Point cloud hole detection method, device and electronic equipment

Abstract

The embodiments of the present application provide a method, device, and electronic device for detecting holes in a point cloud, wherein the method includes: determining a coordinate origin based on three-dimensional point cloud data, establishing a relative coordinate system based on the coordinate origin; constructing a two-dimensional plane based on a first direction and a second direction of the relative coordinate system, rasterizing the two-dimensional plane based on a preset grid size to obtain a grid map; obtaining two-dimensional point cloud data projected onto the two-dimensional plane; and determining hole information in the three-dimensional point cloud data based on the positional relationship between the two-dimensional point cloud data and each grid in the grid map. In this way, hole detection for point cloud data in three-dimensional space is transformed into hole detection for two-dimensional images, simplifying the computational complexity of hole detection and improving the processing efficiency of detecting hole information.

Description

Technical Field

[0001] The present invention relates to the field of point cloud data processing, and in particular to a method, device and electronic equipment for detecting point cloud holes. Background Art

[0002] The rapid development of unmanned vehicles has placed higher demands on digital navigation technology, requiring navigation to rely on complete, high-precision three-dimensional maps built from point cloud data. Point cloud data contains the three-dimensional coordinates of objects, providing accurate location and size information. However, during the scanning process, environmental factors can affect the integrity of the point cloud data, leading to holes in the point cloud data. Currently, point cloud hole detection is time-consuming and complex, resulting in low detection efficiency. Summary of the Invention

[0003] In order to solve the above technical problems, embodiments of the present invention provide a method, device and electronic device for detecting point cloud holes.

[0004] In a first aspect, an embodiment of the present invention provides a method for detecting point cloud holes, comprising:

[0005] Determine a coordinate origin based on the three-dimensional point cloud data, and establish a relative coordinate system based on the coordinate origin;

[0006] constructing a two-dimensional plane based on the first direction and the second direction of the relative coordinate system, and performing a grid processing on the two-dimensional plane based on a preset grid size to obtain a grid map;

[0007] Acquire two-dimensional point cloud data obtained by projecting the three-dimensional point cloud data onto the two-dimensional plane;

[0008] The hole information of the three-dimensional point cloud data is determined based on the positional relationship between the two-dimensional point cloud data and each grid in the grid map.

[0009] In a second aspect, an embodiment of the present invention provides a point cloud hole detection device, comprising:

[0010] a processing module, configured to determine a coordinate origin based on the three-dimensional point cloud data, and establish a relative coordinate system based on the coordinate origin;

[0011] a construction module, configured to construct a two-dimensional plane based on the first direction and the second direction of the relative coordinate system, and perform a gridding process on the two-dimensional plane based on a preset grid size to obtain a grid map;

[0012] An acquisition module, configured to acquire two-dimensional point cloud data obtained by projecting the three-dimensional point cloud data onto the two-dimensional plane;

[0013] The first determining module is configured to determine hole information of the three-dimensional point cloud data based on a positional relationship between the two-dimensional point cloud data and each grid in the grid map.

[0014] In a third aspect, an embodiment of the present invention provides an electronic device comprising a memory and a processor, wherein the memory is used to store a computer program, and when the computer program is run by the processor, the steps of the point cloud hole detection method provided in the first aspect are executed.

[0015] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when running on a processor, executes the steps of the point cloud hole detection method provided in the first aspect.

[0016] The point cloud hole detection method, device, and electronic device provided by the present application determine a coordinate origin based on three-dimensional point cloud data, establish a relative coordinate system based on the coordinate origin, construct a two-dimensional plane based on the first and second directions of the relative coordinate system, rasterize the two-dimensional plane based on a preset grid size to obtain a grid map, obtain two-dimensional point cloud data projected onto the two-dimensional plane, and determine hole information in the three-dimensional point cloud data based on the positional relationship between the two-dimensional point cloud data and each grid in the grid map. In this way, hole detection for point cloud data in three-dimensional space is transformed into hole detection for two-dimensional images, simplifying the computational complexity of hole detection and improving the processing efficiency of detecting hole information. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope of protection of the present invention. In each of the drawings, similar components are numbered similarly.

