A high-energy beam slitting three-dimensional point cloud data processing and manifold modeling method

Abstract

The application provides a high-energy beam slit three-dimensional point cloud data processing and manifold modeling method, and relates to the field of three-dimensional modeling, comprising: splicing two same sample blocks to obtain an experimental block, and cutting the splicing surface of the experimental block by using a high-energy beam; wherein, after cutting, each sample block has a cutting trace at the splicing surface; screening cutting front point cloud data in the point cloud data corresponding to each layer of the cutting layer of the experimental block; fitting the cutting front point cloud data of each layer of the cutting layer of the experimental block by using an ellipse and / or a class ellipse to obtain a cutting contour, and the cutting contour at least comprises contour coordinates and feature information; wherein, the cutting contour corresponding to each layer of the cutting layer is a cross-sectional contour of the manifold of the cutting trace at the layer; introducing the cutting contour corresponding to each layer of the cutting layer into three-dimensional modeling software, and constructing a three-dimensional model of the high-energy beam manifold by using a lofting command.

Description

Technical Field

[0001] This invention relates to the field of 3D modeling, and in particular to a method for processing 3D point cloud data and manifold modeling of high-energy beam slits. Background Technology

[0002] Abrasive waterjet processing involves pressurizing water with a high-pressure pump, then converting the high pressure into high speed through a nozzle, and finally mixing it with abrasive particles to form a complex three-phase flow of liquid, solid, and gas. As the world's only cold-state high-energy beam processing technology, abrasive waterjet technology is a rapidly developing, green, and heat-free high-energy beam processing technology that has been widely applied in an increasing number of fields.

[0003] Unlike traditional cutting tools, water jets are a "soft knife". This "soft knife" has problems such as jet drag and uneven distribution of jet energy in the radial and axial directions during material processing. Its tool shape (that is, jet manifold) will change at any time depending on the parameters and working conditions.

[0004] However, current 3D modeling of water jets often uses simple cylinders instead of 2D cylinders, or builds 3D models from their 2D feature data, which cannot truly reflect the actual manifold and has low accuracy and realism. Summary of the Invention

[0005] In view of this, the present invention aims to propose a method for processing three-dimensional point cloud data and manifold modeling of high-energy beam kerfing, so as to solve the problem of low accuracy and realism of three-dimensional modeling of current high-energy beam flexible cutting tools such as water jets.

[0006] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0007] A method for processing and manifold modeling three-dimensional point cloud data with high-energy beam slits includes:

[0008] Two identical sample blocks are spliced ​​together to obtain an experimental block, and a high-energy beam is used to cut the splicing surface of the experimental block; wherein, after cutting, each sample block has a cutting mark at the splicing surface;

[0009] The cutting front point cloud data of the point cloud data corresponding to the multi-layer cutting of the experimental block are selected. The multi-layer cutting is obtained by dividing the three-dimensional topography point cloud data corresponding to the sample block into layers in the depth direction of the cutting marks. The cutting front point cloud data is the cutting mark topography data that is close to the actual high-energy beam shape.

[0010] Using an elliptical and / or quasi-elliptical shape, the cutting front point cloud data of each cutting layer of the experimental block are fitted to obtain the cutting contour, which includes at least contour coordinates and feature information; wherein, the cutting contour corresponding to each cutting layer is the cross-sectional contour of the manifold of the cutting mark in that cutting layer.

[0011] The cutting contours corresponding to each of the multiple cutting layers are imported into the 3D modeling software, and the 3D model of the high-energy beam shape is constructed using the lofting command.

[0012] Further, the step of filtering out the cutting front point cloud data from the point cloud data corresponding to the multiple cutting layers of the experimental block includes:

[0013] Acquire initial point cloud data for each sample block in the experimental block. The initial point cloud data includes point cloud data on at least multiple surfaces of each sample block, and the multiple surfaces include at least an upper surface, a lower surface, and the splicing surface.

[0014] The initial point cloud data is filtered at least once to obtain the target three-dimensional point cloud model of the experimental block; wherein each subsequent filtering is performed based on the previous filtering, and each filtering is used to remove point cloud data that is irrelevant to the cutting marks.

[0015] In the depth direction of the cutting marks, the target point cloud 3D model is layered according to a preset distance interval to obtain multiple cutting layers;

[0016] Based on the point cloud data of each cutting layer, the corresponding cutting front point cloud data are selected.

[0017] Furthermore, the initial point cloud data of the experimental block is acquired using laser scanning. The initial point cloud data is then filtered at least once to obtain the target 3D point cloud model of the experimental block, including:

[0018] Based on the initial point cloud data, the experimental block is reconstructed in three dimensions to obtain the initial three-dimensional point cloud model of the experimental block.

[0019] Remove point cloud data that is irrelevant to the cutting marks from the initial 3D point cloud model to obtain the filtered 3D point cloud model;

[0020] For each fitted surface of the filtered 3D point cloud model, the point cloud data within a first preset distance from the surface are removed again to remove the error point cloud data obtained by the laser scan.

[0021] The model after undergoing another 3D fitting is used as the target 3D point cloud model.

[0022] The first preset distance is determined based on the resolution of the laser scan.

[0023] Furthermore, the initial point cloud data of the experimental block is acquired using laser scanning. The initial point cloud data is then filtered at least once to obtain the target 3D point cloud model of the experimental block, including:

[0024] Based on the initial point cloud data, the experimental block is reconstructed in three dimensions to obtain the initial three-dimensional point cloud model of the experimental block.

[0025] Remove point cloud data that is irrelevant to the cutting marks from the initial 3D point cloud model to obtain the filtered 3D point cloud model;

[0026] For each fitted surface of the filtered 3D point cloud model, a bounding box with a preset thickness is constructed to surround the surface, and the point cloud data within the bounding box is removed.

[0027] The model after removing the point cloud data within the bounding box is taken as the target 3D point cloud model;

[0028] The preset thickness is determined based on the resolution of the laser scan.

[0029] Furthermore, the cutting includes a first stage and a second stage, wherein the step of filtering out the corresponding cutting front point cloud data based on the point cloud data of each cutting layer includes:

[0030] The point cloud data of each cutting layer is fitted to obtain two non-intersecting fitted straight lines; wherein, the two fitted straight lines of the same cutting layer have a vertical distance, which is the width of the cutting mark in the cutting layer.

[0031] Point cloud data whose distance from the corresponding fitted straight line is greater than a second preset distance are identified as the cutting front point cloud data of the cutting layer; and

[0032] Point cloud data whose distance to the corresponding fitted straight line is less than or equal to the second preset distance are identified as cutting sidewall point cloud data. The cutting sidewall point cloud data is formed in the first stage, wherein the second stage includes the current time and the first stage includes any time before the second stage.

[0033] Furthermore, the preset distance interval is less than or equal to one-fifth of the length of the experimental block in the depth direction of the cutting mark.

[0034] Furthermore, the point cloud data of each of the cutting layers is acquired using laser scanning, and the second preset distance is determined based on the resolution of the laser scan.

