Material pile surface reconstruction method

Abstract

The embodiment of the invention provides a material pile surface reconstruction method, and the method comprises the steps: obtaining a plurality of original scanning points obtained by scanning a material pile through a radar, carrying out the filtering and noise reduction processing of the plurality of original scanning points, and obtaining first point cloud data; generating second point cloud data based on a plurality of first scanning points in the first point cloud data and the radius of the material bin barrel, and determining a set of the first point cloud data and the second point cloud data as third point cloud data; based on a third preset value, a plurality of third scanning points in the third point cloud data and the diameter of the material bin barrel, determining a to-be-generated coordinate, and when the to-be-generated coordinate meets a preset generation condition, generating a plurality of fourth scanning points based on the to-be-generated coordinate; performing fitting processing on the plurality of fourth scanning points to obtain target point cloud data; in this way, interference data and part of noisy point data in the original point cloud data can be eliminated, the point cloud data can be supplemented, and the precision of material pile surface reconstruction is improved.

Description

Technical Field

[0001] The invention relates to the field of material pile survey, and in particular to a material pile surface reconstruction method. Background Art

[0002] Millimeter wave radar is a radar that works in the millimeter wave band. Millimeter wave refers to the 30-300GHz frequency domain (wavelength is 1-10mm). Among them, millimeter wave radar can detect and locate objects by emitting electromagnetic waves and receiving reflected signals. Due to the advantages of millimeter wave radar such as high resolution and strong anti-interference ability, millimeter wave radar is widely used in military, aviation, aerospace and other fields.

[0003] In a warehousing scenario, a millimeter-wave radar can be installed on the top of a material bin to scan the surface of the material pile to form three-dimensional spatial data, that is, to complete the surface reconstruction of the material pile. However, in actual applications, millimeter-wave radar has problems with insufficient target recognition and tracking accuracy, and millimeter-wave radar is easily affected by atmospheric conditions (such as humidity, temperature, and wind speed), resulting in more interference data and noise when the millimeter-wave radar scans the surface of the material pile, which in turn affects the accuracy of subsequent surface reconstruction. Summary of the invention

[0004] The present application provides a material pile surface reconstruction method, which can eliminate interference data and part of noise data in original point cloud data, and can supplement point cloud data to improve the accuracy of material pile surface reconstruction.

[0005] In a first aspect of the present application, a method for reconstructing the surface of a material pile is provided, comprising:

[0006] Obtain scanning data corresponding to each of a plurality of original scanning points obtained by scanning the material pile by the radar and the three-dimensional spatial coordinates of each of the plurality of original scanning points in a three-dimensional spatial coordinate system; wherein the origin of the three-dimensional spatial coordinate system is the radar, the abscissa axis extends along the radius direction of the material bin barrel, the ordinate axis and the abscissa axis are located in the same horizontal plane and extend in a direction perpendicular to the abscissa axis, and the ordinate axis extends along the height direction of the material bin barrel in a direction away from the opening of the material bin barrel;

[0007] Based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, filtering and denoising the multiple original scanning points to obtain first point cloud data;

[0008] Based on the three-dimensional spatial coordinates corresponding to each of the plurality of first scanning points in the first point cloud data and the radius of the material bin, second point cloud data are generated, and a collection of the first point cloud data and the second point cloud data is determined as third point cloud data; the plurality of second scanning points in the second point cloud data are all data points at the junction of the material pile and the material bin;

[0009] Based on the third preset value, the three-dimensional space coordinates corresponding to each of the plurality of third scanning points in the third point cloud data and the diameter of the material bin, the coordinates to be generated are determined, and when the coordinates to be generated meet the preset generation conditions, fourth point cloud data are generated based on the coordinates to be generated;

[0010] Based on the plurality of third scanning points, fitting processing is performed on the plurality of fourth scanning points in the fourth point cloud data respectively to update the fourth point cloud data, and a set of the updated fourth point cloud data and the third point cloud data is determined as the target point cloud data.

[0011] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, filtering and denoising are performed on the multiple original scanning points to obtain first point cloud data, including:

[0012] Based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, a data filtering process is performed on the multiple original scanning points to obtain fifth point cloud data;

[0013] Based on the three-dimensional space coordinates corresponding to each of the plurality of fifth scanning points in the fifth point cloud data, as well as the first preset value and the second preset value, the plurality of fifth scanning points are subjected to at least one noise reduction process to obtain the first point cloud data.

[0014] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, data filtering processing is performed on multiple original scanning points to obtain fifth point cloud data, including:

[0015] Based on the three-dimensional spatial coordinates corresponding to each original scanning point, determine the positional relationship between the original scanning point and the material bin;

[0016] Based on the positional relationship between the original scanning point and the material bin barrel, and the scanning data corresponding to each original scanning point, data filtering processing is performed on the multiple original scanning points to obtain the fifth point cloud data.

[0017] In some embodiments, the scan data corresponding to the original scan point includes the speed of the original scan point and the signal-to-noise ratio of the original scan point;

[0018] Based on the positional relationship between the original scanning point and the material bin, and the scanning data corresponding to each original scanning point, multiple original scanning points are filtered to obtain the fifth point cloud data, including:

[0019] If the positional relationship between the original scanning point and the material silo indicates that the original scanning point is located inside the material silo, and the speed of the original scanning point is the first preset speed threshold, and the signal-to-noise ratio of the original scanning point is within the preset signal-to-noise ratio range, then the original scanning point is determined as the fifth scanning point, thereby obtaining the fifth point cloud data.

[0020] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each of the plurality of fifth scanning points in the fifth point cloud data, as well as the first preset value and the second preset value, performing at least one noise reduction process on the plurality of fifth scanning points to obtain the first point cloud data includes:

[0021] Based on the three-dimensional space coordinates corresponding to each fifth scanning point and the first preset value, performing a primary noise reduction process on the plurality of fifth scanning points to obtain a plurality of candidate first scanning points;

[0022] Based on the three-dimensional space coordinates corresponding to each candidate first scanning point and the second preset value, a secondary noise reduction process is performed on the plurality of candidate first scanning points to obtain first point cloud data.

[0023] In some embodiments, based on the three-dimensional space coordinates corresponding to each fifth scanning point and the first preset value, a plurality of fifth scanning points are subjected to initial noise reduction processing to obtain a plurality of candidate first scanning points, including:

[0024] Constructing a cube based on the three-dimensional space coordinates corresponding to each fifth scanning point, and dividing the cube based on the first preset value to obtain a plurality of cubes;

[0025] Based on the multiple cubes, a three-dimensional array is constructed, and based on the three-dimensional spatial coordinates corresponding to the fifth scanning point and the first preset value, index information of each fifth scanning point is determined; the index information of the fifth scanning point represents the positional relationship between the fifth scanning point and each cube;

[0026] storing the plurality of fifth scanning points into a plurality of elements in the three-dimensional array according to the index information of each fifth scanning point;

[0027] Each element in the three-dimensional array is traversed, and the fifth scanning point with the highest height among each element is determined as a candidate first scanning point.

[0028] In some embodiments, based on the three-dimensional space coordinates corresponding to each candidate first scanning point and the second preset value, a secondary noise reduction process is performed on the plurality of candidate first scanning points to obtain first point cloud data, including:

[0029] Each candidate first scanning point is determined as a target candidate first scanning point in turn. If there is a candidate first scanning point whose height is lower than the target candidate first scanning point, and the distance between the candidate first scanning point and the target candidate second scanning point is less than a second preset value, the target candidate first scanning point is deleted from the multiple candidate first scanning points to obtain first point cloud data.

[0030] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each of the plurality of first scanning points in the first point cloud data and the radius of the material bin, generating the second point cloud data includes:

[0031] Based on the three-dimensional spatial coordinates corresponding to each first scanning point, determining a scanning point among the plurality of first scanning points that is farthest from the vertical coordinate axis of the three-dimensional spatial coordinate system as a reference scanning point;

[0032] Based on the vertical coordinate value of the reference scanning point and the radius of the material bin barrel, as well as the preset incremental angle, a plurality of second scanning points are generated to obtain second point cloud data.