[0018] Figure 1 A schematic diagram of a process for detecting point cloud holes provided by an embodiment of the present application is shown;

[0019] Figure 2 A flow chart of step S104 of the point cloud hole detection method provided in an embodiment of the present application is shown;

[0020] Figure 3 A schematic diagram of a two-dimensional image provided by an embodiment of the present application is shown;

[0021] Figure 4 A structural schematic diagram of a point cloud hole detection device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments.

[0023] The components of the embodiments of the present application generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed application, but rather merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort are within the scope of protection of the present application.

[0024] Hereinafter, the terms "including", "having" and their cognates, which may be used in various embodiments of the present application, are intended only to indicate specific features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be understood as first excluding the existence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items or the possibility of adding one or more features, numbers, steps, operations, elements, components or combinations of the foregoing items.

[0025] Furthermore, the terms “first,” “second,” “third,” etc., are merely used for distinguishing descriptions and are not to be understood as indicating or implying relative importance.

[0026] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by those skilled in the art to which the various embodiments of the present application belong. The terms (such as those defined in generally used dictionaries) will be interpreted as having the same meaning as in the context of the relevant technical field and will not be interpreted as having an idealized meaning or an overly formal meaning unless clearly defined in the various embodiments of the present application.

[0027] Example 1

[0028] The embodiments of the present disclosure provide a method for detecting point cloud holes.

[0029] For details, please refer to Figure 1 ,The detection methods of point cloud holes include:

[0030] Step S101 : determining a coordinate origin based on three-dimensional point cloud data, and establishing a relative coordinate system based on the coordinate origin.

[0031] In this embodiment, initial point cloud data can be obtained through scanning with a device such as radar. This initial point cloud data is dense and may contain holes due to environmental factors. This initial point cloud data can be processed to produce a digital navigation tool such as a three-dimensional map. To ensure the accuracy and completeness of the resulting digital navigation tool, it is necessary to detect holes in the initial point cloud data.

[0032] Because the initial point cloud data is dense and numerous, thinning the initial point cloud data is necessary to improve the speed of hole detection. The resulting 3D point cloud data also has holes, and the distribution of holes in the 3D point cloud data can represent the distribution of holes in the initial point cloud data. Since the number of 3D points in the 3D point cloud data after thinning is much smaller than that in the initial point cloud, the speed of hole detection can be improved.

[0033] In one embodiment, the acquisition of the three-dimensional point cloud data includes:

[0034] The initial point cloud data is divided into multiple three-dimensional grids according to a preset three-dimensional voxel size, the coordinate information and elevation information of the target three-dimensional point of the three-dimensional grid are determined based on the multiple three-dimensional points of each three-dimensional grid, and the three-dimensional point cloud data is generated based on the multiple target three-dimensional points.

[0035] Specifically, the preset 3D voxel size can be user-defined or determined based on a preset filtering principle. The preset filtering principle can be determined based on the actual navigation accuracy of the unmanned device or other point cloud data filtering principles, which are not limited here. For example, the unmanned device actually only needs to ensure that there are 3D points in a 1m×1m×1m 3D voxel to ensure the navigation operation of the unmanned device, so the preset 3D voxel size can be determined as 1m×1m×1m. For example, the original point cloud data is segmented according to the preset 3D voxel size of 1m×1m×1m, divided into multiple 3D grids of size 1m×1m×1m, and the coordinate information and elevation information of the target 3D point of each 3D grid are calculated based on the coordinate information and elevation information of all 3D points in each 1m×1m×1m 3D grid. For example, the target 3D point can be the centroid of each 3D grid, and the coordinate information and elevation information of the centroid of each 3D grid are the average of the coordinate information and the average of the elevation information of all 3D points in each 3D grid.

[0036] In this embodiment, the three-dimensional point cloud data has much fewer three-dimensional points than the initial point cloud data. In order to further simplify the complexity of the calculation, the three-dimensional point cloud data can be determined first, and then point cloud hole detection can be performed based on the three-dimensional point cloud data to reduce the amount of calculation and save computing time and computing resources.