[0035] Furthermore, the step of using a high-energy beam to align and cut the splicing surface of the experimental block includes:

[0036] The high-energy beam was used to cut the splicing surface of the experimental block under various preset cutting conditions. The parameters of the preset cutting conditions included at least one of the following: cutting speed, cutting pressure, abrasive output rate, cutting material, material thickness, nozzle / abrasive tube size, target distance, abrasive type, and abrasive mesh size.

[0037] The method further includes:

[0038] Based on the point cloud data of the experimental block under each preset cutting condition, a three-dimensional model of the high-energy beam shape corresponding to that preset cutting condition is constructed.

[0039] Furthermore, after constructing a three-dimensional model of the high-energy beam shape corresponding to each of the preset cutting conditions based on the point cloud data of the experimental block under each of the preset cutting conditions, the method further includes:

[0040] Obtain the actual cutting conditions at the current moment;

[0041] The actual cutting conditions are matched with a variety of preset cutting conditions;

[0042] Retrieve the 3D model of the high-energy beam shape corresponding to the preset cutting conditions.

[0043] Compared with existing technologies, the high-energy beam slit 3D point cloud data processing and manifold modeling method described in this invention has the following advantages:

[0044] This invention obtains cutting front point cloud data by filtering the point cloud data of multiple cutting layers of each cut sample block. The cutting front point cloud data corresponding to each cutting layer is the morphological data of the cutting mark of the actual high-energy beam at that cutting layer. Ellipses and / or quasi-ellipses are then used to fit the cutting front point cloud data of each cutting layer to obtain the cutting contour, fully considering the actual energy distribution of the high-energy beam during cutting. This ensures that the fitted cutting contour of each cutting layer closely resembles the actual cutting contour of the high-energy beam at that cutting layer. Furthermore, a three-dimensional model of the high-energy beam is constructed based on the cutting contours of each cutting layer using a lofting command. This results in a three-dimensional model that closely resembles the actual high-energy beam, exhibiting high accuracy and realism. Attached Figure Description

[0045] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0046] Figure 1 This diagram illustrates the manifold side profile of a transparent workpiece material cut using a water jet high-energy beam in a related technology.

[0047] Figure 2 A virtual three-dimensional model of a high-energy beam in related technologies is shown;

[0048] Figure 3 The flowchart illustrates the steps of a high-energy beam slit 3D point cloud data processing and manifold modeling method provided by the present invention.

[0049] Figure 4 A flowchart illustrating the steps of processing experimental block point cloud data provided by the present invention is shown.

[0050] Figure 5 This invention provides a flowchart of the first step in obtaining a target 3D point cloud model.

[0051] Figure 6 This invention illustrates a flowchart of the second step in obtaining a target 3D point cloud model.

[0052] Figure 7 A flowchart illustrating the steps for determining the cutting front point cloud data provided by the present invention is shown.

[0053] Figure 8 The flowchart of the steps for retrieving a three-dimensional model matching the actual cutting conditions provided by the present invention is shown.

[0054] Figure 9 A schematic diagram of the high-energy beam cutting experimental block provided by the present invention is shown;

[0055] Figure 10a A schematic diagram of the initial three-dimensional point cloud model provided by the present invention is shown;

[0056] Figure 10b This invention provides a schematic diagram of point cloud data filtering.

[0057] Figure 10c A schematic diagram of the target three-dimensional point cloud model provided by the present invention is shown;

[0058] Figure 10d A schematic diagram of the cutting layer provided by the present invention is shown;

[0059] Figure 11 This diagram illustrates the fitting of point cloud data for the cutting layer provided by the present invention.

[0060] Figure 12a A schematic diagram of the cutting contour fitting provided by the present invention is shown;

[0061] Figure 12b A schematic diagram of the cutting contour corresponding to the multi-layer cutting layer provided by the present invention is shown;

[0062] Figure 13 A schematic diagram of a three-dimensional model for constructing a high-energy beam shape provided by the present invention is shown;

[0063] Figure 14 A three-dimensional model of the manifold of the high-energy beam constructed under various preset cutting conditions provided by the present invention is shown;

[0064] Figure 15 A schematic diagram of the simulation test provided by the present invention is shown;

[0065] Figure 16 The diagram shows a flowchart of the high-energy beam slit 3D point cloud data processing and manifold modeling method provided by the present invention. Detailed Implementation

[0066] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0067] It should be noted that, without conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0068] Abrasive waterjet processing involves pressurizing water with a high-pressure pump, then converting the high pressure into high speed through a nozzle, and finally mixing it with abrasive particles to form a complex three-phase flow of liquid, solid, and gas. As the world's only cold-state high-energy beam processing technology, abrasive waterjet technology is a rapidly developing, green, and heat-free high-energy beam processing technology that has found widespread application in an increasing number of fields. With the development of digital twin technology, the field of intelligent manufacturing is increasingly focusing on virtual simulation technology. Manufacturing processes can be simulated by creating virtual 3D models of corresponding machine tool entities in a virtual environment, allowing for early verification and significant cost savings. However, the most crucial and critical issue lies in whether the virtual 3D model can accurately simulate the process.

[0069] For traditional rigid cutting tools (such as end mills), since their shape is basically fixed, it is easy to create a corresponding virtual 3D model. However, unlike traditional cutting tools, water jet high-energy beams are "soft tools," referring to... Figure 1 , Figure 1This diagram illustrates the manifold side profile of a transparent workpiece material cut using a water jet high-energy beam in a related technology, such as... Figure 1 As shown, this "soft knife" may have problems such as jet drag and uneven distribution of jet energy in the radial and axial directions during material processing. Its tool shape (i.e., jet manifold) will change at any time with different parameters and working conditions, making it difficult to establish a relatively accurate virtual three-dimensional model.

[0070] Furthermore, since water jets contain enormous energy, they can damage various contact-type acquisition devices and are difficult to measure in real time during the cutting process. Therefore, the only way to further extract the water jet manifold features is to preserve the morphology of the cut.

[0071] However, refer to Figure 2 , Figure 2 A virtual three-dimensional model of a high-energy beam in related technologies is shown, such as Figure 2 As shown, commonly used simulation or modeling software often simplifies the virtual 3D model of a high-energy beam into a cylinder, which cannot accurately reflect the actual manifold of the high-energy beam. Furthermore, related technologies often use tools such as dial indicators or coordinate measuring machines to collect 2D morphological information of the material's internal cutting marks. This collected 2D morphological information is then fitted to obtain multiple circular curves, which are then combined into a 3D feature. However, this method of morphological information acquisition can only capture key points, not the entire material, easily leading to missed information. The fitting process does not consider the actual energy distribution of the water jet, resulting in low accuracy and realism.

[0072] Therefore, improving the accuracy and realism of high-energy beam slit 3D point cloud data processing and manifold modeling is now urgent.