[0033] In some embodiments, based on the third preset value, and the three-dimensional space coordinates corresponding to each of the plurality of third scanning points in the third point cloud data and the diameter of the material bin, determining the coordinates to be generated includes:

[0034] Based on the diameter of the material bin barrel, a point cloud generation area is constructed, and based on the third preset value, the point cloud generation area is divided to obtain a plurality of sub-areas; the point cloud generation area is parallel to the horizontal plane;

[0035] Based on the multiple sub-regions, sixth point cloud data is constructed, and based on the third preset value and the three-dimensional space coordinates corresponding to the multiple third scanning points, index information of each third scanning point is determined; the index information of the third scanning point represents the positional relationship between the third scanning point and each sub-region;

[0036] According to the index information of each third scanning point, multiple third scanning points are stored in the sixth point cloud data, and the first index point in the sixth point cloud data is traversed; the first index point represents any area in the multiple sub-areas corresponding to the sixth point cloud data set;

[0037] If the third scanning point does not exist for the first index point, the horizontal axis value and the vertical axis value of the center point of the sub-area corresponding to the first index point and the vertical coordinate value of the reference scanning point are determined as the coordinates to be generated.

[0038] In some embodiments, the preset generation condition includes: the sum of the squares of the horizontal axis value and the vertical axis value in the coordinate to be generated is less than or equal to the square of the radius of the material bin barrel;

[0039] Based on the plurality of third scanning points, fitting processing is performed on the plurality of fourth scanning points in the fourth point cloud data respectively to update the fourth point cloud data, including:

[0040] Determine a first distance between each third scanning point and each fourth scanning point in a plane formed by a horizontal axis and a vertical axis in a three-dimensional space coordinate system, and select a number of third scanning points to be processed equal to a preset number from the plurality of third scanning points in order of the first distance from near to far;

[0041] The vertical coordinate value of each fourth scanning point is fitted using the vertical coordinate value of the third scanning point to be processed, so as to update the fourth point cloud data.

[0042] In a second aspect of the present application, a material pile surface reconstruction device is provided, comprising:

[0043] An acquisition module is used to acquire scanning data corresponding to each of a plurality of original scanning points obtained by scanning a material pile by a radar and the three-dimensional spatial coordinates of each of the plurality of original scanning points in a three-dimensional spatial coordinate system; wherein the origin of the three-dimensional spatial coordinate system is the radar, the horizontal axis extends along the radius direction of the material bin barrel, the vertical axis is located in the same horizontal plane as the horizontal axis and extends in a direction perpendicular to the horizontal axis, and the vertical axis extends along the height direction of the material bin barrel in a direction away from the opening of the material bin barrel;

[0044] A processing module, for filtering and denoising a plurality of original scanning points based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, so as to obtain first point cloud data; and for determining a set of the first point cloud data and the second point cloud data as third point cloud data; the plurality of second scanning points in the second point cloud data are all data points at the junction of the material pile and the material bin; and for determining coordinates to be generated based on a third preset value, and the three-dimensional spatial coordinates corresponding to each of the plurality of third scanning points in the third point cloud data and the diameter of the material bin; and for fitting a plurality of fourth scanning points in the fourth point cloud data based on the plurality of third scanning points, so as to update the fourth point cloud data, and determining the set of the updated fourth point cloud data and the third point cloud data as target point cloud data;

[0045] A generation module is used to generate second point cloud data based on the three-dimensional spatial coordinates corresponding to each of the multiple first scanning points in the first point cloud data and the radius of the material bin barrel; and is also used to generate fourth point cloud data based on the coordinates to be generated when the coordinates to be generated meet preset generation conditions.

[0046] In a third aspect of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in any one of the above embodiments when executing the program.

[0047] In a fourth aspect of the present application, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0048] The embodiment of the present application provides a method for reconstructing the surface of a material pile, comprising obtaining scanning data corresponding to each of a plurality of original scanning points obtained by scanning the material pile by a radar and the three-dimensional spatial coordinates of each of the plurality of original scanning points in a three-dimensional spatial coordinate system; wherein the origin of the three-dimensional spatial coordinate system is the radar, the horizontal axis extends along the radius direction of a material bin barrel, the vertical axis and the horizontal axis are located in the same horizontal plane and extend in a direction perpendicular to the horizontal axis, and the vertical axis extends along the height direction of the material bin barrel in a direction away from an opening of the material bin barrel; based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, filtering and denoising the plurality of original scanning points to obtain first point cloud data; based on the three-dimensional spatial coordinates corresponding to each of the plurality of first scanning points in the first point cloud data and the radius of the material bin barrel, generating second point cloud data, and determining the collection of the first point cloud data and the second point cloud data as third point cloud data; the second point cloud data The multiple second scanning points in the cloud data are all data points at the junction of the material pile and the material silo; based on the third preset value, and the three-dimensional spatial coordinates corresponding to the multiple third scanning points in the third point cloud data and the diameter of the material silo, the coordinates to be generated are determined, and when the coordinates to be generated meet the preset generation conditions, the fourth point cloud data is generated based on the coordinates to be generated; based on the multiple third scanning points, the multiple fourth scanning points in the fourth point cloud data are respectively fitted to update the fourth point cloud data, and the updated fourth point cloud data and the set of the third point cloud data are determined as the target point cloud data; in this way, the interference data and part of the noise data in the multiple original scanning points can be eliminated, and the point cloud data can be supplemented to improve the accuracy of the reconstruction of the material pile surface; in addition, by setting multiple parameters (such as the first preset value, the second preset value, and the third preset value), the precision of the recognition and reconstruction of the surfaces of various material piles under different environments can also be controlled. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0050] Figure 1 A schematic diagram of an application scenario of a material pile surface reconstruction method provided in an embodiment of the present application;

[0051] Figure 2A schematic diagram of a process of a material pile surface reconstruction method provided in an embodiment of the present application;

[0052] Figure 3 A stereogram of the original point cloud data in the material pile surface reconstruction method provided in an embodiment of the present application;

[0053] Figure 4 A stereogram of the first point cloud data in the material pile surface reconstruction method provided in an embodiment of the present application;

[0054] Figure 5 A stereogram of target point cloud data in the material pile surface reconstruction method provided in an embodiment of the present application;

[0055] Figure 6 A schematic diagram of the structure of a material pile surface reconstruction device provided in an embodiment of the present application;

[0056] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application.

[0058] In a warehousing scenario, a millimeter-wave radar can be set on the top of a material silo. The millimeter-wave radar is used to scan the surface of the material pile in the silo to obtain a point cloud, and three-dimensional spatial data is formed based on the point cloud, thereby completing the surface reconstruction of the material pile.

[0059] Among them, millimeter wave radar is a radar that works in the millimeter wave band. Millimeter wave refers to the 30-300GHz frequency domain (wavelength is 1-10mm), that is, the wavelength of millimeter wave is between microwave and centimeter wave. Therefore, millimeter wave radar has some advantages of microwave radar and photoelectric radar, such as high resolution, small size, strong anti-interference ability and good penetration. These advantages make it widely used in military, aviation, aerospace and other fields. In practical applications, millimeter wave radar can detect and locate objects by emitting electromagnetic waves and receiving reflected signals.

[0060] Point cloud is a data set composed of points in three-dimensional space, and the positions of these points in three-dimensional space can express the surface characteristics of the target object. In some embodiments, the point data set of the product appearance surface acquired by the measuring instrument is also called point cloud. According to the different acquisition devices (also called measuring instruments), point clouds can be divided into two categories: sparse point clouds and dense point clouds. For example, if the number of points acquired by the three-dimensional coordinate measuring machine is small and the distance between points is large, a sparse point cloud is formed; if a three-dimensional laser scanner or a photographic scanner is used to obtain a large number of points, the points are dense, that is, the distance between points is small, then a dense point cloud is formed.