[0037] In this embodiment, any 3D point in the 3D point cloud data can be selected as the coordinate origin. This allows relative position information to be determined based on the relative position information between the position information of other points and the coordinate origin. In this implementation, it's impossible to ensure that all relative position information is positive; that is, it's impossible to guarantee that all 3D points in the initial point cloud data have positive 3D information in the relative coordinate system. To ensure that all relative position information is positive, the 3D point with the minimum coordinate value in the 3D point cloud data can be determined as the coordinate origin, and a relative coordinate system can be established based on the 3D point with the minimum coordinate value.

[0038] In one embodiment, establishing a relative coordinate system based on the coordinate origin in step S101 includes the following steps:

[0039] Obtaining a minimum coordinate value in a first direction, a minimum coordinate value in a second direction, and a minimum coordinate value in a third direction in the three-dimensional point cloud data;

[0040] The coordinate origin is determined according to the minimum coordinate value in the first direction, the minimum coordinate value in the second direction, and the minimum coordinate value in the third direction.

[0041] For example, if the minimum coordinate value in the first direction of the 3D point cloud data is Min_x, the minimum coordinate value in the second direction is Min_y, and the minimum coordinate value in the third direction is Min_z, then the 3D point O (Min_x, Min_y, Min_z) is used as the coordinate origin. Assuming any 3D point P (x_k, y_k, z_k) in the 3D point cloud data, the relative position of any 3D point P in the relative coordinate system is Q (x_k-Min_x, y_k-Min_y, z_k-Min_z), and the 3D point Q (x_k-Min_x, y_k-Min_y, z_k-Min_z) is used as the 3D information of the 3D point P in the relative coordinate system. 3D information can include coordinate information and elevation information.

[0042] In this way, it can be ensured that the relative positions of other three-dimensional points with respect to the coordinate origin are all positive, which simplifies the amount of calculation of subsequent void information and saves computing resources and time.

[0043] Step S102 : constructing a two-dimensional plane based on the first direction and the second direction of the relative coordinate system, and performing a gridding process on the two-dimensional plane based on a preset grid size to obtain a grid map.

[0044] In this embodiment, elevation information is not considered, and a two-dimensional plane is established in a first direction and a second direction. The first direction and the second direction can be the x-direction and the y-direction in a three-dimensional rectangular coordinate system. The preset grid size can be determined based on the actual navigation accuracy of the unmanned device. In this embodiment, the preset grid size can be set to 1 meter x 1 meter squares, and the raster image after rasterizing the two-dimensional plane includes multiple 1 meter x 1 meter grids. In other embodiments, the preset grid size can be set to 2 meter x 3 meter grids, which is not limited here.

[0045] In one embodiment, the first and second dimensions of the preset 3D voxel size are respectively the same as the first and second dimensions of the preset grid size. This ensures that the first and second dimensions of each grid in the grid image are the same as the first and second dimensions of the 3D voxel grids divided according to the preset 3D voxel size. This ensures that the 2D point cloud data can subsequently be divided into corresponding grids according to coordinate information.

[0046] Step S103 : obtaining two-dimensional point cloud data obtained by projecting the three-dimensional point cloud data onto the two-dimensional plane.

[0047] In this embodiment, in order to reduce the amount of data calculation, the three-dimensional point cloud data is projected onto a two-dimensional plane to obtain corresponding two-dimensional point cloud data. Specifically, the two-dimensional plane can be a two-dimensional plane where the two-dimensional XOY coordinate system is located. The data dimension is reduced and the elevation information is not considered. The three-dimensional point cloud data is projected onto the two-dimensional plane according to the coordinate information to obtain the two-dimensional point cloud data. Although the two-dimensional point cloud data has no elevation information, the coordinate information is retained on the two-dimensional plane. The sparse feature information, crowded feature information, and hole information of the two-dimensional point cloud data can characterize the sparse feature information, crowded feature information, and hole information of the three-dimensional point cloud data.

[0048] Step S104 : determining hole information of the three-dimensional point cloud data based on a positional relationship between the two-dimensional point cloud data and each grid in the grid map.

[0049] Specifically, each two-dimensional point in the two-dimensional point cloud data includes coordinate information. According to the coordinate information of each two-dimensional point, it can be determined which grid in the grid map each two-dimensional point is located in. According to the coordinate information of each two-dimensional point, the two-dimensional points are distributed to the corresponding grids. According to the distribution of the two-dimensional points in the grid, the hole information in the grid map can be determined, and then the hole information of the three-dimensional point cloud data can be determined according to the hole information of the grid map.