[0073] Reference Figure 3 , Figure 3 The following is a flowchart illustrating the steps of a high-energy beam slit 3D point cloud data processing and manifold modeling method provided by the present invention: Figure 3 As shown, it includes the following steps:

[0074] Step S101: Two identical sample blocks are spliced ​​together to obtain an experimental block, and a high-energy beam is used to cut the splicing surface of the experimental block; wherein, after cutting, each sample block has a cutting mark at the splicing surface.

[0075] For example, refer to Figure 9 , Figure 9 A schematic diagram of the high-energy beam cutting experimental block provided by the present invention is shown, as follows: Figure 9As shown, both samples are cuboids with six surfaces: a top surface, a bottom surface, a joint surface, and other surfaces. The top surface is close to the water jet nozzle and perpendicular to the water jet's direction of extension. The bottom surface is away from the nozzle and opposite to the top surface. The joint surface is the surface where the two samples are joined. When cutting the experimental blocks, the high-energy beam is directed from the top surface to the bottom surface into the joint surface, leaving a cutting mark. After cutting, the two joined samples are separated, resulting in two cut samples.

[0076] It should be noted that the two identical sample blocks here refer to two completely identical sample blocks, including but not limited to materials, dimensions, and preparation methods. The two sample blocks can be joined using any method that can secure them together, such as bolt fixing. The surface where the two sample blocks are joined is the splicing surface.

[0077] It should be noted that when using a high-energy beam to cut the splicing surface of the experimental block, the direction of the high-energy beam's movement is set based on the actual application, such as from left to right or from right to left, and is not restricted here.

[0078] In this embodiment of the invention, a high-energy beam is used to cut the splicing surface of an experimental block obtained by splicing two identical sample blocks, so that the splicing surface retains complete cutting marks, thereby enabling the acquisition of complete cutting mark point cloud data and improving the accuracy of point cloud data.

[0079] Step S102: Filter out the cutting front point cloud data in the point cloud data corresponding to the multi-layer cutting of the experimental block. The multi-layer cutting is obtained by dividing the three-dimensional topography point cloud data corresponding to the sample block into layers in the depth direction of the cutting marks. The cutting front point cloud data is the cutting mark topography data that closely resembles the actual high-energy beam shape.

[0080] Point cloud data refers to a collection of a large number of feature points on the surface of a target, including at least the three-dimensional coordinates of the feature points.

[0081] For the cut experimental blocks, after separating the two spliced ​​blocks, all point cloud data of each cut block is collected. The point cloud data of the spliced ​​surface includes the point cloud data of the cutting marks. Point cloud data acquisition for the blocks can be achieved using laser scanning, depth camera acquisition, industrial computed tomography (CT) technology, etc.

[0082] Based on the point cloud data of the two collected samples, a 3D point cloud model of the experimental block was constructed. The model was then layered along the depth direction of the cutting marks, with preset distance intervals. Multiple cutting layers were selected, and the cutting front point cloud data corresponding to each cutting layer was extracted from the point cloud data of the cutting marks within each layer. The cutting front point cloud data represents the cutting mark morphology data that closely resembles the actual high-energy beam shape. When layering the 3D point cloud model of the experimental block, the preset distance interval between adjacent cutting layers can be set according to actual application requirements. A smaller preset distance interval results in more cutting layers and a more accurate 3D model of the high-energy beam. However, more cutting layers also mean a larger data volume and higher demands on the data processing capabilities for model construction and simulation testing.

[0083] In the process of selecting the cutting point cloud data corresponding to each cutting layer of the experimental block, the splicing surface of the experimental block is used as a reference. The point cloud data on both sides of the splicing surface are fitted respectively. The two non-intersecting fitted lines are located on both sides of the splicing surface. The vertical distance between the point cloud data of each cutting layer and the fitted line on the corresponding side is determined. The point cloud data with a vertical distance greater than a second preset distance is determined as the cutting front point cloud data. The cutting front point cloud data is the data of the cutting trace morphology that is close to the actual high-energy beam shape. The point cloud data with a vertical distance less than or equal to the second preset distance is determined as the cutting sidewall point cloud data.

[0084] In this embodiment of the invention, the point cloud data of the cutting front corresponding to each multi-layer cutting layer of each sample block are selected, so that the selected point cloud data is closer to the cutting trace morphology data of the actual water jet manifold.

[0085] Step S103: Using an ellipse and / or quasi-ellipse shape, fit the cutting front point cloud data of each cutting layer of the experimental block to obtain the cutting contour. The cutting contour includes at least contour coordinates and feature information. The cutting contour corresponding to each cutting layer is the cross-sectional contour of the manifold of the cutting mark in that cutting layer.

[0086] From the perspective of energy distribution, the cross-section of a high-energy beam is considered to be elliptical. Therefore, in this embodiment of the invention, an elliptical and / or elliptical shape is used for fitting, and the fitted curve is the actual cutting profile of the cutting layer when the high-energy beam cuts the experimental block. For example, refer to... Figures 12a to 12b , Figure 12a A schematic diagram of the cutting contour fitting provided by the present invention is shown.

[0087] It should be noted that the contour coordinates here refer to the directions of the major and minor axes of the fitted curve, and the feature information refers to the attribute characteristics of the major and minor axes.

[0088] The embodiments of the present invention obtain the cutting contour by fitting the point cloud data of the cutting front using an ellipse and / or a quasi-ellipse. Compared with the use of a circle for fitting in related technologies, this fully considers the real energy distribution of the high-energy beam, making the cutting contour closer to the actual high-energy beam cutting contour, thus improving the accuracy and realism of the cutting contour.

[0089] Step S104: Import the cutting contours corresponding to each of the multiple cutting layers into the 3D modeling software, and use the lofting command to construct the 3D model of the high-energy beam shape.

[0090] The cutting contours corresponding to each of the multiple cutting layers obtained in step S103 are imported into 3D modeling software to construct a 3D model of the high-energy beam manifold. The 3D modeling software used here can be any software capable of performing 3D modeling.

[0091] The loft command is a common function in 3D modeling software that generates a 3D model by creating transitions between contours.

[0092] For example, refer to Figures 12b to 13 , Figure 12b This diagram illustrates the cutting contour corresponding to the multi-layer cutting structure provided by the present invention. Figure 13 A schematic diagram of a three-dimensional model for constructing a high-energy beam shape provided by the present invention is shown, as follows. Figures 12b to 13 As shown, the cutting contours corresponding to multiple cutting layers are imported into the 3D modeling software in the order of the length direction of the cutting marks. The lofting command is used to transition the cutting contours corresponding to adjacent cutting layers and generate a 3D model.

[0093] For example, refer to Figure 15 , Figure 15 A schematic diagram of the simulation test provided by the present invention is shown, such as... Figure 15 As shown, using a three-dimensional model of the constructed high-energy beam manifold for simulation testing can yield more realistic and accurate simulation results.

