Weld joint identification method, storage medium and computer system

Abstract

The invention relates to a rib plate welding seam identification method, a storage medium and a computer system.The rib plate point cloud of a workpiece is firstly obtained, a welding seam area is determined, then the mass center of the welding seam area is searched for, downsampling is conducted, and each optimized rib plate point cloud is obtained; on this basis, the rib plate point clouds can be divided into single-class rib plate point clouds and multi-class rib plate point clouds according to the number of the end points of each rib plate point cloud in the step S3, in the subsequent step S4, only one welding seam straight line can be obtained through fitting in the single-class rib plate point clouds, and multiple welding seam straight lines can be obtained through fitting in the multi-class rib plate point clouds; according to the method, redundant welding line straight lines can be prevented from being mistakenly extracted from single-class rib plate point clouds, and essential welding line straight lines can be prevented from being missed from being extracted from multi-class rib plate point clouds, so that the accuracy and precision of the welding line straight lines are improved; on this basis, the step S5 determines the end points of each weld straight line as the start-stop point and the end point of the weld, and a weld line segment is obtained.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing, and in particular to a weld identification method, a storage medium and a computer system. Background Art

[0002] In the field of modern shipbuilding, small group welding is a production management method and production process adopted to speed up shipbuilding, improve shipbuilding quality, and form large-scale production. It mainly includes the assembly, welding, repair and polishing of the bottom plate and rib plate. The traditional small group welding method mainly relies on manual operation, which is not only labor-intensive, but also the welding quality and efficiency depend on the experience and skill level of the operator. In order to improve welding quality and efficiency, intelligent robot welding technology based on visual information has gradually been introduced into shipbuilding. At present, small group robot welding mostly adopts welding trolley welding, robot teaching welding, and importing the workpiece model and then aligning with the workpiece point cloud collected by the camera to obtain the weld position.

[0003] However, the quality of the welding carriage is unstable, its work efficiency is low and it cannot handle workpieces with complex structures; robot teaching welding relies on the robot engineer to teach the weld trajectory and the robot's motion path, which is not only inefficient but also cannot guarantee the accuracy of the weld position; after importing the workpiece model and aligning it with the workpiece point cloud collected by the camera to obtain the weld position, it depends on the accuracy of the correspondence between the model and the workpiece, and workers are required to operate the computer and manually select the weld, which increases the complexity of the operation.

[0004] Therefore, how to visually identify this type of weld, automatically guide the welding process, and improve work efficiency and weld quality is a technical problem that needs to be solved urgently in this field. Summary of the invention

[0005] In order to solve at least one of the above technical problems, the present invention provides a weld identification method for a rib plate, comprising:

[0006] Obtain the rib point cloud of the workpiece;

[0007] For each rib plate point cloud, find the centroid for downsampling to obtain the optimized rib plate point cloud;

[0008] Determine the endpoints of each rib plate point cloud, and classify the rib plate point cloud into a single-type rib plate point cloud and a multi-type rib plate point cloud according to the number of endpoints;

[0009] For each single-type rib plate point cloud, a weld straight line is fitted; for each multi-type rib plate point cloud, the longest straight line is fitted and extracted in sequence to obtain multiple weld straight lines;

[0010] Determine the endpoints of each weld line as the start and end points of the weld to obtain the weld line segment.

[0011] Furthermore, the step of obtaining the rib point cloud of the workpiece includes:

[0012] Obtaining a workpiece point cloud, extracting the largest plane in the workpiece point cloud to separate the base plate point cloud, and determining the base plate plane;

[0013] The remaining point clouds in the workpiece point cloud except the base plate point cloud are projected onto the base plate plane, and each rib plate point cloud is clustered and segmented.

[0014] Further, the steps of determining the endpoints of each rib plate point cloud and classifying the rib plate point cloud into a single-type rib plate point cloud and a multi-type rib plate point cloud according to the number of endpoints include:

[0015] Count the number of endpoints of each rib point cloud;

[0016] If the number of endpoints is equal to 2, it is determined to be a single-type rib plate point cloud;

[0017] If the number of endpoints is greater than 2, it is determined to be a multi-type rib plate point cloud.

[0018] Furthermore, for each single type of rib point cloud, a weld line is fitted, including:

[0019] For any two points in each single-type stiffener point cloud, a straight line is fitted to determine whether the remaining number of points is less than the set threshold. If so, a weld straight line is obtained. Otherwise, refitting is performed until the remaining number of points is less than the set threshold to obtain a weld straight line.