[0050] In one embodiment, see Figure 2 , step S104 includes:

[0051] Step S1041, based on the position coordinates of each two-dimensional point and the grid map position information, determining whether each grid includes at least one two-dimensional point;

[0052] Step S1042: when the grid includes at least one two-dimensional point, adding a first mark to the grid;

[0053] Step S1043: when the grid does not include a two-dimensional point, adding a second mark to the grid;

[0054] Step S1044 : determining hole information of the three-dimensional point cloud data according to the grid containing the second mark in the grid map.

[0055] In this embodiment, the first and second markers can be letters or numbers. For example, the first marker is 1 and the second marker is 0. Other values are also possible and are not limited here. In this way, the grids in the grid image are distinguished by the first and second markers. The second marker is added to the grids that do not contain a two-dimensional point. Based on the number of grids containing the second marker and the size of each grid, the hole information of the two-dimensional plane in the grid image can be determined, and the hole information of the two-dimensional plane in the grid image can be converted into hole information of the three-dimensional point cloud data. In this way, relatively accurate hole information of the three-dimensional point cloud data can be obtained.

[0056] In another embodiment, step S104 includes: the above steps S1041, S1042, and S1043, and further includes the following steps:

[0057] Step S1045 , mapping each grid of the grid image to a pixel point, and determining a first pixel value or a second pixel value of the corresponding pixel point according to the first mark or the second mark of each grid, to obtain a two-dimensional image;

[0058] Step S1046: Determine hole information of the three-dimensional point cloud data based on the two-dimensional image.

[0059] Specifically, each pixel of the two-dimensional image is adjusted based on the first or second marker of each grid. The first marker can be pre-set to correspond to a first pixel value, and the second marker can be pre-set to correspond to a second pixel value. For example, the first pixel value is (255, 255, 255), and the second pixel value is (0, 0, 0). In this way, hole information can be determined based on each pixel of the two-dimensional image.

[0060] In this way, in order to facilitate the subsequent point cloud hole detection, the two-dimensional plane is converted into a two-dimensional image, and each grid of the two-dimensional plane is mapped to a pixel point according to the first mark or the second mark of each grid. The pixel value is set for the pixel point corresponding to each grid, and the area occupied by the pixel point of the second pixel value is detected. The product of the detected area and the grid size is used as the hole information of the two-dimensional point cloud data, so that the hole information of the three-dimensional point cloud data is converted based on the hole information of the two-dimensional point cloud data, thereby improving the efficiency of subsequent three-dimensional point cloud hole detection.

[0061] In one embodiment, the cavity information includes cavity area information, and step S1046 includes the following steps:

[0062] Acquire continuous pixel points whose pixel values are second pixel values in the two-dimensional image;

[0063] Based on the size information of the continuous pixel points and each grid, the area information of the hole is determined.

[0064] Specifically, the continuous pixel points in the two-dimensional image whose pixel values are the second pixel value represent a plurality of continuous grids that do not contain two-dimensional points, so that the area information occupied by the continuous plurality of grids corresponding to the continuous pixel points whose pixel values are the second pixel value can be used as the area information of the holes in the three-dimensional point cloud data.

[0065] In one embodiment, the area information occupied by multiple consecutive grids can be calculated using the following formula 1: S = N × L, where S represents the hole area, N represents the number of consecutive pixels with the second pixel value, and L represents the grid size. In this way, the hole information of the three-dimensional point cloud data can be quickly and accurately calculated, reducing the amount of calculation and shortening the calculation time.

[0066] In one embodiment, in order to facilitate users to understand the quality of the three-dimensional point cloud data, the quality of the point cloud data can be judged based on the three-dimensional point cloud data. Specifically, after step S104, the point cloud hole detection method can further include the following steps:

[0067] The point cloud quality of the three-dimensional point cloud data is determined based on the hole information of the three-dimensional point cloud data, where the hole information includes at least the hole area and the number of holes.

