtain the set of points of interest (POIs) in the current scene, and obtain the height value of each POI in the POI set.

[0078] The point cloud of interest (ROI) is the set of point clouds located within a preset region of interest (ROI) in the current scene. Optionally, the ROI can be a local area of ​​the current scene or a panoramic area of ​​the current scene. In this embodiment, the location and size of the ROI are not specifically limited. The height value of the point cloud of ROI is the Z-axis coordinate in the three-dimensional coordinate system of the point cloud.

[0079] Optionally, the terminal can acquire a set of original point clouds in the current scene using LiDAR, match the coordinates of each original point cloud in the set with the coverage location of a preset region of interest (ROI), and determine that the original point cloud covered by the ROI is the point cloud of interest, thus obtaining a set of point clouds of interest. Further, the height value of each point cloud in this set of point clouds of interest is obtained by traversing each point cloud in the set of point clouds of interest to obtain the Z-axis coordinate in the three-dimensional coordinate system of each point cloud of interest.

[0080] S120. Obtain the background height threshold corresponding to each point cloud of interest in different background query tables.

[0081] Among them, different background lookup tables correspond to raster images of different specifications. Each raster image of a certain specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specifications of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image.

[0082] It should be noted that the background height threshold of the raster is the minimum height value of the point cloud of interest falling within the raster after the current scene is rasterized in two dimensions. Generally, the minimum height value of the point cloud falling within each raster is considered to be the height of the ground point cloud. Therefore, the background point cloud determined in this application is the ground point cloud. Different specifications of raster maps essentially refer to the different sizes of the raster cells within the raster map; there are raster maps that include large-sized cells, and there are raster maps that include small-sized cells, such as... Figure 2 As shown, these are two different raster charts, A and B.

[0083] Different background lookup tables are also constructed based on the point cloud set in the current scene, which is the same scene as the acquired point cloud set of interest. Continuing the example above, the raster map of specification A includes four large-sized graticles A1 to A4, and the raster map of specification B includes six small-sized graticles B1 to B6. Correspondingly, the background lookup table 1 corresponding to the raster map of specification A includes the background height thresholds corresponding to each of graticles A1 to A4, and the background lookup table 2 corresponding to the raster map of specification B includes the background height thresholds corresponding to each of graticles B1 to B6.

[0084] Optionally, the terminal traverses the point clouds of interest (POIs) in the set of interest and performs 2D rasterization on the 3D coordinates of the POIs using rasteres of different sizes. This projects each POI onto a raster map of different sizes, and then obtains the background height threshold of the raster corresponding to each POI from different background lookup tables. Continuing the example above, if a POI is projected onto a raster map of size A, corresponding to raster A1, then the background height threshold H corresponding to raster A1 is obtained. A-1; if the point cloud of interest is then mapped to a B-size raster image, corresponding to raster B2, then the background height threshold H corresponding to raster B2 is obtained. B-2 .

[0085] S130. Determine the background point cloud in the set of point clouds of interest based on the height value of each point cloud of interest and the background height threshold corresponding to each point cloud of interest in each background lookup table.

[0086] Optionally, for each point cloud of interest in the set of interest, the terminal compares the height value of the point cloud of interest with the background height threshold corresponding to the point cloud of interest in each background lookup table, and determines whether the point cloud of interest is a background point cloud based on the comparison result, thereby obtaining the background point cloud in the set of point clouds of interest.

[0087] Optionally, after determining the background point cloud, the terminal can further filter out the background point cloud from the set of point clouds of interest to obtain the target point cloud, which is the point cloud of the target object in the corresponding region of interest.

[0088] In this embodiment, the terminal acquires a set of point clouds of interest (POIs) in the current scene, and obtains the height value of each POI in the set, as well as the background height threshold corresponding to each POI in different background lookup tables. Then, based on the height values ​​of each POI and the corresponding background height thresholds in each background lookup table, the background POIs in the set are determined. Different background lookup tables correspond to raster images of different specifications. Each raster image of a specific specification is a planar image obtained by rasterizing the current scene using the same specification of raster, and each background lookup table includes the background height threshold for each raster in the corresponding raster image. Since different raster images of different specifications delineate different scene areas, some may cover ground POIs, while others may not. Therefore, using background lookup tables corresponding to multiple raster image specifications to determine the background POIs in the set improves the accuracy of the determination and avoids misjudgment of background POIs based on a background lookup table corresponding to a single raster image specification. This improves the accuracy of background POI determination and ultimately enhances the accuracy of subsequent target recognition.

[0089] The point cloud of interest (POI) is point cloud data in three-dimensional space, while the background lookup table corresponds to a raster image of the current scene in two-dimensional plane. To determine the background height threshold corresponding to the POI in different background lookup tables, a process of converting the POI from three-dimensional space to a two-dimensional plane is involved. Based on this, in one embodiment, such as... Figure 3 As shown, S120 above, obtaining the background height threshold corresponding to each point cloud of interest in different background query tables, includes:

[0090] S310. Determine the target raster in the raster map corresponding to each point of interest cloud in different background lookup tables.

[0091] Among them, different background lookup tables correspond to raster maps containing multiple grids, and each grid corresponds to a background height threshold determined based on the height value of the point cloud of interest defined by itself.

[0092] Optionally, the terminal first determines the raster (i.e., the target raster) to which the corresponding point of interest cloud belongs in the raster map of different background lookup tables based on the 3D coordinates of each point of interest cloud. As in the example above, based on the point of interest cloud C... i The three-dimensional coordinates determine the point cloud of interest C. i In a raster map of specification A, it belongs to raster A1, and raster A1 is the point cloud of interest C. i The target raster in the raster map of specification A; based on the point cloud of interest C. i The three-dimensional coordinates determine the point cloud of interest C. i In a B-size raster map, it belongs to raster B2, and raster B2 is the point cloud of interest C. i The target raster in a B-size raster map.

[0093] S320. Determine the background height threshold corresponding to the target raster of each point cloud of interest as the background height threshold corresponding to each point cloud of interest in different background lookup tables.