[0020] Furthermore, for each multi-type rib point cloud, the longest straight line is sequentially fitted and extracted to obtain multiple weld straight lines, including:

[0021] For any two points of each multi-category stiffener point cloud, a collimator line is fitted, and the longest line in the collimator line is extracted to obtain a weld line. The inner points of the longest line are removed to obtain the remaining point cloud. The longest line is repeatedly fitted to the remaining point cloud to obtain another weld line until the iteration termination condition is reached, and multiple weld lines are fitted.

[0022] Furthermore, after obtaining a plurality of weld straight lines, the method further includes:

[0023] Calculate the angle between every two weld lines;

[0024] Calculate the distance between the endpoints of every two weld lines;

[0025] According to the angle and distance, the redundancy of the weld straight line is eliminated to obtain the updated weld straight line.

[0026] Further, the step of determining the endpoint of each weld line includes:

[0027] Calculate the distance from the point in the rib point cloud to which each weld line belongs to the weld line;

[0028] Determine whether the distance is greater than the set threshold, if so, it is an interior point;

[0029] After projecting the inner points of each weld line onto the line, clustering removes outliers to determine the endpoints of each weld line.

[0030] Furthermore, it also includes:

[0031] Traverse the straight lines where the weld line segments are located and calculate the angle between every two straight lines;

[0032] If the angle is less than the set threshold, no processing is performed;

[0033] If the angle is not less than the set threshold, the intersection of the two corresponding straight lines and the distance between the intersection and the endpoints of the two straight lines are calculated, and the weld line segment is optimized accordingly.

[0034] On the other hand, the present invention also provides a computer-readable storage medium having executable program code stored thereon, which, when executed by a processor, is used to execute any of the above-mentioned weld identification methods.

[0035] On the other hand, the present invention also provides a computer system, including a memory and a processor; the memory stores program codes executable by the processor, and when executed by the processor, is used to implement any of the above-mentioned weld identification methods.

[0036] The weld identification method, storage medium and computer system of the rib plate of the present invention first obtain the rib plate point cloud of the workpiece, determine the weld area, and then find the centroid and perform downsampling to obtain each optimized rib plate point cloud. On this basis, step S3 can divide the rib plate point cloud into a single-type rib plate point cloud and multiple-type rib plate point clouds according to the number of endpoints of each rib plate point cloud. In the subsequent step S4, only one weld straight line can be fitted in the single-type rib plate point cloud, while multiple weld straight lines can be fitted in the multiple-type rib plate point cloud. This can avoid erroneously extracting redundant weld straight lines in the single-type rib plate point cloud, and avoid missing the extraction of substantial weld straight lines in the multiple-type rib plate point cloud, thereby improving the accuracy and precision of the weld straight lines. On this basis, step S5 determines that the endpoints of each weld straight line are the starting and ending points of the weld, thereby obtaining a weld line segment. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A flow chart of an embodiment of a weld identification method of the present invention;

[0038] Figure 2 A schematic diagram of an embodiment of a workpiece point cloud;

[0039] Figure 3A schematic diagram of an embodiment of a rib plate point cloud;

[0040] Figure 4 A schematic diagram of an embodiment of a multi-type rib plate point cloud;

[0041] Figure 5 A schematic diagram of an embodiment of a weld end point;

[0042] Figure 6 is a schematic diagram of an embodiment of an optimized weld end point;

[0043] Figure 7 The flowchart of another embodiment of the weld identification method of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] It should be noted that if the embodiments of the present invention involve directional indications, such as up, down, left, right, front, back, etc., then the directional indication is only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly. In addition, if the embodiments of the present invention involve descriptions of "first, second", "S1, S2", "step one, step two", etc., then such descriptions are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of indicated technical features or indicating the execution order of the method, etc. Those skilled in the art can understand that anything that does not violate the gist of the invention under the technical concept of the invention should be included in the protection scope of the present invention.

[0046] like Figure 1 As shown, a rib plate weld identification method is not limited to being used in the welding application of the above-mentioned ship, and includes:

[0047] S1: Obtain the rib plate point cloud of the workpiece; specifically, the workpiece point cloud may be first obtained through an acquisition device, and then the position of each rib plate may be separated and extracted through image recognition to obtain the rib plate point cloud of the workpiece.