[0061] A material pile refers to the storage of mined materials (such as coal, ore and waste rock) on the surface of a mine for a longer period of time to balance the fluctuations in fuel supply and demand, or for long-term storage. In some embodiments, the material pile includes not only a pile of materials formed after the mined materials are stored in the mining industry, but also piles of coal, mineral materials, grain storage, earthwork, etc. The embodiments of the present application are not limited to this. Among them, the cross-sectional shape of the material pile can be a herringbone or a trapezoid. Material silos are used to store various materials, that is, the materials are stored in the material silos to form a material pile.

[0062] In the process of scanning the surface of the material pile with millimeter-wave radar to obtain a point cloud and reconstructing the surface of the material pile based on the point cloud, due to the problem of insufficient target recognition and tracking accuracy of millimeter-wave radar, and the millimeter-wave radar is easily affected by atmospheric conditions (such as humidity, temperature and wind speed), the millimeter-wave radar will generate more interference data and noise when scanning the surface of the material pile, which in turn affects the accuracy of subsequent surface reconstruction.

[0063] In order to solve the above technical problems, the present application provides a material pile surface reconstruction method, which can eliminate interference data and some noise data in the original point cloud data, and can generate new scanning points to supplement the point cloud data, thereby improving the accuracy of material pile surface reconstruction.

[0064] Before describing the technical solution of the embodiments of the present application, the application scenarios of the embodiments of the present application are first described with reference to the accompanying drawings. Figure 1 Schematic diagram of application scenario of the tally method provided in the embodiment of the present application. Figure 1 As shown, the application scenario includes a radar 1 (also called a millimeter wave radar), a material pile 2 and a material silo 3.

[0065] In some embodiments, the material pile 2 is stored in a material silo barrel 3, and the radar 1 is arranged directly above the material silo barrel to emit transmission waves toward the surface of the material pile 2 in the material silo barrel 3 to scan the surface of the material pile 2 and receive reflected waves to obtain corresponding point cloud data.

[0066] based on Figure 1 The application scenario shown, Figure 2 A schematic flow chart of a material pile surface reconstruction method is provided. Figure 2 , an embodiment of the present application provides a material pile surface reconstruction method, including S201-S205.

[0067] S201, obtaining scanning data corresponding to each of a plurality of original scanning points obtained by scanning a material pile by a radar and the three-dimensional spatial coordinates of each of the plurality of original scanning points in a three-dimensional spatial coordinate system.

[0068] Among them, the origin of the three-dimensional space coordinate system is the radar, the horizontal axis extends along the radius direction of the material silo, the vertical axis and the horizontal axis are located in the same horizontal plane and extend in a direction perpendicular to the horizontal axis, and the vertical axis extends along the height direction of the material silo away from the opening of the material silo.

[0069] In some embodiments, the original scanning point is a scanning point obtained by scanning the material pile by radar. Each scanning data corresponds to the relevant information of an original scanning point. The scanning data corresponding to the original scanning point may include the distance from the original scanning point to the radar, the azimuth formed between the original scanning point and the radar, the elevation formed between the original scanning point and the radar, and the scanning data corresponding to the original scanning point also includes the speed of the original scanning point and the signal-to-noise ratio of the original scanning point. The embodiment of the present application does not limit the content of the scanning data corresponding to the original scanning point.

[0070] All original scanning points obtained by radar scanning are located in polar coordinates. In order to facilitate the subsequent reconstruction of the material pile surface, the original scanning points can be converted from polar coordinates to Figure 1 In the three-dimensional space coordinate system shown.

[0071] In some embodiments, a three-dimensional spatial coordinate system is constructed based on the radar and the material bin barrels corresponding to the material pile.

[0072] For example, Figure 1 As shown, the radar is determined as the coordinate origin, the direction parallel to the height direction of the material bin barrel and extending in the direction away from the opening of the material bin barrel is determined as the y-axis (also called the vertical coordinate axis) direction, the direction perpendicular to the y-axis and parallel to the radius direction of the material bin barrel is determined as the x-axis (also called the horizontal coordinate axis) direction, and the direction located in the same horizontal plane as the x-axis and extending in the direction perpendicular to the x-axis is determined as the z-axis (also called the vertical coordinate axis) to establish three-dimensional space coordinates.

[0073] In some embodiments, the three-dimensional spatial coordinates of each original scanning point in the three-dimensional spatial coordinate system can be determined based on the distance from the original scanning point on the surface of the material pile to the radar, the elevation angle formed by the original scanning point on the surface of the material pile and the radar, the azimuth angle formed by the original scanning point on the surface of the material pile and the radar, and a preset polar coordinate conversion formula.

[0074] For example, let the three-dimensional space coordinates be (x, y, z) and the polar coordinates be The formula for converting polar coordinates to three-dimensional coordinates is shown in Formula 1-1:

[0075]

[0076] Where x is the horizontal coordinate of the three-dimensional space coordinate system, y is the vertical coordinate of the three-dimensional space coordinate system, and z is the ordinate of the three-dimensional space coordinate system; R is the distance from the original scanning point on the surface of the material pile to the radar, and θ is the elevation angle between the original scanning point on the surface of the material pile and the radar. It is the azimuth formed by the original scanning point on the surface of the material pile and the radar.

[0077] It should be noted that, since the formula for converting polar coordinates to three-dimensional space coordinates is an existing technology, the method for converting each original scanning point from the polar coordinate system to the three-dimensional space coordinate system is not described in detail here.

[0078] S202 , based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, filter and reduce noise on the multiple original scanning points to obtain first point cloud data.

[0079] In some embodiments, the first point cloud data includes multiple first scanning points, wherein the first scanning points refer to the scanning points remaining after the multiple original scanning points complete filtering and noise reduction processing, and the first point cloud data is the point cloud data after the filtering and noise reduction processing, and the filtering and noise reduction processing includes data filtering processing and noise reduction processing.

[0080] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, filtering and denoising are performed on multiple original scanning points to obtain first point cloud data, including: first, based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, data filtering is performed on multiple original scanning points to obtain fifth point cloud data; then, based on the three-dimensional spatial coordinates corresponding to each of the multiple fifth scanning points in the fifth point cloud data, as well as the first preset value and the second preset value, denoising is performed on the multiple fifth scanning points at least once to obtain first point cloud data.

[0081] In some embodiments, the fifth point cloud data includes a plurality of fifth scanning points, and the fifth scanning points are the scanning points remaining after data filtering processing is performed on the plurality of original scanning points, that is, the fifth point cloud data is the point cloud data obtained after data filtering processing is performed on the original point cloud data composed of the plurality of original scanning points, and the relationship between the fifth point cloud data and the original point cloud data is that the original point cloud data includes the fifth point cloud data. The first point cloud data is the point cloud data obtained after at least one noise reduction processing is performed on the fifth point cloud data, that is, the relationship between the first point cloud data and the fifth point cloud data is that the fifth point cloud data includes the first point cloud data.

[0082] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, data filtering processing is performed on multiple original scanning points to obtain fifth point cloud data, which may include: first, based on the three-dimensional spatial coordinates corresponding to each original scanning point, determining the positional relationship between the original scanning point and the material bin; then, based on the positional relationship between the original scanning point and the material bin, and the scanning data corresponding to each original scanning point, data filtering processing is performed on multiple original scanning points to obtain fifth point cloud data.

[0083] In some embodiments, each original scanning point in the original point cloud data is traversed, and the position of the original scanning point is determined according to the three-dimensional spatial coordinates corresponding to each original scanning point; then, according to the position of the original scanning point, the positional relationship between the original scanning point and the material bin is determined.

[0084] Exemplarily, the positional relationship between the original scanning point and the material bin barrel includes at least one of: the original scanning point is located inside the material bin barrel; and the original scanning point is located outside the material bin barrel.