[0094] Specifically, after determining the target raster for each point cloud of interest, the corresponding background height threshold is obtained, which serves as the background height threshold for each point cloud of interest in different background lookup tables. Continuing the example above, point cloud of interest C... i In a raster image of specification A, the target raster is raster A1, and the corresponding background height threshold in background lookup table 1 is H. A-1 Point cloud of interest C i In the B-size raster map, the target raster is raster B1, and the corresponding background height threshold in background lookup table 2 is H. B-2 .

[0095] In an alternative embodiment, such as Figure 4 As shown, S310 above, determining the target raster in the raster map corresponding to each point cloud of interest in different background lookup tables, includes:

[0096] S410. Map each point cloud of interest to a raster map of different sizes, and obtain the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes.

[0097] Optionally, the terminal uses different grid sizes to perform two-dimensional rasterization on each point cloud of interest in the current scene set, so as to map each point cloud of interest to a raster map of different sizes, and then obtain the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes.

[0098] S420. Obtain the raster area range of each raster in the raster map corresponding to each background lookup table.

[0099] The raster map contains multiple grids, each with a different raster region range. To determine the target grid to which the point cloud of interest belongs, the terminal further obtains the raster region range of each grid in the raster map corresponding to each background lookup table.

[0100] S430. The grid cells within the grid regions to which the two-dimensional coordinates of each point cloud of interest belong in the grid maps of different sizes are determined as the target grid cells of each point cloud of interest in the corresponding grid maps.

[0101] Specifically, the terminal traverses each point of interest (POI) in the set of POIs and performs position matching between the two-dimensional coordinates of the POI in each grid map of different sizes and the grid area range to which each grid belongs in the corresponding grid map, so as to obtain the target grid to which the POI belongs. That is, the grid range to which the two-dimensional coordinates of each POI in each grid map of different sizes belong is determined as the target grid of each POI in the corresponding grid map.

[0102] In this embodiment, the terminal determines the target raster in the corresponding raster map of each point of interest (POI) in different background lookup tables. The background height threshold corresponding to the target raster of each POI is then determined as the background height threshold for that POI. Specifically, the raster within the raster area of ​​each POI in different raster maps of varying sizes can be determined using the two-dimensional coordinates of each POI and the raster region of each raster in the corresponding background lookup table raster map. This raster is then used as the target raster for each POI in the corresponding raster map. This method effectively transforms the POI from three-dimensional space to a two-dimensional plane, facilitating the subsequent determination of the background height threshold for the target raster to which the POI belongs, thus ensuring the accuracy of the background point cloud determination.

[0103] The two-dimensional coordinates of the point cloud of interest in a raster map of different sizes depend on the position of the origin in the raster map and the size of the raster map. In one embodiment, such as Figure 5 As shown, S410 above, mapping each point cloud of interest to a raster map of different sizes, and obtaining the two-dimensional coordinates of each point cloud of interest in the corresponding raster map, includes:

[0104] S510. Obtain the two-dimensional coordinates of the preset reference point.

[0105] The preset reference point is the origin of the raster image of different specifications. Optionally, the two-dimensional coordinates of the origin in the raster images of different specifications can be the same or different. In this embodiment, the two-dimensional coordinates of the origin in the raster images of different specifications are the same, that is, the two-dimensional coordinates of the preset reference point are the same in the raster images of different specifications.

[0106] S520. Based on the two-dimensional coordinates of the preset reference point and the size of the raster images of different specifications, map each point of interest cloud to obtain the two-dimensional coordinates of each point of interest cloud in the raster images of different specifications.

[0107] The dimensions of a raster image are the same as the dimensions of the grid cells within the raster image.

[0108] Specifically, the terminal acquires the two-dimensional coordinates of a preset reference point, and then determines the two-dimensional coordinates of each point of interest in the corresponding raster based on the two-dimensional coordinates of the preset reference point, the size of each raster map of different specifications, and the three-dimensional coordinates of each point of interest cloud.

[0109] In an alternative embodiment, such as Figure 6 As shown, in step S520 above, based on the two-dimensional coordinates of a preset reference point and the dimensions of different raster sizes, each point cloud of interest is mapped to obtain the two-dimensional coordinates of each point cloud of interest in the corresponding raster, including:

[0110] S610. Based on the X-axis coordinates of the preset reference point in the two-dimensional coordinates, the X-axis coordinates of each point of interest cloud, and the length of the grid in the X-axis direction of each grid in the grid map of different specifications, determine the X-axis coordinate value of each point of interest cloud in the corresponding grid map.

[0111] Specifically, the X-axis coordinate value (index) of the point cloud of interest within raster maps of different sizes can be determined using the following formula. x :

[0112] index x =floor(x-min) x / grid_size x )

[0113] Where x is the X-axis coordinate in the 3D coordinate system of the point cloud of interest, min x The grid_size is the X-axis coordinate of the preset reference point in a two-dimensional coordinate system. x This represents the length of the grid along the X-axis.

[0114] S620. Based on the Y-axis coordinates of the preset reference point in the two-dimensional coordinates, the Y-axis coordinates of each point of interest cloud, and the length of the grid in the Y-axis direction of each grid in the grid map of different specifications, determine the Y-axis coordinate value of each point of interest cloud in the corresponding grid map.

[0115] Specifically, the Y-axis coordinate value (index) of the point cloud of interest within raster maps of different sizes can be determined using the following formula. y :

[0116] Index y =floor(y-min) y / grid_size y )

[0117] Where y is the Y-axis coordinate in the 3D coordinate system of the point cloud of interest, min y The grid_size is the Y-axis coordinate of the preset reference point in a two-dimensional coordinate system. y This represents the length of the grid along the Y-axis.

[0118] S630. Based on the X-axis and Y-axis coordinates of each point cloud of interest in the corresponding raster image, obtain the two-dimensional coordinates of each point cloud of interest in raster images of different sizes.

[0119] Specifically, based on the X-axis coordinate value (index) of each point cloud of interest within the corresponding raster image. y and Y-axis coordinate value index y This allows us to obtain the two-dimensional coordinates (index) of each point cloud of interest in raster maps of different sizes. y index y ).