[0048] Preferably, step S1 comprises:

[0049] S11: acquiring a workpiece point cloud, extracting a maximum plane in the workpiece point cloud, separating a bottom plate point cloud, and determining a bottom plate plane;

[0050] Specifically, a system can be set up for the complex structural parts composed of bottom plates and reinforcement ribs in the ship group, which consists of a welding robot, a ground rail cantilever, a line scan camera, a weld tracking sensor, a welding machine, a PLC and an industrial computer. The robot is hung upside down on the cantilever, and the line scan camera is installed on the cantilever and its relative position with the robot base coordinate system remains unchanged. The ground rail moves at a uniform speed to drive the line scan camera to scan the workpiece and obtain the workpiece point cloud, such as Figure 2 As shown; the point cloud is then converted to the robot base coordinate system through visual calibration parameters; the largest plane in the fitted workpiece point cloud is extracted through the RANSAC algorithm, the workpiece bottom plate is separated from it as the bottom plate point cloud, and the remaining point clouds are the upper surface of the workpiece and interference points.

[0051] S12: Project the remaining point clouds in the workpiece point cloud except the bottom plate point cloud onto the bottom plate plane, and cluster and segment each rib plate point cloud;

[0052] Specifically, the remaining point clouds can be projected onto the bottom plate plane, and statistical filtering can be used for noise reduction, uniform downsampling can be used to reduce the point cloud density, and cluster segmentation can be used to segment the rib plate point clouds one by one to obtain each rib plate point cloud.

[0053] S2: For each rib plate point cloud, find the centroid and perform downsampling to obtain the optimized rib plate point cloud;

[0054] Since each rib plate has a certain length, width, thickness, etc., and is an object with a certain volume and shape, it is not a straight line. On this basis, if step S3 is operated, many types of rib plate point clouds may be misjudged. One of the core points of the present invention is to further downsample each segmented rib plate point cloud by finding the centroid to obtain its similar center line, representing each rib plate point cloud, such as Figure 3 shown.

[0055] S3: Determine the endpoints of each rib plate point cloud, and classify the rib plate point cloud into single-type rib plate point cloud and multi-type rib plate point cloud according to the number of endpoints;

[0056] Specifically, the rib plate point cloud is traversed, and points in the radius neighborhood are searched based on the point, and the distance between each point in the neighborhood and the current point is calculated. The point with the farthest distance is selected to construct the basis vector, and the vector formed by other points and the current point is traversed to construct the cosine of the angle with the basis vector. If the cosine value is less than 0, the point is not an endpoint. If the cosine value is greater than 0, the point is an endpoint. In this way, the endpoint of each rib plate point cloud is determined, and the openness is determined. If the endpoint is 0, it indicates that it is a closed curve, so that the rib plate point cloud can be classified according to its specific situation. Preferably, it includes:

[0057] S31: Count the number of endpoints of each rib point cloud;

[0058] S32: If the number of endpoints is equal to 2, it is determined to be a single-type rib plate point cloud;

[0059] S33: If the number of endpoints is greater than 2, it is determined to be a multi-type rib plate point cloud.

[0060] Specifically, if there are two endpoints, it can be determined that a straight line can be fitted in each rib plate point cloud, and it is determined to be a single-type rib plate point cloud; if there are more than two endpoints, it can be determined that multiple straight lines can be fitted in each rib plate point cloud, and it is determined to be a multi-type rib plate point cloud, such as Figure 4 As shown, relative to Figure 2 After discarding the single-type ribbed plate point clouds, only the multi-type ribbed plate point clouds with more than 2 endpoints remain.

[0061] S4: For each single-type rib plate point cloud, a weld straight line is fitted; for each multi-type rib plate point cloud, the longest straight line is sequentially fitted to obtain multiple weld straight lines;

[0062] Specifically, for each single-type rib plate point cloud and multi-type rib plate point cloud, any fitting algorithm, such as the RANSAC algorithm, can be used to fit and extract the corresponding straight line, which is the preset weld straight line.

[0063] Preferably,

[0064] A: For each single type of rib point cloud, a weld line is fitted, including:

[0065] For any two points in each single-type rib plate point cloud, a straight line is obtained by fitting, and it is determined whether the remaining number of points is less than the set threshold. If so, a weld straight line is obtained. If not, refitting is performed until the remaining number of points is less than the set threshold, and a weld straight line is obtained.

