Material volume detection method, device and equipment of quantitative bin and detection system

Abstract

The embodiment of the invention provides a material volume detection method, device and equipment for a quantitative bin and a detection system. The method comprises the following steps: acquiring at least two pieces of first point cloud information acquired by at least two laser radars respectively; carrying out fusion processing on the at least two pieces of first point cloud information to obtain fusion point cloud information of the materials in the quantitative bin; and determining volume information of the materials in the quantitative bin according to the fused point cloud information. According to the scheme, the problem that shielding exists in a single radar or manual inspection accuracy is poor is effectively solved, real-time monitoring of the material volume in the dynamic discharging process is achieved, and reliable data support is provided for quantitative control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of loading technology, and in particular to a material volume detection method, device, equipment and detection system for a quantitative bin. BACKGROUND

[0002] The bulk density of bulk materials (such as ore, coal, etc.) is affected by many factors such as grade, particle size composition, moisture, etc. When the material is loaded according to the specified weight, the volume of the material is affected by the bulk density, making it difficult to control the loading height and shape, and prone to unbalanced loading, overloading and even material scattering.

[0003] At present, the artificial inspection method is mainly used to complete the volume evaluation of the material through visual inspection and simple tool measurement.

[0004] However, under the current method, not only a large amount of manpower needs to be invested, but also the detection result is greatly affected by the experience of the operator, and the inaccurate detection result may occur. SUMMARY

[0005] The embodiments of the present application provide a material volume detection method, device, equipment and detection system for a quantitative bin, so as to achieve the condition of lower cost and more accurate determination of the material volume.

[0006] In a first aspect, the embodiments of the present application provide a material volume detection method for a quantitative bin, comprising:

[0007] Obtaining at least two first point cloud information collected by at least two laser radars respectively, the at least two laser radars being arranged at a discharge port of the quantitative bin;

[0008] Performing fusion processing on the at least two first point cloud information to obtain fusion point cloud information of the material in the quantitative bin;

[0009] According to the fusion point cloud information, determining the volume information of the material in the quantitative bin.

[0010] In one or more embodiments, the fusion processing on the at least two first point cloud information to obtain the fusion point cloud information of the material in the quantitative bin comprises:

[0011] Performing global alignment processing and redundant point cloud removal processing on the at least two first point cloud information to obtain second point cloud information;

[0012] Performing pile foot hollow filling processing and curved surface reconstruction processing on the second point cloud information to obtain the fusion point cloud information.

[0013] In one or more embodiments, before the fusion processing of the at least two first point cloud information to obtain the fusion point cloud information of the material in the quantitative bin, the method further comprises:

[0014] According to the size information of the quantitative bin, filtering processing is performed on the at least two first point cloud information to obtain at least two first point cloud information of filtered material;

[0015] Filtering and denoising processing is performed on the at least two first point cloud information to obtain the at least two first point cloud information after processing.

[0016] In one or more embodiments, the determining of the volume information of the material in the quantitative bin according to the fusion point cloud information comprises:

[0017] Voxelization processing is performed on the fusion point cloud information to obtain a plurality of point cloud information blocks;

[0018] Volume calculation processing is performed on each point cloud information block to obtain the volume information of the material in the quantitative bin.

[0019] In one or more embodiments, the number of the at least two laser radars is five, and the number of the discharge ports is four;

[0020] The first laser radar is arranged at the center position of the four discharge ports;

[0021] The remaining four laser radars are respectively arranged at a first distance from the center position, and the distance between two adjacent laser radars is equal.

[0022] In one or more embodiments, the obtaining of the at least two first point cloud information respectively collected by the at least two laser radars comprises:

[0023] For each laser radar, according to the installation information of the laser radar, the coordinate system of the laser radar is unified into the same coordinate system;

[0024] According to the time synchronization information, the first point cloud information collected by each laser radar is obtained.

[0025] In one or more embodiments, the obtaining of the at least two first point cloud information respectively collected by the at least two laser radars comprises:

[0026] The counterweight state of the quantitative bin is obtained;

[0027] When the counterweight state indicates that the quantitative bin is in a completed counterweight state, the at least two first point cloud information respectively collected by the at least two laser radars is obtained.

[0028] In a second aspect, the embodiments of the present application provide a material volume detection device for a quantitative bin, comprising:

[0029] An acquisition module is configured to acquire at least two first point cloud information collected by at least two laser radars respectively, and the at least two laser radars are arranged at discharge ports of the quantitative bin;

[0030] A fusion module is configured to perform fusion processing on the at least two first point cloud information to obtain fusion point cloud information of the material in the quantitative bin;

[0031] A determination module is configured to determine volume information of the material in the quantitative bin according to the fusion point cloud information.

[0032] In one or more embodiments, the fusion module is specifically configured to:

[0033] perform global alignment processing and redundant point cloud removal processing on the at least two first point cloud information to obtain second point cloud information;

[0034] perform pile foot cavity filling processing and curved surface reconstruction processing on the second point cloud information to obtain the fusion point cloud information.