[0085] In some embodiments, based on the positional relationship between the original scanning point and the material bin barrel, and the scanning data corresponding to each original scanning point, data filtering processing is performed on multiple original scanning points to obtain fifth point cloud data, including: if the positional relationship between the original scanning point and the material bin barrel indicates that the original scanning point is located in the material bin barrel, and the speed of the original scanning point is a first preset speed threshold, and the signal-to-noise ratio of the original scanning point is within the preset signal-to-noise ratio range, then the original scanning point is determined as the fifth scanning point, thereby obtaining the fifth point cloud data.

[0086] The original scanning point being located in the material bin means that the x, y and z in the three-dimensional space coordinates corresponding to the original scanning point are all located in the material bin. The fifth scanning point is an original scanning point located in the material bin among the multiple original scanning points, and the speed is the first preset speed threshold, and the signal-to-noise ratio is within the preset signal-to-noise ratio range.

[0087] In some embodiments, the first preset speed threshold is a preset value, for example, the first preset speed threshold may be 0. The preset signal-to-noise ratio interval may be determined according to actual needs. The embodiment of the present application does not limit the first preset speed threshold and the preset signal-to-noise ratio interval.

[0088] Exemplarily, if the absolute value of y (also called the vertical coordinate value) in the three-dimensional space coordinate corresponding to the original scanning point is smaller than the absolute value of y corresponding to the bottom wall of the material silo, the absolute value of x (also called the horizontal coordinate value) is smaller than the absolute value of x corresponding to the side wall of the material silo, and the absolute value of z (also called the vertical coordinate value) is smaller than the absolute value of z corresponding to the side wall of the material silo, and the speed of the original scanning point is 0, and the signal-to-noise ratio of the original scanning point is within the preset signal-to-noise ratio range, then it is determined that the original scanning point is located in the material silo, and the original scanning point is determined as the fifth scanning point.

[0089] In some embodiments, after traversing a plurality of original scanning points, a plurality of fifth scanning points are obtained, and fifth point cloud data can be obtained based on the plurality of fifth scanning points.

[0090] In some embodiments, it is possible to simultaneously detect whether the three-dimensional spatial coordinates corresponding to the original scanning point are located in the material bin, whether the speed of the original scanning point is a first preset value, and whether the signal-to-noise ratio of the original scanning point is within a preset signal-to-noise ratio range; it is also possible to detect in sequence whether the three-dimensional spatial coordinates corresponding to the original scanning point are located in the material bin, whether the speed of the original scanning point is a first preset value, and whether the signal-to-noise ratio of the original scanning point is within a preset signal-to-noise ratio range. During the sequential detection process, if the current detection result of the original scanning point is yes, the original scanning point continues to be detected; if the current detection result of the original scanning point is no, the next original scanning point is directly detected.

[0091] Exemplarily, it is possible to first detect whether the three-dimensional space coordinates corresponding to the original scanning point are located in the material bin barrel. If the three-dimensional space coordinates corresponding to the original scanning point are not located in the material bin barrel, then detect whether the three-dimensional space coordinates corresponding to the next original scanning point are located in the material bin barrel; if the three-dimensional space coordinates corresponding to the original scanning point are located in the material bin barrel, then detect whether the speed of the original scanning point is the first preset value. If the speed of the original scanning point is not the first preset value, then detect whether the three-dimensional space coordinates corresponding to the next original scanning point are located in the material bin barrel; if the speed of the original scanning point is the first preset value, then detect whether the signal-to-noise ratio of the original scanning point is within the preset signal-to-noise ratio interval. If the signal-to-noise ratio of the original scanning point is not within the preset signal-to-noise ratio interval, then detect whether the three-dimensional space coordinates corresponding to the next original scanning point are located in the material bin barrel; if the signal-to-noise ratio of the original scanning point is within the preset signal-to-noise ratio interval, then determine that the original scanning point is the fifth scanning point. In this way, the calculation time can be shortened and the speed of material pile surface construction can be improved.

[0092] It can be understood that the multiple fifth scanning points in the fifth point cloud data are valid point clouds obtained after filtering the multiple original scanning points. When the multiple fifth scanning points are subsequently processed to reconstruct the surface of the material pile, the reconstruction accuracy of the material pile surface can be improved.

[0093] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each of the multiple fifth scanning points in the fifth point cloud data, as well as the first preset value and the second preset value, the multiple fifth scanning points are subjected to at least one noise reduction process to obtain the first point cloud data, including: first, based on the three-dimensional spatial coordinates corresponding to each of the fifth scanning points, as well as the first preset value, the multiple fifth scanning points are subjected to initial noise reduction process to obtain multiple candidate first scanning points; then, based on the three-dimensional spatial coordinates corresponding to each of the candidate first scanning points, as well as the second preset value, the multiple candidate first scanning points are subjected to secondary noise reduction process to obtain the first point cloud data.

[0094] Among them, the first preset value and the second preset value are both preset values, which can be determined according to actual needs, and the embodiments of the present application do not limit this.

[0095] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each fifth scanning point and a first preset value, a plurality of fifth scanning points are initially denoised to obtain a plurality of candidate first scanning points, including: first, based on the three-dimensional spatial coordinates corresponding to each fifth scanning point, a quadrilateral is constructed, and based on the first preset value, the quadrilateral is divided to obtain a plurality of cubes; then, based on the plurality of cubes, a three-dimensional array is constructed, and based on the three-dimensional spatial coordinates corresponding to the fifth scanning point and the first preset value, index information of each fifth scanning point is determined; then, based on the index information of each fifth scanning point, the plurality of fifth scanning points are stored in a plurality of elements in the three-dimensional array; finally, each element in the three-dimensional array is traversed, and the fifth scanning point with the highest height in each element is determined as a candidate first scanning point.

[0096] In some embodiments, each fifth scanning point in the fifth point cloud data can be traversed, and the maximum and minimum values ​​of x, y, and z in the fifth point cloud data can be determined according to the three-dimensional space coordinates corresponding to each fifth scanning point; then, a cube is constructed using the maximum and minimum values ​​of x, y, and z in the fifth point cloud data.

[0097] Exemplarily, multiple fifth scanning points in the fifth point cloud data are traversed to determine the maximum value x of x max and the minimum value x min , the maximum value of y max and the minimum value y min The maximum value of z max and the minimum value z min ; After that, it is easy to construct a cube based on the maximum and minimum values ​​of x, y, and z, so that the 8 vertices of the cube are: min ,y min ,z min )、(x min ,y min ,z max )、(x max ,y min ,z min )、(x max ,y min ,z max )、(x min ,y max ,z min )、(x min ,y max ,z max )、(x max ,y max ,z min )、(x max ,y max ,z max); finally, the quadrilateral is divided into a plurality of cubes of equal size based on the first preset value. For example, if the first preset value is w, the quadrilateral is divided into a plurality of cubes with a side length of w.

[0098] In some embodiments, the three-dimensional array is used to store scan points located in each cube; the multiple dimensions in the three-dimensional array include row dimension, column dimension and vertical dimension. Each dimension in the three-dimensional array includes multiple elements, and each element corresponds to a cube.

[0099] Exemplarily, the three-dimensional array may be M[i][j][k], where the column dimension in the three-dimensional array is i=(y max -y min ) / w+((y max -y min )%w=0?0:1), the row dimension is j=(x max -x min ) / w+((x max -x min )%w=0?0:1), the vertical dimension is k=(z max -z min ) / w+((z max -z min )%w=0?0:1).

[0100] In some embodiments, the index information of the fifth scanning point represents the positional relationship between the fifth scanning point and each cube. The positional relationship between the fifth scanning point and each cube (also referred to as an element) includes whether the fifth scanning point is located in each cube, and in which of the multiple cubes the fifth scanning point is located, that is, the index information of the fifth scanning point can be used to determine whether the fifth scanning point is located in the cube and in which of the multiple cubes the fifth scanning point is located.