[0068] It is understood that smaller the hole area and smaller the number of holes, the more evenly distributed the 3D point cloud data is, and the better its quality is. A preset point cloud quality evaluation method can be used to assess whether the point cloud data is qualified. For example, the preset point cloud quality evaluation method can be set to define 3D point cloud data as qualified if the hole area is less than or equal to a preset hole area threshold, and the number of holes is less than or equal to a preset hole number threshold. Other point cloud quality evaluation methods are also possible and are not limited here.

[0069] In one embodiment, determining the point cloud quality of the three-dimensional point cloud data based on the hole information of the three-dimensional point cloud data may include the following steps:

[0070] Determining a preset operating area corresponding to the three-dimensional point cloud data, and selecting a portion of the preset operating area as a reference area;

[0071] Determining first cavity information within the preset operation area and second cavity information within the reference area;

[0072] The point cloud quality of the three-dimensional point cloud data is evaluated based on the first hole information and / or the second hole information.

[0073] See also Figure 3 ,exist Figure 3 In the two-dimensional image 3, each pixel corresponds to a grid, and a preset operation area 31 is determined in the two-dimensional image 3. The preset operation area 31 can be a polygonal operation area or a circular operation area, which is not limited here. Figure 3 In the example, the preset working area 31 is formed by connecting a plurality of vertices.

[0074] In this embodiment, a breadth-first search algorithm or a depth-first search algorithm can be used to detect the number of holes in the preset operation area, and to detect hole information such as the area of each hole and the total area of the holes. Figure 3 The preset operation area 31 includes a first hole area 301, a second hole area 302, a third hole area 303, a fourth hole area 304, and a fifth hole area 305. The first hole area 301 is a polygon, the second hole area 302 is a diamond, the third hole area 303 is a circle, the fourth hole area 304 is a triangle, and the fifth hole area 305 is a triangle. The area and shape of each hole can be different, and can also be irregular images. The area and shape are determined based on the actual search results and are not limited here. In this embodiment, the hole area of the point cloud can be determined based on the pixels in the hole area and the preset grid size.

[0075] In one embodiment, the preset operation area can be reduced according to a preset ratio to obtain the reference area. For example, the preset ratio can be set by the user or by default, for example, 70%.

[0076] In one embodiment, the selecting of a portion of the preset operating area as a reference area may include the following steps:

[0077] The preset operation area is reduced by a preset ratio to determine a reference area in the two-dimensional image.

[0078] For example, please see again Figure 3 , the preset operation area 31 is reduced by a preset ratio to obtain the reference area 32. A breadth-first search algorithm or a depth-first search algorithm can be used to detect the number of holes in the reference area 32, and to detect hole information such as the area of each hole and the total area of the holes. Figure 3 The reference region 32 includes a portion of the first hole region 301 , the second hole region 302 , the third hole region 303 , and the fifth hole region 305 .

[0079] In another embodiment, the selecting a portion of the area from the preset operation area as the reference area may include the following steps:

[0080] The midpoint of the preset operation area is determined, and a reference area of a preset area is generated based on the midpoint; or, an area corresponding to a preset selected graphic is determined from the preset operation area as the reference area.

[0081] In this way, the reference region can be obtained in multiple ways, thereby improving the flexibility of obtaining the reference region.

[0082] In this embodiment, the user may plan the scene boundary of the operation, and determine the preset operation area based on the scene boundary.

[0083] To further determine whether the point cloud data meets the void assessment requirements, the number of voids and the total area of voids in the preset working area can be judged against the preset void assessment conditions to obtain the point cloud void assessment results.

[0084] In one embodiment, the evaluating the point cloud quality of the three-dimensional point cloud data based on the first hole information and / or the second hole information includes:

[0085] When the hole area of the second hole information is greater than a first preset area threshold, and the ratio of the hole area of the second hole information to the hole area of the first hole information is greater than a preset ratio, determining that the point cloud quality of the three-dimensional point cloud data is unqualified; or

[0086] When the hole area of the first hole information is greater than a second preset area threshold, it is determined that the point cloud quality of the three-dimensional point cloud data is unqualified.

[0087] In one embodiment, to ensure the accuracy of the void assessment results, the first preset area threshold, the second preset area threshold, and the preset ratio may be determined based on experimental data, or may be set by the user. For example, the first preset area threshold is 150 square meters, the second preset area threshold is 500 square meters, and the preset ratio is 50%. The first preset area threshold, the second preset area threshold, and the preset ratio may also be other values, without limitation herein.