[0035] In one or more embodiments, before the fusion processing on the at least two first point cloud information to obtain the fusion point cloud information of the material in the quantitative bin, the fusion module is further configured to:

[0036] perform filtering processing on the at least two first point cloud information according to size information of the quantitative bin to obtain filtered at least two first point cloud information of the material;

[0037] perform filtering and denoising processing on the at least two first point cloud information to obtain processed at least two first point cloud information.

[0038] In one or more embodiments, the determination module is specifically configured to:

[0039] perform voxelization processing on the fusion point cloud information to obtain a plurality of point cloud information blocks;

[0040] perform volume calculation processing on each point cloud information block to obtain the volume information of the material in the quantitative bin.

[0041] In one or more embodiments, the number of the at least two laser radars is five, and the number of the discharge ports is four;

[0042] The first laser radar is arranged at a center position of the four discharge ports;

[0043] The remaining four laser radars are arranged at a first distance from the center position, and the distance between each two adjacent laser radars is equal.

[0044] In one or more embodiments, the acquisition module is specifically configured to:

[0045] For each laser radar, the coordinate system of the laser radar is unified into the same coordinate system according to the installation information of the laser radar.

[0046] According to the time synchronization information, the first point cloud information collected by each laser radar is acquired.

[0047] In one or more embodiments, the acquisition module is specifically configured to:

[0048] The weight state of the quantitative bin is acquired.

[0049] When the weight state indicates that the quantitative bin is in a full weight state, at least two first point cloud information collected by at least two laser radars is acquired.

[0050] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor;

[0051] The memory stores computer execution instructions.

[0052] The processor executes the computer execution instructions stored in the memory, so that the processor executes the first aspect and / or various possible implementation manners of the first aspect.

[0053] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, the computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by a processor to implement the first aspect and / or various possible implementation manners of the first aspect.

[0054] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising a computer program, which is executed by a processor to implement the first aspect and / or various possible implementation manners of the first aspect.

[0055] In a sixth aspect, an embodiment of the present application provides a material volume detection system, comprising: a quantitative bin, at least two laser radars arranged on the quantitative bin, and an electronic device.

[0056] The at least two laser radars are used to collect at least two first point cloud information of the quantitative bin.

[0057] The electronic device is used to execute the first aspect and / or various possible implementation manners of the first aspect.

[0058] The method provided by the embodiment of the application comprises the following steps: acquiring at least two first point cloud information collected by at least two laser radars respectively, the at least two laser radars being arranged at a discharge port of the quantitative bin; performing fusion processing on the at least two first point cloud information to obtain fusion point cloud information of the material in the quantitative bin; and determining volume information of the material in the quantitative bin according to the fusion point cloud information. The scheme can effectively overcome the problems of occlusion of a single radar or poor accuracy of artificial inspection by deploying at least two laser radars at the discharge port of the quantitative bin to collect multi-angle point clouds and using multi-source data fusion technology to significantly improve the integrity and spatial resolution of three-dimensional point cloud data. Meanwhile, the multi-laser-radar redundant design enhances the system robustness, and the system can still maintain stable measurement in a vibration or dust environment, so that real-time monitoring of the volume of the material in a dynamic discharge process is finally realized, and reliable data support is provided for quantitative control. BRIEF DESCRIPTION OF DRAWINGS

[0059] The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate the embodiments of the application and, together with the description, serve to explain the principles of the application.

[0060] Figure 1 A radar arrangement front view schematic diagram provided by the embodiment of the application;

[0061] Figure 2 A radar arrangement top view schematic diagram provided by the embodiment of the application;

[0062] Figure 3 A flowchart of a material volume detection method of a quantitative bin provided by the embodiment of the application Figure 1 ;

[0063] Figure 4 A flowchart of a material volume detection method of a quantitative bin provided by the embodiment of the application Figure 2 ;

[0064] Figure 5 A flowchart of a material volume detection method of a quantitative bin provided by the embodiment of the application Figure 3 ;

[0065] Figure 6 A flowchart of a material volume detection method of a quantitative bin provided by the embodiment of the application Figure 4 ;

[0066] Figure 7 A flowchart of a material volume detection method of a quantitative bin provided by the embodiment of the application Figure 5 ;

[0067] Figure 8 A flowchart of a material volume detection method of a quantitative bin provided by the embodiment of the application Figure 6;

[0068] Figure 9 Flowchart of the method for detecting the volume of material in the quantitative bin provided by the embodiment of the present application Figure 7 ;

[0069] Figure 10 Flowchart of the method for detecting the volume of material in the quantitative bin provided by the embodiment of the present application Figure 8 ;

[0070] Figure 11 Structural diagram of the device for detecting the volume of material in the quantitative bin provided by the embodiment of the present application

[0071] Figure 12 Structural diagram of the electronic device provided by the embodiment of the present application

[0072] Figure 13 Structural diagram of the material volume detection system provided by the embodiment of the present application

[0073] Explanation of reference signs:

[0074] 1 - Bin wall of the quantitative bin

[0075] 2 - Material inlet of the quantitative bin

[0076] 3 - Laser radar

[0077] The specific embodiments of the present application have been shown by the above-mentioned drawings, and will be described in more detail hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of the present application by any means, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0078] The exemplary embodiments will be described in detail herein with reference to the accompanying drawings. The following description is presented with reference to the drawings to explain the present application. The present application can be implemented in various manners, and is not limited to the exemplary embodiments described herein. Therefore, specific structural examples described in the following exemplary embodiments do not represent all the embodiments consistent with the present application. Rather, they are merely examples of devices and methods consistent with some aspects of the present application, as detailed in the appended claims.