[0120] In this embodiment, the terminal maps each point of interest (POI) cloud to the grid images based on the two-dimensional coordinates of a preset reference point and the dimensions of each grid image, thus obtaining the two-dimensional coordinates of each POI cloud in the grid images of different sizes. Specifically, the X-axis coordinates of each POI cloud in the corresponding grid image are determined based on the X-axis coordinates of the preset reference point, the X-axis coordinates of each POI cloud, and the length of the grid in the grid images of different sizes along the X-axis. Similarly, the Y-axis coordinates of each POI cloud in the corresponding grid image are determined based on the Y-axis coordinates of the preset reference point, the Y-axis coordinates of each POI cloud, and the length of the grid in the grid images of different sizes along the Y-axis. This process yields the two-dimensional coordinates of each POI cloud in the grid images of different sizes. This method accurately obtains the two-dimensional coordinates of each POI cloud in the grid images of different sizes, thereby improving the accuracy of the determined target grid, the accuracy of the background height threshold corresponding to the target grid, and ultimately, the accuracy of background point cloud determination based on the background height threshold.

[0121] In an optional embodiment, the different background lookup tables include a first background lookup table corresponding to a first raster image and a second background lookup table corresponding to a second raster image. The size of the grid cells in the first raster image is smaller than the size of the grid cells in the second raster image. Figure 7 As shown, S130 above, determining the background point cloud in the set of point clouds of interest based on the height value of each point cloud of interest and the corresponding background height threshold in each background lookup table, includes:

[0122] S710. Based on the height value of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table, determine the candidate background point cloud from the set of point clouds of interest.

[0123] Optionally, the terminal traverses the point clouds of interest in the set of point clouds of interest, compares the height value of the point cloud of interest with the first background height threshold corresponding to the point cloud of interest in the first background lookup table, and determines whether the point cloud of interest is a candidate background point cloud based on the comparison result, and then determines the candidate background point cloud from the set of point clouds of interest.

[0124] Optionally, if the comparison result shows that the height value of the point cloud of interest is greater than the first background height threshold, the point cloud of interest is determined to be the target point cloud and not the candidate background point cloud; if the comparison result shows that the height value of the point cloud of interest is less than or equal to the first background height threshold, the point cloud of interest is determined to be the candidate background point cloud.

[0125] S720. Determine the background point cloud from the candidate background point clouds based on the height value of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table.

[0126] Optionally, after the terminal determines the candidate background point cloud in the set of point clouds of interest, it further compares the height value of each candidate background point cloud with the second background height threshold corresponding to the candidate background point cloud in the second background lookup table, and determines whether each candidate background point cloud is a background point cloud based on the comparison result, and then determines the background point cloud from the candidate background point cloud.

[0127] Optionally, if the comparison result shows that the height value of the candidate background point cloud is greater than the second background height threshold, the corresponding candidate background point cloud is determined to be the target point cloud and not the background point cloud; if the comparison result shows that the height value of the candidate background point cloud is less than or equal to the second background height threshold, the corresponding candidate background point cloud is determined to be the background point cloud.

[0128] In practical applications, considering that there is a certain measurement error in the height value of point cloud, in order to reduce the inaccuracy of background point cloud determination caused by measurement error, the above process of determining the background point cloud based on the comparison result of the height value and the corresponding background height threshold can be relaxed to determine it based on the comparison result of the height difference between the height value and the corresponding background height threshold and the height difference threshold.

[0129] In an alternative embodiment, such as Figure 8 As shown, S710 above, determining candidate background point clouds from the set of point clouds of interest based on the height values ​​of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table, includes:

[0130] S810, Obtain the first height difference between the height value of each point cloud of interest and the first background height threshold.

[0131] S820. Identify the point clouds of interest whose first height difference is less than or equal to the first threshold as candidate background point clouds.

[0132] Optionally, the terminal performs the following judgment (1) for each point cloud of interest:

[0133]

[0134] in, Let i be the height value of point cloud i (Z-axis coordinate in 3D coordinates), min s Thresh is the first background height threshold. s This is the first threshold.

[0135] Specifically, the terminal obtains the first height difference between the height value of each point cloud of interest and the first background height threshold to perform the above judgment (1). If the first height difference is greater than the first threshold, the above judgment (1) is satisfied, and the terminal determines the corresponding point cloud of interest as the target point cloud. Conversely, if the first height is less than or equal to the first threshold, the above judgment (1) is not satisfied, and the terminal determines the corresponding point cloud of interest as the candidate background point cloud and continues to execute subsequent steps to further determine the background point cloud in the candidate background point cloud.

[0136] In one embodiment, such as Figure 9 As shown, S720 above, determining the background point cloud from the candidate background point cloud based on the height value of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table, includes:

[0137] S910. Obtain the second height difference between the height value of each candidate background point cloud and the second background height threshold.

[0138] S920. Candidate background point clouds whose second height difference is less than or equal to the second threshold are identified as background point clouds.

[0139] Optionally, the terminal performs the following judgment (2) for each candidate background point cloud:

[0140]

[0141] in, Let i be the height value of point cloud i (Z-axis coordinate in 3D coordinates), min b Thresh is the second background height threshold. b This is the second threshold.

[0142] Specifically, the terminal obtains the second height difference between the height value of each candidate background point cloud and the second background height threshold to perform the above judgment (2). If the second height difference is greater than the second threshold, the above judgment (2) is satisfied, and the terminal determines the corresponding candidate background point cloud as the target point cloud. Conversely, if the second height is less than or equal to the second threshold, the above judgment (2) is not satisfied, and the terminal determines the corresponding candidate background point cloud as the background point cloud, thereby obtaining the background point cloud in the candidate background point cloud.