[0066] B: For each multi-type rib point cloud, the longest straight line is extracted in sequence to obtain multiple weld lines, including:

[0067] For any two points of each multi-category stiffener point cloud, a direct line is fitted, and the longest straight line in the direct line is extracted to obtain a weld straight line. The inner points of the longest straight line are removed to obtain the remaining point cloud. The remaining point cloud is repeatedly fitted to extract the longest straight line to obtain another weld straight line until the iteration termination condition is met, and multiple weld straight lines are fitted. Specifically, for each multi-category stiffener point cloud, downsampling is first performed, and then the longest fitted straight line is extracted through the RANSAC algorithm. The inner points of the straight line are removed to obtain the remaining point cloud. The RANSAC algorithm is repeatedly used to extract straight lines for the remaining point cloud until the number of remaining points is less than the set threshold and fitting is stopped, and the straight lines obtained are several weld straight lines.

[0068] More preferably, since there may be multiple single-type rib plate point clouds, a straight weld is obtained for each single-type rib plate point cloud. Due to the segmentation error of the rib plate point cloud, there may be misjudgment; multiple types of rib plate point clouds may also have redundancy, so it is also necessary to eliminate the redundancy of each weld line according to the specific parameters of the weld line to eliminate the weld lines that are essentially one but repeated due to segmentation, recognition and other errors. Therefore, step S4 also includes:

[0069] S41: Calculate the angle between every two weld lines;

[0070] S42: Calculate the distance between the endpoints of every two weld lines;

[0071] S43: Eliminate redundancy of the weld straight line according to the angle and the distance to obtain an updated weld straight line.

[0072] Specifically, an angle threshold and a distance threshold may be optionally set. According to the accuracy requirement, when the angle is less than the angle threshold and the distance is less than the distance threshold, the weld straight line is redundantly eliminated, such as merging multiple welds that meet the requirements into one weld, retaining only the weld with the longest length, etc., to obtain an updated weld straight line, such as Figure 5 shown.

[0073] S5: Determine the endpoints of each weld line as the start and end points of the weld, and obtain the weld line segment.

[0074] Specifically, the upper end point and the lower end point of each weld line are determined as the starting and ending points of the weld to obtain each weld line segment. Preferably, the step of determining the end points of each weld line includes:

[0075] S51: Calculate the distance from the point in the rib point cloud to which each weld line belongs to the weld line;

[0076] S52: Determine whether the distance is greater than a set threshold, if so, it is an interior point;

[0077] S53: After projecting the inner points of each weld line onto the line, clustering is performed to remove outliers to determine the endpoints of each weld line.

[0078] Specifically, for the weld line obtained by fitting the point cloud of a single type of rib plate, the distance from the point in the point cloud of the single type of rib plate to the weld line is calculated. When it is less than the threshold, it is determined to be the inner point of the line. After the inner point is projected onto the line, the outliers are removed by clustering. Then, the endpoints are found in the same way as step S3 to determine the starting point and the ending point of the weld. For the weld line obtained by fitting the point cloud of multiple types of rib plates, the distance from the point of the weld line obtained by fitting the point cloud of a single type of rib plate to each weld line is calculated. When it is less than the threshold, it is determined to be the inner point of the line. After the inner point is projected onto the line, the outliers are removed by clustering. The endpoints of the line are found and temporarily determined as the starting point and the ending point of the weld.

[0079] In this embodiment, a weld identification method for a rib plate of the present invention is provided, which first obtains a rib plate point cloud of a workpiece, determines the weld area, and then finds the centroid for downsampling to obtain each optimized rib plate point cloud, on this basis, step S3 can divide the rib plate point cloud into a single-type rib plate point cloud and multiple-type rib plate point clouds according to the number of endpoints of each rib plate point cloud, in the subsequent step S4, only one weld straight line can be fitted in the single-type rib plate point cloud, while multiple weld straight lines can be fitted in the multiple-type rib plate point cloud; it can avoid erroneously extracting redundant weld straight lines in the single-type rib plate point cloud, and avoid missing the extraction of substantial weld straight lines in the multiple-type rib plate point cloud, thereby improving the accuracy and precision of the weld straight lines; on this basis, step S5 determines that the endpoints of each weld straight line are the starting and ending points of the weld, and obtains a weld line segment.

