Automatic pile volume measuring method and device

Abstract

The invention relates to an automatic stack volume measurement method and device, and the method comprises the steps: scanning a target stack and a container environment for storing the target stack, so as to collect scanning point cloud data and sampling point cloud data; carrying out point cloud registration operation on the scanning point cloud data and the sampling point cloud data to obtain registration point cloud data, and carrying out cutting and adsorption matching on the registration point cloud data to obtain a boundary point set at the boundary of the target heap body and the container environment; and based on the boundary point set, carrying out denoising and point cloud segmentation on the clipped registration point cloud data to obtain a segmentation point cloud, and extracting complete surface information of the target heap according to the boundary point set and the segmentation point cloud to carry out surface reconstruction operation so as to generate a closed triangular mesh corresponding to the target heap. And calculating the volume of the pile body corresponding to the target pile body by using the closed triangular mesh. Therefore, the problems that an existing point cloud volume measurement technology is insufficient in accuracy, difficult in boundary processing, high in data acquisition requirement, limited in application scene and the like are solved.

Description

Technical Field

[0001] The present application relates to the technical field of point cloud data processing, and in particular to an automated pile volume measurement method and device. Background Art

[0002] Traditional volume measurement methods are mostly based on manual measurement or two-dimensional image analysis, which are not only time-consuming and labor-intensive, but also difficult to guarantee accuracy. In recent years, with the development of laser scanning technology, volume measurement methods based on Lidar point cloud data have gradually become mainstream, among which 2.5D point cloud volume measurement technology has been widely used.

[0003] 2.5D point cloud volume measurement mainly relies on projecting point cloud data in three-dimensional space onto a two-dimensional plane to form a so-called "height map". In this method, each projection point retains its corresponding height information, so that the approximate volume of the object can be calculated. The advantage of this method is that its calculation is relatively simple and it is suitable for piles with relatively regular geometric shapes. 2.5D technology can effectively reduce the computational complexity by reducing the dimension of point cloud data, so it can also operate well in resource-constrained environments.

[0004] However, 2.5D point cloud volume measurement technology also has the following significant limitations and defects:

[0005] 1. Insufficient accuracy: 2.5D technology ignores the complex geometric structure of the pile. Especially when the pile surface is undulating or there are suspended areas, the accuracy of the measurement results is significantly reduced;

[0006] 2. Difficulty in handling boundary issues: In actual measurement scenarios, the surface of the pile may be partially blocked by other objects or in contact with containers (such as the inner wall of a warehouse). 2.5D technology cannot accurately handle these complex situations, especially in boundary processing, which further affects the reliability of the measurement results.

[0007] 3. High data collection requirements: Although 2.5D technology is relatively simple to calculate, its accuracy still depends on high-quality point cloud data, which means that during the data collection process, the point cloud density must be high enough and the coverage must be comprehensive. If there are omissions or noise in the data, the volume measurement results may be inaccurate.

[0008] 4. Limited application scenarios: 2.5D technology is only applicable to piles with relatively regular boundaries and surfaces, and cannot meet the measurement requirements in some complex scenarios. For example, the surface shape and stacking pattern of materials piled in a container may be very complex. In these cases, 2.5D technology is difficult to provide accurate measurement results.

[0009] In summary, the existing point cloud volume measurement technology lacks accuracy, is difficult to handle boundary problems, has high data collection requirements, and has limited application scenarios, which need to be solved urgently. Summary of the invention

[0010] The present application provides an automated pile volume measurement method and device to solve the problems of existing point cloud volume measurement technology, such as insufficient accuracy, difficulty in handling boundary problems, high data collection requirements, and limited application scenarios.

[0011] The first aspect of the present application provides an automated pile volume measurement method, comprising the following steps: scanning a target pile and a container environment storing the target pile to collect scanning point cloud data corresponding to the target pile and sampling point cloud data corresponding to the container environment; performing point cloud registration operations on the scanning point cloud data and the sampling point cloud data to obtain registration point cloud data, and performing cropping and adsorption matching operations on the registration point cloud data to obtain a boundary point set at the boundary of the target pile and the container environment; based on the boundary point set, denoising and point cloud segmenting the cropped registration point cloud data to obtain a segmented point cloud, and extracting complete surface information of the target pile based on the boundary point set and the segmented point cloud, and performing surface reconstruction operations based on the complete surface information to generate a closed triangular mesh corresponding to the target pile, and using the closed triangular mesh to calculate the pile volume corresponding to the target pile.

[0012] Optionally, in one embodiment of the present application, the scanning of the target pile and the container environment for storing the target pile to collect scanning point cloud data corresponding to the target pile and sampling point cloud data corresponding to the container environment includes: determining the container environment of the target pile, and constructing a container three-dimensional model corresponding to the container environment based on a preset Poisson reconstruction strategy or an Alpha Wrap reconstruction strategy; sampling the container three-dimensional model to generate sampling point cloud data of the container three-dimensional model; and performing a preset comprehensive scanning operation on the target pile by a preset laser scanner to obtain the scanning point cloud data corresponding to the target pile.

[0013] Optionally, in one embodiment of the present application, the point cloud registration operation is performed on the scanned point cloud data and the sampled point cloud data to obtain the registered point cloud data, and the registered point cloud data is cropped and adsorbed and matched to obtain a boundary point set at the boundary of the target pile and the container environment, including: performing a coarse registration operation on the scanned point cloud data and the sampled point cloud data to obtain a coarse registration result; based on the coarse registration result, performing a fine registration operation on the scanned point cloud data and the sampled point cloud data to obtain the registered point cloud data; constructing a spatial index tree through the aligned bounding box of the three-dimensional model of the container, and using the spatial index tree to crop the registered point cloud data to obtain point cloud cropped data; based on a preset threshold matching algorithm, matching the point cloud cropped data and the three-dimensional model of the container until a preset number of times is reached to obtain the boundary point set.