[0094] This invention, through lofting commands, fits the cutting contours corresponding to each of multiple cutting layers to obtain a three-dimensional model of the high-energy beam manifold. Data processing fills in the transitions between cutting contours on different cutting layers, further improving the accuracy and realism of the high-energy beam three-dimensional model. The three-dimensional model of the high-energy beam manifold constructed using the high-energy beam kerf three-dimensional point cloud data processing and manifold modeling method provided in this invention closely resembles the actual manifold of the high-energy beam and can be used in technologies such as virtual manufacturing and digital twins. This allows virtual cutting simulation to replace some physical cutting, overcoming various limitations and constraints of physical cutting, such as high cost and difficulty in observation.

[0095] The high-energy beam kerf 3D point cloud data processing and manifold modeling method provided in this invention is applicable to various cutting conditions. It is not only suitable for abrasive waterjet, but can also be extended to laser beams and plasma beams with similar "soft blade" characteristics to construct virtual 3D models corresponding to laser beams and plasma beams, respectively.

[0096] In some alternative embodiments, refer to Figure 4 , Figure 4 A flowchart illustrating the steps of the experimental block point cloud data processing provided by the present invention is shown, as follows: Figure 4 As shown, step S102 above includes the following steps S201 to S204:

[0097] Step S201: Obtain the initial point cloud data of each sample block in the experimental block. The initial point cloud data includes point cloud data on at least multiple surfaces of each sample block, and the multiple surfaces include at least an upper surface, a lower surface, and the splicing surface (e.g., Figure 9 (As shown).

[0098] After cutting the experimental block with a high-energy beam, the two spliced ​​blocks are separated. For each cut block, the initial point cloud data is obtained, and the initial three-dimensional point cloud model of the cut experimental block is constructed based on the initial point cloud data of the two blocks.

[0099] Step S202: The initial point cloud data is filtered at least once to obtain the target three-dimensional point cloud model of the experimental block; wherein the next filtering is performed based on the previous filtering, and each filtering is used to remove point cloud data that is not related to the cutting marks.

[0100] Based on the initial 3D point cloud model of the experimental block constructed in step S201, at least one screening is performed to obtain the target 3D point cloud model of the experimental block. The target 3D point cloud model includes at least multiple surfaces of the experimental block, such as the upper surface, lower surface, and splicing surface of the experimental block.

[0101] In the case of a single screening, the 3D point cloud model obtained from the first screening is used as the target 3D point cloud model for the experimental block. In the case of multiple screenings, for example, the point cloud data unrelated to the cutting marks is further simplified based on the 3D point cloud model obtained from the first screening, which is the second screening, to retain the point cloud data only related to the cutting marks, and the 3D point cloud model after multiple simplifications is used as the target 3D point cloud model.

[0102] During the filtering process, users can manually divide the area into deletion and retention regions. The retention region includes the cutting marks. Point cloud data within the deletion region is then removed, thus filtering out the point cloud data at the cutting front. Cutting marks can also be marked; the software identifies these marks and automatically removes point cloud data outside of the cutting marks.

[0103] Step S203: In the depth direction of the cutting marks, the target point cloud 3D model is layered according to a preset distance interval to obtain multiple cutting layers.

[0104] Based on the target 3D point cloud model of the experimental block obtained in step S202, the target 3D point cloud model is layered along the depth direction F1 of the cutting marks according to a preset distance interval, resulting in the following: Figure 10d The cut layers are shown.

[0105] It should be noted that the depth direction F1 of the cutting mark here refers to the direction in which the high-energy beam is directed from the nozzle towards the bottom of the experimental block. For example, refer to... Figure 10a , Figure 10a A schematic diagram of the initial three-dimensional point cloud model provided by the present invention is shown, as follows. Figure 10a As shown, the depth direction F1 of the cutting mark is from A to A'.

[0106] When layering a target 3D point cloud model, a virtual plane can be formed at the target cutting depth. This virtual plane is then fitted with point cloud data at a third preset distance from it to obtain the cutting layer plane. A bounding box with a preset thickness can also be formed around the virtual plane. The third preset distance is slightly greater than or equal to the resolution of the laser scan. For example, if the laser scan resolution is 0.05, the third preset distance could be ±0.05mm, ±0.06mm, ±0.07mm, etc.

[0107] Step S204: Based on the point cloud data of each cutting layer, filter out the corresponding cutting front point cloud data.

[0108] When filtering the point cloud data of the cutting front corresponding to each cutting layer, the point cloud data at the final cutting position of the high-energy beam can be selected as the cutting front point cloud data based on the distribution of the point cloud data of each cutting layer. Alternatively, non-linearly distributed point cloud data can be selected as the cutting front point cloud data based on the distribution of the point cloud data of each cutting layer. Furthermore, the point cloud data of each cutting layer can be fitted, and point cloud data at a second preset distance from the fitted line can be selected as the cutting front point cloud data based on the fitted line.

[0109] This invention, through at least one filtering step based on the initial point cloud data of the experimental block, identifies point cloud data related to the cutting marks, thus obtaining the target 3D point cloud model of the experimental block. This ensures that the target 3D point cloud model retains only point cloud data related to the cutting marks to the greatest extent possible. This reduces the amount of data that needs to be processed when layering the cutting layers and makes the filtering of point cloud data at the cutting front more accurate. Furthermore, by setting preset distance intervals, the refinement of the high-energy beam 3D model can be adjusted, improving the flexibility of model construction.

[0110] In some optional embodiments, the preset distance interval is less than or equal to one-fifth of the length of the experimental block in the depth direction of the cutting mark.

[0111] For example, if the length of the experimental block in the depth direction F1 of the cutting mark is 40mm, then the preset distance interval is less than or equal to 8mm, such as 8mm, 5mm, 3mm, etc., which can be flexibly set according to the actual modeling accuracy requirements.

[0112] In some optional embodiments, the initial point cloud data of the experimental block is acquired using laser scanning. The parameters of the laser scanning instrument include resolution, which to some extent characterizes the error of the scanning results.

[0113] Laser scanning is a method that uses the principle of laser ranging to quickly reconstruct the three-dimensional model of the target object and various data such as lines, surfaces, and volumes by recording a large number of dense points on the surface of the object being measured, including their three-dimensional coordinates, reflectivity, and texture.

[0114] In specific implementation, the above step S202 can be achieved in the following two ways:

[0115] In the first implementation, refer to Figure 5 , Figure 5 The following is a flowchart illustrating the first step in obtaining a target 3D point cloud model of a sample block according to the present invention: Figure 5 As shown, it includes the following steps:

[0116] Step S301: Based on the initial point cloud data, perform three-dimensional reconstruction on the experimental block to obtain the initial three-dimensional point cloud model of the experimental block.

[0117] For the cut experimental block, the initial point cloud data of two sample blocks in the experimental block are obtained by laser scanning. Based on the initial point cloud data of the two sample blocks, the initial three-dimensional point cloud model of the experimental block is reconstructed. The initial three-dimensional point cloud model includes the fitted upper surface, the fitted lower surface, and the fitted splicing surface of the experimental block.

[0118] Step S302: Remove point cloud data that is unrelated to the cutting marks from the initial three-dimensional point cloud model to obtain the filtered three-dimensional point cloud model.