[0079] The bulk density of bulk material (e.g., ore, coal, etc.) is affected by grade, particle size composition, moisture, and many other factors. When the material is loaded according to the specified weight, the volume of the material is affected by the bulk density, resulting in difficulty in grasping the loading height and shape during loading, and easy production of unbalanced load, overloading, and even material scattering.

[0080] Therefore, it has practical value to detect the volume of the material in the quantitative bin before loading. In a current implementation, the volume of the material is often evaluated by visual inspection and simple tool measurement through manual inspection. In another implementation, a laser radar or a millimeter wave radar is combined with a pan-tilt head to form a dynamic scanning scheme.

[0081] That is, the current technical implementation has the following technical problems:

[0082] 1) Low detection efficiency, unable to adapt to high-frequency and large-batch loading operation requirements;

[0083] 2) High labor cost, non-uniform detection standard, and strong subjectivity of results;

[0084] 3) In extreme weather or complex environment, the detection quality is difficult to guarantee, and there are problems such as safety hazards and data lag;

[0085] 4) The speed of the pan-tilt head is limited, and the detection cannot be completed in a short time, generally more than 30s.

[0086] 5) The scanning range of a single radar is limited, and when the material in the bin is full, it cannot cover the whole bin; a single radar cannot work after being contaminated, and the maintenance or cleaning is large.

[0087] 6) The explosion-proof requirement of the electric components is high, and the installation is complex.

[0088] Based on the above-mentioned technical problems, the technical concept of the inventor is as follows: First, it is observed that manual detection of the volume of the material has the problems of large subjective error, low efficiency, and difficulty in real-time monitoring, and a single laser radar is prone to point cloud loss due to obstruction or viewing angle limitation, affecting the measurement accuracy; further inspired by the successful application of multi-sensor data fusion technology in automatic driving and other fields, it is thought that multiple laser radars are symmetrically deployed at the discharge port of the quantitative bin, data complementation is achieved through multi-angle synchronous acquisition, point cloud registration and three-dimensional reconstruction technology are combined, and discrete point cloud information is fused into a complete material surface model, thereby breaking through the limitation of a single sensor. The essence of this design is to convert manual experience judgment into accurate quantitative calculation based on multi-source heterogeneous data, which not only inherits the advantages of high-precision ranging of laser radars, but also overcomes environmental interference through system-level redundancy design, and finally realizes automatic and high-precision volume detection.

[0089] The technical solutions of the present application and how the technical solutions of the present application solve the above-mentioned technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0090] Figure 1 The radar arrangement front view schematic diagram provided for the embodiments of the present application,Figure 2 This is a top view of the radar arrangement provided in an embodiment of this application, combined with... Figure 1 and Figure 2 As shown:

[0091] In this diagram, label 1 indicates the location of the metering bin wall (usually an upper cylinder with a lower cone, or an upper cylinder with a lower prismatic cone), label 2 indicates the location of the metering bin inlet (the number is arbitrary, up to 4), and label 3 indicates the location of the lidar (at least two, for example, 5).

[0092] And, A( Figure 1 granular matter, and Figure 2 The dashed line indicates the material accumulation pattern inside the quantitative storage bin.

[0093] The installation locations of the lidar can be: suspended from the top of the quantitative silo with an opening, evenly distributed, with a lidar overlap rate ≥ the first threshold (e.g., 70%); one lidar in the center of the quantitative silo, and the other four lidars evenly distributed around the discharge port.

[0094] Furthermore, the execution entity of this application is an electronic device, specifically an edge computing unit such as a Graphics Processing Unit (GPU) or a Field-Programmable Gate Array (FPGA), which acquires point cloud information collected by various LiDARs via fiber optic / network cables. One implementation involves: determining that the quantitative bin weight distribution is complete, triggering the start of volume calculation, receiving point cloud information, fusing it, performing volume calculation, and then sending the result to staff via fiber optic / network cables or displaying it on an electronic device.

[0095] Based on the radar layout diagram above, the technical solutions involved in the embodiments of this application will be described in detail below:

[0096] Figure 3 A flowchart illustrating the material volume detection method for a quantitative silo provided in this application embodiment. Figure 1 ,like Figure 3 As shown, the method includes the following steps:

[0097] Step 31: Obtain at least two first point cloud information samples collected by at least two lidar sensors;

[0098] At least two lidar sensors are installed at the discharge port of the quantitative silo.