[0014] Optionally, in one embodiment of the present application, based on the boundary point set, denoising and point cloud segmenting are performed on the cropped registered point cloud data to obtain a segmented point cloud, and the complete surface information of the target pile is extracted according to the boundary point set and the segmented point cloud, including: denoising the point cloud cropped data to obtain denoised point cloud data, and segmenting the denoised point cloud data to obtain the segmented point cloud; drawing a perpendicular line from each point in the boundary point set to the nearest surface of the container inner wall of the container environment, extending the perpendicular line beyond the container space, and sampling the perpendicular line at equal intervals to obtain an extended point set; merging the boundary point set, the extended point set and the segmented point cloud to obtain a first merged point cloud, and performing Alpha Wrap three-dimensional reconstruction on the first merged point cloud to generate a cropping surface corresponding to the target pile; based on the cropping surface, extracting the contact area between the container inner wall of the container environment and the pile surface of the target pile, and obtaining the complete surface information of the target pile through the contact area.

[0015] Optionally, in one embodiment of the present application, the surface reconstruction operation is performed using the complete surface information to generate a closed triangular mesh corresponding to the target body, and the closed triangular mesh is used to calculate the body volume corresponding to the target body, including: sampling the contact surface to obtain contact surface point cloud data, and merging the boundary point set, the extension point set, the segmentation point cloud and the contact surface point cloud data to obtain a second merged point cloud; performing Alpha Wrap reconstruction on the second merged point cloud to obtain a first closed three-dimensional model volume; merging the boundary point set and the segmentation point cloud to obtain a third merged point cloud, and performing Alpha Wrap reconstruction on the third merged point cloud to obtain a second closed three-dimensional model volume; based on the first closed three-dimensional model volume and the second closed three-dimensional model volume, calculating the body volume corresponding to the target body.

[0016] A second aspect of the present application provides an automated pile volume measurement device, comprising: an acquisition module, used to scan a target pile and a container environment for storing the target pile, so as to acquire scanning point cloud data corresponding to the target pile and sampling point cloud data corresponding to the container environment; an adsorption matching module, used to perform point cloud registration operations on the scanning point cloud data and the sampling point cloud data to obtain registration point cloud data, and to perform cropping and adsorption matching operations on the registration point cloud data to obtain a boundary point set at the boundary between the target pile and the container environment; a calculation module, used to perform denoising and point cloud segmentation on the cropped registration point cloud data based on the boundary point set to obtain a segmented point cloud, and to extract complete surface information of the target pile based on the boundary point set and the segmented point cloud, and to perform surface reconstruction operations based on the complete surface information to generate a closed triangular mesh corresponding to the target pile, and to calculate the pile volume corresponding to the target pile using the closed triangular mesh.

[0017] Optionally, in one embodiment of the present application, the acquisition module includes: a first reconstruction unit, used to determine the container environment of the target pile, and construct a container three-dimensional model corresponding to the container environment based on a preset Poisson reconstruction strategy or an Alpha Wrap reconstruction strategy; a generation unit, used to sample the container three-dimensional model and generate sampling point cloud data of the container three-dimensional model; a scanning unit, used to perform a preset comprehensive scanning operation on the target pile through a preset laser scanner to obtain scanning point cloud data corresponding to the target pile.

[0018] Optionally, in one embodiment of the present application, the adsorption matching module includes: a coarse registration unit, used to perform a coarse registration operation on the scanned point cloud data and the sampled point cloud data to obtain a coarse registration result; a fine registration unit, used to perform a fine registration operation on the scanned point cloud data and the sampled point cloud data based on the coarse registration result to obtain the registered point cloud data; a cropping unit, used to construct a spatial index tree through the aligned bounding box of the container three-dimensional model, and use the spatial index tree to crop the registered point cloud data to obtain point cloud cropped data; a matching unit, used to perform a matching operation on the point cloud cropped data and the container three-dimensional model based on a preset threshold matching algorithm until a preset number of times is reached to obtain the boundary point set.

[0019] Optionally, in one embodiment of the present application, the computing module includes: a denoising unit, configured to perform denoising processing on the point cloud cropping data to obtain denoised point cloud data, and to segment the denoised point cloud data to obtain the segmented point cloud; an acquisition unit, configured to draw a perpendicular line from each point in the boundary point set to the nearest surface of the container inner wall of the container environment, and extend the perpendicular line beyond the container space, and perform equal-interval sampling on the perpendicular line to obtain an extended point set; a first merging unit, configured to merge the boundary point set, the extended point set and the segmented point cloud to obtain a first merged point cloud, and perform Alpha Wrap three-dimensional reconstruction on the first merged point cloud to generate a cropping surface corresponding to the target pile; an extraction unit, configured to extract a contact area between the container inner wall of the container environment and the pile surface of the target pile based on the cropping surface, and obtain complete surface information of the target pile through the contact area.

[0020] Optionally, in one embodiment of the present application, the calculation module also includes: a second merging unit, used to sample the contact surface to obtain contact surface point cloud data, and merge the boundary point set, the extension point set, the segmentation point cloud and the contact surface point cloud data to obtain a second merged point cloud; a second reconstruction unit, used to perform Alpha Wrap reconstruction on the second merged point cloud to obtain a first closed three-dimensional model volume; a third merging unit, used to merge the boundary point set and the segmentation point cloud to obtain a third merged point cloud, and perform Alpha Wrap reconstruction on the third merged point cloud to obtain a second closed three-dimensional model volume; a computing unit, used to calculate the pile volume corresponding to the target pile based on the first closed three-dimensional model volume and the second closed three-dimensional model volume.

[0021] A third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the automated stack volume measurement method as described in the above embodiment.