[0119] The initial 3D point cloud model is coarsely segmented. Using the cutting marks as a reference, the lower surface of the fitted surface is indented in the direction pointing from the lower to the upper surface. Additionally, the left and right sides of the fitted stitching surface are indented inwards. The indentation distance is flexibly set according to the actual application. This process removes point cloud data unrelated to the cutting marks from the initial 3D model, resulting in the following: Figure 10a The filtered 3D point cloud model is shown.

[0120] Step S303: For each fitted surface of the filtered 3D point cloud model, the point cloud data within a first preset distance from the surface is removed again to remove the error point cloud data obtained by the laser scan; wherein, the first preset distance is determined according to the resolution of the laser scan.

[0121] Step S304: Use the model after the third-dimensional fitting is performed again as the target three-dimensional point cloud model.

[0122] For example, such as Figure 9 As shown, the sample block used in this invention is a cuboid, and the sample block entity includes 6 surfaces. Therefore, the fitted 3D model of the sample block also includes 6 surfaces.

[0123] The selected 3D point cloud model is then finely segmented. For example, refer to... Figure 10b , Figure 10b This illustrates a point cloud data filtering method provided by the present invention, such as... Figure 10b As shown, for each surface of the fitted 3D model, the point cloud data within a first preset distance of that surface is removed again. This first preset distance is slightly greater than or equal to the resolution of the laser scan. For example, if the laser scan resolution is 0.05, the first preset distance can be ±0.05 mm. This process removes all point cloud data within the laser scan error range that could potentially be the fitted surface, resulting in the following... Figure 10c The target 3D point cloud model is shown.

[0124] This invention provides a standardized and applicable point cloud segmentation method for complete 3D point cloud data of cutting marks. This method accurately filters and removes point cloud data unrelated to the cutting marks, retaining only the point cloud data relevant to the cutting marks for subsequent processing. By repeatedly filtering the initial 3D point cloud model of the experimental block, the resulting target 3D point cloud model retains only point cloud data related to the cutting marks, reducing the amount of data processing in subsequent steps. Furthermore, by fitting based on laser scanning resolution to remove erroneous point cloud data, the accuracy of the data is further improved.

[0125] In the second implementation, refer to Figure 6 , Figure 6 The flowchart illustrating the second step in obtaining the target 3D point cloud model of the sample block provided by the present invention is shown below. Figure 6 As shown, it includes the following steps:

[0126] Step S401: Based on the initial point cloud data, perform three-dimensional reconstruction on the experimental block to obtain the initial three-dimensional point cloud model of the experimental block.

[0127] The implementation method of step S401 is the same as that of step S301 described above, and will not be repeated here.

[0128] Step S402: Remove point cloud data that is unrelated to the cutting marks from the initial three-dimensional point cloud model to obtain the filtered three-dimensional point cloud model.

[0129] The implementation method of step S402 is the same as that of step S302 described above, and will not be repeated here.

[0130] Step S403: For each fitted surface of the filtered 3D point cloud model, construct a bounding box surrounding the surface with a preset thickness, and remove the point cloud data within the bounding box; wherein, the preset thickness is determined according to the resolution of the laser scan.

[0131] Step S404: The model after removing the point cloud data within the bounding box is taken as the target 3D point cloud model.

[0132] The selected 3D point cloud model is finely segmented. For each surface of the selected 3D point cloud model, a bounding box with a preset thickness is constructed around the surface. This preset thickness is determined by the resolution of the laser scan. For example, if the laser scan resolution is 0.05, the preset thickness is 0.1 mm. This is equivalent to constructing virtual planes at a distance of 0.05 mm from the surface on both sides. The bounding box is then constructed based on these virtual planes, thereby removing point cloud data that might be from that surface within the laser scan error range. The result is as follows: Figure 10c The target 3D model is shown.

[0133] The embodiments of the present invention further improve the accuracy of the data by fitting based on the laser scanning resolution to remove erroneous point cloud data.

[0134] In some alternative embodiments, the process of cutting the experimental block with a high-energy beam includes a first stage and a second stage, wherein the second stage includes the current moment and the first stage includes any moment prior to the second stage.

[0135] For example, the high-energy beam moves from left to right to cut the experimental block. At the current time T1, the high-energy beam moves to position D1. Then the second stage is the current time T1. The first stage includes any time when the high-energy beam moves from the starting point to position D1.

[0136] In some alternative embodiments, refer to Figure 7 , Figure 7 The flowchart illustrating the steps for determining the cutting front point cloud data provided by the present invention is shown below. Figure 7 As shown, step S204 above includes the following steps:

[0137] Step S501: Fit the point cloud data of each cutting layer to obtain two fitting straight lines that do not intersect; wherein, the two fitting straight lines of the same cutting layer have a vertical distance, and the vertical distance is the width of the cutting mark in the cutting layer.

[0138] Reference Figure 11 , Figure 11 This invention provides a schematic diagram of point cloud data fitting for a cutting layer. Figure 11 As shown, taking the intersection of the splicing surface of the experimental block and the cutting layer plane as the reference, the point cloud data on both sides of the splicing surface are fitted respectively, and the two non-intersecting fitted lines are located on both sides of the splicing surface.

[0139] For example, the point cloud data of the cutting layer 1 on the left side of the experimental block splicing surface is fitted to obtain the fitting line A, and the point cloud data of the cutting layer 1 on the right side of the experimental block splicing surface is fitted to obtain the fitting line B.

[0140] Step S502: The point cloud data whose distance from the corresponding fitted line is greater than the second preset distance is determined as the cutting front point cloud data of the cutting layer.

[0141] In each cutting layer, point cloud data whose distance to the fitted line on its corresponding side is greater than a second preset distance are identified as the cutting front point cloud data of that cutting layer. The cutting front point cloud data is formed in the second stage of the cutting process, i.e., at the current moment. Since the cutting marks of the high-energy beam are the result of the continuous accumulation of high-energy beams at multiple moments, the cutting front point cloud is not interfered with by the water jet energy at other moments, thus representing the cutting morphology that best approximates the actual water jet manifold. For example, for the point cloud data located on the left side of the experimental block splicing surface in cutting layer 1, the distance between it and the fitted line A on the left is determined, and point cloud data with a distance greater than the second preset distance are identified as the cutting front point cloud data on the left side of the splicing surface in cutting layer 1; for the point cloud data located on the right side of the splicing surface, the distance between it and the fitted line B on the right is determined, and point cloud data with a distance greater than the second preset distance are identified as the cutting front point cloud data on the right side of the splicing surface in cutting layer 1. The cutting front point cloud data on the left side of the splicing surface and the cutting front point cloud data on the right side together constitute the cutting front point cloud data of the cutting layer 1 of the experimental block, and their distribution satisfies the elliptical / elliptical front shape.

[0142] Step S503: Point cloud data whose distance to the corresponding fitted straight line is less than or equal to the second preset distance are identified as cutting sidewall point cloud data, wherein the cutting sidewall point cloud data is formed in the first stage.

