A workpiece point cloud reverse modeling method, a computer device and a readable storage medium
By combining dynamic indexing and preprocessing, point cloud files can be identified and distinguished, enabling partial stitching and model merging. This solves the problem of stitching failures and breakpoints in point cloud data management, and improves modeling efficiency and stability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FAIR INNOVATION (SUZHOU) ROBOTIC SYSTEM CO LTD
- Filing Date
- 2026-04-27
- Publication Date
- 2026-07-21
AI Technical Summary
The existing point cloud data management and stitching operations lack dynamic perception capabilities, making it difficult to locate breakpoints in stitching failures. This requires full data backtracking, resulting in a large amount of redundant computation, which affects modeling efficiency. Furthermore, abnormal steps trigger a full process restart, which seriously affects modeling stability and efficiency.
By dynamically assigning index identifiers and combining them with point cloud preprocessing, invalid and valid point cloud files can be identified, enabling local point cloud stitching and model merging, effectively reusing intermediate modeling results, and improving stitching reliability and modeling efficiency.
It enables effective differentiation and recognition of point cloud files, reduces the amount of modeling computation, improves the stability and overall efficiency of the modeling process, and ensures good point cloud management and stitching fault tolerance capabilities.
Smart Images
Figure CN122435148A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of point cloud scanning modeling technology, and more specifically, to a reverse modeling method for workpiece point clouds, a computer device, and a readable storage medium. Background Technology
[0002] With the continuous development of science and technology, the application of 3D point cloud scanning technology in the industrial manufacturing field is becoming increasingly widespread. It often requires the use of point cloud reverse modeling technology to construct a 3D point cloud model of the target workpiece, so that various subsequent analysis tasks can be performed based on this 3D workpiece point cloud model (e.g., weld distribution analysis, welding torch obstacle avoidance trajectory planning, etc.). For point cloud reverse modeling technology, point cloud data management and stitching operations are crucial aspects of its technical implementation.
[0003] Currently, traditional point cloud data management and stitching operations rely on a static point cloud indexing mechanism. This mechanism uses a linear sequence index architecture to adaptively generate matching index identifiers following the point cloud file generation sequence to achieve the physical storage mapping of point cloud files and the file stitching and calling effect. However, it is worth noting that this point cloud file processing mechanism inherently lacks dynamic awareness of the reverse modeling process status (e.g., point cloud stitching progress, stitching failure breakpoints, point cloud incremental correlation, etc.). It often cannot locate stitching failure breakpoints caused by factors such as point cloud scanning interruption or single-point scanning failure, requiring a large amount of computing resources to perform full data backtracking to effectively remove stitching failure breakpoints. In addition, this point cloud file processing mechanism requires re-stitching the new point cloud data with all historical point cloud data every time it faces an incremental modeling scenario. The generated intermediate modeling results cannot be effectively reused, resulting in redundant computing load growing exponentially with the scale of point cloud data, seriously affecting modeling efficiency. Even more seriously, any abnormal link in this point cloud file processing mechanism will trigger a complete restart of the entire point cloud reverse modeling process, seriously affecting modeling efficiency. Summary of the Invention
[0004] In view of this, the purpose of this application is to provide a workpiece point cloud reverse modeling method, computer equipment, and readable storage medium, which can dynamically assign an index identifier to the organic combination between operations and point cloud preprocessing operations, realize the effective differentiation and identification of invalid point cloud files (i.e., preprocessed point cloud files with failed splicing breakpoints) and valid point cloud files, facilitate the improvement of subsequent point cloud splicing reliability, and effectively reuse the generated intermediate modeling results for workpiece reverse modeling in incremental modeling scenarios by utilizing local point cloud splicing operations and point cloud model merging operations, thereby significantly reducing the modeling computation load, effectively improving the stability of the modeling process and the overall modeling efficiency, and ensuring good point cloud management and splicing fault tolerance capabilities.
[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: In a first aspect, this application provides a method for reverse modeling of workpiece point clouds, the method comprising: Acquire multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the execution of this scanning task; Based on recent index identifier record data, the multiple local point cloud files are dynamically assigned index identifiers and preprocessed to obtain the actual index number of each of the successfully preprocessed valid point cloud files, wherein the actual index numbers of each of the valid point cloud files are continuously distributed among each other. Based on the actual index number of each of the valid point cloud files, the point cloud files are stitched together to obtain the corresponding local point cloud model. Detect whether this scanning task is the first scanning task for the workpiece to be modeled; If the current scanning task is detected as the first scanning task, the local point cloud model is directly used as the target workpiece point cloud model after the workpiece to be modeled is completed in this scanning task. Otherwise, the local point cloud model is merged with the latest workpiece point cloud model to obtain the target workpiece point cloud model, wherein the latest workpiece point cloud model is the workpiece point cloud model of the workpiece to be modeled after the previous scanning task was completed.
[0006] In an optional implementation, the recent index identifier record data includes mutually related latest index number and latest index name, mutually related second-newest index number and second-newest index name, and mutually related third-newest index number and third-newest index name. Then, the step of dynamically assigning index identifiers to the multiple local point cloud files based on the recent index identifier record data and performing point cloud preprocessing to obtain the actual index number of each of the successfully preprocessed valid point cloud files includes: The multiple local point cloud files are traversed sequentially, and for each local point cloud file traversed, it is detected whether the local point cloud file belongs to the first point cloud file in the first scan task execution process; If it is detected that the local point cloud file belongs to the first point cloud file, the second new index number and the third new index number are set to the default index values respectively, and the second new index name and the third new index name are set to empty strings respectively; The theoretical index value and the actual file name corresponding to the local point cloud file are assigned one-to-one to the latest index number and the latest index name, and point cloud preprocessing is performed on the local point cloud file, wherein the theoretical index value corresponding to the local point cloud file is obtained by adding one to the default index value; The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number after assignment is taken as the actual index number of the valid point cloud file.