[0088] In this embodiment, if the point cloud quality is unqualified, it means that the point cloud data has large holes and the point cloud quality is poor, and it cannot be used as basic data for navigation data. The unqualified point cloud quality can be fed back to the user and the unqualified data can be discarded.

[0089] The point cloud hole detection method provided in this embodiment determines a coordinate origin based on three-dimensional point cloud data, establishes a relative coordinate system based on the coordinate origin, constructs a two-dimensional plane based on the first and second directions of the relative coordinate system, rasterizes the two-dimensional plane based on a preset grid size to obtain a grid map, obtains two-dimensional point cloud data projected onto the two-dimensional plane, and determines hole information in the three-dimensional point cloud data based on the positional relationship between the two-dimensional point cloud data and each grid in the grid map. In this way, hole detection for three-dimensional point cloud data is transformed into hole detection for two-dimensional images, simplifying the computational complexity of hole detection and improving the processing efficiency of detecting hole information.

[0090] Example 2

[0091] In addition, an embodiment of the present disclosure provides a device for detecting point cloud holes.

[0092] Specifically, such as Figure 4 As shown, the point cloud hole detection device 400 includes:

[0093] Processing module 401, configured to determine a coordinate origin based on the three-dimensional point cloud data, and establish a relative coordinate system based on the coordinate origin;

[0094] A construction module 402 is configured to construct a two-dimensional plane based on the first direction and the second direction of the relative coordinate system, and perform a gridding process on the two-dimensional plane based on a preset grid size to obtain a grid map;

[0095] An acquisition module 403 is configured to acquire two-dimensional point cloud data obtained by projecting the three-dimensional point cloud data onto the two-dimensional plane;

[0096] The first determining module 404 is configured to determine hole information of the three-dimensional point cloud data based on a positional relationship between the two-dimensional point cloud data and each grid in the grid map.

[0097] In one embodiment, the first determining module 404 is further configured to determine whether each grid includes at least one two-dimensional point based on the position coordinates of each two-dimensional point and the grid map position information;

[0098] When the grid includes at least one two-dimensional point, adding a first mark to the grid;

[0099] When the grid does not include the two-dimensional point, adding a second mark to the grid;

[0100] The hole information of the three-dimensional point cloud data is determined according to the grid containing the second mark in the grid map.

[0101] In one embodiment, the first determining module 404 is further configured to determine whether each grid includes at least one two-dimensional point based on the position coordinates of each two-dimensional point and the grid map position information;

[0102] When the grid includes at least one two-dimensional point, adding a first mark to the grid;

[0103] When the grid does not include the two-dimensional point, adding a second mark to the grid;

[0104] Mapping each grid of the grid image to a pixel point, and determining a first pixel value or a second pixel value of the corresponding pixel point according to the first mark or the second mark of each grid to obtain a two-dimensional image;

[0105] Cavity information of the three-dimensional point cloud data is determined based on the two-dimensional image.

[0106] In one embodiment, the first determining module 404 is further configured to obtain continuous pixel points in the two-dimensional image whose pixel values are the second pixel value;

[0107] Based on the size information of the continuous pixel points and each grid, the area information of the hole is determined.

[0108] In one embodiment, the processing module 401 is further configured to obtain a minimum coordinate value in a first direction, a minimum coordinate value in a second direction, and a minimum coordinate value in a third direction in the three-dimensional point cloud data;

[0109] The coordinate origin is determined according to the minimum coordinate value in the first direction, the minimum coordinate value in the second direction, and the minimum coordinate value in the third direction.

[0110] In one embodiment, the point cloud hole detection device 400 further includes:

[0111] A division module is used to divide the initial point cloud data into multiple three-dimensional grids according to a preset three-dimensional voxel size, determine the coordinate information and elevation information of the target three-dimensional point of the three-dimensional grid based on the multiple three-dimensional points of each three-dimensional grid, and generate the three-dimensional point cloud data based on the multiple target three-dimensional points.

[0112] In one embodiment, the first dimension and the second dimension of the preset three-dimensional voxel size are respectively the same as the first dimension and the second dimension of the preset grid size.