[0143] Point cloud data in each cutting layer whose distance to the fitted line on its corresponding side is less than or equal to a second preset distance are identified as the cutting sidewall point cloud data of that cutting layer. The cutting sidewall point cloud is formed in the first stage, accumulating over a period prior to the current moment. For example, for point cloud data located on the left side of the experimental block splicing surface in cutting layer 1, the distance between it and the fitted line A on the left is determined, and point cloud data with a distance less than or equal to the second preset distance are identified as the cutting sidewall point cloud data on the left side of the splicing surface in cutting layer 1. Similarly, for point cloud data located on the right side of the splicing surface, the distance between it and the fitted line B on the right is determined, and point cloud data with a distance less than or equal to the second preset distance are identified as the cutting sidewall point cloud data on the right side of the splicing surface in cutting layer 1.

[0144] This invention provides a standardized and applicable point cloud segmentation method for the point cloud data of each cutting layer. It filters and removes point cloud data from the cutting sidewalls, retaining the cutting front point cloud data for subsequent processing. The cutting front point cloud data and the cutting sidewall point cloud data are determined by fitting a straight line, thus filtering out the cutting front point cloud data formed at the current moment, resulting in higher accuracy of the subsequently constructed three-dimensional model of the high-energy beam.

[0145] In some optional embodiments, the point cloud data of each cutting layer is acquired by laser scanning, and the second preset distance is determined according to the resolution of the laser scan.

[0146] For example, the value of the second preset distance is the same as the resolution value of the laser scan. For instance, if the resolution of the laser scan is 0.05, then the second preset distance can be 0.05mm. Of course, other resolutions can also be used in practice, and the second preset distance can also be the same value as other resolutions.

[0147] The embodiments of the present invention further improve the accuracy of the data by determining a second preset distance based on the laser scanning resolution to remove erroneous point cloud data.

[0148] In some optional embodiments, step S101 above includes the following sub-step A1:

[0149] Sub-step A1: Under various preset cutting conditions, the high-energy beam is used to cut the splicing surface of the experimental block; the parameters of the preset cutting conditions include at least one of the following: cutting speed, cutting pressure, abrasive output rate, cutting material, material thickness, nozzle / abrasive tube size, target distance, abrasive type, and abrasive mesh size.

[0150] Preset cutting conditions refer to cutting conditions with pre-set parameter values, which may include one or more of the above parameters. For example, the parameters of preset cutting condition 1 include cutting speed, and preset cutting condition 2 includes cutting speed and cutting pressure.

[0151] Furthermore, each preset cutting condition includes multiple levels of parameters, with the parameter values ​​decreasing sequentially from low to high. For example, the cutting speed condition includes cutting speed levels one, two, three, four, and five, with the cutting speed decreasing sequentially from level one to level five. The preset cutting condition can also be further refined to include parameters for one or more target levels. For instance, preset cutting condition 3 includes a cutting speed of level one, while preset cutting condition 2 includes both a cutting speed of level one and a cutting pressure of level one.

[0152] In some optional embodiments, the high-energy beam slit 3D point cloud data processing and manifold modeling method provided in this disclosure further includes the following step A2:

[0153] Step A2: Based on the point cloud data of the experimental block under each preset cutting condition, construct a three-dimensional model of the high-energy beam shape corresponding to that preset cutting condition.

[0154] Multiple experimental blocks are provided, each obtained by splicing two identical sample blocks. The splicing surface of each experimental block is cut using a preset cutting condition. Initial point cloud data is obtained for each cut sample block, and a 3D model of the corresponding high-energy beam manifold is constructed based on this initial point cloud data. The method for constructing the 3D model of the corresponding high-energy beam manifold based on this initial point cloud data is the same as the high-energy beam slit 3D point cloud data processing and manifold modeling method described in any of the above embodiments, and will not be repeated here.

[0155] Reference Figure 14 , Figure 14 The present invention illustrates a three-dimensional model of the manifold of a high-energy beam constructed under various preset cutting conditions, as provided in this invention. Figure 14 As shown, based on the point cloud data of the experimental block corresponding to each preset cutting condition, a three-dimensional model of the high-energy beam shape corresponding to that preset cutting condition is constructed. The constructed three-dimensional models corresponding to each preset cutting condition are stored in a database for later retrieval.

[0156] This disclosure overcomes the shortcomings of traditional modeling methods in terms of poor versatility by constructing three-dimensional models of high-energy beam shapes corresponding to various preset cutting conditions. The pre-constructed models can be used at any time, improving the efficiency of simulation testing.

[0157] In an alternative embodiment, refer to Figure 8 , Figure 8 The flowchart illustrating the steps of retrieving a 3D model matching the actual cutting conditions provided by the present invention is shown below. Figure 8 As shown, after step A2 above, the following steps are also included:

[0158] Step S601: Obtain the actual cutting conditions at the current moment.

[0159] Step S602: Match the actual cutting conditions with a variety of preset cutting conditions.

[0160] Step S603: Retrieve the three-dimensional model of the high-energy beam shape corresponding to the matching preset cutting condition.

[0161] Actual cutting conditions refer to the real-time cutting conditions during actual cutting experiments or cutting operations.

[0162] Based on pre-constructed 3D models of the high-energy beam manifolds corresponding to various preset cutting conditions, during cutting experiments or operations, the manifold shape of the water jet varies under different conditions and is time-varying. Therefore, it is necessary to obtain the target parameter values ​​of the actual cutting condition in real time. These target parameter values ​​are then matched with various preset cutting conditions. If the parameters included in the preset cutting condition are consistent with the target parameters, and the parameter levels are consistent with the target parameter values, then the preset cutting condition is considered to match the actual cutting condition. Finally, the 3D model of the high-energy beam manifold corresponding to the matched preset cutting condition is retrieved for use.

[0163] In this way, the versatility and flexibility of the model can be improved on the one hand, and the accuracy of experimental or test results can be improved on the other hand.

[0164] The following example illustrates in detail the high-energy beam slit 3D point cloud data processing and manifold modeling method provided by this invention:

[0165] Reference Figure 16 , Figure 16 The flowchart illustrating the high-energy beam slit 3D point cloud data processing and manifold modeling method provided by the present invention is shown below. Figure 16 As shown, a high-energy beam is used to cut the splicing surface of an experimental block obtained by splicing two identical samples. The splicing surface includes cutting marks. For each cut sample, initial point cloud data of the sample is obtained by laser scanning. Based on the initial point cloud data, an initial three-dimensional point cloud model of the experimental block is reconstructed. The initial three-dimensional point cloud model includes a fitted upper surface, a fitted lower surface, and a fitted splicing surface.

[0166] The initial 3D point cloud model is coarsely segmented. Based on the cutting marks, the lower surface of the fitting is indented in the direction from the lower surface to the upper surface of the fitting. In addition, the left and right sides of the fitting splicing surface are indented inward. The indentation distance is flexibly set according to the actual application. In this way, point cloud data that is not related to the cutting marks in the initial 3D point cloud model is deleted, and the filtered 3D point cloud model is obtained.