[0007] In an optional implementation, the step of dynamically assigning index identifiers to the plurality of local point cloud files based on recent index identifier record data and performing point cloud preprocessing to obtain the actual index number of each of the valid point cloud files that have been successfully preprocessed further includes: If it is detected that the local point cloud file does not belong to the first point cloud file, it is checked whether the target point cloud file pointed to by the latest index name has been successfully preprocessed; When the preprocessing of the target point cloud file is detected to be successful, the second new index number and the second new index name are assigned to the next new index number and the next new index name in a one-to-one correspondence, and the latest index number and the latest index name are assigned to the second new index number and the second new index name in a one-to-one correspondence. The theoretical index value and the actual file name corresponding to the local point cloud file are assigned one-to-one to the latest index number and the latest index name, and point cloud preprocessing is performed on the local point cloud file. The theoretical index value corresponding to the local point cloud file is obtained by adding one to the latest index number before the assignment. The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number after assignment is taken as the actual index number of the valid point cloud file.
[0008] In an optional implementation, the step of dynamically assigning index identifiers to the plurality of local point cloud files based on recent index identifier record data and performing point cloud preprocessing to obtain the actual index number of each of the valid point cloud files that have been successfully preprocessed further includes: When the preprocessing of the target point cloud file is detected to be unsuccessful, the second new index number and the second new index name are assigned to the third new index number and the third new index name in a one-to-one correspondence. Assign the actual filename of the local point cloud file to the latest index name, and perform point cloud preprocessing on the local point cloud file; The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number is taken as the actual index number of the valid point cloud file.
[0009] In an optional implementation, the step of stitching together all valid point cloud files according to their respective actual index numbers to obtain a corresponding local point cloud model includes: Based on the maximum modeled index number before the execution of this scanning task, target valid point cloud files whose corresponding actual index number exceeds the maximum modeled index number are screened among all valid point cloud files, wherein the maximum modeled index number is the default index value before the execution of the first scanning task. All selected target valid point cloud files are stitched together, and the stitched 3D point cloud model is used as the local point cloud model. Then, the maximum modeled index number is updated according to the maximum actual index number of all target valid point cloud files.
[0010] In an optional implementation, the method further includes: Perform model integrity detection on the point cloud model of the target workpiece; If the point cloud model of the target workpiece is found to be incomplete, a supplementary scanning task for the workpiece to be modeled is issued to the point cloud scanning device, and the above steps of obtaining multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the execution of this scanning task continue to be performed. If the target workpiece point cloud model is found to be complete, point cloud filtering downsampling and maximum connected component extraction are performed on the target workpiece point cloud model to obtain the three-dimensional workpiece point cloud model of the workpiece to be modeled.
[0011] In an optional implementation, the method further includes: Based on the three-dimensional workpiece point cloud model, workpiece weld analysis is performed to obtain the corresponding weld database. The weld database is subjected to a database integrity check. If the weld database is found to be incomplete, a supplementary scanning task for the workpiece to be modeled is issued to the point cloud scanning device. The steps described above, which involve obtaining multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the current scanning task, are continued until the weld database is complete.
[0012] In an optional implementation, the method includes: Construct a collision model of the workpiece to be modeled based on the three-dimensional workpiece point cloud model; The collision model of the workpiece is subjected to model integrity detection. If the collision model of the workpiece is found to be incomplete, a supplementary scanning task for the workpiece to be modeled is issued to the point cloud scanning device. The steps described above, which involve obtaining multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the execution of this scanning task, are continued until the collision model of the workpiece is complete.
[0013] Secondly, this application provides a computer device, including a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and the processor can execute the computer program to implement the workpiece point cloud reverse modeling method described in any of the foregoing embodiments.
[0014] Thirdly, this application provides a readable storage medium storing a computer program thereon, which, when executed by a computer device, implements the workpiece point cloud reverse modeling method described in any of the foregoing embodiments.
[0015] In this case, the beneficial effects of the embodiments of this application may include the following: This application, after obtaining multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the current scanning task, dynamically assigns index identifiers and performs point cloud preprocessing on these multiple local point cloud files based on recent index identifier record data (which records the recent index identifier change history during the index identifier assignment process). This organic combination of dynamic index identifier assignment and point cloud preprocessing effectively distinguishes between invalid point cloud files (those with failed preprocessing points involving stitching failures) and valid point cloud files (i.e., successfully preprocessed point cloud files). This yields the actual index numbers of all valid point cloud files, which are continuously distributed, thus improving the reliability of subsequent point cloud stitching. Then, based on the actual index numbers of each valid point cloud file, point cloud processing is performed on these valid point cloud files. By stitching together the data, a corresponding local point cloud model is obtained. At this point, if the current scanning task is the first scanning task for the workpiece to be modeled, this local point cloud model can be directly used as the target workpiece point cloud model after the completion of this scanning task. Alternatively, if the current scanning task is not the first scanning task, this local point cloud model can be merged with the latest workpiece point cloud model (which is the workpiece point cloud model after the completion of the previous scanning task) to obtain the target workpiece point cloud model after the completion of this scanning task. This allows for the effective reuse of the generated intermediate modeling results for reverse modeling of the workpiece in incremental modeling scenarios by utilizing local point cloud stitching operations and point cloud model merging operations. This significantly reduces the amount of modeling computation, thereby effectively improving the stability of the modeling process and the overall modeling efficiency, and ensuring good point cloud management and stitching fault tolerance.