[0099] In this step, it can be based on Figure 1 and Figure 2 The lidar installation method shown involves each lidar performing a downward data acquisition operation on the top of the quantitative chamber, thereby obtaining the first point cloud information collected by each lidar.

[0100] Optionally, one implementation of this step 31 can be:

[0101] Step 1, for each lidar, according to the installation information of the lidar, the coordinate system of the lidar is unified into the same coordinate system;

[0102] In this implementation, for each lidar, its installation information (including three-dimensional coordinates, pitch angle, yaw angle, and other pose information) can be read first, and the local coordinate system of each radar is converted to a unified global coordinate system (for example, the origin can be the geometric center of the quantitative bin) through a homogeneous coordinate transformation matrix.

[0103] The above implementation can use a calibration board assisted hand-eye calibration method (such as Eye-on-Hand calibration) or an automatic target matching algorithm to ensure the accuracy of coordinate conversion and compensate for mechanical installation errors (such as pose shift caused by vibration), providing a basis for spatial alignment of multi-source point clouds.

[0104] Step 2, according to the time synchronization information, the first point cloud information collected by each lidar is obtained.

[0105] In this implementation, in order to ensure the time consistency of the first point cloud information collected by each lidar, the scanning clock deviation of each lidar can be controlled at the microsecond level through a hardware trigger signal or a software level time synchronization protocol.

[0106] Within the synchronization time window corresponding to the time synchronization information, each radar collects the first point cloud information of the discharge port area at the current time at a predetermined frequency (such as 10Hz), and through a timestamp alignment mechanism to ensure the time consistency of multiple frames of point cloud information, and eliminate dynamic deformation errors caused by material flow or equipment vibration.

[0107] Optionally, the number of at least two lidars is five, and the number of discharge ports is four; the first lidar is arranged at the center position of the four discharge ports; the remaining four lidars are arranged at a first distance from the center position and have equal distances between each other.

[0108] In this implementation, a five-laser-radar array configuration can be used, in which the main radar (the first lidar) is deployed at the geometric center of the four discharge ports as the global coordinate reference and core coverage node.

[0109] The remaining four auxiliary lidars are uniformly distributed at equal angles of 90° in the horizontal plane with the center as the origin, and each auxiliary lidar is at a first distance (which can be set according to the diameter of the discharge port, for example, 1.5-2* the width of the discharge port) from the center.

[0110] The above layout mode can have the following advantages: the main radar vertically downward covers all the common monitoring areas of the discharge ports, and can solve the problem of certain pile foot blind area; the four auxiliary radars are installed at an inclination angle of 45° (the pitch angle is adjustable), and respectively point to the edge areas of the corresponding discharge ports, and through multi-view cross scanning, the point cloud coverage of each discharge port is ensured to be large, and the beam interference of adjacent radars is avoided.

[0111] Further, before step 32, the following steps can also be performed:

[0112] Step 1, filtering at least two first point cloud information according to the size information of the quantitative bin, to obtain at least two first point cloud information of the filtered material;

[0113] In this implementation, according to the size information of the quantitative bin (including length, width, height, discharge port position, and bin wall thickness, etc.), a virtual bin boundary model (i.e., a spatial model) can be established in point cloud processing.

[0114] Then, by matching the first point cloud data collected by each radar with the bin boundary model, a geometric clipping algorithm based on an axial bounding box can be used to automatically remove all invalid point cloud data outside the actual physical boundary of the bin, such as stray points generated by reflection of equipment outside the bin or environmental interference.

[0115] Step 2, filtering and denoising at least two first point cloud information to obtain at least two first point cloud information after processing

[0116] In this implementation, for the filtered at least two first point cloud information, a straight-through filter and a statistical outlier removal (SOR) algorithm can be used to analyze the point density distribution within a first predetermined distance (e.g., 5 cm) radius neighborhood around each data point, and remove isolated noise points (e.g., dust reflection points) deviating from the mean value.

[0117] Then, denoising processing is performed, such as through time series analysis of the spatial distribution characteristics of continuous multiple frames of point cloud data, a dynamic point cloud stability model based on KD tree is first established, and the occurrence frequency and position offset of each spatial point in the time dimension are statistically analyzed (the time window can be set to 5-10 frames); then, an outlier detection algorithm based on a probability density function is used to remove transient noise points (such as random floating particles or transient reflection noise) that are unstable in continuous frames, while retaining effective material point clouds that exist stably in multiple timestamps.

[0118] Step 32, fusion processing of at least two first point cloud information to obtain fusion point cloud information of the material in the quantitative bin;

[0119] In this step, after obtaining the first point cloud information collected by the plurality of laser radars, the plurality of first point cloud information can be fused to obtain the fusion point cloud information capable of describing the material in the quantitative bin.

[0120] For example, based on the installation position information of the five laser radars, the five first point cloud information is fused into one point cloud information, i.e., the fusion point cloud information.