[0167] The selected 3D point cloud model is finely segmented. Fine segmentation can be achieved by fitting each surface of the selected 3D point cloud model with point cloud data at a distance of ±0.05mm from that surface; or by constructing a bounding box with a thickness of 0.1mm around each surface.

[0168] For the target 3D point cloud model of the experimental block, the model is layered at 5mm intervals along the depth direction F1 of the cutting marks. During layering, a virtual plane can be formed at the target cutting depth. This virtual plane is then fitted with point cloud data at a third preset distance from it to obtain the cutting layer plane. A bounding box with a preset thickness can also be formed around the virtual plane.

[0169] like Figure 11 As shown, for each layer cut, the point cloud data on the left side of the experimental block splicing surface is fitted to obtain a fitted line A, and the point cloud data on the right side of the experimental block splicing surface is fitted to obtain a fitted line B. Point cloud data on the left side of the splicing surface with a distance greater than 0.05 mm from the fitted line A are identified as the cutting front point cloud data on the left side of that cutting layer. Point cloud data on the left side of the splicing surface with a distance less than or equal to 0.05 mm from the fitted line A are identified as the cutting sidewall point cloud data on the left side of that cutting layer. Point cloud data on the right side of the splicing surface with a distance greater than 0.05 mm from the fitted line B are identified as the cutting front point cloud data on the right side of that cutting layer. Point cloud data on the right side of the splicing surface with a distance less than or equal to 0.05 mm from the fitted line B are identified as the cutting sidewall point cloud data on the right side of that cutting layer.

[0170] Elliptical and / or elliptical shapes were used to fit the cutting front point cloud data of each cutting layer of the experimental block to obtain the cutting contour corresponding to each cutting layer. The cutting contours corresponding to each cutting layer were then imported into 3D modeling software, and the lofting command was used to construct a 3D model of the high-energy beam manifold.

[0171] Based on the same inventive concept, embodiments of the present invention also provide a device for processing high-energy beam slit three-dimensional point cloud data and manifold modeling, comprising:

[0172] A cutting module is used to splice two identical sample blocks to obtain an experimental block, and to cut the splicing surface of the experimental block using a high-energy beam; wherein, after cutting, each sample block has a cutting mark at the splicing surface;

[0173] The first screening module is used to screen out the cutting front point cloud data in the point cloud data corresponding to the multi-layer cutting of the experimental block. The multi-layer cutting is obtained by dividing the three-dimensional topography point cloud data corresponding to the sample block into layers in the depth direction of the cutting marks. The cutting front point cloud data is the cutting mark topography data that closely resembles the actual high-energy beam shape.

[0174] The contour fitting module is used to fit the cutting front point cloud data of each cutting layer of the experimental block using an ellipse and / or a near-ellipse shape to obtain a cutting contour. The cutting contour includes at least contour coordinates and feature information. The cutting contour corresponding to each cutting layer is the cross-sectional contour of the manifold of the cutting mark in that cutting layer.

[0175] The import module is used to import the cutting contours corresponding to each of the multiple cutting layers into the 3D modeling software, and to construct the 3D model of the high-energy beam shape using the lofting command.

[0176] In some optional embodiments, the first screening module includes:

[0177] The first acquisition module is used to acquire the initial point cloud data of each sample block in the experimental block. The initial point cloud data includes at least point cloud data on multiple surfaces of each sample block, and the multiple surfaces include at least an upper surface, a lower surface and the splicing surface.

[0178] The second filtering module is used to filter the initial point cloud data at least once to obtain the target three-dimensional point cloud model of the experimental block; wherein, the next filtering is performed based on the previous filtering, and each filtering is used to remove point cloud data that is unrelated to the cutting marks;

[0179] The layering module is used to layer the target point cloud 3D model in the depth direction of the cutting marks according to a preset distance interval to obtain multiple cutting layers;

[0180] The third filtering module is used to filter out the corresponding cutting front point cloud data based on the point cloud data of each cutting layer.

[0181] In some optional embodiments, the second screening module includes:

[0182] The first 3D reconstruction module is used to perform 3D reconstruction on the experimental block based on the initial point cloud data to obtain the initial 3D point cloud model of the experimental block.

[0183] The first removal module is used to remove point cloud data that is unrelated to the cutting marks from the initial three-dimensional point cloud model to obtain the filtered three-dimensional point cloud model.

[0184] The second removal module is used to remove point cloud data within a first preset distance from each fitted surface of the filtered 3D point cloud model, so as to remove the error point cloud data obtained by the laser scanning.

[0185] The first determining module is used to use the model after the third-dimensional fitting is performed again as the target three-dimensional point cloud model;

[0186] The first preset distance is determined based on the resolution of the laser scan.

[0187] In some optional embodiments, the second screening module includes:

[0188] The second 3D reconstruction module is used to perform 3D reconstruction on the experimental block based on the initial point cloud data to obtain the initial 3D point cloud model of the experimental block.

[0189] The third removal module is used to remove point cloud data that is unrelated to the cutting marks from the initial three-dimensional point cloud model, so as to obtain the filtered three-dimensional point cloud model.

[0190] The bounding box construction module is used to construct a bounding box with a preset thickness for each fitted surface of the filtered 3D point cloud model, and to remove the point cloud data within the bounding box.

[0191] The second determining module is used to take the model after removing the point cloud data within the bounding box as the target 3D point cloud model.

[0192] The preset thickness is determined based on the resolution of the laser scan.

[0193] In some optional embodiments, the third screening module includes:

[0194] The straight line fitting module is used to fit the point cloud data of each of the cutting layers to obtain two fitting straight lines that do not intersect; wherein, the two fitting straight lines of the same cutting layer have a vertical distance, which is the width of the cutting mark in the cutting layer.

[0195] The third determining module is used to determine point cloud data whose distance to the corresponding fitted straight line is greater than a second preset distance as the cutting front point cloud data of the cutting layer; and

[0196] The fourth determining module is used to determine point cloud data whose distance to the corresponding fitted straight line is less than or equal to the second preset distance as cutting sidewall point cloud data. The cutting sidewall point cloud data is formed in the first stage, wherein the second stage includes the current time and the first stage includes any time before the second stage.

[0197] In some optional embodiments, the cutting module includes:

[0198] The cutting submodule is used to cut the splicing surface of the experimental block with the high-energy beam under various preset cutting conditions. The parameters of the preset cutting conditions include at least one of the following: cutting speed, cutting pressure, abrasive output rate, cutting material, material thickness, nozzle / abrasive tube size, target distance, abrasive type, and abrasive mesh size.

[0199] In some alternative embodiments, the apparatus further includes:

[0200] The construction module is used to construct a three-dimensional model of the high-energy beam shape corresponding to each preset cutting condition based on the point cloud data of the experimental block under each preset cutting condition.