[0022] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned automated pile volume measurement method.

[0023] A fifth aspect of the present application provides a computer program product, including a computer program, which is executed to implement the above-mentioned automated stack volume measurement method.

[0024] Therefore, the embodiments of the present application have the following beneficial effects:

[0025] The embodiments of the present application can collect the scanning point cloud data corresponding to the target pile and the sampling point cloud data corresponding to the container environment by scanning the target pile and the container environment; perform point cloud registration operation on the scanning point cloud data and the sampling point cloud data to obtain the registration point cloud data, and perform cropping and adsorption matching operation on the registration point cloud data to obtain the boundary point set at the boundary of the target pile and the container environment; based on the boundary point set, perform denoising and point cloud segmentation on the cropped registration point cloud data to obtain the segmented point cloud, and extract the complete surface information of the target pile according to the boundary point set and the segmented point cloud, and perform surface reconstruction operation through the complete surface information to generate a closed triangular mesh corresponding to the target pile, and use the closed triangular mesh to calculate the pile volume corresponding to the target pile. The present application realizes the accurate measurement of the pile volume through data preprocessing, point cloud registration, contact surface extraction and surface reconstruction, and significantly improves the automation and accuracy of the measurement. Therefore, the existing point cloud volume measurement technology solves the problems of insufficient accuracy, difficulty in handling boundary problems, high data collection requirements, and limited application scenarios.

[0026] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0028] Figure 1 A flowchart of an automated stack volume measurement method provided according to an embodiment of the present application;

[0029] Figure 2 A schematic diagram of the relationship between various targets involved in an adsorption matching operation provided in one embodiment of the present application;

[0030] Figure 3 A schematic diagram of execution logic of an automated stack volume measurement method provided for one embodiment of the present application;

[0031] Figure 4 This is an exemplary diagram of an automated stack volume measuring device according to an embodiment of the present application;

[0032] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0033] Among them, 10 is an automated stack volume measuring device; 100 is a collection module, 200 is an adsorption matching module, 300 is a calculation module; 501 is a memory, 502 is a processor, and 503 is a communication interface. DETAILED DESCRIPTION

[0034] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0035] The following describes the automated pile volume measurement method and device of the embodiment of the present application with reference to the accompanying drawings. In view of the problems mentioned in the above background technology, the present application provides an automated pile volume measurement method, in which the target pile and the container environment storing the target pile are scanned to collect the scanning point cloud data corresponding to the target pile and the sampling point cloud data corresponding to the container environment; the scanning point cloud data and the sampling point cloud data are subjected to point cloud registration operation to obtain the registration point cloud data, and the registration point cloud data are clipped and adsorbed and matched to obtain the boundary point set at the boundary of the target pile and the container environment; based on the boundary point set, the clipped registration point cloud data is denoised and point cloud segmented to obtain the segmented point cloud, and the complete surface information of the target pile is extracted according to the boundary point set and the segmented point cloud, and the surface reconstruction operation is performed through the complete surface information to generate the closed triangular mesh corresponding to the target pile, and the pile volume corresponding to the target pile is calculated using the closed triangular mesh. The present application realizes the accurate measurement of the pile volume through data preprocessing, point cloud registration, contact surface extraction and surface reconstruction operations, and significantly improves the automation and accuracy of the measurement. This solves the problems of existing point cloud volume measurement technology, such as insufficient accuracy, difficulty in handling boundary problems, high data collection requirements, and limited application scenarios.

[0036] Specifically, Figure 1 A flow chart of an automated stack volume measurement method provided in an embodiment of the present application.

[0037] like Figure 1 As shown, the automatic pile volume measurement method includes the following steps:

[0038] In step S101 , a target pile and a container environment storing the target pile are scanned to collect scanning point cloud data corresponding to the target pile and sampling point cloud data corresponding to the container environment.

[0039] In the embodiment of the present application, the target pile and the container environment where the pile is stored can be scanned by using Lidar, wherein the specific requirements for data collection by Lidar scanning are as follows:

[0040] 1. The scanning of the pile must cover the entire visible surface of the object being measured, as well as the junction between the visible surface and the inner wall of the container;

[0041] 2. The point density on the surface of the pile in the point cloud data is uniform and without omissions;

[0042] 3. The target pile does not come into contact with irrelevant objects;

[0043] 4. The closure of the container model. Taking the warehouse scene as an example, the 3D model of the inner wall of the warehouse needs to be topologically closed, the doors and windows are filled with geometric shapes, and the outer surface of the column is part of the inner wall of the container.

[0044] In addition, the embodiments of the present application can also construct a three-dimensional model of the inner wall of the container through engineering drawing modeling or point cloud scanning modeling, and sample the inner wall point cloud data (i.e., sampling point cloud data) from the three-dimensional model of the inner wall of the container.

[0045] It should be noted that the pile in the embodiment of the present application is the processing object, and its surface information is obtained through laser scanning to form scanning point cloud data; the container is a closed geometric body where the pile is located, and its inner wall is attached to the invisible surface of the pile. The container model involved in the embodiment of the present application is the three-dimensional model data of the container, which is used to limit the boundary of the pile to ensure that the measurement result is within the closed range of the container.

[0046] Optionally, in one embodiment of the present application, a target pile and a container environment for storing the target pile are scanned to collect scanning point cloud data corresponding to the target pile and sampling point cloud data corresponding to the container environment, including: determining the container environment of the target pile, and constructing a container three-dimensional model corresponding to the container environment based on a preset Poisson reconstruction strategy or an Alpha Wrap reconstruction strategy; sampling the container three-dimensional model to generate sampling point cloud data of the container three-dimensional model; and performing a preset comprehensive scanning operation on the target pile through a preset laser scanner to obtain scanning point cloud data corresponding to the target pile.