[0143] In this embodiment, the terminal first determines a first background height threshold for the point cloud of interest (POI) based on a first background lookup table corresponding to a small-sized raster image. Then, it determines whether the corresponding POI is a candidate background point cloud based on a comparison between the POI's height value and the first background height threshold, or a comparison between the difference and the threshold. If it is a candidate background point cloud, it further determines a second background height threshold for the POI based on a second background lookup table corresponding to a large-sized raster image. This allows it to determine whether the candidate background point cloud is a background point cloud based on a comparison between the candidate background point cloud's height value and the second background height threshold, or a comparison between the difference and the threshold, thus obtaining the background point cloud. Because the small size of the raster in the small-sized raster image reduces the probability of identifying ground point clouds, the reliability of determining whether a point cloud is a background point cloud solely based on the first background lookup table of the small-sized raster image is poor. In contrast, the large size of the raster in the large-sized raster image effectively increases the probability of identifying ground point clouds. By using the above method of determining the small size first and then the large size, the accuracy of background point cloud identification can be effectively improved.

[0144] In one embodiment, the method further includes the process of constructing a background query table based on the point cloud set in the current scene. For example... Figure 10 As shown, the above method also includes:

[0145] S1010. Obtain the reference point cloud set in the current scene, and convert the reference point clouds in the reference point cloud set to raster maps of different sizes to obtain the two-dimensional coordinates of each reference point cloud in the corresponding raster map.

[0146] In this scenario, the target object is not present in the reference point cloud set. Optionally, the terminal may acquire the reference point cloud set for the current scene using LiDAR when the target object is not present in the current scene.

[0147] Specifically, the reference point cloud set is rasterized using grids of different sizes to transform the reference point clouds in the set into corresponding raster images, obtaining the two-dimensional coordinates of each reference point cloud in the raster images of different sizes. For example, the reference point cloud set is rasterized using a first-size grid to transform the reference point clouds in the set into a first raster image, obtaining the two-dimensional coordinates of each reference point cloud in the first raster image; the reference point cloud set is rasterized using a second-size grid to transform the reference point clouds in the set into a second raster image, obtaining the two-dimensional coordinates of each reference point cloud in the second raster image. The first size is smaller than the second size.

[0148] S1020. Based on the two-dimensional coordinates of each reference point cloud in the corresponding raster map, obtain the height value of the reference point cloud within the region of interest in each raster map of different specifications.

[0149] It should be noted that the background lookup table does not include the background height thresholds corresponding to all rasters in the raster tables of different sizes; it only includes the background height thresholds corresponding to rasters within the region of interest. Therefore, the terminal needs to determine the reference point clouds belonging to the region of interest based on the two-dimensional coordinates of each reference point cloud in the corresponding raster map, and then obtain the height value of each reference point cloud within the region of interest in each raster map of different sizes.

[0150] S1030. Based on the height values ​​of the reference point cloud within the region of interest in each raster map of different specifications and the rasters within the region of interest in each raster map of different specifications, determine different background lookup tables.

[0151] In this context, a background lookup table is determined by the height values ​​of all reference point clouds within the region of interest in a raster image of one specification and the corresponding graticules within the region of interest in the raster image. For raster images of multiple specifications, multiple different background lookup tables are determined accordingly.

[0152] In an alternative embodiment, such as Figure 11 As shown, in S1030 above, based on the height values ​​of the reference point clouds within the regions of interest in raster images of different specifications and the raster cells within the regions of interest in raster images of different specifications, different background lookup tables are determined, including:

[0153] S1110. Obtain the minimum height value of the reference point cloud in each grid cell within the region of interest in grid maps of different specifications.

[0154] Specifically, the terminal can determine the minimum height value corresponding to each cell within the region of interest in a raster map of different specifications by traversing and comparing updated height values. For example, for each type of raster map, the reference point clouds belonging to the same cell within the region of interest are traversed. The height value of the first reference point cloud is taken as the initial height value. The height value of the second reference point cloud is compared with the initial height value. If the height value of the second reference point is less than the initial height value, the height value of the second reference point is updated to the new initial height value. Conversely, if the height value of the second reference point is greater than or equal to the initial height value, the initial height value is maintained. The comparison continues with the height value of the third reference point and the new initial height value / the maintained initial height value, and so on, until the last reference point cloud is reached. This yields the minimum height value of the reference point cloud within each cell of the corresponding raster map.

[0155] S1120. Determine the background height threshold of the corresponding raster based on the minimum height value of the reference point cloud in each raster within the region of interest in the raster map of different specifications, and obtain the background lookup table corresponding to the raster map of different specifications.

[0156] Specifically, the terminal determines the background height threshold of the corresponding raster based on the minimum height value of the reference point cloud in each raster within the region of interest in each raster image of different specifications, thereby obtaining the background lookup table corresponding to each raster image of different specifications.

[0157] In an alternative embodiment, such as Figure 12 As shown, obtaining the reference point cloud set for the current scene in S1010 above includes:

[0158] S1210. Obtain the initial point cloud set in the current scene and extract the ground point cloud from the initial point cloud set.

[0159] Optionally, the RANSAC algorithm can be used to fit the ground equation, and the ground point cloud in the initial point cloud set can be obtained by setting a threshold.

[0160] S1220. Expand the ground point cloud based on its three-dimensional coordinates and interpolation algorithm to obtain a reference point cloud set.

[0161] Specifically, given the sparsity of point clouds, the terminal can use an interpolation algorithm to interpolate the three-dimensional coordinates of the ground point cloud in order to expand the ground point cloud and obtain the expanded ground point cloud as a reference point cloud set.

[0162] In this embodiment, the terminal obtains a set of reference point clouds where no target object exists in the current scene. It then converts these reference point clouds into raster maps of different sizes, obtaining the two-dimensional coordinates of each reference point cloud in its corresponding raster map. Based on these coordinates, the terminal obtains the height values ​​of the reference point clouds within the regions of interest (ROIs) in each raster map of different sizes. Subsequently, based on the height values ​​of the reference point clouds within their respective ROIs and the individual raster cells within those ROIs, different background lookup tables are determined. This allows for the construction of different background lookup tables for the current scene, obtaining the background raster height (i.e., the height of the ground point cloud) corresponding to each raster cell in each raster map of different sizes. This facilitates the subsequent determination of the background point clouds within the ROI set for the current scene and improves the accuracy of background point cloud identification.