[0201] In some alternative embodiments, the building module includes:

[0202] The second acquisition module is used to acquire the actual cutting conditions at the current moment;

[0203] A matching module is used to match the actual cutting conditions with a variety of preset cutting conditions;

[0204] The retrieval module is used to retrieve the 3D model of the high-energy beam shape corresponding to the preset cutting conditions.

[0205] Based on the same inventive concept, embodiments of the present invention also provide an electronic device, comprising:

[0206] The processor, the memory, and the computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the high-energy beam slit three-dimensional point cloud data processing and manifold modeling method as described in any of the above embodiments.

[0207] Based on the same inventive concept, embodiments of the present invention also provide a computer-readable storage medium, comprising: when the instructions in the storage medium are executed by a processor of an electronic device, enabling the electronic device to perform the high-energy beam slit three-dimensional point cloud data processing and manifold modeling method as described in any of the above embodiments.

[0208] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0209] For the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, as some steps can be performed in other orders or simultaneously according to the present invention. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and components involved are not necessarily essential to the present invention.

[0210] The above provides a detailed description of a high-energy beam slit 3D point cloud data processing and manifold modeling method provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for processing and manifold modeling three-dimensional point cloud data with high-energy beam slits, characterized in that, The method includes: Two identical sample blocks are spliced ​​together to obtain an experimental block, and a high-energy beam is used to cut the splicing surface of the experimental block; wherein, after cutting, each sample block has a cutting mark at the splicing surface; The cutting front point cloud data of the point cloud data corresponding to the multi-layer cutting of the experimental block are selected. The multi-layer cutting is obtained by dividing the three-dimensional topography point cloud data corresponding to the sample block into layers in the depth direction of the cutting marks. The cutting front point cloud data is the cutting mark topography data that is close to the actual high-energy beam shape. Using an elliptical and / or quasi-elliptical shape, the cutting front point cloud data of each cutting layer of the experimental block are fitted to obtain the cutting contour, which includes at least contour coordinates and feature information; wherein, the cutting contour corresponding to each cutting layer is the cross-sectional contour of the manifold of the cutting mark in that cutting layer. The cutting contours corresponding to each of the multiple cutting layers are imported into the 3D modeling software, and the lofting command is used to construct the 3D model of the high-energy beam shape. The step of selecting the cutting front point cloud data from the point cloud data corresponding to the multiple cutting layers of the experimental block includes: Acquire initial point cloud data for each sample block in the experimental block. The initial point cloud data includes point cloud data on at least multiple surfaces of each sample block, and the multiple surfaces include at least an upper surface, a lower surface, and the splicing surface. The initial point cloud data is filtered at least once to obtain the target three-dimensional point cloud model of the experimental block; wherein each subsequent filtering is performed based on the previous filtering, and each filtering is used to remove point cloud data that is irrelevant to the cutting marks. In the depth direction of the cutting marks, the target three-dimensional point cloud model is layered according to a preset distance interval to obtain multiple cutting layers; Based on the point cloud data of each cutting layer, the corresponding cutting front point cloud data are selected.

2. The method for processing high-energy beam slit 3D point cloud data and manifold modeling according to claim 1, characterized in that, The initial point cloud data of the experimental block is acquired using laser scanning. The initial point cloud data is then filtered at least once to obtain the target 3D point cloud model of the experimental block, including: Based on the initial point cloud data, the experimental block is reconstructed in three dimensions to obtain the initial three-dimensional point cloud model of the experimental block. Remove point cloud data that is irrelevant to the cutting marks from the initial 3D point cloud model to obtain the filtered 3D point cloud model; For each fitted surface of the filtered 3D point cloud model, the point cloud data within a first preset distance from the surface are removed again to remove the error point cloud data obtained by the laser scan. The model after undergoing another 3D fitting is used as the target 3D point cloud model. The first preset distance is determined based on the resolution of the laser scan.

3. The method for processing high-energy beam slit 3D point cloud data and manifold modeling according to claim 1, characterized in that, The initial point cloud data of the experimental block is acquired using laser scanning. The initial point cloud data is then filtered at least once to obtain the target 3D point cloud model of the experimental block, including: Based on the initial point cloud data, the experimental block is reconstructed in three dimensions to obtain the initial three-dimensional point cloud model of the experimental block. Remove point cloud data that is irrelevant to the cutting marks from the initial 3D point cloud model to obtain the filtered 3D point cloud model; For each fitted surface of the filtered 3D point cloud model, a bounding box with a preset thickness is constructed to surround the surface, and the point cloud data within the bounding box is removed. The model after removing the point cloud data within the bounding box is taken as the target 3D point cloud model; The preset thickness is determined based on the resolution of the laser scan.

4. The method for processing high-energy beam slit three-dimensional point cloud data and manifold modeling according to claim 1, characterized in that, The cutting process includes a first stage and a second stage, wherein the step of filtering out the corresponding cutting front point cloud data based on the point cloud data of each cutting layer includes: The point cloud data of each cutting layer is fitted to obtain two non-intersecting fitted straight lines; wherein, the two fitted straight lines of the same cutting layer have a vertical distance, which is the width of the cutting mark in the cutting layer. Point cloud data whose distance from the corresponding fitted straight line is greater than a second preset distance are identified as the cutting front point cloud data of the cutting layer; and Point cloud data whose distance to the corresponding fitted straight line is less than or equal to the second preset distance are identified as cutting sidewall point cloud data. The cutting sidewall point cloud data is formed in the first stage, wherein the second stage includes the current time and the first stage includes any time before the second stage.

5. The method for processing high-energy beam slit 3D point cloud data and manifold modeling according to claim 1, characterized in that, The preset distance interval is less than or equal to one-fifth of the length of the experimental block in the depth direction of the cutting mark.

6. The method according to claim 4, characterized in that, The point cloud data of each cutting layer is acquired by laser scanning, and the second preset distance is determined based on the resolution of the laser scan.

7. The method for processing high-energy beam slit 3D point cloud data and manifold modeling according to claim 1, characterized in that, The step of cutting the experimental block by aligning a high-energy beam with the splicing surface includes: The high-energy beam was used to cut the splicing surface of the experimental block under various preset cutting conditions. The parameters of the preset cutting conditions included at least one of the following: cutting speed, cutting pressure, abrasive output rate, cutting material, material thickness, nozzle / abrasive tube size, target distance, abrasive type, and abrasive mesh size. The method further includes: Based on the point cloud data of the experimental block under each preset cutting condition, a three-dimensional model of the high-energy beam shape corresponding to that preset cutting condition is constructed.

8. The method for processing three-dimensional point cloud data and modeling manifolds of high-energy beam slits according to claim 7, characterized in that, After constructing a three-dimensional model of the high-energy beam shape corresponding to each preset cutting condition based on the point cloud data of the experimental block under each preset cutting condition, the method further includes: Obtain the actual cutting conditions at the current moment; The actual cutting conditions are matched with a variety of preset cutting conditions; Retrieve the 3D model of the high-energy beam shape corresponding to the preset cutting conditions.

Images