[0016] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description
[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 A schematic diagram of the composition of a computer device provided in the embodiments of this application; Figure 2 This is one of the flowcharts illustrating the reverse modeling method for workpiece point clouds provided in the embodiments of this application; Figure 3 for Figure 2 A flowchart illustrating the sub-steps included in step S220; Figure 4 The second flowchart illustrates the reverse modeling method for workpiece point clouds provided in this application embodiment. Figure 5 The third flowchart illustrates the reverse modeling method for workpiece point clouds provided in this application embodiment; Figure 6 The fourth flowchart illustrates the reverse modeling method for workpiece point clouds provided in this application embodiment.
[0019] Icons: 10-Computer equipment; 11-Memory; 12-Processor; 13-Communication unit. Detailed Implementation
[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0021] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0022] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.
[0023] In the description of this application, it should be understood that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product is in use, or the orientation or positional relationship commonly understood by those skilled in the art. They are used only for the convenience of describing this application and simplifying the description, and are not intended to indicate or imply that the equipment or component referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.
[0024] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "set up," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.
[0025] In the description of this application, it should be understood that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Those skilled in the art will understand the specific meaning of the above terms in this application based on the specific circumstances.
[0026] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.
[0027] Please refer to the reference. Figure 1 , Figure 1This is a schematic diagram of the computer device 10 provided in an embodiment of this application. In this embodiment, the computer device 10 can be communicatively connected to a point cloud scanning device to control the point cloud scanning device to perform three-dimensional point cloud scanning on the workpiece to be modeled, thereby obtaining the workpiece surface point cloud data of the workpiece to be modeled, and performing reverse modeling processing on the workpiece surface point cloud data to obtain a three-dimensional workpiece point cloud model that can be used to perform various analysis tasks (e.g., weld distribution analysis task, welding torch motion obstacle avoidance trajectory planning task, etc.). The computer device 10 can be integrated with the point cloud scanning device, or it can be an electronic device independent of the point cloud scanning device. The electronic device can be, but is not limited to, a server, a personal computer, a laptop, etc. The point cloud scanning device can be composed of multiple point cloud scanners (which can be implemented using point cloud scanning devices such as depth cameras and LiDAR) deployed around the workpiece to be modeled. Each point cloud scanner corresponds to a scanning point around the workpiece to be modeled. The point cloud scanning device can also be implemented as a robotic arm equipped with point cloud scanning devices, so that by adjusting the pose of the end effector of the robotic arm, the point cloud scanning devices (located at the end effector of the robotic arm) can realize the point cloud acquisition function at each scanning point around the workpiece to be modeled.
[0028] In this embodiment, the computer device 10 may include a memory 11, a processor 12, and a communication unit 13. The memory 11, the processor 12, and the communication unit 13 are electrically connected to each other directly or indirectly to achieve data transmission or interaction. For example, these components can be electrically connected to each other via one or more communication buses or signal lines.
[0029] In this embodiment, the memory 11 may be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory 11 is used to store computer programs, and the processor 12 can execute the computer programs accordingly after receiving execution instructions.
[0030] In this embodiment, the processor 12 can be an integrated circuit chip with signal processing capabilities. The processor 12 can be a general-purpose processor, including at least one of a central processing unit (CPU), graphics processing unit (GPU), network processor (NP), digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.
[0031] In this embodiment, the communication unit 13 is used to establish a communication connection between the computer device 10 and other electronic devices through a network, and to send and receive data through the network, wherein the network includes wired communication networks and wireless communication networks. For example, the computer device 10 can obtain the workpiece surface point cloud data (i.e., the real point cloud data of the workpiece surface in three-dimensional space) from the point cloud scanning device through the communication unit 13, and perform reverse modeling of the workpiece based on the workpiece surface point cloud data.
[0032] In this embodiment, the computer device 10 can pre-store a specific computer program related to the reverse modeling function of workpiece point cloud in the memory 11. By driving the processor 12 to execute the specific computer program, the operation and point cloud preprocessing operation are dynamically assigned by the index identifier, which realizes the effective differentiation and identification of invalid point cloud files and valid point cloud files. This facilitates the improvement of the reliability of subsequent point cloud stitching. Furthermore, by utilizing the local point cloud stitching operation and the point cloud model merging operation, the generated intermediate modeling results can be effectively reused for workpiece reverse modeling in the incremental modeling scenario, thereby significantly reducing the amount of modeling calculations. This effectively improves the stability of the modeling process and the overall modeling efficiency, and ensures good point cloud management and stitching fault tolerance.
[0033] Understandable, Figure 1 The block diagram shown is only a schematic diagram of one configuration of the computer device 10. The computer device 10 may also include components such as... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown. Figure 1 The components shown can be implemented using hardware, software, or a combination thereof.
[0034] In this application, to ensure that the computer device 10 can effectively improve the stability of the modeling process and the overall modeling efficiency in the reverse modeling of workpiece point clouds, and to ensure good point cloud management and stitching fault tolerance, this application embodiment achieves the aforementioned objective by providing a workpiece point cloud reverse modeling method. The workpiece point cloud reverse modeling method provided in this application will be described in detail below.