[0163] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0164] In one embodiment, such as Figure 13 As shown, a background point cloud determination device is provided, including: a height determination module 1301, a threshold determination module 1302, and a background determination module 1303, wherein:

[0165] The height determination module 1301 is used to obtain the set of points of interest clouds in the current scene and obtain the height value of each point of interest cloud in the set of points of interest clouds;

[0166] The threshold determination module 1302 is used to obtain the background height threshold corresponding to each point cloud of interest in different background lookup tables; different background lookup tables correspond to raster images of different specifications, and each raster image of a specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specification of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image.

[0167] Background determination module 1303 is used to determine the background point cloud in the set of point clouds of interest based on the height value of each point cloud of interest and the background height threshold corresponding to each point cloud of interest in each background lookup table.

[0168] In one embodiment, the threshold determination module 1302 is specifically used for:

[0169] Identify the target raster in the raster map corresponding to each point cloud of interest in different background lookup tables; determine the background height threshold corresponding to the target raster of each point cloud of interest as the background height threshold corresponding to each point cloud of interest in different background lookup tables.

[0170] In one embodiment, the threshold determination module 1302 is specifically used for:

[0171] Each point cloud of interest is mapped to a raster map of different sizes to obtain the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes; the raster region range of each raster in the raster map corresponding to each background lookup table is obtained; the raster of the raster region range to which the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes belong is determined as the target raster of each point cloud of interest in the corresponding raster map.

[0172] In one embodiment, the threshold determination module 1302 is specifically used for:

[0173] Obtain the two-dimensional coordinates of a preset reference point; where the preset reference point is the origin of a raster image of different specifications; based on the two-dimensional coordinates of the preset reference point and the specifications of each raster image of different specifications, map each point of interest cloud to obtain the two-dimensional coordinates of each point of interest cloud in each raster image of different specifications.

[0174] In one embodiment, the threshold determination module 1302 is specifically used for:

[0175] Based on the X-axis coordinates of the preset reference point in the 2D coordinate system, the X-axis coordinates of each point of interest (POI) cloud, and the length of the grid cells in the raster images of different sizes along the X-axis, determine the X-axis coordinates of each POI cloud within the corresponding raster image. Based on the Y-axis coordinates of the preset reference point in the 2D coordinate system, the Y-axis coordinates of each POI cloud, and the length of the grid cells in the raster images of different sizes along the Y-axis, determine the Y-axis coordinates of each POI cloud within the corresponding raster image. Based on the X-axis and Y-axis coordinates of each POI cloud within the corresponding raster image, obtain the 2D coordinates of each POI cloud in the raster images of different sizes.

[0176] In one embodiment, the different background lookup tables include a first background lookup table corresponding to a first raster image and a second background lookup table corresponding to a second raster image; wherein, the size of the raster in the first raster image is smaller than the size of the raster in the second raster image; the background determination module 1303 is specifically used for:

[0177] Candidate background point clouds are determined from the set of point clouds of interest based on the height value of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table; background point clouds are determined from the candidate background point clouds based on the height value of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table.

[0178] In one embodiment, the background determination module 1303 is specifically used for:

[0179] Obtain the first height difference between the height value of each point cloud of interest and the first background height threshold; determine the point clouds of interest whose first height difference is less than or equal to the first threshold as candidate background point clouds.

[0180] In one embodiment, the background determination module 1303 is specifically used for:

[0181] Obtain the second height difference between the height value of each candidate background point cloud and the second background height threshold; determine the candidate background point clouds whose second height difference is less than or equal to the second threshold as background point clouds.

[0182] In one embodiment, the above-described apparatus further includes a background construction module for:

[0183] Obtain a set of reference point clouds for the current scene, and convert the reference point clouds in the set to raster maps of different sizes to obtain the two-dimensional coordinates of each reference point cloud in the corresponding raster map; the target object does not exist in the set of reference point clouds; based on the two-dimensional coordinates of each reference point cloud in the corresponding raster map, obtain the height value of the reference point cloud in the region of interest in each raster map of different sizes; based on the height value of the reference point cloud in the region of interest in each raster map of different sizes and the raster in the region of interest in each raster map of different sizes, determine different background lookup tables.

[0184] In one embodiment, the background construction module is specifically used for:

[0185] Obtain the minimum height value of the reference point cloud in each cell within the region of interest in raster images of different specifications; determine the background height threshold of the corresponding cell based on the minimum height value of the reference point cloud in each cell within the region of interest in raster images of different specifications, and obtain the background lookup table corresponding to each raster image of different specifications.

[0186] In one embodiment, the height determination module 1301 is specifically used for:

[0187] Obtain the initial point cloud set for the current scene and extract the ground point cloud from the initial point cloud set; expand the ground point cloud according to the three-dimensional coordinates of the ground point cloud and the interpolation algorithm to obtain the reference point cloud set.

[0188] Each module in the aforementioned background point cloud determination device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0189] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 14 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for determining background point clouds. The display screen can be an LCD screen or an e-ink display screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0190] Those skilled in the art will understand that Figure 14 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0191] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0192] Obtain the set of points of interest (POIs) in the current scene, and obtain the height value of each POI in the set. Obtain the background height threshold corresponding to each POI in different background lookup tables. Different background lookup tables correspond to raster images of different specifications. Each raster image of a specific specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specifications of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image. Based on the height value of each POI and the background height threshold corresponding to each POI in each background lookup table, determine the background point cloud in the set of POIs.

[0193] In one embodiment, the processor further performs the following steps when executing the computer program:

[0194] Identify the target raster in the raster map corresponding to each point cloud of interest in different background lookup tables; determine the background height threshold corresponding to the target raster of each point cloud of interest as the background height threshold corresponding to each point cloud of interest in different background lookup tables.

[0195] In one embodiment, the processor further performs the following steps when executing the computer program:

[0196] Each point cloud of interest is mapped to a raster map of different sizes to obtain the two-dimensional coordinates of each point cloud of interest in the corresponding raster map of different sizes; the raster region range of each raster in the raster map corresponding to each background lookup table is obtained; the raster of the raster region range to which the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes belong is determined as the target raster of each point cloud of interest in the corresponding raster map.