[0101] Exemplarily, if the three-dimensional spatial coordinates of the fifth scanning point are (x, y, z), and the first preset value is w, the index information of the fifth scanning point may be i_n=(yy min ) / w or j_n=(xx min ) / w or k_n=(zz min ) / w; that is, the vertical coordinate y and the minimum vertical coordinate y in the three-dimensional space coordinates of the fifth scanning point can be determined first min Then, by determining the vertical coordinate y and the minimum vertical coordinate y min The quotient between the difference between the first preset value and the first preset value determines whether the fifth scanning point is located in the element corresponding to the column dimension i. Where n is the serial number of the fifth scanning point. Similarly, the horizontal coordinate x of the fifth scanning point is compared with the minimum horizontal coordinate x minThe quotient between the difference between the first preset value and the fifth scanning point can determine whether the fifth scanning point is located in the element corresponding to the row dimension j. min The quotient of the difference between and the first preset value can determine whether the fifth scanning point is located in the element corresponding to the vertical dimension k.

[0102] In some embodiments, if the index information of the fifth scanning point indicates that the fifth scanning point is located in a first cube, and the first cube is an element in a row dimension in a three-dimensional array, the fifth scanning point is stored in the row dimension i of the three-dimensional array.

[0103] In some embodiments, the candidate first scanning points are the scanning points remaining after the multiple fifth scanning points are subjected to the initial noise reduction processing. It should be noted that by performing the initial noise reduction processing on the multiple fifth scanning points, the noise points in the fifth point cloud data can be removed, and when the surface of the material pile is subsequently reconstructed based on the multiple candidate first scanning points obtained after the initial noise reduction processing on the multiple fifth scanning points, the reconstruction accuracy can be improved.

[0104] Exemplarily, the elements corresponding to the row dimension, the elements corresponding to the column dimension, and the elements corresponding to the vertical dimension in the three-dimensional array can be traversed in sequence based on the horizontal axis direction, the vertical axis direction, and the vertical axis direction in the three-dimensional space coordinates, thereby completing the traversal of multiple elements in the three-dimensional array. For example, the column dimension Mp[][][i] of the p-level in the three-dimensional array can be traversed along the y-axis (also called the vertical axis). If there is no fifth scanning point in Mp[][][i], the column dimension of the next level below the p-level in the three-dimensional array is traversed along the x-axis (also called the horizontal axis). If there is no fifth scanning point in the column dimensions of multiple levels in the three-dimensional array, the row dimension Mp[][j][] of the p-level in the three-dimensional array is traversed. If there is a fifth scanning point in Mp[][][i], the fifth scanning points in Mp[][][i] are sorted by y (vertical coordinate), and the fifth scanning point with the smallest y (vertical coordinate value) is selected from the sorted fifth scanning points as the candidate first scanning point. Then, the column dimension of the next level below the p level in the three-dimensional array is traversed. After the column dimension in the three-dimensional array is traversed, Mp[][j][] is traversed along the x-axis. If there is a fifth scanning point in Mp[][j][], the fifth scanning point in Mp[][j][] is sorted by x (horizontal coordinate), and the fifth scanning point with the smallest x (horizontal coordinate value) is selected from the sorted fifth scanning points as the candidate first scanning point. This cycle is repeated until the vertical dimension M[k][][] of the next level in the three-dimensional array is traversed. In this way, multiple candidate first scanning points can be determined from multiple fifth scanning points.

[0105] In some embodiments, based on the three-dimensional spatial coordinates corresponding to each candidate first scanning point and a second preset value, a secondary noise reduction process is performed on multiple candidate first scanning points to obtain first point cloud data, including: determining each candidate first scanning point as a target candidate first scanning point in turn, if there is a candidate first scanning point whose height is lower than the target candidate first scanning point, and the distance between the candidate first scanning point and the target candidate first scanning point is less than the second preset value, then deleting the target candidate first scanning point from the multiple candidate first scanning points to obtain first point cloud data.

[0106] The target candidate first scanning point is any candidate first scanning point among the multiple candidate first scanning points.

[0107] In some embodiments, the first horizontal coordinate value, the first vertical coordinate value and the first ordinate value of each candidate first scanning point can be determined with any candidate first scanning point among the plurality of candidate first scanning points as the origin. Figure 1 In the three-dimensional space coordinate system shown, if the first vertical coordinate value of the candidate first scanning point is less than the vertical coordinate value of the target candidate first scanning point, it means that the height of the candidate first scanning point is lower than the target candidate first scanning point. If the sum of the squares of the first horizontal coordinate value and the first vertical coordinate value of the candidate first scanning point is less than the square of the second preset value, it means that the candidate first scanning point is located within a circular range defined along the xz plane in the three-dimensional space coordinates with the target candidate first scanning point as the center and the second preset value as the radius; that is, the distance between the candidate first scanning point and the target candidate first scanning point is less than the second preset value.

[0108] For example, let the vertical coordinate value of the first scanning point of the target candidate be i , the second preset value is r, and the first vertical coordinate value of the candidate first scanning point determined with the target candidate first scanning point as the origin is y j , the first horizontal coordinate value is x j and the first ordinate value is z j Satisfy the condition: j <y i &&(x j ) 2 +(z j ) 2 <r 2 , it is determined that the height of the candidate first scanning point is lower than the target candidate first scanning point, and the distance between the candidate first scanning point and the target candidate first scanning point is less than the second preset value. At this time, the candidate first scanning point is deleted from the multiple candidate first scanning points until the multiple candidate first scanning points are traversed to obtain the first point cloud data.

[0109] S203. Generate second point cloud data based on the three-dimensional spatial coordinates corresponding to each of the multiple first scanning points in the first point cloud data and the radius of the material bin, and determine the set of the first point cloud data and the second point cloud data as third point cloud data.

[0110] The second point cloud data includes multiple first scanning points and multiple second scanning points.

[0111] In some embodiments, second point cloud data is generated based on the three-dimensional spatial coordinates corresponding to each of the multiple first scanning points in the first point cloud data and the radius of the material bin, including: first, based on the three-dimensional spatial coordinates corresponding to each first scanning point, determining the scanning point among the multiple first scanning points that is farthest from the vertical coordinate axis of the three-dimensional spatial coordinate system as a reference scanning point; then, based on the vertical coordinate value of the reference scanning point and the radius of the material bin, as well as a preset incremental angle, generating multiple second scanning points to obtain second point cloud data.

[0112] In some embodiments, each second scanning point is generated by the following formula 1-2:

[0113]

[0114] Among them, P y is the vertical coordinate value of the reference scanning point, α is the preset incremental angle, r is the radius of the material bin, and n is the index of the second scanning point.

[0115] In some embodiments, it can be determined by formula 1-2 that the generated multiple second scanning points are located at Figure 1 The y-axis of the three-dimensional space coordinate system is the center of the circle, and the radius of the material bin is the radius of the circle, that is, the multiple second scanning points generated are all located at the intersection of the material pile and the material bin.

[0116] In some embodiments, the index n of the second scanning point can be used as a counter to loop through the above steps to generate multiple second scanning points through formula 1-2. When the product of the number of loops and the preset increment angle is greater than 2π, that is, when n*α>2*π, the loop is stopped. The initial value of n is 0, and n increases by 1 each time the loop is repeated.

[0117] S204. Based on the third preset value, the three-dimensional space coordinates corresponding to each of the plurality of third scanning points in the third point cloud data and the diameter of the material bin, determine the coordinates to be generated, and generate fourth point cloud data based on the coordinates to be generated when the coordinates to be generated meet the preset generation conditions.

[0118] Among them, since the third point cloud data is a collection of the first point cloud data and the second point cloud data, the multiple third scanning points in the third point cloud data include multiple first scanning points and multiple second scanning points. The fourth point cloud data includes multiple fourth scanning points, and the fourth scanning points are used to supplement the third point cloud data.

[0119] In some embodiments, the third preset value is a preset value, which may be equal to the first preset value or may not be equal to the first preset value, and may also be determined according to actual needs, which is not limited in the embodiments of the present application.