[0121] Step 33, determining the volume information of the material in the quantitative bin according to the fusion point cloud information.

[0122] In this step, after obtaining the fusion point cloud information of the material, the volume information of the material can be determined in combination with the empty bin volume and other information of the quantitative bin.

[0123] The method for detecting the volume of the material in the quantitative bin provided by the embodiments of the present application comprises the following steps: acquiring at least two first point cloud information collected by at least two laser radars, the at least two laser radars being arranged at the discharge port of the quantitative bin; performing fusion processing on the at least two first point cloud information to obtain fusion point cloud information of the material in the quantitative bin; and determining the volume information of the material in the quantitative bin according to the fusion point cloud information. The scheme deploys at least two laser radars at the discharge port of the quantitative bin to collect point cloud information from multiple angles, and effectively overcomes the problems of occlusion of single radar and poor accuracy of manual inspection by using multi-source data fusion technology, thereby significantly improving the integrity and spatial resolution of three-dimensional point cloud data. Meanwhile, the redundant design of multiple laser radars enhances the robustness of the system, and the system can still maintain stable measurement in a vibrating or dusty environment, thereby realizing real-time monitoring of the volume of the material in the dynamic discharge process and providing reliable data support for quantitative control.

[0124] On the basis of the above-mentioned embodiments, Figure 4 The flowchart of the method for detecting the volume of the material in the quantitative bin provided by the embodiments of the present application is shown in Figure 2 As shown in Figure 4 Step 32 can comprise the following steps:

[0125] Step 41, performing global alignment processing and redundant point cloud removal processing on the at least two first point cloud information to obtain second point cloud information.

[0126] In this step, the global alignment processing can be performed in a manner based on feature matching or an iterative closest point algorithm, the first point cloud information collected from multiple perspectives is unified in the same coordinate system to minimize the spatial error between the point clouds and eliminate the data deviation caused by the difference in the poses of the sensors.

[0127] Further, redundant point cloud removal is performed, which can utilize statistical outlier removal or voxel grid filtering technology to eliminate repeated scanning points (such as repeated points in the overlapping area of adjacent radar fields of view) and noise points (such as outlier points generated by dust reflection), reduce data redundancy, and finally generate second point cloud information with high consistency.

[0128] Step 42: The second point cloud information is subjected to pile foot hole filling processing and surface reconstruction processing to obtain fused point cloud information.

[0129] In this step, for the pile foot area hole in the second point cloud caused by dust and other obstructions or radar angle blind area (for example, the junction of the material and the warehouse wall), a filling algorithm based on radial basis function interpolation or morphological dilation can be used to reconstruct the missing surface according to the spatial gradient characteristics of the surrounding effective point cloud, so as to restore the bottom contour of the material.

[0130] Further, the discrete point cloud can be converted into a continuous, watertight triangular mesh surface (such as a mesh edge length that can be adjusted, usually 3-5 mm) through Poisson reconstruction or moving least squares surface reconstruction, while retaining the subtle undulating features of the material surface (such as the grain accumulation morphology), and finally outputting high-fidelity fused point cloud information.

[0131] The method for detecting the volume of the material in the quantitative warehouse provided by the embodiments of the present application obtains second point cloud information by performing global alignment processing and redundant point cloud removal processing on at least two first point cloud information; and obtains fused point cloud information by performing pile foot hole filling processing and surface reconstruction processing on the second point cloud information. This scheme eliminates the differences between multiple radar coordinate systems through global alignment processing, ensuring the spatial consistency of the point cloud; then, redundant point cloud removal eliminates repeated noise points, improving the rationality of data density; subsequently, the local missing caused by obstructions is repaired through pile foot hole filling, improving the integrity of the material surface; finally, the continuous topological structure is generated through surface reconstruction, converting the discrete point cloud into a high-precision three-dimensional model, reducing the volume calculation error, and effectively solving the problems of incomplete traditional single-radar point cloud and low artificial detection efficiency.

[0132] On the basis of the above embodiments, Figure 5 The flowchart of the method for detecting the volume of the material in the quantitative warehouse provided by the embodiments of the present application is shown in Figure 3 As shown in Figure 6 Step 33 can include the following steps:

[0133] Step 51: The fused point cloud information is subjected to voxelization processing to obtain a plurality of point cloud information blocks.

[0134] In this step, the fused point cloud information is subjected to voxelization processing based on octree or fixed resolution grid, so as to divide the fused point cloud information into a plurality of point cloud information blocks.

[0135] The plurality of point cloud information blocks can be a plurality of uniform voxel units after discretization of the fused point cloud information in continuous three-dimensional space.

[0136] In this way, the data amount is reduced (for example, the compression rate can reach 80%) by downsampling while the key geometric features in the fused point cloud information are retained.

[0137] Step 52, volume calculation processing is performed on each point cloud information block to obtain volume information of the material in the quantitative bin.