[0197] In one embodiment, the processor further performs the following steps when executing the computer program:

[0198] Obtain the two-dimensional coordinates of a preset reference point; where the preset reference point is the origin of a raster image of different specifications; based on the two-dimensional coordinates of the preset reference point and the specifications of each raster image of different specifications, map each point of interest cloud to obtain the two-dimensional coordinates of each point of interest cloud in each raster image of different specifications.

[0199] In one embodiment, the processor further performs the following steps when executing the computer program:

[0200] Based on the X-axis coordinates of the preset reference point in the 2D coordinate system, the X-axis coordinates of each point of interest (POI) cloud, and the length of the grid cells in the raster images of different sizes along the X-axis, determine the X-axis coordinates of each POI cloud within the corresponding raster image. Based on the Y-axis coordinates of the preset reference point in the 2D coordinate system, the Y-axis coordinates of each POI cloud, and the length of the grid cells in the raster images of different sizes along the Y-axis, determine the Y-axis coordinates of each POI cloud within the corresponding raster image. Based on the X-axis and Y-axis coordinates of each POI cloud within the corresponding raster image, obtain the 2D coordinates of each POI cloud in the raster images of different sizes.

[0201] In one embodiment, the different background lookup tables include a first background lookup table corresponding to a first raster image and a second background lookup table corresponding to a second raster image; wherein the size of the raster in the first raster image is smaller than the size of the raster in the second raster image; the processor further implements the following steps when executing the computer program:

[0202] Candidate background point clouds are determined from the set of point clouds of interest based on the height value of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table; background point clouds are determined from the candidate background point clouds based on the height value of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table.

[0203] In one embodiment, the processor further performs the following steps when executing the computer program:

[0204] Obtain the first height difference between the height value of each point cloud of interest and the first background height threshold; determine the point clouds of interest whose first height difference is less than or equal to the first threshold as candidate background point clouds.

[0205] In one embodiment, the processor further performs the following steps when executing the computer program:

[0206] Obtain the second height difference between the height value of each candidate background point cloud and the second background height threshold; determine the candidate background point clouds whose second height difference is less than or equal to the second threshold as background point clouds.

[0207] In one embodiment, the processor further performs the following steps when executing the computer program:

[0208] Obtain a set of reference point clouds for the current scene, and convert the reference point clouds in the set to raster maps of different sizes to obtain the two-dimensional coordinates of each reference point cloud in the corresponding raster map; the target object does not exist in the set of reference point clouds; based on the two-dimensional coordinates of each reference point cloud in the corresponding raster map, obtain the height value of the reference point cloud in the region of interest in each raster map of different sizes; based on the height value of the reference point cloud in the region of interest in each raster map of different sizes and the raster in the region of interest in each raster map of different sizes, determine different background lookup tables.

[0209] In one embodiment, the processor further performs the following steps when executing the computer program:

[0210] Obtain the minimum height value of the reference point cloud in each cell within the region of interest in raster images of different specifications; determine the background height threshold of the corresponding cell based on the minimum height value of the reference point cloud in each cell within the region of interest in raster images of different specifications, and obtain the background lookup table corresponding to each raster image of different specifications.

[0211] In one embodiment, the processor further performs the following steps when executing the computer program:

[0212] Obtain the initial point cloud set for the current scene and extract the ground point cloud from the initial point cloud set; expand the ground point cloud according to the three-dimensional coordinates of the ground point cloud and the interpolation algorithm to obtain the reference point cloud set.

[0213] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0214] Obtain the set of points of interest (POIs) in the current scene, and obtain the height value of each POI in the set. Obtain the background height threshold corresponding to each POI in different background lookup tables. Different background lookup tables correspond to raster images of different specifications. Each raster image of a specific specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specifications of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image. Based on the height value of each POI and the background height threshold corresponding to each POI in each background lookup table, determine the background point cloud in the set of POIs.

[0215] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0216] Identify the target raster in the raster map corresponding to each point cloud of interest in different background lookup tables; determine the background height threshold corresponding to the target raster of each point cloud of interest as the background height threshold corresponding to each point cloud of interest in different background lookup tables.

[0217] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0218] Each point cloud of interest is mapped to a raster map of different sizes to obtain the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes; the raster region range of each raster in the raster map corresponding to each background lookup table is obtained; the raster of the raster region range to which the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes belong is determined as the target raster of each point cloud of interest in the corresponding raster map.

[0219] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0220] Obtain the two-dimensional coordinates of a preset reference point; where the preset reference point is the origin of a raster image of different specifications; based on the two-dimensional coordinates of the preset reference point and the specifications of each raster image of different specifications, map each point of interest cloud to obtain the two-dimensional coordinates of each point of interest cloud in each raster image of different specifications.

[0221] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0222] Based on the X-axis coordinates of the preset reference point in the 2D coordinate system, the X-axis coordinates of each point of interest (POI) cloud, and the length of the grid cells in the raster images of different sizes along the X-axis, determine the X-axis coordinates of each POI cloud within the corresponding raster image. Based on the Y-axis coordinates of the preset reference point in the 2D coordinate system, the Y-axis coordinates of each POI cloud, and the length of the grid cells in the raster images of different sizes along the Y-axis, determine the Y-axis coordinates of each POI cloud within the corresponding raster image. Based on the X-axis and Y-axis coordinates of each POI cloud within the corresponding raster image, obtain the 2D coordinates of each POI cloud in the raster images of different sizes.

[0223] In one embodiment, the different background lookup tables include a first background lookup table corresponding to a first raster image and a second background lookup table corresponding to a second raster image; wherein the size of the raster in the first raster image is smaller than the size of the raster in the second raster image; when the computer program is executed by the processor, it further implements the following steps:

[0224] Candidate background point clouds are determined from the set of point clouds of interest based on the height value of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table; background point clouds are determined from the candidate background point clouds based on the height value of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table.

[0225] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0226] Obtain the first height difference between the height value of each point cloud of interest and the first background height threshold; determine the point clouds of interest whose first height difference is less than or equal to the first threshold as candidate background point clouds.