[0035] Please refer to Figure 2 , Figure 2 This is one of the flowcharts illustrating the reverse modeling method for workpiece point clouds provided in this application embodiment. In this application embodiment, Figure 2 The reverse modeling method for workpiece point cloud shown may include steps S210 to S260.
[0036] Step S210: Obtain multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the execution of this scanning task.
[0037] In this embodiment, any scanning task issued to the point cloud scanning device can record multiple point cloud acquisition points (each point cloud acquisition point is distributed around the workpiece to be modeled), so that the point cloud scanning device completes the point cloud scanning operation on the workpiece to be modeled at each point cloud acquisition point to obtain multiple local point cloud files, wherein each local point cloud file corresponds to a point cloud acquisition point around the workpiece to be modeled; each local point cloud file records the local workpiece surface point cloud data observed by the point cloud scanning device at the corresponding point cloud acquisition point, as well as the actual observed pose data of the corresponding local workpiece surface point cloud data.
[0038] In one embodiment of this invention, when the point cloud scanning device is implemented using a robotic arm equipped with a point cloud scanning device, the point cloud scanning device can be compatible with multiple point cloud scanning modes (e.g., "manual traction scanning mode (which requires manual dragging of the robotic arm to drive the point cloud scanning device to perform point cloud scanning at different point cloud acquisition points)", "fixed viewing angle scanning mode (which requires the robotic arm to move the point cloud scanning device to different point cloud acquisition points to perform point cloud scanning at a fixed viewing angle)", "viewpoint swing scanning mode (which requires the robotic arm to move the point cloud scanning device to different point cloud acquisition points to perform viewpoint swing scanning)", etc.). Any scanning task assigned to the point cloud scanning device can also record a desired scanning mode (which belongs to any of the aforementioned multiple point cloud scanning modes; the desired scanning modes recorded for different scanning tasks can be the same or different), so that the point cloud scanning device performs point cloud scanning operations according to the corresponding desired scanning mode. This reduces the missed scanning rate of the workpiece to be modeled through the coordinated operation of multiple point cloud scanning modes.
[0039] Step S220: Dynamically assign index identifiers and perform point cloud preprocessing on the multiple local point cloud files based on recent index identifier record data to obtain the actual index number of each of the successfully preprocessed valid point cloud files.
[0040] In this embodiment, the recent index identifier record data records the recent index identifier change history during the index identifier assignment process. It may include the latest index number and latest index name that are mutually related, the second newest index number and second newest index name that are mutually related, and the next newest index number and next newest index name that are mutually related. The computer device 10 can sequentially assign index identifiers to the multiple local point cloud files. For the current local point cloud file that needs to be assigned an index identifier, it will update the index number and index name of the current local point cloud file based on the recent index identifier records and the success or failure results of the preprocessing of previous local point cloud files. This is to achieve effective differentiation and identification of invalid and valid point cloud files through the organic combination of the dynamic index identifier assignment operation and the point cloud preprocessing operation. This ensures that the actual index numbers of all valid point cloud files are continuously distributed, thus guaranteeing the orderliness of the reverse modeling process of the workpiece (i.e., the actual index number of each valid point cloud file can represent its preprocessing sequence and point cloud splicing sequence). The actual index names of all valid point cloud files are used to ensure data traceability (i.e., the actual index name of each valid point cloud file can be used to realize the function of fast search of original files). In addition, after determining the actual index number of each of the valid point cloud files involved in this scanning task, the computer device 10 will store all the valid point cloud files in the stitched point cloud data storage space to achieve the local storage effect of all the valid point cloud files. The stitched point cloud data storage space is used to store the valid point cloud files involved in each scanning task for the workpiece to be modeled.
[0041] Alternatively, please refer to Figure 3 , Figure 3 yes Figure 2 A flowchart illustrating the sub-steps included in step S220. In this embodiment, step S220 may include sub-steps S221 to S2211 to effectively distinguish between invalid and valid point cloud files and establish a continuous index relationship between valid point cloud files, thereby ensuring the orderliness of the reverse modeling process and improving the reliability of subsequent point cloud stitching.
[0042] Sub-step S221: Iterate through multiple local point cloud files in sequence, and for each local point cloud file iterates through, check whether the local point cloud file belongs to the first point cloud file in the first scan task execution process.
[0043] In this embodiment, the computer device 10 can execute the sub-step S221 according to the point cloud scanning order of multiple local point cloud files belonging to the same scanning task. When a local point cloud file is the first point cloud file scanned for the workpiece to be modeled, the computer device 10 will execute sub-step S222 for the local point cloud file; and when the local point cloud file is not the first point cloud file, the computer device 10 will execute sub-step S225 for the local point cloud file.
[0044] Sub-step S222: Set the second new index number and the third new index number to the default index values respectively, and set the second new index name and the third new index name to empty strings respectively.
[0045] In this embodiment, the default index value is used to indicate that the computer device 10 has not yet processed any local point cloud file of the workpiece to be modeled, and it can be represented by "-1" or "0".
[0046] Sub-step S223 assigns the theoretical index value and actual file name corresponding to the local point cloud file to the latest index number and latest index name in a one-to-one correspondence, and performs point cloud preprocessing on the local point cloud file.