[0047] Specifically, for containers that can be emptied, the embodiments of the present application can use a laser scanner to scan the inner wall of the container, based on Poisson reconstruction or Alpha Wrap reconstruction, and after processing by modeling software, a usable closed container three-dimensional model (i.e., container model) can be obtained; if the container information cannot be obtained through scanning, the embodiments of the present application can obtain the corresponding container model by manual modeling from BIM or engineering drawings.

[0048] It should be noted that if the stack is located in an open space, the embodiment of the present application needs to determine the virtual container that can accommodate the stack based on the surface in contact with the stack.

[0049] Secondly, the embodiments of the present application can sample the container model by selecting a sampling algorithm such as random uniform sampling or grid sampling, and the sampling density matches the density of the scanned point cloud to facilitate subsequent point cloud alignment.

[0050] Afterwards, the embodiment of the present application may use a laser scanner to perform a comprehensive scan on the pile to ensure that the visible surface of the pile is covered, and the generated scanning point cloud data should ensure that the point distribution density is uniform to avoid omissions.

[0051] Therefore, the embodiments of the present application reduce manual intervention, improve the degree of automation of volume measurement, and enhance measurement efficiency and accuracy through laser scanning and point cloud data processing.

[0052] In step S102, a point cloud registration operation is performed on the scanned point cloud data and the sampled point cloud data to obtain registered point cloud data, and a cropping and adsorption matching operation is performed on the registered point cloud data to obtain a boundary point set at the boundary of the target stack and the container environment.

[0053] Furthermore, the embodiments of the present application also need to register the scanned point cloud data with the sampled point cloud data obtained by sampling the container model, so that the scanned point cloud is aligned with the container model to obtain the registered point cloud data; and then the registered point cloud data is cropped and adsorption matched to remove the point set located outside the container in the scanned point cloud, and obtain the boundary point set at the junction of the pile and the container, so as to ensure the accuracy of the boundary information of the pile surface for subsequent volume calculation.

[0054] Optionally, in one embodiment of the present application, a point cloud registration operation is performed on the scanned point cloud data and the sampled point cloud data to obtain registered point cloud data, and the registered point cloud data is cropped and adsorbed and matched to obtain a boundary point set at the boundary of the target pile and the container environment, including: performing a coarse registration operation on the scanned point cloud data and the sampled point cloud data to obtain a coarse registration result; based on the coarse registration result, performing a fine registration operation on the scanned point cloud data and the sampled point cloud data to obtain registered point cloud data; constructing a spatial index tree through the alignment bounding box of the container three-dimensional model, and using the spatial index tree to crop the registered point cloud data to obtain point cloud cropped data; based on a preset threshold matching algorithm, performing a matching operation on the point cloud cropped data and the container three-dimensional model until a preset number of times is reached to obtain a boundary point set.

[0055] It should be noted that the embodiment of the present application can register the scanned point cloud data with the point cloud data sampled from the container model to align the scanned point cloud data to the container. The specific registration steps are as follows:

[0056] Step 1: Rough registration

[0057] The embodiment of the present application uses the 4PCS (4-point coplanar consistency) algorithm to perform preliminary alignment of the scanned point cloud data and the container model sampling point cloud (i.e., the sampling point cloud data), that is, in the rough registration stage, the embodiment of the present application mainly ensures that the two are roughly aligned in three-dimensional space;

[0058] Step 2: Fine alignment

[0059] After completing the rough registration, the embodiment of the present application may use an ICP (Iterative Closest Point) algorithm to perform precise registration to obtain registration point cloud data, further improving the alignment accuracy of the two to ensure the accuracy of the measurement results.

[0060] Afterwards, the embodiment of the present application can also use the closed space determined by the container model as a boundary to crop the registration point cloud data to remove the point set outside the container in the scanned point cloud. The specific cropping steps are:

[0061] Step 1: Build a spatial index tree

[0062] Use the AABB (Axis Aligned Bounding Box) of the container model to build a spatial index tree to provide support for subsequent clipping operations;

[0063] Step 2: Point cloud clipping

[0064] The constructed spatial index tree is used to traverse the scanned point cloud data and determine whether each point in the scanned point cloud data is located inside the container model. The point cloud data outside the container is discarded, and only the point cloud within the closed space of the container is retained to obtain the point cloud clipping data, thereby ensuring the accuracy of the final volume measurement.

[0065] Furthermore, the embodiment of the present application also needs to perform adsorption matching operation on the point cloud clipping data, such as Figure 2 In order to facilitate those skilled in the art to understand the process of adsorption matching operation, the embodiment of the present application may first define a threshold matching algorithm, and the input parameters accepted by the algorithm are shown in Table 1:

[0066] Table 1

[0067] Parameter name Parameter Type Remark Source point cloud Point Cloud Target point cloud or 3D model Point cloud or 3D model Threshold Floating point This parameter is greater than 0

[0068] It should be noted that the output result of the threshold matching algorithm is the point set in the source point cloud that matches the target point cloud or 3D model. When the matching target is a point cloud, the output point set is the points in the source point cloud whose distance to the nearest point in the target point cloud is less than or equal to the threshold; when the matching target is a 3D model, the output point set is the points in the source point cloud whose distance to the nearest surface of the target 3D model is less than or equal to the threshold.

[0069] In order to determine the nearest point or surface between each point in the source point cloud and the target, the embodiment of the present application needs to first construct a KD tree for the target point cloud, or construct an AABB tree for the target three-dimensional model.

[0070] During the actual execution of the adsorption matching operation, in order to obtain the boundary point set, the embodiment of the present application can first use the threshold matching algorithm to match the scanned point cloud (i.e., the point cloud cropping data) after the cropping in the previous step with the container model, and obtain a new point cloud composed of the boundary point set and the container inner wall points obtained in the scanning point cloud process; then, the new point cloud is matched again with the remaining point set in the source point cloud, and the obtained point set is the boundary point set.