[0227] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0228] Obtain the second height difference between the height value of each candidate background point cloud and the second background height threshold; determine the candidate background point clouds whose second height difference is less than or equal to the second threshold as background point clouds.

[0229] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0230] Obtain a set of reference point clouds for the current scene, and convert the reference point clouds in the set to raster maps of different sizes to obtain the two-dimensional coordinates of each reference point cloud in the corresponding raster map; the target object does not exist in the set of reference point clouds; based on the two-dimensional coordinates of each reference point cloud in the corresponding raster map, obtain the height value of the reference point cloud in the region of interest in each raster map of different sizes; based on the height value of the reference point cloud in the region of interest in each raster map of different sizes and the raster in the region of interest in each raster map of different sizes, determine different background lookup tables.

[0231] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0232] Obtain the minimum height value of the reference point cloud in each cell within the region of interest in raster images of different specifications; determine the background height threshold of the corresponding cell based on the minimum height value of the reference point cloud in each cell within the region of interest in raster images of different specifications, and obtain the background lookup table corresponding to each raster image of different specifications.

[0233] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0234] Obtain the initial point cloud set for the current scene and extract the ground point cloud from the initial point cloud set; expand the ground point cloud according to the three-dimensional coordinates of the ground point cloud and the interpolation algorithm to obtain the reference point cloud set.

[0235] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0236] Obtain the set of points of interest (POIs) in the current scene, and obtain the height value of each POI in the set. Obtain the background height threshold corresponding to each POI in different background lookup tables. Different background lookup tables correspond to raster images of different specifications. Each raster image of a specific specification is a planar image obtained by rasterizing the current scene in two dimensions using the same specifications of raster. Each background lookup table includes the background height threshold of each raster in the corresponding raster image. Based on the height value of each POI and the background height threshold corresponding to each POI in each background lookup table, determine the background point cloud in the set of POIs.

[0237] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0238] Identify the target raster in the raster map corresponding to each point cloud of interest in different background lookup tables; determine the background height threshold corresponding to the target raster of each point cloud of interest as the background height threshold corresponding to each point cloud of interest in different background lookup tables.

[0239] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0240] Each point cloud of interest is mapped to a raster map of different sizes to obtain the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes; the raster region range of each raster in the raster map corresponding to each background lookup table is obtained; the raster of the raster region range to which the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes belong is determined as the target raster of each point cloud of interest in the corresponding raster map.

[0241] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0242] Obtain the two-dimensional coordinates of a preset reference point; where the preset reference point is the origin of a raster image of different specifications; based on the two-dimensional coordinates of the preset reference point and the specifications of each raster image of different specifications, map each point of interest cloud to obtain the two-dimensional coordinates of each point of interest cloud in each raster image of different specifications.

[0243] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0244] Based on the X-axis coordinates of the preset reference point in the 2D coordinate system, the X-axis coordinates of each point of interest (POI) cloud, and the length of the grid cells in the raster images of different sizes along the X-axis, determine the X-axis coordinates of each POI cloud within the corresponding raster image. Based on the Y-axis coordinates of the preset reference point in the 2D coordinate system, the Y-axis coordinates of each POI cloud, and the length of the grid cells in the raster images of different sizes along the Y-axis, determine the Y-axis coordinates of each POI cloud within the corresponding raster image. Based on the X-axis and Y-axis coordinates of each POI cloud within the corresponding raster image, obtain the 2D coordinates of each POI cloud in the raster images of different sizes.

[0245] In one embodiment, the different background lookup tables include a first background lookup table corresponding to a first raster image and a second background lookup table corresponding to a second raster image; wherein the size of the raster in the first raster image is smaller than the size of the raster in the second raster image; when the computer program is executed by the processor, it further implements the following steps:

[0246] Candidate background point clouds are determined from the set of point clouds of interest based on the height value of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table; background point clouds are determined from the candidate background point clouds based on the height value of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table.

[0247] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0248] Obtain the first height difference between the height value of each point cloud of interest and the first background height threshold; determine the point clouds of interest whose first height difference is less than or equal to the first threshold as candidate background point clouds.

[0249] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0250] Obtain the second height difference between the height value of each candidate background point cloud and the second background height threshold; determine the candidate background point clouds whose second height difference is less than or equal to the second threshold as background point clouds.

[0251] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0252] Obtain a set of reference point clouds for the current scene, and convert the reference point clouds in the set to raster maps of different sizes to obtain the two-dimensional coordinates of each reference point cloud in the corresponding raster map; the target object does not exist in the set of reference point clouds; based on the two-dimensional coordinates of each reference point cloud in the corresponding raster map, obtain the height value of the reference point cloud in the region of interest in each raster map of different sizes; based on the height value of the reference point cloud in the region of interest in each raster map of different sizes and the raster in the region of interest in each raster map of different sizes, determine different background lookup tables.

[0253] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0254] Obtain the minimum height value of the reference point cloud in each cell within the region of interest in raster images of different specifications; determine the background height threshold of the corresponding cell based on the minimum height value of the reference point cloud in each cell within the region of interest in raster images of different specifications, and obtain the background lookup table corresponding to each raster image of different specifications.

[0255] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:

[0256] Obtain the initial point cloud set for the current scene and extract the ground point cloud from the initial point cloud set; expand the ground point cloud according to the three-dimensional coordinates of the ground point cloud and the interpolation algorithm to obtain the reference point cloud set.

[0257] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0258] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0259] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for determining background point clouds, characterized in that, The method includes: Obtain the set of points of interest (POIs) in the current scene, and obtain the height value of each POI in the set of POIs; Obtain the background height threshold corresponding to each point cloud of interest in different background lookup tables; different background lookup tables correspond to raster images of different specifications, each raster image of a specification is a planar image obtained by rasterizing the current scene in two dimensions using raster images of the same specification, and each background lookup table includes the background height threshold of each raster in the corresponding raster image; the different background lookup tables include a first background lookup table corresponding to a first raster image and a second background lookup table corresponding to a second raster image; wherein, the size of the raster in the first raster image is smaller than the size of the raster in the second raster image. Candidate background point clouds are determined from the set of point clouds of interest based on the height value of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table. The background point cloud is determined from the candidate background point clouds based on the height value of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table.