[0047] In this embodiment, for a local point cloud file belonging to the first point cloud file, its corresponding theoretical index value is obtained by adding one to the default index value. Point cloud preprocessing operations for any local point cloud file may include: filtering out null points and NAN (Not a Number) points from the corresponding local workpiece surface point cloud data to obtain the corresponding valid surface point cloud data; and based on the actual observed pose data corresponding to the valid surface point cloud data, transforming and mapping the valid surface point cloud data to the world coordinate system to obtain valid real point cloud data in the world coordinate system. Specifically, when the valid real point cloud data belongs to an empty set, it indicates that the preprocessing of the local point cloud file has failed. In this case, the modeling failure flag configured for the local point cloud file can be activated to indicate that the preprocessing of the local point cloud file has failed. Conversely, when the valid real point cloud data belongs to a valid dataset, it indicates that the preprocessing of the local point cloud file has succeeded. In this case, the modeling failure flag of the local point cloud file will be set to an inactive state to indicate that the preprocessing of the local point cloud file has succeeded.
[0048] Sub-step S224: The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number after assignment is taken as the actual index number of the valid point cloud file.
[0049] In this embodiment, when the preprocessing of a local point cloud file belonging to the first point cloud file is successful, the preprocessed local point cloud file can be used as a valid point cloud file for this scanning task. At this time, the latest index number after being assigned is directly used as the actual index number of the valid point cloud file, and the latest index number after being assigned is directly used as the actual index name of the valid point cloud file, so as to complete the effect of assigning an index identifier to the valid point cloud file.
[0050] Sub-step S225: Check whether the target point cloud file pointed to by the latest index name has been successfully preprocessed.
[0051] In this embodiment, when a local point cloud file encountered during traversal does not belong to the first point cloud file, the preprocessing success of the target point cloud file can be determined by detecting whether the modeling failure flag of the target point cloud file (which is essentially a local point cloud file traversed before the local point cloud file) pointed to by the latest index name recorded at that time is inactive. Specifically, if the target point cloud file is successfully preprocessed, it indicates that the target point cloud file is essentially a valid point cloud file. In this case, the theoretical index value of the local point cloud file (obtained by adding one to the latest index number recorded at that time) needs to be calculated using the latest index number recorded at that time. The computer device 10 will then execute sub-step S226 for the local point cloud file encountered during traversal. Conversely, if the target point cloud file fails to preprocess, it indicates that the target point cloud file is essentially an invalid point cloud file. In this case, the latest index number recorded at that time needs to be reassigned to the local point cloud file. The computer device 10 will then execute sub-step S229 for the local point cloud file encountered during traversal.
[0052] Sub-step S226: Assign the second new index number and the second new index name to the next new index number and the next new index name in a one-to-one correspondence, and assign the latest index number and the latest index name to the second new index number and the second new index name in a one-to-one correspondence.
[0053] In this embodiment, for any local point cloud file that is traversed, when the previous target point cloud file traversed by the local point cloud file is successfully preprocessed, the second new index number and the second new index name recorded at that time can be directly assigned to the next new index number and the next new index name in a one-to-one correspondence, and the latest index number and the latest index name recorded at that time can be assigned to the second new index number and the second new index name in a one-to-one correspondence.
[0054] Sub-step S227 assigns the theoretical index value and actual file name corresponding to the local point cloud file to the latest index number and latest index name in a one-to-one correspondence, and performs point cloud preprocessing on the local point cloud file.
[0055] In this embodiment, the theoretical index value corresponding to the local point cloud file after executing sub-step S226 is obtained by adding one to the latest index number before the assignment.
[0056] Sub-step S228: The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number after assignment is taken as the actual index number of the valid point cloud file.
[0057] In this embodiment, when any partial point cloud file that does not belong to the first point cloud file is successfully preprocessed, the preprocessed partial point cloud file can be used as a valid point cloud file for this scanning task. At this time, the latest index number after being assigned is directly used as the actual index number of the valid point cloud file, and the latest index number after being assigned is directly used as the actual index name of the valid point cloud file, so as to complete the effect of assigning index identification to the valid point cloud file.
[0058] Sub-step S229: Assign the second new index number and the second new index name to the third new index number and the third new index name in a one-to-one correspondence.
[0059] In this embodiment, for any local point cloud file traversed, when the preprocessing of the previous target point cloud file traversed for that local point cloud file fails, it is necessary to retain the second new index number and the second new index name recorded at that time, and assign the second new index number and the second new index name to the third new index number and the third new index name recorded at that time in a one-to-one correspondence, so as to automatically fall back to the previous valid point cloud file, and then execute sub-step S2210 for that local point cloud file.
[0060] Sub-step S2210 assigns the actual filename of the local point cloud file to the latest index name and performs point cloud preprocessing on the local point cloud file.
[0061] In this embodiment, the actual filename of the local point cloud file can be assigned to the latest index name recorded at that time, thereby achieving an effective association between the actual filename of the local point cloud file and the latest index number recorded at that time (which is essentially obtained by adding one to the second newest index number recorded at that time).
[0062] Sub-step S2211: The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number is taken as the actual index number of the valid point cloud file.
[0063] In this embodiment, when any partial point cloud file that does not belong to the first point cloud file is successfully preprocessed, the preprocessed partial point cloud file can be used as a valid point cloud file for this scanning task. At this time, the latest index number recorded at that time is directly used as the actual index number of the valid point cloud file, and the latest index number after being assigned is directly used as the actual index name of the valid point cloud file, so as to complete the effect of assigning the index identifier to the valid point cloud file.
[0064] Therefore, by executing the above sub-steps S221 to S2211, this application can effectively distinguish and identify invalid point cloud files and establish a continuous index relationship between valid point cloud files, thereby ensuring the orderliness of the reverse modeling process of the workpiece and facilitating the improvement of the reliability of subsequent point cloud splicing.