[0080] More preferably, Figure 5 As shown, it can be seen that the extracted weld line segments are cross-shaped, resulting in inaccurate positioning of the weld endpoints. Therefore, the method preferably includes: S6: Optimization step

[0081] S61: traverse the straight lines where the weld line segments are located, and calculate the angle between every two straight lines;

[0082] S62: If the angle is less than the set threshold, no processing is performed; specifically, if the angle between the two straight lines is less than the set threshold, such as less than 5°, it is considered that the angle is caused by recognition error, and it is assumed that they do not intersect, and no processing is performed;

[0083] S63: If the angle is not less than the set threshold, the intersection of the two corresponding straight lines and the distance between the intersection and the endpoints of the two straight lines are calculated, and the weld line segment is optimized accordingly.

[0084] Preferably, step S63 includes:

[0085] If equation (1) is satisfied, no processing is performed. Specifically, if equation (1) is satisfied, it means that the intersection point is outside the two line segments, and it is determined that the two actually have no intersection point, but are just different weld line segments located at two positions, and no processing is performed.

[0086]

[0087] Wherein, L1 and L2 are the lengths of the two weld line segments respectively; a1, a2, b1, and b2 are the distances between the intersection and the two end points of the weld line segments L1 and L2 respectively;

[0088] If equation (2) is satisfied, if the distance between the intersection point and the endpoint of the line segment is less than the intersection threshold, the endpoint is determined to coincide with the intersection point, and the intersection point is the new endpoint. If it is greater than the threshold, the intersection point is used as the reference point to divide the line segment into 4 weld line segments. Specifically, if equation (2) is satisfied, it means that the intersection point is in the middle of the two line segments, and the above optimization needs to be performed on it.

[0089]

[0090] If formula (2) is not satisfied, then determine whether formula (3) is satisfied. If so, it means that the intersection is located in the middle of a line segment. Similarly, the distance between the intersection and the endpoint of the line segment is used to determine whether the intersection and the endpoint coincide. If they coincide, the endpoint is replaced by the intersection. Otherwise, the intersection is used as the reference point and divided into 4 weld line segments.

[0091] (a1<L1&&a2<L1)||(b1<L2&&b2<L2)(3)

[0092] The weld endpoints extracted after optimization are as follows: Figure 6 As shown. Furthermore, the extracted weld endpoints are transmitted to the robot after subsequent path planning, and the robot drives the weld tracking sensor to accurately locate the weld before welding. The specific algorithm flow is as follows Figure 7 shown.

[0093] In summary, this application targets small groups of complex structural parts. It obtains workpiece point clouds by scanning the workpieces with a line scan camera, and separates single-type rib plate point clouds (linear rib plate point clouds) and multi-type rib plate point clouds (special rib plate point clouds) based on clustering and endpoint finding algorithms. The linear rib plate point clouds can directly extract weld lines through the RANSAC algorithm, and the special rib plate point clouds can successively realize weld positioning of complex structural parts through steps such as straight line fitting, selection and merging of weld line segments, and optimization of weld line segment endpoints. There is no need to import workpiece models and manually select welds, which reduces the difficulty of workers' operations and provides a prerequisite for path planning for subsequent robot scanning of welds for precise weld positioning.

[0094] The preferred embodiment also includes at least the following beneficial effects:

[0095] 1. For the special rib plate point cloud, first extract the longest straight line, remove its inner points, and then continue to fit the straight line in the remaining point cloud to avoid the situation where two rib plate point clouds that are close and connected are fitted into a straight line.

[0096] 2. The rib plate has an irregular shape and a certain thickness, so multiple line segments may be extracted from one rib plate. The line segments on the same rib plate are merged by calculating the angle between the line segments and the distance between the endpoints. The line segment with the longest length is selected as the final weld line segment to ensure that the line segment can represent the entire rib plate weld.

[0097] 3. For the intersection of weld line segments, the length of the line segment, the angle of the line segment, and the distance between the intersection point and the line segment endpoint are combined to determine whether there is an intersection and the position of the intersection if there is an intersection. Finally, based on the position of the intersection and the distance between the intersection and the endpoint, it is determined whether the intersection coincides with the endpoint to complete the optimization of the weld line segment endpoint.

[0098] On the other hand, the present invention also provides a computer storage medium having executable program code stored thereon, which, when executed by a processor, is used to implement any of the above-mentioned weld identification methods.

[0099] On the other hand, the present invention also provides a terminal device, including a memory and a processor; the memory stores program codes that can be executed by the processor, and when executed by the processor, is used to implement any of the above-mentioned weld identification methods.