[0120] In some embodiments, the preset generation condition includes: the sum of the squares of x (abscissa value) and z (ordinate value) in the coordinate to be generated is less than or equal to the square of the radius of the material bin, that is, the coordinate to be generated is located in the material bin.

[0121] Exemplarily, the preset generation condition may be x 2 +z 2 <=r 2 , that is, x and z in the coordinates to be generated satisfy x 2 +z 2 <=r 2 When the fourth scanning point is generated, the fourth point cloud data is created by using the coordinates to be generated.

[0122] In some embodiments, when generating the second scanning points, each time a second scanning point is generated, the generated second scanning point is additionally stored in the third point cloud data.

[0123] In some embodiments, based on the third preset value, the three-dimensional spatial coordinates corresponding to each of the multiple third scanning points in the third point cloud data, and the diameter of the material silo, determining the coordinates to be generated includes: first, based on the diameter of the material silo, constructing a point cloud generation area, and dividing the point cloud generation area based on the third preset value to obtain multiple sub-areas; then, based on the multiple sub-areas, constructing the sixth point cloud data, and based on the third preset value and the three-dimensional spatial coordinates corresponding to the multiple third scanning points, determining the index information of each third scanning point; then, according to the index information of each third scanning point, storing the multiple third scanning points in the sixth point cloud data, and traversing the first index point in the sixth point cloud data; if there is no third scanning point at the first index point, the horizontal axis value and the vertical axis value of the center point of the sub-area corresponding to the first index point, and the vertical coordinate value of the reference scanning point are determined as the coordinates to be generated.

[0124] Among them, the point cloud generation area is parallel to the horizontal plane.

[0125] In some embodiments, a square area is constructed along the xz plane in the three-dimensional space coordinate system with the y axis as the center and the diameter of the material bin as the side length to obtain a point cloud generation area. After obtaining the point cloud generation area, the area of ​​the xz plane covered by the point cloud generation area in the three-dimensional space coordinate system is divided into a plurality of square areas of the same size based on a third preset value to obtain a plurality of sub-areas.

[0126] For example, the area of ​​the xz plane covered by the point cloud generation area in the three-dimensional space coordinate system can be divided into a plurality of square areas with the same side length as the third preset value to obtain a plurality of sub-areas; the area of ​​the xz plane covered by the point cloud generation area in the three-dimensional space coordinate system can also be divided into square areas with the same size and a number equal to the third preset value to obtain a plurality of sub-areas. This embodiment of the application is not limited to this.

[0127] In some embodiments, the index information of the third scanning point represents the positional relationship between the third scanning point and each sub-region. In some embodiments, the positional relationship between the third scanning point and each sub-region includes whether the third scanning point is located in each sub-region and in which of the multiple sub-regions the third scanning point is located, that is, the index information of the third scanning point can be used to determine whether the third scanning point is located in the sub-region and in which of the multiple sub-regions the third scanning point is located. The sixth point cloud data is used to store the third scanning point located in the sub-region; the sixth point cloud data can cover multiple sub-regions, and therefore, the length of the sixth point cloud data can be equal to the quotient of the diameter of the material bin barrel and the third preset value.

[0128] Exemplarily, the index information of each third scanning point can be determined according to the three-dimensional space coordinates corresponding to the third scanning point, and each third scanning point can be stored at the corresponding index point in the sixth point cloud data according to the index information of each third scanning point. For example, if the index information of the third scanning point indicates that the third scanning point is located in the sub-region with an index position of 2, the third scanning point is stored at the index point with an index position of 2 in the sixth point cloud data.

[0129] In some embodiments, the first index point represents any area among a plurality of sub-areas corresponding to the sixth point cloud data set, and the coordinates to be generated indicate the center point of the sub-area corresponding to the first index point.

[0130] Exemplarily, if the third scanning point does not exist at the first index point in the sixth point cloud data, it means that the sub-region corresponding to the first index point lacks scanning points. At this time, the horizontal axis value and the vertical axis value of the center point of the sub-region corresponding to the first index point, as well as the vertical coordinate value of the reference scanning point, can be determined as the coordinates to be generated. When the coordinates to be generated meet the preset generation conditions (e.g., the coordinates to be generated are in the material bin), the fourth scanning point is generated based on the coordinates to be generated to supplement the scanning points of the sub-region. If the third scanning point exists at the first index point in the sixth point cloud data, the next index point of the sixth point cloud data is traversed.

[0131] In some embodiments, the coordinates to be generated are determined by the following formula 1-3:

[0132]

[0133] Among them, P y is the ordinate value of the reference scanning point, q is the third preset value, r is the radius of the material bin, 2r is the diameter of the material bin, and i is the index of the coordinate to be generated.

[0134] It can be understood that the embodiments of the present application can control the precision of surface recognition and reconstruction of various material piles in different environments by setting multiple parameters (e.g., a first preset value, a second preset value, and a third preset value). For example, the distribution density of the generated scanning points can be controlled. The higher the distribution density of the generated scanning points, the higher the precision of surface recognition and reconstruction of the material pile.

[0135] S205. Based on the plurality of third scanning points, respectively perform fitting processing on the plurality of fourth scanning points in the fourth point cloud data to update the fourth point cloud data, and determine a set of the updated fourth point cloud data and the third point cloud data as the target point cloud data.

[0136] The target point cloud data is the point cloud data after the surface reconstruction of the material pile is completed. The target point cloud data includes a plurality of third scanning points and the fourth scanning point in the updated fourth point cloud data set.

[0137] In some embodiments, based on multiple third scanning points, multiple fourth scanning points in the fourth point cloud data are respectively fitted to update the fourth point cloud data, including: first determining a first distance between each third scanning point and each fourth scanning point in a plane formed by a horizontal axis and a vertical axis in a three-dimensional space coordinate system, and selecting a number of third scanning points to be processed equal to a preset number from multiple third scanning points in order of the first distance from near to far; thereafter, fitting the vertical coordinate value of each fourth scanning point using the vertical coordinate value of the third scanning point to be processed to update the fourth point cloud data.

[0138] Among them, the preset number is a preset value.

[0139] Exemplarily, if the preset number is k, then based on multiple third scanning points, when fitting each fourth scanning point separately, the first distance between each third scanning point and the target fourth scanning point in the xz plane of the three-dimensional space coordinate system can be first determined, and according to the first distance, k third scanning points to be processed are selected from the multiple third scanning points in the order of the first distance from near to far. Among them, the target fourth scanning point is any fourth scanning point among the multiple fourth scanning points. After obtaining k third scanning points to be processed, the third scanning point to be processed closest to the target fourth scanning point is determined as the seventh scanning point from the k third scanning points to be processed, and the y (vertical coordinate) of the seventh scanning point is recorded as P_y, and the first distance between the seventh scanning point and the target fourth scanning point is recorded as P_xz. After that, determine the average value of y in the k third scanning points to be processed, recorded as a_y; determine the average value of the first distance between the k third scanning points to be processed and the target fourth scanning point, recorded as A_xz; finally, determine the difference 1-P_xz between 1 and the first distance P_xz between the seventh scanning point and the target fourth scanning point, and according to the ratio of 1-P_xz to A_xz: 1-P_xz / A_xz, update the y-axis coordinate of the target fourth scanning point to: a_y+(1-P_xz / A_xz)*(P_y-a_y), and obtain the updated fourth scanning point, and update the fourth point cloud data according to the updated fourth scanning point, until multiple fourth scanning points are updated to obtain updated fourth point cloud data.

[0140] In some embodiments, Figure 3 shows a stereogram of the original point cloud data, Figure 4 shows a stereogram of the first point cloud data after filtering and denoising the original point cloud data, Figure 5 Figure 2 shows a stereogram of the target point cloud data. Figure 3 and Figure 4 As shown in Figure 3 After filtering and denoising, the interference data and most of the noise data in the original point cloud data can be reduced to obtain the first point cloud data (as shown in FIG. Figure 4 In this way, the error between the reconstructed surface and the actual surface can be guaranteed, and the spatial position of the original point cloud will not be changed; Figure 5 As shown, based on Figure 4 After executing S203-S205, the target point cloud data (such as Figure 5 In this way, by supplementing (generating) new scanning points, the precision of the material pile surface recognition and reconstruction can be improved.