[0065] Step S230: Based on the actual index number of each of the valid point cloud files, perform point cloud stitching on all valid point cloud files to obtain the corresponding local point cloud model.
[0066] In this embodiment, the computer device 10 pre-configures a maximum modeled index number for the workpiece to be modeled (which represents the maximum existing index number after each reverse modeling operation for the workpiece to be modeled, wherein each scanning task corresponds to a separate reverse modeling operation), so as to extract all valid point cloud files strongly related to the current scanning task from the stitched point cloud data storage space to realize the local point cloud stitching operation, wherein the maximum modeled index number is the default index value before the first scanning task is executed. Therefore, step S230 may include: based on the maximum modeled index number before the current scanning task is executed, filtering out target valid point cloud files whose corresponding actual index number exceeds the maximum modeled index number from all valid point cloud files stored in the stitched point cloud data storage space; performing point cloud stitching on all the filtered target valid point cloud files, and using the stitched 3D point cloud model as the local point cloud model, and then updating the maximum modeled index number according to the maximum actual index number of all target valid point cloud files.
[0067] Therefore, this application can achieve the effect of local stitching modeling of effective incremental point cloud data in the incremental modeling scenario through the specific details of the above step S230.
[0068] Step S240: Detect whether this scanning task is the first scanning task for the workpiece to be modeled.
[0069] In this embodiment, when the current scanning task is the first scanning task for the workpiece to be modeled, it indicates that the current scanning task is not a supplementary scanning task in the incremental modeling scenario, and step S250 can be executed directly; while when the current scanning task is not the first scanning task for the workpiece to be modeled, it indicates that the current scanning task is actually a supplementary scanning task in the incremental modeling scenario, and step S260 can be executed directly.
[0070] Step S250: The local point cloud model is directly used as the target workpiece point cloud model after the workpiece to be modeled is completed in this scanning task.
[0071] Step S260: Merge the local point cloud model with the latest workpiece point cloud model to obtain the target workpiece point cloud model.
[0072] In this embodiment, the latest workpiece point cloud model is the workpiece point cloud model of the workpiece to be modeled after the completion of the last scanning task.
[0073] Therefore, by executing the above steps S210 to S260, this application utilizes the organic combination of index identification dynamic assignment operation and point cloud preprocessing operation to effectively distinguish and identify invalid and valid point cloud files, and establishes a continuous index relationship between valid point cloud files. This ensures the orderliness of the reverse modeling process for the workpiece, facilitates the improvement of the reliability of subsequent point cloud stitching, and then uses local point cloud stitching operation and point cloud model merging operation to effectively reuse the generated intermediate modeling results for reverse modeling of the workpiece in the incremental modeling scenario. This significantly reduces the amount of modeling computation, thereby effectively improving the stability of the modeling process and the overall modeling efficiency, and ensuring good point cloud management and stitching fault tolerance.
[0074] Alternatively, please refer to Figure 4 , Figure 4 This is the second flowchart illustrating the reverse modeling method for workpiece point clouds provided in this application embodiment. In this application embodiment, with... Figure 2 Compared to the reverse modeling method of workpiece point cloud shown, Figure 4 The reverse modeling method for workpiece point cloud shown may also include steps S270 to S290, in conjunction with the steps S210 to S280 above, to ensure that the final constructed three-dimensional workpiece point cloud model can reflect the workpiece surface details of the workpiece to be modeled as comprehensively as possible.
[0075] Step S270: Perform model integrity detection on the point cloud model of the target workpiece.
[0076] In this embodiment, after obtaining the target workpiece point cloud model after the completion of this scanning task, a model integrity check can be performed on the target workpiece point cloud model to determine the comprehensiveness of its exposure of the surface details of the workpiece to be modeled. Specifically, if the target workpiece point cloud model is incomplete, it indicates that the target workpiece point cloud model cannot fully expose the surface details of the workpiece to be modeled, and step S280 will be executed accordingly; conversely, if the target workpiece point cloud model is complete, it indicates that the target workpiece point cloud model can fully expose the surface details of the workpiece to be modeled, and step S290 will be executed accordingly.
[0077] Step S280: Issue a supplementary scanning task for the workpiece to be modeled to the point cloud scanning device.
[0078] In this embodiment, the missing point cloud region at the workpiece to be modeled can be determined based on the target workpiece point cloud model. Then, according to the actual distribution location and shape characteristics of the missing point cloud region at the workpiece to be modeled, a desired scanning mode and a desired set of acquisition points (including at least one point cloud acquisition point) adapted to the missing point cloud region are selected to generate a scanning task. The generated scanning task is then issued to the point cloud scanning device as a supplementary scanning task for execution, so that the computer device 10 can jump to the above step S210 to continue execution during or after the execution of the supplementary scanning task, until the final constructed target workpiece point cloud model is complete.
[0079] Step S290: Perform point cloud filtering downsampling and maximum connected component extraction on the target workpiece point cloud model to obtain the three-dimensional workpiece point cloud model of the workpiece to be modeled.
[0080] In this embodiment, the point cloud data size of the target workpiece point cloud model can be reduced by point cloud filtering downsampling operation, and the three-dimensional workpiece point cloud model that is strongly correlated with the workpiece to be modeled can be directly extracted from the target workpiece point cloud model by the maximum connected component extraction operation (which can be implemented by KD-Tree (K-Dimensional Tree) nearest neighbor search algorithm), so that it can reflect the workpiece surface details of the workpiece to be modeled as comprehensively as possible.