2. The method according to claim 1, characterized in that, The step of obtaining the background height threshold corresponding to each of the point clouds of interest in different background lookup tables includes: Determine the target raster in the raster map corresponding to each point cloud of interest in different background lookup tables; The background height threshold corresponding to the target raster of each point cloud of interest is determined as the background height threshold corresponding to each point cloud of interest in different background lookup tables.

3. The method according to claim 2, characterized in that, Determining the target raster in the raster map corresponding to each of the point clouds of interest in different background lookup tables includes: Each point cloud of interest is mapped to a raster image of a different size to obtain the two-dimensional coordinates of each point cloud of interest in the raster image of a different size. Obtain the raster region range of each raster in the raster image corresponding to each of the background query tables; The grid cells within the grid region to which the two-dimensional coordinates of each point cloud of interest belong in each of the different grid sizes are determined as the target grid cells of each point cloud of interest in the corresponding grid cells.

4. The method according to claim 3, characterized in that, The step of mapping each point cloud of interest to a raster map of different sizes, and obtaining the two-dimensional coordinates of each point cloud of interest in the raster map of different sizes, includes: Obtain the two-dimensional coordinates of a preset reference point; wherein the preset reference point is the origin of the raster images of different specifications; Based on the two-dimensional coordinates of the preset reference point and the size of each of the different grid images, each point cloud of interest is mapped to obtain the two-dimensional coordinates of each point cloud of interest in each of the different grid images.

5. The method according to claim 4, characterized in that, The step of mapping each point cloud of interest to the raster image of different sizes based on the two-dimensional coordinates of the preset reference point and the size of the raster image includes: Based on the X-axis coordinates of the preset reference point in the two-dimensional coordinates, the X-axis coordinates of each point of interest cloud, and the length of the grid in the X-axis direction of each grid in the different sizes of the grid, the X-axis coordinate values ​​of each point of interest cloud in the corresponding grid are determined. Based on the Y-axis coordinates of the preset reference point in the two-dimensional coordinates, the Y-axis coordinates of each point of interest cloud, and the length of the grid in the Y-axis direction of each grid of different specifications, the Y-axis coordinate values ​​of each point of interest cloud in the corresponding grid are determined. Based on the X-axis and Y-axis coordinates of each point cloud of interest within the corresponding raster image, the two-dimensional coordinates of each point cloud of interest in each raster image of different sizes are obtained.

6. The method according to any one of claims 1-5, characterized in that, The step of determining candidate background point clouds from the set of point clouds of interest based on the height values ​​of each point cloud of interest and the first background height threshold corresponding to each point cloud of interest in the first background lookup table includes: Obtain the first height difference between the height values ​​of each point cloud of interest and the first background height threshold; Point clouds of interest whose first height difference is less than or equal to a first threshold are identified as candidate background point clouds.

7. The method according to any one of claims 1-5, characterized in that, The step of determining the background point cloud from the candidate background point clouds based on the height values ​​of each candidate background point cloud and the second background height threshold corresponding to each candidate background point cloud in the second background lookup table includes: Obtain the second height difference between the height value of each candidate background point cloud and the second background height threshold; Candidate background point clouds with a second height difference less than or equal to a second threshold are identified as the background point clouds.

8. The method according to any one of claims 1-5, characterized in that, The method further includes: Obtain a set of reference point clouds for the current scene, and convert the reference point clouds in the set of reference point clouds to raster maps of different sizes to obtain the two-dimensional coordinates of each reference point cloud in the corresponding raster map; wherein, there is no target object in the set of reference point clouds. Based on the two-dimensional coordinates of each reference point cloud in the corresponding raster image, obtain the height value of the reference point cloud in the region of interest in each raster image of different specifications; Based on the height values ​​of the reference point cloud within the region of interest in each of the different raster sizes and the raster within the region of interest in each of the different raster sizes, the different background lookup tables are determined.

9. The method according to claim 8, characterized in that, The step of determining the different background lookup tables based on the height values ​​of reference point clouds within the regions of interest in each of the different raster sizes and the raster cells within the regions of interest in each of the different raster sizes includes: Obtain the minimum height value of the reference point cloud in each grid cell within the region of interest in each of the different grid sizes described; The minimum height value of the reference point cloud in each grid cell within the region of interest of each of the different grid sizes is determined as the background height threshold of the corresponding grid cell, thus obtaining the background lookup table corresponding to each of the different grid sizes.

10. The method according to claim 8, characterized in that, The step of obtaining the reference point cloud set in the current scene includes: Obtain the initial point cloud set for the current scene, and extract the ground point cloud from the initial point cloud set; The ground point cloud is expanded based on its three-dimensional coordinates and an interpolation algorithm to obtain the reference point cloud set.

11. A device for determining background point clouds, characterized in that, The device includes: The height determination module is used to obtain a set of point clouds of interest in the current scene and to obtain the height value of each point cloud of interest in the set of point clouds of interest. A threshold determination module is used to obtain the background height threshold corresponding to each of the points of interest in different background lookup tables; different background lookup tables correspond to raster images of different specifications, and each specification of raster image is a planar image obtained by rasterizing the current scene in two dimensions using raster images of the same specification. Each background lookup table includes the background height threshold of each raster in the corresponding raster image; the different background lookup tables include a first background lookup table corresponding to a first raster image and a second background lookup table corresponding to a second raster image; wherein, the size of the raster in the first raster image is smaller than the size of the raster in the second raster image. The background determination module is used to determine candidate background point clouds from the set of point clouds of interest based on the height values ​​of each point cloud of interest and a first background height threshold corresponding to each point cloud of interest in the first background lookup table, and to determine the background point cloud from the candidate background point clouds based on the height values ​​of each candidate background point cloud and a second background height threshold corresponding to each candidate background point cloud in the second background lookup table.

12. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

14. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

Images