[0138] In this step, independent volume calculation is performed on each segmented point cloud information block, effective material point sets are extracted through point cloud density analysis and spatial distribution feature extraction, and noise and abnormal value interference are excluded; then, a volume integral algorithm based on a voxel filling method or surface triangulation is used to calculate the three-dimensional space volume occupied by each information block; finally, the volume results of the subblocks are weighted and accumulated, and the volume information of the whole material in the quantitative bin is output in combination with the point cloud resolution and the measurement error compensation coefficient.

[0139] In the overload described above, for each voxel block, effective material points and noise points can also be distinguished through point cloud attribution judgment (for example, normal vector analysis or density clustering) to exclude fly point interference; then, in combination with the empty bin calibration volume (pre-scanned empty bin point cloud reference data), a voxel difference method or Monte Carlo integral is used to calculate the effective voxel proportion, and finally the subblock volumes are accumulated and the system error (for example, the adhesion layer of the bin wall) is deducted to obtain the net volume of the material.

[0140] The material volume detection method of the quantitative bin provided in the embodiments of the present application obtains a plurality of point cloud information blocks through voxelization processing of fused point cloud information; and volume calculation processing is performed on each point cloud information block to obtain volume information of the material in the quantitative bin. The scheme discretizes the fused point cloud information into uniformly distributed three-dimensional voxel units through voxelization processing, effectively reduces the data complexity and improves the calculation efficiency; then, local volume calculation is performed on each point cloud information block to realize high-precision real-time measurement of the material volume, which is significantly better than the traditional overall point cloud processing method, and is especially suitable for dynamic monitoring scenes of large-scale and irregular materials.

[0141] On the basis of the above-described embodiments, Figure 4 The flowchart of the material volume detection method of the quantitative bin provided in the embodiments of the present application is shown in Figure 6 As shown in Figure 7 Step 31 can include the following steps:

[0142] Step 61, obtain the counterweight state of the quantitative bin;

[0143] In this step, the counterweight status of the quantitative bin can be monitored in real time using a weighing sensor, pressure detection device, or PLC signal to determine whether the material loading has reached the preset weight threshold.

[0144] Step 62: When the counterweight status indicator indicates that the quantitative bin is in the counterweight completion state, acquire at least two first point cloud information collected by at least two lidars respectively.

[0145] In this step, when the counterweight status indicator indicates that the quantitative bin is in the counterweight complete state, multiple lidars deployed at the discharge port (e.g., 2-5 16-line lidars) are simultaneously activated to collect the first point cloud information in the quantitative bin from different perspectives with a millisecond time difference.

[0146] The material volume detection method for a quantitative silo provided in this application acquires the counterweight status of the quantitative silo; when the counterweight status indicates that the quantitative silo is in the stage of complete counterweighting, it acquires at least two first point cloud information collected by at least two lidar sensors. This scheme initiates multi-lidar point cloud acquisition after the quantitative silo is completed, effectively avoiding measurement interference caused by material flow or vibration; at the same time, the combination of counterweight status and point cloud acquisition timing coordination reduces invalid data storage and ensures the matching of measurement timing with the static distribution of materials, thereby reducing the dynamic error of subsequent volume calculation and significantly improving reliability and measurement efficiency in complex industrial environments.

[0147] Based on the above embodiments, Figure 5 A flowchart illustrating the material volume detection method for a quantitative silo provided in this application embodiment. Figure 8 ,like Figure 6 As shown, the complete implementation process of this method may include:

[0148] Step 71, Begin;

[0149] Step 72: The quantitative silo configuration is complete;

[0150] Step 73: Trigger scanning and acquisition of radar point cloud;

[0151] Step 74: Point cloud preprocessing;

[0152] Step 75: Point cloud fusion;

[0153] Step 76: Volume calculation;

[0154] Step 77: Output the volume value;

[0155] Step 78, End.

[0156] Based on the above embodiments, Figure 9 A flowchart illustrating the material volume detection method for a quantitative silo provided in this application embodiment. Figure 7 ,likeFigure 10 As shown, point cloud preprocessing in this method may include:

[0157] Step 81: Preprocessing begins;

[0158] Step 82: Spatial model filtering;

[0159] Step 83: Pass-through filtering;

[0160] Step 84: Outlier filtering;

[0161] Step 85: Multi-frame consistent noise reduction;

[0162] Step 86: Preprocessing complete.

[0163] Based on the above embodiments, Figure 8 A flowchart illustrating the material volume detection method for a quantitative silo provided in this application embodiment. Figure 11 ,like Figure 9 As shown, point cloud fusion in this method may include:

[0164] Step 91, Fusion begins;

[0165] Step 92: Multi-radar coordinate calibration;

[0166] Step 93: Radar clock synchronization;

[0167] Step 94: Global alignment based on feature matching;

[0168] Step 95: Redundancy removal from point clouds;

[0169] Step 96: Filling and correcting corner voids based on the angle of repose;

[0170] Step 97: Surface reconstruction based on triangulation;

[0171] Step 98: Fusion complete.