[0141] It can be understood that in the embodiment of the present application, interference data and some noise data in multiple original scanning points can be eliminated through S202, and point cloud data can be supplemented through S203-S204. In this way, the accuracy of material pile surface reconstruction can be improved by eliminating interference data and some noise data and supplementing point cloud data.

[0142] Corresponding to the above-mentioned embodiment of the material pile surface reconstruction method, the present application also provides an embodiment of the material pile surface reconstruction device.

[0143] Reference Figure 6 , the present application embodiment provides a material pile surface reconstruction device, comprising:

[0144] The acquisition module 601 is used to acquire scanning data corresponding to each of the multiple original scanning points obtained by scanning the material pile by the radar and the three-dimensional spatial coordinates of each of the multiple original scanning points in the three-dimensional spatial coordinate system; wherein the origin of the three-dimensional spatial coordinate system is the radar, the horizontal axis extends along the radius direction of the material bin barrel, the vertical axis and the horizontal axis are located in the same horizontal plane and extend in a direction perpendicular to the horizontal axis, and the vertical axis extends along the height direction of the material bin barrel in a direction away from the opening of the material bin barrel;

[0145] The processing module 602 is used to perform filtering and noise reduction processing on the multiple original scanning points based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point to obtain first point cloud data; based on the multiple third scanning points, respectively perform fitting processing on the multiple fourth scanning points in the fourth point cloud data to update the fourth point cloud data, and determine the set of the updated fourth point cloud data and the third point cloud data as the target point cloud data;

[0146] Generation module 603 is used to generate second point cloud data based on the three-dimensional spatial coordinates corresponding to each of the multiple first scanning points in the first point cloud data and the radius of the material silo, and determine the collection of the first point cloud data and the second point cloud data as third point cloud data; the multiple second scanning points in the second point cloud data are all data points at the junction of the material pile and the material silo; based on the third preset value, and the three-dimensional spatial coordinates corresponding to each of the multiple third scanning points in the third point cloud data and the diameter of the material silo, determine the coordinates to be generated, and when the coordinates to be generated meet the preset generation conditions, generate fourth point cloud data based on the coordinates to be generated.

[0147] In some embodiments, the processing module 602 is also used to perform data filtering processing on multiple original scanning points based on the three-dimensional spatial coordinates corresponding to each original scanning point and the scanning data corresponding to each original scanning point, so as to obtain fifth point cloud data; based on the three-dimensional spatial coordinates corresponding to each of the multiple fifth scanning points in the fifth point cloud data, as well as the first preset value and the second preset value, perform noise reduction processing on the multiple fifth scanning points at least once to obtain first point cloud data.

[0148] In some embodiments, the processing module 602 is also used to determine the positional relationship between the original scanning point and the material bin barrel based on the three-dimensional spatial coordinates corresponding to each original scanning point; based on the positional relationship between the original scanning point and the material bin barrel, and the scanning data corresponding to each original scanning point, the multiple original scanning points are subjected to data filtering processing to obtain fifth point cloud data.

[0149] In some embodiments, the scanning data corresponding to the original scanning point also includes the speed of the original scanning point and the signal-to-noise ratio of the original scanning point; the processing module 602 is also used to determine the original scanning point as the fifth scanning point if the positional relationship between the original scanning point and the material bin barrel indicates that the original scanning point is located in the material bin barrel, and the speed of the original scanning point is the first preset speed threshold, and the signal-to-noise ratio of the original scanning point is within the preset signal-to-noise ratio range, thereby obtaining the fifth point cloud data.

[0150] In some embodiments, the processing module 602 is also used to perform initial noise reduction processing on multiple fifth scanning points based on the three-dimensional spatial coordinates corresponding to each fifth scanning point and a first preset value to obtain multiple candidate first scanning points; based on the three-dimensional spatial coordinates corresponding to each candidate first scanning point and a second preset value, perform secondary noise reduction processing on multiple candidate first scanning points to obtain first point cloud data.

[0151] In some embodiments, the processing module 602 is also used to construct a cube based on the three-dimensional space coordinates corresponding to each fifth scanning point, and divide the cube based on a first preset value to obtain a plurality of cubes; construct a three-dimensional array based on the plurality of cubes, and determine the index information of each fifth scanning point based on the three-dimensional space coordinates corresponding to the fifth scanning point and the first preset value; the index information of the fifth scanning point represents the positional relationship between the fifth scanning point and each cube; according to the index information of each fifth scanning point, store the plurality of fifth scanning points in a plurality of elements in the three-dimensional array; traverse each element in the three-dimensional array, and determine the fifth scanning point with the highest height in each element as a candidate first scanning point.

[0152] In some embodiments, the processing module 602 is also used to determine each candidate first scanning point as a target candidate first scanning point in turn. If there is a candidate first scanning point whose height is lower than the target candidate first scanning point, and the distance between the candidate first scanning point and the target candidate second scanning point is less than a second preset value, the target candidate first scanning point is deleted from the multiple candidate first scanning points to obtain the first point cloud data.

[0153] In some embodiments, the processing module 602 is further configured to determine, based on the three-dimensional spatial coordinates corresponding to each first scanning point, a scanning point among the plurality of first scanning points that is farthest from a vertical coordinate axis of the three-dimensional spatial coordinate system as a reference scanning point;

[0154] The generation module 603 is also used to generate multiple second scanning points based on the vertical coordinate value of the reference scanning point and the radius of the material bin barrel, as well as the preset incremental angle, to obtain second point cloud data.

[0155] In some embodiments, the processing module 602 is also used to construct a point cloud generation area based on the diameter of the material silo barrel, and divide the point cloud generation area based on a third preset value to obtain multiple sub-areas; the point cloud generation area is parallel to the horizontal plane; based on the multiple sub-areas, the sixth point cloud data is constructed, and based on the third preset value and the three-dimensional space coordinates corresponding to the multiple third scanning points, the index information of each third scanning point is determined; the index information of the third scanning point represents the positional relationship between the third scanning point and each sub-area; according to the index information of each third scanning point, the multiple third scanning points are stored in the sixth point cloud data, and the first index point in the sixth point cloud data is traversed; the first index point represents any area of ​​the multiple sub-areas corresponding to the sixth point cloud data set; if there is no third scanning point at the first index point, the horizontal axis value and the vertical axis value of the center point of the sub-area corresponding to the first index point, and the vertical coordinate value of the reference scanning point are determined as the coordinates to be generated.

[0156] In some embodiments, the processing module 602 is also used to determine the first distance between each third scanning point and each fourth scanning point in the plane formed by the horizontal axis and the vertical axis in the three-dimensional space coordinate system, and select a number of third scanning points to be processed equal to a preset number from multiple third scanning points in order of the first distance from near to far; the vertical coordinate value of each fourth scanning point is fitted with the vertical coordinate value of the third scanning point to be processed to update the fourth point cloud data.

[0157] like Figure 7As shown, an electronic device provided in an embodiment of the present application may include: a processor (processor) 710, a communication interface (Communications Interface) 720, a memory (memory) 730 and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the above methods.

[0158] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the switchgear mechanical state monitoring method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0159] In yet another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented to perform the above methods when executed by a processor.