[0113] In one embodiment, the point cloud hole detection device 400 further includes:

[0114] The second determination module is configured to determine the point cloud quality of the three-dimensional point cloud data based on hole information of the three-dimensional point cloud data, where the hole information includes at least a hole area and a hole number.

[0115] In one embodiment, the second determining module is further configured to determine a preset operation area corresponding to the three-dimensional point cloud data, and select a portion of the preset operation area as a reference area;

[0116] Determining first cavity information within the preset operation area and second cavity information within the reference area;

[0117] The point cloud quality of the three-dimensional point cloud data is evaluated based on the first hole information and / or the second hole information.

[0118] In one embodiment, the second determining module is further configured to reduce the preset operating area by a preset ratio to determine a reference area in the two-dimensional image.

[0119] In one embodiment, the second determining module is further configured to determine that the point cloud quality of the three-dimensional point cloud data is unqualified when the hole area of the second hole information is greater than a first preset area threshold and the ratio of the hole area of the second hole information to the hole area of the first hole information is greater than a preset ratio; or

[0120] When the hole area of the first hole information is greater than a second preset area threshold, it is determined that the point cloud quality of the three-dimensional point cloud data is unqualified.

[0121] The point cloud hole detection device provided in this embodiment determines a coordinate origin based on three-dimensional point cloud data, establishes a relative coordinate system based on the coordinate origin, constructs a two-dimensional plane based on the first and second directions of the relative coordinate system, rasterizes the two-dimensional plane based on a preset grid size to obtain a grid map, obtains two-dimensional point cloud data projected onto the two-dimensional plane, and determines hole information in the three-dimensional point cloud data based on the positional relationship between the two-dimensional point cloud data and each grid in the grid map. In this way, hole detection for three-dimensional point cloud data is transformed into hole detection for two-dimensional images, simplifying the computational complexity of hole detection and improving the processing efficiency of detecting hole information.

[0122] Example 3

[0123] In addition, an embodiment of the present disclosure provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program runs on the processor, the steps of the point cloud hole detection method provided in Example 1 are executed.

[0124] It should be noted that the specific implementation steps of this embodiment can refer to the description of the corresponding content of the point cloud hole detection method in the above embodiment 1, and will not be repeated here.

[0125] Example 4

[0126] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the point cloud hole detection method described in Example 1 are implemented.

[0127] It should be noted that the specific implementation steps of this embodiment can refer to the description of the corresponding content of the point cloud hole detection method in the above embodiment 1, and will not be repeated here.

[0128] In this embodiment, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0129] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or terminal comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or terminal comprising the element.

[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0131] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A method for detecting holes in point clouds, characterized in that: The method comprises: Determine a coordinate origin based on the three-dimensional point cloud data, and establish a relative coordinate system based on the coordinate origin; constructing a two-dimensional plane based on the first direction and the second direction of the relative coordinate system, and performing a grid processing on the two-dimensional plane based on a preset grid size to obtain a grid map; Acquire two-dimensional point cloud data obtained by projecting the three-dimensional point cloud data onto the two-dimensional plane; Determining hole information of the three-dimensional point cloud data based on a positional relationship between the two-dimensional point cloud data and each grid in the grid map; The determining of the hole information of the three-dimensional point cloud data based on the positional relationship between the two-dimensional point cloud data and each grid in the grid map includes: Determining whether each grid includes at least one two-dimensional point based on the position coordinates of each two-dimensional point and the grid map position information; When the grid includes at least one two-dimensional point, adding a first mark to the grid; When the grid does not include the two-dimensional point, adding a second mark to the grid; The hole information of the three-dimensional point cloud data is determined according to the grid containing the second mark in the grid map.

2. The method according to claim 1, characterized in that The determining of the hole information of the three-dimensional point cloud data based on the positional relationship between the two-dimensional point cloud data and each grid in the grid map includes: Determining whether each grid includes at least one two-dimensional point based on the position coordinates of each two-dimensional point and the grid map position information; When the grid includes at least one two-dimensional point, adding a first mark to the grid; When the grid does not include the two-dimensional point, adding a second mark to the grid; Mapping each grid of the grid image to a pixel point, and determining a first pixel value or a second pixel value of the corresponding pixel point according to the first mark or the second mark of each grid to obtain a two-dimensional image; Cavity information of the three-dimensional point cloud data is determined based on the two-dimensional image.