[0071] In actual data processing, the above matching process needs to be executed 3 to 5 times, each based on different thresholds, and the final boundary point set is merged from the results of multiple matching. The initial threshold can be the average point distance of the scanned point cloud after cropping, and n times the initial threshold is used as the threshold for this matching in the loop.

[0072] In addition, the adsorption matching step also requires exporting the remaining point set after removing the container inner wall points and boundary points from the scanned point cloud, which usually includes the main part of the target pile point cloud, noise points and the main part of irrelevant object surface points.

[0073] Therefore, the embodiment of the present application obtains a boundary point set at the boundary of the target pile and the container environment through point cloud registration, clipping and adsorption matching operations, thereby providing reliable data support for the calculation of the pile volume.

[0074] In step S103, based on the boundary point set, the cropped registered point cloud data is denoised and point cloud segmented to obtain a segmented point cloud, and the complete surface information of the target pile is extracted based on the boundary point set and the segmented point cloud, and a surface reconstruction operation is performed using the complete surface information to generate a closed triangular mesh corresponding to the target pile, and the closed triangular mesh is used to calculate the pile volume corresponding to the target pile.

[0075] Furthermore, the embodiments of the present application also need to ensure the quality of the remaining point cloud by statistically analyzing the density of points in the neighborhood and eliminating isolated points and abnormal point cloud data; secondly, the denoised point cloud is segmented based on the DBSCAN (Density-based spatial clustering of applications with noise) algorithm to distinguish the point cloud of the pile from that of irrelevant objects.

[0076] Afterwards, if Figure 3 As shown, the embodiment of the present application can extract the contact area between the pile surface and the inner wall of the container by analyzing the boundary point set to obtain the complete surface information of the pile.

[0077] Finally, the embodiment of the present application can perform Alpha Wrap surface reconstruction on the complete pile point cloud to obtain a closed triangular mesh, and calculate the pile volume (ie, the spatial volume occupied by the irregular geometric body of the pile) through the voxel integration method and the closed triangular mesh.

[0078] Therefore, the embodiments of the present application achieve accurate measurement of the volume of the pile through innovative data preprocessing, point cloud registration, cropping, adsorption matching, denoising, segmentation, contact surface extraction and surface reconstruction operations, and significantly improve the degree of automation and accuracy of the measurement.

[0079] Optionally, in one embodiment of the present application, based on the boundary point set, the cropped registered point cloud data is denoised and the point cloud is segmented to obtain a segmented point cloud, and the complete surface information of the target pile is extracted based on the boundary point set and the segmented point cloud, including: denoising the point cloud cropped data to obtain denoised point cloud data, and segmenting the denoised point cloud data to obtain a segmented point cloud; drawing a perpendicular line from each point in the boundary point set to the nearest surface of the container inner wall of the container environment, and extending it beyond the container space, and sampling the perpendicular line at equal intervals to obtain an extended point set; merging the boundary point set, the extended point set and the segmented point cloud to obtain a first merged point cloud, and performing Alpha Wrap three-dimensional reconstruction on the first merged point cloud to generate a cropping surface corresponding to the target pile; based on the cropping surface, extracting the contact area between the container inner wall of the container environment and the pile surface of the target pile, and obtaining the complete surface information of the target pile through the contact area.

[0080] Specifically, the embodiments of the present application may also use SOR (Statistical Outlier Removal, a denoising algorithm based on spatial distribution) to denoise the point cloud cropping data to eliminate noise points that may be generated during scanning by a laser point cloud scanner to obtain denoised point cloud data.

[0081] Secondly, the embodiment of the present application also needs to separate the main part of the target pile point cloud from the main part of the surface of irrelevant objects based on the DBSCAN clustering algorithm. The neighborhood radius eps selects 8 to 16 times the average point distance, and the minimum number of points minPts selects 1 / 2 of the average number of points within the eps radius; in order to determine which clusters belong to the target pile, the embodiment of the present application can first screen out the point clusters with too high center of gravity and too few points. These two parameters are pre-configured according to the empirical value of the measurement scene, and then accept user input, and select the target pile through the graphical interface.

[0082] Afterwards, the embodiment of the present application can measure the close contact between the invisible surface portion (ie, the contact surface) of the target pile and the container by cutting out the three-dimensional model of the inner wall of the container.

[0083] In the specific implementation process, the pile point cloud needs to be preprocessed before the cropping step; the boundary point set obtained in the adsorption matching operation may contain the boundary between irrelevant objects and the inner wall of the container, which needs to be eliminated first, and then a threshold matching is performed. The boundary point set is used as the source point cloud, and the pile point cloud obtained in the segmentation step (i.e., the segmentation point cloud) is used as the target point cloud. The threshold is selected as 3 to 5 times the average point distance to obtain the boundary point set of the target pile and the inner wall of the container.

[0084] In addition, the specific steps of creating the cutting surface in the embodiment of the present application are as follows:

[0085] Step 1: Draw a perpendicular line from each point in the boundary point set to the nearest surface of the inner wall of the container and extend it beyond the container space, and sample the perpendicular lines at equal intervals to obtain the extended point set;

[0086] Step 2: Merge the boundary point set, the extended point set and the target body point cloud to obtain a first merged point cloud;

[0087] Step 3: Perform Alpha Wrap 3D reconstruction on the first merged point cloud to obtain an appropriate clipping surface (actually a clipping volume). Among the Alpha Wrap algorithm parameters, the alpha value can be selected as about 2 times the average point distance, and the offset can be selected as 1 / 2 of the average point distance.