[0160] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0161] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0162] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for reconstructing the surface of a material pile, characterized in that: include: Obtain scanning data corresponding to each of a plurality of original scanning points obtained by scanning the material pile by a radar and the three-dimensional spatial coordinates of each of the plurality of original scanning points in a three-dimensional spatial coordinate system; wherein the origin of the three-dimensional spatial coordinate system is the radar, the horizontal axis extends along the radius direction of the material bin barrel, the vertical axis is located in the same horizontal plane as the horizontal axis and extends in a direction perpendicular to the horizontal axis, and the vertical axis extends along the height direction of the material bin barrel in a direction away from the opening of the material bin barrel; Based on the three-dimensional spatial coordinates corresponding to each of the original scanning points and the scanning data corresponding to each of the original scanning points, filtering and denoising the plurality of original scanning points to obtain first point cloud data; Based on the three-dimensional spatial coordinates corresponding to each of the plurality of first scanning points in the first point cloud data and the radius of the material bin, second point cloud data are generated, and a collection of the first point cloud data and the second point cloud data is determined as third point cloud data; the plurality of second scanning points in the second point cloud data are all data points at the junction of the material pile and the material bin; Based on the third preset value, the three-dimensional space coordinates corresponding to each of the plurality of third scanning points in the third point cloud data and the diameter of the material bin, the coordinates to be generated are determined, and when the coordinates to be generated meet the preset generation conditions, fourth point cloud data are generated based on the coordinates to be generated; Based on the plurality of third scanning points, fitting processing is performed on the plurality of fourth scanning points in the fourth point cloud data respectively to update the fourth point cloud data, and a set of the updated fourth point cloud data and the third point cloud data is determined as the target point cloud data.

2. The method according to claim 1, characterized in that The filtering and denoising process is performed on the plurality of original scanning points based on the three-dimensional spatial coordinates corresponding to each of the original scanning points and the scanning data corresponding to each of the original scanning points to obtain first point cloud data, including: Based on the three-dimensional space coordinates corresponding to each of the original scanning points and the scanning data corresponding to each of the original scanning points, a data filtering process is performed on the plurality of original scanning points to obtain fifth point cloud data; Based on the three-dimensional space coordinates corresponding to each of the plurality of fifth scanning points in the fifth point cloud data, as well as the first preset value and the second preset value, at least one noise reduction process is performed on the plurality of fifth scanning points to obtain the first point cloud data.

3. The method according to claim 2, characterized in that The step of performing data filtering processing on the plurality of original scanning points based on the three-dimensional spatial coordinates corresponding to each of the original scanning points and the scanning data corresponding to each of the original scanning points to obtain fifth point cloud data includes: Based on the three-dimensional spatial coordinates corresponding to each of the original scanning points, determining the positional relationship between the original scanning point and the material bin barrel; Based on the positional relationship between the original scanning point and the material bin barrel, and the scanning data corresponding to each of the original scanning points, data filtering processing is performed on the multiple original scanning points to obtain the fifth point cloud data.

4. The method according to claim 3, characterized in that The scanning data corresponding to the original scanning point includes the speed of the original scanning point and the signal-to-noise ratio of the original scanning point; The data filtering process is performed on the plurality of original scanning points based on the positional relationship between the original scanning point and the material bin barrel, and the scanning data corresponding to each of the original scanning points, to obtain the fifth point cloud data, including: If the positional relationship between the original scanning point and the material bin barrel indicates that the original scanning point is located in the material bin barrel, and the speed of the original scanning point is a first preset speed threshold, and the signal-to-noise ratio of the original scanning point is within a preset signal-to-noise ratio range, then the original scanning point is determined as the fifth scanning point, thereby obtaining the fifth point cloud data.

5. The method according to any one of claims 2 to 4, characterized in that: The step of performing at least one noise reduction process on the plurality of fifth scanning points based on the three-dimensional space coordinates respectively corresponding to the plurality of fifth scanning points in the fifth point cloud data, as well as the first preset value and the second preset value, to obtain the first point cloud data comprises: Based on the three-dimensional space coordinates corresponding to each of the fifth scanning points and the first preset value, performing a primary noise reduction process on the plurality of the fifth scanning points to obtain a plurality of candidate first scanning points; Based on the three-dimensional space coordinates corresponding to each of the candidate first scanning points and the second preset value, a secondary noise reduction process is performed on the plurality of candidate first scanning points to obtain the first point cloud data.

6. The method according to claim 5, characterized in that The method of performing a primary noise reduction process on the plurality of fifth scanning points based on the three-dimensional space coordinates corresponding to each of the fifth scanning points and the first preset value to obtain a plurality of candidate first scanning points includes: constructing a cube based on the three-dimensional space coordinates corresponding to each of the fifth scanning points, and dividing the cube based on the first preset value to obtain a plurality of cubes; Based on the plurality of cubes, a three-dimensional array is constructed, and based on the three-dimensional spatial coordinates corresponding to the fifth scanning point and the first preset value, index information of each of the fifth scanning points is determined; the index information of the fifth scanning point represents a positional relationship between the fifth scanning point and each of the cubes; According to the index information of each of the fifth scanning points, storing the plurality of the fifth scanning points into a plurality of elements in the three-dimensional array; Each of the elements in the three-dimensional array is traversed, and a fifth scanning point with the highest height among the elements is determined as the candidate first scanning point.

7. The method according to claim 5, characterized in that The step of performing secondary noise reduction processing on the plurality of candidate first scanning points based on the three-dimensional space coordinates corresponding to each of the candidate first scanning points and the second preset value to obtain the first point cloud data includes: Determine each of the candidate first scanning points as a target candidate first scanning point in turn; if there is a candidate first scanning point whose height is lower than the target candidate first scanning point, and the distance between the candidate first scanning point and the target candidate second scanning point is less than the second preset value, delete the target candidate first scanning point from the multiple candidate first scanning points to obtain the first point cloud data.

8. The method according to claim 1, characterized in that: The generating of the second point cloud data based on the three-dimensional space coordinates corresponding to each of the plurality of first scanning points in the first point cloud data and the radius of the material bin barrel comprises: Based on the three-dimensional space coordinates corresponding to each of the first scanning points, determining a scanning point that is farthest from the vertical coordinate axis of the three-dimensional space coordinate system among the plurality of first scanning points as a reference scanning point; Based on the vertical coordinate value of the reference scanning point, the radius of the material bin barrel, and a preset incremental angle, a plurality of the second scanning points are generated to obtain the second point cloud data.

9. The method according to claim 8, characterized in that The determining of the coordinates to be generated based on the third preset value, the three-dimensional space coordinates corresponding to each of the plurality of third scanning points in the third point cloud data, and the diameter of the material bin barrel includes: Based on the diameter of the material bin barrel, a point cloud generation area is constructed, and based on the third preset value, the point cloud generation area is divided to obtain a plurality of sub-areas; the point cloud generation area is parallel to the horizontal plane; Based on the multiple sub-regions, sixth point cloud data is constructed, and based on the third preset value and the three-dimensional space coordinates corresponding to the multiple third scanning points, index information of each of the third scanning points is determined; the index information of the third scanning point represents the positional relationship between the third scanning point and each of the sub-regions; According to the index information of each of the third scanning points, the plurality of the third scanning points are stored in the sixth point cloud data, and the first index point in the sixth point cloud data is traversed; the first index point represents any area in the plurality of the sub-areas corresponding to the sixth point cloud data set; If the third scanning point does not exist at the first index point, the horizontal axis value and the vertical axis value of the center point of the sub-area corresponding to the first index point and the vertical coordinate value of the reference scanning point are determined as the coordinates to be generated.

10. The method according to claim 9, characterized in that The preset generation condition includes: the sum of the squares of the horizontal coordinate value and the vertical coordinate axis value in the coordinate to be generated is less than or equal to the square of the radius of the material bin barrel; The fitting process is performed on the plurality of fourth scanning points in the fourth point cloud data based on the plurality of third scanning points to update the fourth point cloud data, including: Determine a first distance between each of the third scanning points and each of the fourth scanning points in a plane formed by the abscissa axis and the ordinate axis in the three-dimensional space coordinate system, and select a number of third scanning points to be processed equal to a preset number from the plurality of third scanning points in order of the first distance from near to far; The vertical coordinate value of each of the fourth scanning points is fitted using the vertical coordinate value of the third scanning point to be processed, so as to update the fourth point cloud data.

Images