3. The method according to claim 2, characterized in that The hole information includes area information of the hole, and determining the hole information based on the two-dimensional image includes: Acquire continuous pixel points whose pixel values are second pixel values in the two-dimensional image; Based on the size information of the continuous pixel points and each grid, the area information of the hole is determined.

4. The method according to claim 1, wherein Determining the coordinate origin based on the three-dimensional point cloud data includes: Obtaining a minimum coordinate value in a first direction, a minimum coordinate value in a second direction, and a minimum coordinate value in a third direction in the three-dimensional point cloud data; The coordinate origin is determined according to the minimum coordinate value in the first direction, the minimum coordinate value in the second direction, and the minimum coordinate value in the third direction.

5. The method according to claim 1, wherein The acquisition of the three-dimensional point cloud data includes: The initial point cloud data is divided into multiple three-dimensional grids according to a preset three-dimensional voxel size, the coordinate information and elevation information of the target three-dimensional point of the three-dimensional grid are determined based on the multiple three-dimensional points of each three-dimensional grid, and the three-dimensional point cloud data is generated based on the multiple target three-dimensional points.

6. The method according to claim 5, characterized in that The first dimension and the second dimension of the preset three-dimensional voxel size are respectively the same as the first dimension and the second dimension of the preset grid size.

7. The method according to claim 3, characterized in that The method further comprises: The point cloud quality of the three-dimensional point cloud data is determined based on the hole information of the three-dimensional point cloud data, where the hole information includes at least the hole area and the number of holes.

8. The method according to claim 7, characterized in that The determining the point cloud quality of the three-dimensional point cloud data based on the hole information of the three-dimensional point cloud data includes: Determining a preset operating area corresponding to the three-dimensional point cloud data, and selecting a portion of the preset operating area as a reference area; Determining first cavity information within the preset operation area and second cavity information within the reference area; The point cloud quality of the three-dimensional point cloud data is evaluated based on the first hole information and / or the second hole information.

9. The method according to claim 8, characterized in that The selecting a portion of the area from the preset operation area as a reference area includes: The preset operation area is reduced by a preset ratio to determine a reference area in the two-dimensional image.

10. The method according to claim 8, characterized in that The evaluating the point cloud quality of the three-dimensional point cloud data based on the first hole information and / or the second hole information includes: When the hole area of the second hole information is greater than a first preset area threshold, and the ratio of the hole area of the second hole information to the hole area of the first hole information is greater than a preset ratio, determining that the point cloud quality of the three-dimensional point cloud data is unqualified; or When the hole area of the first hole information is greater than a second preset area threshold, it is determined that the point cloud quality of the three-dimensional point cloud data is unqualified.

11. A point cloud hole detection device, characterized in that: The device comprises: a processing module, configured to determine a coordinate origin based on the three-dimensional point cloud data, and establish a relative coordinate system based on the coordinate origin; a construction module, configured to construct a two-dimensional plane based on the first direction and the second direction of the relative coordinate system, and perform a gridding process on the two-dimensional plane based on a preset grid size to obtain a grid map; An acquisition module, configured to acquire two-dimensional point cloud data obtained by projecting the three-dimensional point cloud data onto the two-dimensional plane; A first determining module is configured to determine hole information of the three-dimensional point cloud data based on a positional relationship between the two-dimensional point cloud data and each grid in the grid map; The first determining module is further configured to determine whether each grid includes at least one two-dimensional point based on the position coordinates of each two-dimensional point and the grid map position information; When the grid includes at least one two-dimensional point, adding a first mark to the grid; When the grid does not include the two-dimensional point, adding a second mark to the grid; The hole information of the three-dimensional point cloud data is determined according to the grid containing the second mark in the grid map.

12. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory is used to store a computer program, and when the computer program is run on the processor, the method for detecting point cloud holes according to any one of claims 1 to 10 is executed.

13. A computer-readable storage medium, characterized in that The device stores a computer program, which, when running on a processor, executes the point cloud hole detection method according to any one of claims 1 to 10.

Images