[0088] After obtaining the clipping plane, the container model is segmented using the clipping plane to obtain three connected parts located above the target pile, inside the clipping body, and below the target pile (in fact, there may be more than three connected parts). In the actual execution process, strictly speaking, the boundaries of all three parts are not on the clipping body. The embodiments of the present application can distinguish these three parts through the following steps:

[0089] Step 1: Traverse all vertices of the connected part. For vertex v, determine the relative position of v and the clipping volume. If v is inside the clipping volume, the part is inside the clipping volume and the algorithm ends. If v is outside the clipping volume, execute step 3. If v is exactly on the surface of the clipping volume, continue the loop.

[0090] Step 2: If the algorithm in step 1 traverses until the last vertex, the part is inside the clipping volume;

[0091] Step 3: Treat all vertices of the connected part as point clouds, perform threshold matching with the boundary point set, and accumulate the vectors composed of each boundary point-vertex sequence pair to obtain the feature vector of the connected part;

[0092] Step 4: The connected part corresponding to the small z value (generally less than 0) is located below the target pile, while the connected part corresponding to the large z value (generally greater than 0) is located above the target pile.

[0093] Finally, the embodiment of the present application can retain the connected portion located below the target pile and use it as the contact surface (ie, complete surface information).

[0094] Therefore, the embodiments of the present application can effectively handle the relationship between the pile and the complex environment by introducing the steps of registration, cropping, and segmentation of the container model and the pile, so as to ensure that the measurement results are limited to the target pile and avoid interference with the volume calculation by the external environment of the container or irrelevant objects.

[0095] Optionally, in one embodiment of the present application, a surface reconstruction operation is performed using complete surface information to generate a closed triangular mesh corresponding to the target body, and the closed triangular mesh is used to calculate the body volume corresponding to the target body, including: sampling the contact surface to obtain contact surface point cloud data, and merging the boundary point set, extension point set, segmentation point cloud and contact surface point cloud data to obtain a second merged point cloud; performing Alpha Wrap reconstruction on the second merged point cloud to obtain a first closed three-dimensional model volume; merging the boundary point set and the segmentation point cloud to obtain a third merged point cloud, and performing Alpha Wrap reconstruction on the third merged point cloud to obtain a second closed three-dimensional model volume; based on the first closed three-dimensional model volume and the second closed three-dimensional model volume, the body volume corresponding to the target body is calculated.

[0096] It should be noted that due to the characteristics of Alpha Wrap reconstruction, part of the seam with the target pile will be lost, and the embodiment of the present application also needs to extend the boundary point set to fill the seam.

[0097] Specifically, the embodiments of the present application need to merge the extended point set, the boundary point set, the main part of the target pile point cloud (i.e., the segmentation point cloud) and the point cloud obtained by sampling the contact surface (i.e., the contact surface point cloud data) to obtain a second merged point cloud, and perform Alpha Wrap reconstruction on the second merged point cloud to obtain a closed three-dimensional model with a volume of (i.e., the first closed three-dimensional model volume) V1.

[0098] In the ideal scanning data, the surface of the target object is a layer without thickness, but in reality it often has thickness, which causes errors in volume measurement. Therefore, the embodiment of the present application also needs to merge the boundary point set and the main part of the target body point cloud (i.e., the third merged point cloud) to perform Alpha Wrap reconstruction, and the calculated closed 3D model volume is (i.e., the second closed 3D model volume) V2.

[0099] Furthermore, based on the first closed three-dimensional model volume and the second closed three-dimensional model volume, the volume of the target stack corresponding to the target stack, that is, the volume V1-1 / 2*V2 of the target stack is calculated.

[0100] Therefore, in the processes of registration, cropping and segmentation, the embodiments of the present application ensure the integrity of the point cloud data and filter irrelevant points, thereby avoiding the accumulation of errors as much as possible and providing a more reliable volume measurement result.

[0101] According to the automated pile volume measurement method proposed in the embodiment of the present application, the target pile and the container environment storing the target pile are scanned to collect the scanning point cloud data corresponding to the target pile and the sampling point cloud data corresponding to the container environment; the scanning point cloud data and the sampling point cloud data are subjected to point cloud registration operations to obtain the registration point cloud data, and the registration point cloud data are clipped and adsorbed and matched to obtain the boundary point set at the boundary of the target pile and the container environment; based on the boundary point set, the clipped registration point cloud data is denoised and point cloud segmented to obtain the segmented point cloud, and the complete surface information of the target pile is extracted according to the boundary point set and the segmented point cloud, and the surface reconstruction operation is performed through the complete surface information to generate a closed triangular mesh corresponding to the target pile, and the pile volume corresponding to the target pile is calculated using the closed triangular mesh. The present application realizes the accurate measurement of the pile volume through operations such as data preprocessing, point cloud registration, contact surface extraction and surface reconstruction, and significantly improves the degree of automation and accuracy of measurement.

[0102] Next, an automated stack volume measuring device according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0103] Figure 4 It is a block diagram of an automated stack volume measuring device according to an embodiment of the present application.

[0104] like Figure 4 As shown, the automatic stack volume measuring device 10 includes: a collection module 100 , an adsorption matching module 200 and a calculation module 300 .

[0105] The acquisition module 100 is used to scan the target pile and the container environment storing the target pile, so as to acquire the scanning point cloud data corresponding to the target pile and the sampling point cloud data corresponding to the container environment.

[0106] The adsorption matching module 200 is used to perform point cloud registration operations on the scanned point cloud data and the sampled point cloud data to obtain registered point cloud data, and to perform cropping and adsorption matching operations on the registered point cloud data to obtain a boundary point set at the boundary of the target pile and the container environment.

[0107] The computing module 300 is used to denoise and segment the cropped registered point cloud data based on the boundary point set to obtain the segmented point cloud, and to extract the complete surface information of the target pile based on the boundary point set and the segmented point cloud, and to perform surface reconstruction operations based on the complete surface information to generate a closed triangular mesh corresponding to the target pile, and to calculate the pile volume corresponding to the target pile using the closed triangular mesh.