[0081] Therefore, by executing the above steps S210 to S290, this application can ensure the orderly nature of the reverse modeling process, the stability of the modeling process, the overall modeling efficiency, and the good point cloud management and splicing fault tolerance capabilities, so that the final constructed three-dimensional workpiece point cloud model can reflect the workpiece surface details of the workpiece to be modeled as comprehensively as possible.
[0082] Alternatively, please refer to Figure 5 , Figure 5This is the third flowchart illustrating the reverse modeling method for workpiece point clouds provided in this application embodiment. In this application embodiment, with... Figure 4 Compared to the reverse modeling method of workpiece point cloud shown, Figure 5 The reverse modeling method for workpiece point cloud shown may also include steps S2100 to S2110, in conjunction with the above steps S210 to S290, to ensure that the final weld database can reflect the effective weld distribution at the workpiece to be modeled as comprehensively as possible.
[0083] Step S2100: Perform weld seam analysis on the workpiece based on the 3D workpiece point cloud model to obtain the corresponding weld seam database.
[0084] In this embodiment, for the three-dimensional workpiece point cloud model determined by the above step S290, weld data extraction strategies such as surface structure lookup association method based on explicit plane detection operation, global point cloud registration and semantic segmentation method, and end-to-end deep learning model can be used to extract the weld database involved in the three-dimensional workpiece point cloud model.
[0085] Step S2110: Perform a database integrity check on the weld database.
[0086] In this embodiment, after determining the weld database based on step S2100, a database integrity check can be performed on the weld database to determine the comprehensiveness of its disclosure of weld distribution details. Specifically, if the weld database is incomplete, it indicates that the weld database cannot fully disclose the weld distribution details at the workpiece to be modeled. In this case, step S280 will be executed, and then the process will jump back to step S210 to continue until the final constructed weld database remains complete, ensuring that the corresponding weld database can fully disclose the weld distribution details at the workpiece to be modeled.
[0087] Therefore, by executing the above steps S210 to S2110, this application can ensure the orderly nature of the reverse modeling process, the stability of the modeling process, and the overall modeling efficiency, as well as good point cloud management and splicing fault tolerance capabilities, so that the final weld database can reflect the effective weld distribution at the workpiece to be modeled as comprehensively as possible.
[0088] Alternatively, please refer to Figure 6 , Figure 6 This is the fourth flowchart illustrating the reverse modeling method for workpiece point clouds provided in this application embodiment. In this application embodiment, with... Figure 4 Compared to the reverse modeling method of workpiece point cloud shown, Figure 6The reverse modeling method for workpiece point cloud shown may also include steps S2120 to S2130, in conjunction with the above steps S210 to S290, to ensure that the final constructed workpiece collision model can achieve reliable welding torch motion obstacle avoidance trajectory planning effect.
[0089] Step S2120: Construct a workpiece collision model of the workpiece to be modeled based on the 3D workpiece point cloud model.
[0090] In this embodiment, for the three-dimensional workpiece point cloud model determined by the above step S290, collision model generation strategies such as spherical surface reconstruction algorithm, Poisson reconstruction algorithm, greedy projection triangulation algorithm or moving cube algorithm can be used to construct the workpiece collision model of the workpiece to be modeled, which is used to realize the obstacle avoidance trajectory planning function of welding torch movement.
[0091] Step S2130: Perform model integrity detection on the workpiece collision model.
[0092] In this embodiment, after determining the workpiece collision model based on step S2120, a model integrity check can be performed on the workpiece collision model to determine the comprehensiveness of its exposure of the welding torch's movable space. Specifically, if the workpiece collision model is incomplete, it indicates that the model cannot fully expose the details of the welding torch's movable space at the workpiece to be modeled. In this case, step S280 will be executed, followed by a jump to step S210 until the final constructed workpiece collision model remains complete. This ensures that the workpiece collision model can fully expose the details of the welding torch's movable space at the workpiece to be modeled, facilitating reliable welding torch motion obstacle avoidance trajectory planning in the future.
[0093] Therefore, by executing the above steps S210 to S290 and S2120 to S2130, this application can ensure the orderly nature of the reverse modeling process, the stability of the modeling process, the overall modeling efficiency, and the good point cloud management and splicing fault tolerance capabilities, so that the final constructed workpiece collision model can achieve reliable welding torch motion obstacle avoidance trajectory planning effect.
[0094] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.
[0095] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part. If the various functions provided in this application are implemented in the form of software functional modules and sold or used as independent products, they can be stored in a readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium and includes several instructions to cause a computer device (e.g., a laptop computer, server, etc., communicatively connected to a point cloud scanning device) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned readable storage medium includes: USB flash drives, mobile hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.
[0096] The above descriptions are merely various embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A method for reverse modeling of workpiece point clouds, characterized in that, The method includes: Acquire multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the execution of this scanning task; Based on recent index identifier record data, the multiple local point cloud files are dynamically assigned index identifiers and preprocessed to obtain the actual index number of each of the successfully preprocessed valid point cloud files, wherein the actual index numbers of each of the valid point cloud files are continuously distributed among each other. Based on the actual index number of each of the valid point cloud files, the point cloud files are stitched together to obtain the corresponding local point cloud model. Detect whether this scanning task is the first scanning task for the workpiece to be modeled; If the current scanning task is detected as the first scanning task, the local point cloud model is directly used as the target workpiece point cloud model after the workpiece to be modeled is completed in this scanning task. Otherwise, the local point cloud model is merged with the latest workpiece point cloud model to obtain the target workpiece point cloud model, wherein the latest workpiece point cloud model is the workpiece point cloud model of the workpiece to be modeled after the previous scanning task was completed.