[0172] Based on the above embodiments, Figure 10 A flowchart illustrating the material volume detection method for a quantitative silo provided in this application embodiment. Figure 8 ,like Figure 11 As shown, the volume calculation in this method may include:

[0173] Step 101: Volume calculation begins;

[0174] Step 102, spatial voxelization;

[0175] Step 103: Point cloud attribution determination;

[0176] Step 104: Spatial volume statistics;

[0177] Step 105, calculate the material volume based on the empty bin calibration volume;

[0178] Step 106, volume calculation is completed.

[0179] The above Figure 12 The material volume detection method for the quantitative bin provided by the embodiment has the following technical effects:

[0180] 1) The detection efficiency is greatly improved, and the detection is completed within 2-3 seconds.

[0181] 2) Regardless of the height of the material pile, full bin coverage scanning is always performed

[0182] 3) No gimbal, no electric parts, pure electronic product

[0183] 4) The entire system has greatly improved dirt resistance.

[0184] 5) The detection volume precision error is not more than 2%.

[0185] Figure 13 The structure diagram of the material volume detection device for the quantitative bin provided by the embodiment is shown in the figure Figure 13 The material volume detection device for the quantitative bin comprises:

[0186] The acquisition module 111 is configured to acquire at least two first point cloud information collected by at least two laser radars arranged at the discharge port of the quantitative bin.

[0187] The fusion module 112 is configured to perform fusion processing on the at least two first point cloud information to obtain fusion point cloud information of the material in the quantitative bin.

[0188] The determination module 113 is configured to determine volume information of the material in the quantitative bin according to the fusion point cloud information.

[0189] In one or more embodiments, the fusion module 112 is specifically configured to:

[0190] perform global alignment processing and redundant point cloud removal processing on the at least two first point cloud information to obtain second point cloud information;

[0191] perform pile foot hole filling processing and curved surface reconstruction processing on the second point cloud information to obtain the fusion point cloud information.

[0192] In one or more embodiments, before performing fusion processing on the at least two first point cloud information to obtain the fusion point cloud information of the material in the quantitative bin, the fusion module 112 is further configured to:

[0193] perform filtering processing on the at least two first point cloud information according to size information of the quantitative bin to obtain filtered at least two first point cloud information of the material.

[0194] Filter and denoise the at least two first point cloud information to obtain processed at least two first point cloud information.

[0195] In one or more embodiments, the determining module 113 is specifically configured to:

[0196] Perform voxelization processing on the fused point cloud information to obtain a plurality of point cloud information blocks.

[0197] Perform volume calculation processing on each point cloud information block to obtain volume information of the material in the quantitative bin.

[0198] In one or more embodiments, the number of the at least two laser radars is five, and the number of the discharge ports is four.

[0199] The first laser radar is arranged at a central position of the four discharge ports.

[0200] The remaining four laser radars are respectively arranged at a first distance from the central position, and the distance between each two adjacent laser radars is equal.

[0201] In one or more embodiments, the obtaining module 111 is specifically configured to:

[0202] For each laser radar, the coordinate system of the laser radar is unified into the same coordinate system according to the installation information of the laser radar.

[0203] According to the time synchronization information, the first point cloud information collected by each laser radar is obtained.

[0204] In one or more embodiments, the obtaining module 111 is specifically configured to:

[0205] Obtain the counterweight state of the quantitative bin.

[0206] When the counterweight state indicates that the quantitative bin is in a completed counterweight state, obtain at least two first point cloud information collected by at least two laser radars.

[0207] The material volume detection device for the quantitative bin provided by the embodiments of the present application can be used to execute the material volume detection method for the quantitative bin in any of the above embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0208] Figure 13 The structural schematic diagram of the electronic device provided by the embodiments of the present application is shown in FIG. 1. ​ As shown in FIG. 1, the electronic device can include a processor 121, a memory 122, and computer program instructions stored in the memory 122 and executable on the processor 121, and the processor 121 executes the computer program instructions to implement the method provided by any of the preceding embodiments.

[0209] Optionally, the above-mentioned various devices of the electronic device can be connected through a system bus 123.

[0210] The memory 122 can be a separate storage unit or can be integrated into the processor 121. The number of processors 121 is one or more.

[0211] It should be understood that the processor 121 can be a central processing unit (CPU), and can also be other general-purpose processors 121, digital signal processors 121 (DSP), application specific integrated circuits (ASIC), etc. The general-purpose processor 121 can be a microprocessor or can also be any conventional processor 121, etc. The steps of the method disclosed in the present application can be directly embodied as execution by a hardware processor 121, or by a combination of hardware and software modules in the processor 121.

[0212] The system bus 123 can be a peripheral component interconnect (PCI) bus 123 or an extended industry standard architecture (EISA) bus 123, etc. The system bus 123 can be divided into an address bus 123, a data bus 123, a control bus 123, etc. For ease of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus 123 or only one type of bus 123. The memory 122 can include a random access memory 122 (RAM) and can also include a non-volatile memory 122 (NVM), such as at least one disk memory 122.