[0108] Optionally, in one embodiment of the present application, the acquisition module 100 includes: a first reconstruction unit, a generation unit and a scanning unit.

[0109] The first reconstruction unit is used to determine the container environment of the target stack, and to construct a three-dimensional container model corresponding to the container environment based on a preset Poisson reconstruction strategy or an Alpha Wrap reconstruction strategy.

[0110] The generating unit is used to sample the three-dimensional model of the container and generate sampling point cloud data of the three-dimensional model of the container.

[0111] The scanning unit is used to perform a preset comprehensive scanning operation on the target pile body through a preset laser scanner to obtain scanning point cloud data corresponding to the target pile body.

[0112] Optionally, in one embodiment of the present application, the adsorption matching module 200 includes: a coarse registration unit, a fine registration unit, a cropping unit and a matching unit.

[0113] The coarse registration unit is used to perform a coarse registration operation on the scanned point cloud data and the sampled point cloud data to obtain a coarse registration result.

[0114] The fine registration unit is used to perform a fine registration operation on the scanned point cloud data and the sampled point cloud data based on the rough registration result to obtain the registered point cloud data.

[0115] The clipping unit is used to construct a spatial index tree through the aligned bounding box of the container three-dimensional model, and to clip the registration point cloud data using the spatial index tree to obtain point cloud clipping data.

[0116] The matching unit is used to perform a matching operation on the point cloud clipping data and the container three-dimensional model based on a preset threshold matching algorithm until a preset number of times is reached to obtain a boundary point set.

[0117] Optionally, in one embodiment of the present application, the calculation module 300 includes: a denoising unit, an acquisition unit, a first merging unit and an extraction unit.

[0118] The denoising unit is used to perform denoising on the point cloud cropping data to obtain denoised point cloud data, and to segment the denoised point cloud data to obtain segmented point clouds.

[0119] The acquisition unit is used to draw a perpendicular line from each point in the boundary point set to the nearest surface of the container inner wall of the container environment, extend the perpendicular line beyond the container space, and perform equal-interval sampling on the perpendicular line to obtain an extended point set.

[0120] The first merging unit is used to merge the boundary point set, the extended point set and the segmented point cloud to obtain a first merged point cloud, and perform Alpha Wrap three-dimensional reconstruction on the first merged point cloud to generate a clipping surface corresponding to the target pile.

[0121] The extraction unit is used to extract the contact area between the inner wall of the container environment and the pile surface of the target pile based on the clipping surface, and obtain the complete surface information of the target pile through the contact area.

[0122] Optionally, in one embodiment of the present application, the calculation module 300 further includes: a second merging unit, a second reconstruction unit, a third merging unit and a calculation unit.

[0123] The second merging unit is used to sample the contact surface to obtain contact surface point cloud data, and merge the boundary point set, the extended point set, the segmentation point cloud and the contact surface point cloud data to obtain a second merged point cloud;

[0124] A second reconstruction unit is used to perform Alpha Wrap reconstruction on the second merged point cloud to obtain a first closed three-dimensional model volume;

[0125] a third merging unit, used for merging the boundary point set and the segmentation point cloud to obtain a third merged point cloud, and performing Alpha Wrap reconstruction on the third merged point cloud to obtain a second closed three-dimensional model volume;

[0126] The computing unit is used to calculate the volume of the target pile body based on the first closed three-dimensional model volume and the second closed three-dimensional model volume.

[0127] It should be noted that the above explanations and descriptions of the embodiment of the automated pile volume measurement method are also applicable to the automated pile volume measurement device of this embodiment, and will not be described in detail here.

[0128] The automatic pile volume measurement device proposed in the embodiment of the present application includes an acquisition module for scanning the target pile and the container environment for storing the target pile to collect the scanning point cloud data corresponding to the target pile and the sampling point cloud data corresponding to the container environment; an adsorption matching module for performing point cloud registration operations on the scanning point cloud data and the sampling point cloud data to obtain the registration point cloud data, and performing cropping and adsorption matching operations on the registration point cloud data to obtain the boundary point set at the boundary of the target pile and the container environment; a calculation module for denoising and point cloud segmentation of the cropped registration point cloud data based on the boundary point set to obtain the segmented point cloud, and extracting the complete surface information of the target pile according to the boundary point set and the segmented point cloud, and performing surface reconstruction operations through the complete surface information to generate a closed triangular mesh corresponding to the target pile, and using the closed triangular mesh to calculate the pile volume corresponding to the target pile. The present application realizes the accurate measurement of the pile volume through operations such as data preprocessing, point cloud registration, contact surface extraction and surface reconstruction, and significantly improves the automation and accuracy of the measurement.

[0129] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0130] A memory 501 , a processor 502 , and a computer program stored in the memory 501 and executable on the processor 502 .

[0131] When the processor 502 executes the program, the automatic stack volume measurement method provided in the above embodiment is implemented.

[0132] Furthermore, the electronic device further comprises:

[0133] The communication interface 503 is used for communication between the memory 501 and the processor 502 .

[0134] The memory 501 is used to store computer programs that can be executed on the processor 502 .

[0135] The memory 501 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0136] If the memory 501, the processor 502 and the communication interface 503 are implemented independently, the communication interface 503, the memory 501 and the processor 502 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0137] Optionally, in a specific implementation, if the memory 501, the processor 502 and the communication interface 503 are integrated on a chip, the memory 501, the processor 502 and the communication interface 503 can communicate with each other through an internal interface.

[0138] The processor 502 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0139] The embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned automated pile volume measurement method when executed by a processor.

[0140] The embodiment of the present application further provides a computer program product, including a computer program, and when the computer program is executed, it is used to implement the above-mentioned automatic stack volume measurement method.