2. The method according to claim 1, characterized in that, The recent index identifier record data includes interrelated latest index number and latest index name, interrelated second-newest index number and second-newest index name, and interrelated third-newest index number and third-newest index name. The step of dynamically assigning index identifiers to the multiple local point cloud files based on the recent index identifier record data and performing point cloud preprocessing to obtain the actual index number of each of the successfully preprocessed valid point cloud files includes: The multiple local point cloud files are traversed sequentially, and for each local point cloud file traversed, it is detected whether the local point cloud file belongs to the first point cloud file in the first scan task execution process; If it is detected that the local point cloud file belongs to the first point cloud file, the second new index number and the third new index number are set to the default index values respectively, and the second new index name and the third new index name are set to empty strings respectively; The theoretical index value and the actual file name corresponding to the local point cloud file are assigned one-to-one to the latest index number and the latest index name, and point cloud preprocessing is performed on the local point cloud file, wherein the theoretical index value corresponding to the local point cloud file is obtained by adding one to the default index value; The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number after assignment is taken as the actual index number of the valid point cloud file.
3. The method according to claim 2, characterized in that, The step of dynamically assigning index identifiers to the multiple local point cloud files based on recent index identifier record data and performing point cloud preprocessing to obtain the actual index number of each of the valid point cloud files that have been successfully preprocessed also includes: If it is detected that the local point cloud file does not belong to the first point cloud file, it is checked whether the target point cloud file pointed to by the latest index name has been successfully preprocessed; When the preprocessing of the target point cloud file is detected to be successful, the second new index number and the second new index name are assigned to the next new index number and the next new index name in a one-to-one correspondence, and the latest index number and the latest index name are assigned to the second new index number and the second new index name in a one-to-one correspondence. The theoretical index value and the actual file name corresponding to the local point cloud file are assigned one-to-one to the latest index number and the latest index name, and point cloud preprocessing is performed on the local point cloud file. The theoretical index value corresponding to the local point cloud file is obtained by adding one to the latest index number before the assignment. The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number after assignment is taken as the actual index number of the valid point cloud file.
4. The method according to claim 3, characterized in that, The step of dynamically assigning index identifiers to the multiple local point cloud files based on recent index identifier record data and performing point cloud preprocessing to obtain the actual index number of each of the valid point cloud files that have been successfully preprocessed also includes: When the preprocessing of the target point cloud file is detected to be unsuccessful, the second new index number and the second new index name are assigned to the third new index number and the third new index name in a one-to-one correspondence. Assign the actual filename of the local point cloud file to the latest index name, and perform point cloud preprocessing on the local point cloud file; The preprocessed local point cloud file is taken as a valid point cloud file, and the latest index number is taken as the actual index number of the valid point cloud file.
5. The method according to claim 1, characterized in that, The step of stitching together all valid point cloud files according to their respective actual index numbers to obtain the corresponding local point cloud model includes: Based on the maximum modeled index number before the execution of this scanning task, target valid point cloud files whose corresponding actual index number exceeds the maximum modeled index number are screened among all valid point cloud files, wherein the maximum modeled index number is the default index value before the execution of the first scanning task. All selected target valid point cloud files are stitched together, and the stitched 3D point cloud model is used as the local point cloud model. Then, the maximum modeled index number is updated according to the maximum actual index number of all target valid point cloud files.
6. The method according to any one of claims 1-5, characterized in that, The method further includes: Perform model integrity detection on the point cloud model of the target workpiece; If the point cloud model of the target workpiece is found to be incomplete, a supplementary scanning task for the workpiece to be modeled is issued to the point cloud scanning device, and the above steps of obtaining multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the execution of this scanning task continue to be performed. If the target workpiece point cloud model is found to be complete, point cloud filtering downsampling and maximum connected component extraction are performed on the target workpiece point cloud model to obtain the three-dimensional workpiece point cloud model of the workpiece to be modeled.
7. The method according to claim 6, characterized in that, The method further includes: Based on the three-dimensional workpiece point cloud model, workpiece weld analysis is performed to obtain the corresponding weld database. The weld database is subjected to a database integrity check. If the weld database is found to be incomplete, a supplementary scanning task for the workpiece to be modeled is issued to the point cloud scanning device. The steps described above, which involve obtaining multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the current scanning task, are continued until the weld database is complete.
8. The method according to claim 6, characterized in that, The method includes: Construct a collision model of the workpiece to be modeled based on the three-dimensional workpiece point cloud model; The collision model of the workpiece is subjected to model integrity detection. If the collision model of the workpiece is found to be incomplete, a supplementary scanning task for the workpiece to be modeled is issued to the point cloud scanning device. The steps described above, which involve obtaining multiple local point cloud files continuously scanned by the point cloud scanning device for the workpiece to be modeled during the execution of this scanning task, are continued until the collision model of the workpiece is complete.
9. A computer device, characterized in that, It includes a processor and a memory, the memory storing a computer program that can be executed by the processor, the processor being able to execute the computer program to implement the workpiece point cloud reverse modeling method according to any one of claims 1-8.
10. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a computer device, it implements the reverse modeling method for workpiece point clouds as described in any one of claims 1-8.