[0213] All or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a readable memory 122. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned memory 122 (storage medium) includes a read-only memory 122 (ROM), a RAM, a flash memory 122, a hard disk, a solid state disk, a magnetic tape (English: magnetic tape), a floppy disk (English: floppy disk), an optical disc (English: optical disc), and any combination thereof.

[0214] The electronic device provided by the embodiments of the present application can be used to execute the method provided by any of the method embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0215] ​ A structure diagram of the material volume detection system provided by the embodiments of the present application is shown in FIG. 1, which includes a quantitative bin 131, at least two laser radars 132 (only two are shown in the figure, and the actual number is two or more, which is determined based on actual needs) arranged on the quantitative bin 131, and an electronic device 133. ​ ​ The at least two laser radars 132 are used to collect at least two first point cloud information of the quantitative bin 131.

[0216] The at least two laser radars 132 are used to collect at least two first point cloud information of the quantitative bin 131.

[0217] The electronic device 133 is used to execute the material volume detection method as described above.

[0218] The electronic device provided by the embodiments of the present application can be used to execute the method provided by any of the method embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0219] The computer readable storage medium provided by the embodiments of the present application stores computer instructions, and when the computer instructions run on a computer, the computer executes the above method.

[0220] The computer readable storage medium described above can be realized by any type of volatile or non-volatile storage device or their combination, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0221] Optionally, the readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0222] ​The embodiment of the present application further provides a computer program product, which comprises a computer program stored in a computer readable storage medium, at least one processor can read the computer program from the computer readable storage medium, and the at least one processor executes the computer program to realize the method.

[0223] Finally, it should be noted that: other embodiments of the application will be readily apparent to those skilled in the art in view of the description and practice of the application disclosed herein. The application is intended to cover any variations, uses or adaptive changes of the application following the general principles of the application and including common knowledge or conventional technical means in the art not disclosed by the application, and is not limited to the precise structure described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the application is only limited by the appended claims.

Claims

1. A method for detecting the volume of materials in a quantitative silo, characterized in that, include: Acquire at least two first point cloud information collected by at least two lidars respectively, wherein the at least two lidars are set at the discharge port of the quantitative silo; The at least two first point cloud information are fused to obtain the fused point cloud information of the material in the quantitative warehouse; Based on the fused point cloud information, the volume information of the material in the quantitative bin is determined.

2. The method according to claim 1, characterized in that, The process of fusing the at least two first point cloud information sets to obtain the fused point cloud information of the material in the quantitative storage bin includes: Global alignment and redundant point cloud removal are performed on the at least two first point cloud information to obtain second point cloud information; The second point cloud information is subjected to hole filling and surface reconstruction processing to obtain the fused point cloud information.

3. The method according to claim 1, characterized in that, Before fusing the at least two first point cloud information sets to obtain the fused point cloud information of the material in the quantitative storage bin, the method further includes: Based on the size information of the quantitative bin, the at least two first point cloud information are filtered to obtain at least two first point cloud information of the filtered material. The at least two first point cloud information are filtered and denoised to obtain the processed at least two first point cloud information.

4. The method according to any one of claims 1-3, characterized in that, Determining the volume information of the material in the quantitative bin based on the fused point cloud information includes: The fused point cloud information is voxelized to obtain multiple point cloud information blocks; The volume of each point cloud information block is calculated to obtain the volume information of the material in the quantitative bin.

5. The method according to any one of claims 1-3, characterized in that, The number of the at least two lidars is five, and the number of the feeding ports is four; The first lidar is positioned at the center of the four feed ports; The remaining four lidars are respectively set at a first distance from the center position, and the distance between any two adjacent lidars is equal.

6. The method according to any one of claims 1-3, characterized in that, The acquisition of at least two first point cloud information items obtained by at least two lidar sensors includes: For each lidar, based on the lidar's installation information, the lidar's coordinate system is aligned to the same coordinate system; Based on the time synchronization information, obtain the first point cloud information collected by each lidar.

7. The method according to any one of claims 1-3, characterized in that, The acquisition of at least two first point cloud information items obtained by at least two lidar sensors includes: Obtain the weight status of the quantitative bin; When the counterweight status indicates that the quantitative bin is in the counterweight complete state, at least two first point cloud information collected by at least two lidars are acquired.

8. A material volume detection device for a quantitative silo, characterized in that, include: An acquisition module is used to acquire at least two first point cloud information collected by at least two lidars respectively, wherein the at least two lidars are set at the discharge port of the quantitative silo; The fusion module is used to fuse the at least two first point cloud information to obtain the fused point cloud information of the material in the quantitative warehouse; The determination module is used to determine the volume information of the material in the quantitative bin based on the fused point cloud information.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A material volume detection system, characterized in that, include: A quantitative chamber, at least two lidar sensors mounted on the quantitative chamber, and electronic equipment; The at least two lidars are used to collect at least two first point cloud information of the quantitative chamber; The electronic device is used to perform the method as described in any one of claims 1-7.

Images