[0141] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0142] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0143] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0144] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0145] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0146] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0147] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0148] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. An automated pile volume measurement method, characterized in that: The following steps are involved: Scanning a target pile and a container environment storing the target pile to collect scanning point cloud data corresponding to the target pile and sampling point cloud data corresponding to the container environment; Performing a point cloud registration operation on the scanned point cloud data and the sampled point cloud data to obtain registered point cloud data, and performing a cropping and adsorption matching operation on the registered point cloud data to obtain a boundary point set at the boundary of the target pile and the container environment; Based on the boundary point set, the cropped registered point cloud data is denoised and point cloud segmented to obtain a segmented point cloud, and the complete surface information of the target pile is extracted according to the boundary point set and the segmented point cloud, and a surface reconstruction operation is performed using the complete surface information to generate a closed triangular mesh corresponding to the target pile, and the closed triangular mesh is used to calculate the pile volume corresponding to the target pile.

2. The method according to claim 1, characterized in that The scanning of the target pile and the container environment storing the target pile to collect scanning point cloud data corresponding to the target pile and sampling point cloud data corresponding to the container environment includes: Determine the container environment of the target stack, and construct a three-dimensional container model corresponding to the container environment based on a preset Poisson reconstruction strategy or an Alpha Wrap reconstruction strategy; Sampling the three-dimensional model of the container to generate sampling point cloud data of the three-dimensional model of the container; A preset comprehensive scanning operation is performed on the target pile body by a preset laser scanner to obtain scanning point cloud data corresponding to the target pile body.

3. The method according to claim 2, characterized in that The step of performing a point cloud registration operation on the scanned point cloud data and the sampled point cloud data to obtain registered point cloud data, and performing a clipping and adsorption matching operation on the registered point cloud data to obtain a boundary point set at a boundary between the target pile and the container environment includes: Performing a coarse registration operation on the scanned point cloud data and the sampled point cloud data to obtain a coarse registration result; Based on the rough registration result, performing a fine registration operation on the scanned point cloud data and the sampled point cloud data to obtain the registered point cloud data; Building a spatial index tree through the aligned bounding box of the container three-dimensional model, and using the spatial index tree to clip the registered point cloud data to obtain point cloud clipping data; Based on a preset threshold matching algorithm, a matching operation is performed on the point cloud clipping data and the container three-dimensional model until a preset number of times is reached to obtain the boundary point set.

4. The method according to claim 3, characterized in that Based on the boundary point set, the cropped registration point cloud data is denoised and segmented to obtain a segmented point cloud, and complete surface information of the target pile is extracted according to the boundary point set and the segmented point cloud, including: De-noising the point cloud cropped data to obtain de-noised point cloud data, and segmenting the de-noised point cloud data to obtain the segmented point cloud; Draw a perpendicular line from each point in the boundary point set to the nearest surface of the container inner wall of the container environment, and extend the perpendicular line beyond the container space, and perform equal-interval sampling on the perpendicular line to obtain an extended point set; Merging the boundary point set, the extended point set and the segmented point cloud to obtain a first merged point cloud, and performing Alpha Wrap three-dimensional reconstruction on the first merged point cloud to generate a clipping surface corresponding to the target pile body; Based on the clipping surface, a contact area between the container inner wall of the container environment and the body surface of the target body is extracted, and complete surface information of the target body is acquired through the contact area.

5. The method according to claim 4, characterized in that The performing of a surface reconstruction operation using the complete surface information to generate a closed triangular mesh corresponding to the target pile body, and calculating a pile body volume corresponding to the target pile body using the closed triangular mesh, comprises: Sampling the contact surface to obtain contact surface point cloud data, and merging the boundary point set, the extended point set, the segmentation point cloud and the contact surface point cloud data to obtain a second merged point cloud; Performing Alpha Wrap reconstruction on the second merged point cloud to obtain a first closed three-dimensional model volume; Merging the boundary point set and the segmented point cloud to obtain a third merged point cloud, and performing Alpha Wrap reconstruction on the third merged point cloud to obtain a second closed three-dimensional model volume; The volume of the target stack is calculated based on the first closed three-dimensional model volume and the second closed three-dimensional model volume.

6. An automated pile volume measuring device, characterized in that: include: An acquisition module, used for scanning a target pile and a container environment storing the target pile, so as to acquire scanning point cloud data corresponding to the target pile and sampling point cloud data corresponding to the container environment; An adsorption matching module is used to perform a point cloud registration operation on the scanned point cloud data and the sampled point cloud data to obtain registered point cloud data, and to perform a cropping and adsorption matching operation on the registered point cloud data to obtain a boundary point set at the boundary of the target pile and the container environment; A computing module is used to perform denoising and point cloud segmentation on the cropped registered point cloud data based on the boundary point set to obtain a segmented point cloud, and to extract complete surface information of the target body according to the boundary point set and the segmented point cloud, and to perform a surface reconstruction operation through the complete surface information to generate a closed triangular mesh corresponding to the target body, and to calculate the body volume corresponding to the target body using the closed triangular mesh.

7. The device according to claim 6, characterized in that The acquisition module comprises: A first reconstruction unit is used to determine the container environment of the target stack, and construct a three-dimensional container model corresponding to the container environment based on a preset Poisson reconstruction strategy or Alpha Wrap reconstruction strategy; A generating unit, used for sampling the three-dimensional model of the container to generate sampling point cloud data of the three-dimensional model of the container; The scanning unit is used to perform a preset comprehensive scanning operation on the target pile body through a preset laser scanner to obtain scanning point cloud data corresponding to the target pile body.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the automated stack volume measurement method according to any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the automatic stack volume measurement method according to any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that The computer program is executed to implement the automated stack volume measurement method according to any one of claims 1 to 5.

Images