[0100] Exemplarily, the program code may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the program code in the terminal device.

[0101] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the terminal device may also include an input / output device, a network access device, a bus, etc.

[0102] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0103] The memory may be an internal storage unit of the terminal device, such as a hard disk or a memory. The memory may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device. Further, the memory may also include both an internal storage unit of the terminal device and an external storage device. The memory is used to store the program code and other programs and data required by the terminal device. The memory may also be used to temporarily store data that has been output or is to be output.

[0104] The above-mentioned computer storage medium and terminal device are created based on the above-mentioned weld identification method, and their technical functions and beneficial effects are no longer repeated here. The various technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the various technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0105] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A method for identifying a weld of a rib plate, characterized in that: include: Obtain the rib point cloud of the workpiece; For each rib plate point cloud, find the centroid for downsampling to obtain the optimized rib plate point cloud; Determine the endpoints of each rib plate point cloud, and classify the rib plate point cloud into a single-type rib plate point cloud and a multi-type rib plate point cloud according to the number of endpoints; For each single-type rib plate point cloud, a weld straight line is fitted; for each multi-type rib plate point cloud, the longest straight line is fitted and extracted in sequence to obtain multiple weld straight lines; Determine the endpoints of each weld line as the start and end points of the weld to obtain the weld line segment.

2. The weld identification method according to claim 1, characterized in that: The steps for obtaining the rib point cloud of the workpiece include: Obtaining a workpiece point cloud, extracting the largest plane in the workpiece point cloud to separate the base plate point cloud, and determining the base plate plane; The remaining point clouds in the workpiece point cloud except the base plate point cloud are projected onto the base plate plane, and each rib plate point cloud is clustered and segmented.

3. The weld identification method according to claim 1, characterized in that: The steps of determining the endpoints of each rib plate point cloud and classifying the rib plate point cloud into a single-type rib plate point cloud and a multi-type rib plate point cloud according to the number of endpoints include: Count the number of endpoints of each rib point cloud; If the number of endpoints is equal to 2, it is determined to be a single-type rib plate point cloud; If the number of endpoints is greater than 2, it is determined to be a multi-type rib plate point cloud.

4. The weld identification method according to claim 1, characterized in that: For each single type of rib point cloud, a weld line is fitted, including: For any two points in each single-type stiffener point cloud, a straight line is fitted to determine whether the remaining number of points is less than the set threshold. If so, a weld straight line is obtained. Otherwise, refitting is performed until the remaining number of points is less than the set threshold to obtain a weld straight line.

5. The weld identification method according to claim 1, characterized in that: For each multi-type rib point cloud, the longest straight line is fitted and extracted in sequence to obtain multiple weld straight lines, including: For any two points of each multi-category stiffener point cloud, a collimator line is fitted, and the longest line in the collimator line is extracted to obtain a weld line. The inner points of the longest line are removed to obtain the remaining point cloud. The longest line is repeatedly fitted to the remaining point cloud to obtain another weld line until the iteration termination condition is reached, and multiple weld lines are fitted.

6. The weld identification method according to claim 1, characterized in that: After obtaining multiple weld lines, it also includes: Calculate the angle between every two weld lines; Calculate the distance between the endpoints of every two weld lines; According to the angle and distance, the redundancy of the weld straight line is eliminated to obtain the updated weld straight line.

7. The weld identification method according to claim 1, characterized in that: The steps to determine the endpoints of each weld line include: Calculate the distance from the point in the rib point cloud to which each weld line belongs to the weld line; Determine whether the distance is greater than the set threshold, if so, it is an interior point; After projecting the inner points of each weld line onto the line, clustering removes outliers to determine the endpoints of each weld line.

8. The weld identification method according to any one of claims 1 to 7, characterized in that: Also includes: Traverse the straight lines where the weld line segments are located and calculate the angle between every two straight lines; If the angle is less than the set threshold, no processing is performed; If the angle is not less than the set threshold, the intersection of the two corresponding straight lines and the distance between the intersection and the endpoints of the two straight lines are calculated, and the weld line segment is optimized accordingly.

9. A computer-readable storage medium, characterized in that: An executable program code is stored thereon, which, when executed by a processor, is used to implement the weld identification method described in any one of claims 1-8.

10. A computer system, characterized in that: It comprises a memory and a processor; the memory stores program codes executable by the processor, and when executed by the processor, is used to implement the weld identification method described in any one of claims 1-8.

Images