Method, device and equipment for identifying welding seam in steel bar and steel plate lap joint workpiece and medium

Abstract

The invention relates to a method, a device, equipment and a medium for identifying a welding seam in a steel bar and steel plate lap joint workpiece, and the method comprises the steps: obtaining a welding image of the steel bar and steel plate lap joint workpiece, and extracting a steel bar main body point cloud and a steel plate plane main body point cloud corresponding to the welding image; constructing a steel plate main plane and a steel plate auxiliary plane according to the steel bar main body point cloud and the steel plate plane main body point cloud, and obtaining an intersection line of the steel plate auxiliary plane and the steel plate main plane; the steel plate auxiliary plane is a plane perpendicular to the steel plate; screening point clouds of which the distance from the intersection line is smaller than the radius of the steel bar from the steel bar main body point clouds to obtain target point clouds; calculating a first projection segment formed after the target point cloud is projected on the intersecting line and a second projection segment formed after the steel plate plane main body point cloud is projected on the intersecting line; and based on the first projection segment and the second projection segment, determining a welding seam between the reinforcing steel bar and the steel plate in the reinforcing steel bar and steel plate lap joint workpiece. According to the whole scheme, efficient and accurate weld joint recognition can be achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image recognition, in particular to a steel bar and steel plate lap joint workpiece weld identification method and device, computer equipment, storage medium and computer program product. BACKGROUND

[0002] Steel structure has been widely used in many industries such as bridges and marine equipment due to its high strength, light weight, good seismic performance, high degree of industrial assembly, excellent comprehensive economic benefits, and green environmental protection. In the manufacturing process of steel structure, welding as a core process and key technology plays a decisive role in the overall efficiency and product quality of steel structure production.

[0003] In actual application, steel structure often needs to be connected by welding, and in many scenarios of steel structure welding, weld identification is a key prerequisite for efficient and accurate welding. For example, steel bar and steel plate lap joint is widely used in steel structure and building fields. However, for such welding scenarios, the industry mainly relies on offline programming or manual teaching-based weld identification, which cannot achieve efficient and accurate weld identification.

[0004] Therefore, there are problems of low efficiency and inaccuracy in traditional steel bar and steel plate lap joint workpiece weld identification. SUMMARY

[0005] Therefore, it is necessary to provide a steel bar and steel plate lap joint workpiece weld identification method, device, computer equipment, computer readable storage medium and computer program product which are efficient and accurate.

[0006] In a first aspect, the present application provides a steel bar and steel plate lap joint workpiece weld identification method. The method comprises:

[0007] Obtaining a welding image of a steel bar and steel plate lap joint workpiece, extracting a steel bar main point cloud and a steel plate plane main point cloud corresponding to the welding image;

[0008] According to the steel bar main point cloud and the steel plate plane main point cloud, a main plane of the steel plate and an auxiliary plane of the steel plate are constructed, and an intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate is obtained; the auxiliary plane of the steel plate is a plane perpendicular to the steel plate;

[0009] From the steel bar main point cloud, a point cloud with a distance less than the radius of the steel bar from the intersection line is selected to obtain a target point cloud;

[0010] The first projection segment formed after the target point cloud is projected on the intersection line, and the second projection segment formed after the steel plate plane main point cloud is projected on the intersection line are calculated;

[0011] determine a weld between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment.

[0012] In one of the embodiments, the determining the weld between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment comprises:

[0013] calculating an intersection between the first projection segment and the second projection segment, and intercepting a straight line segment on the intersection line based on the intersection;

[0014] solving a vector V perpendicular to the intersection line on the main plane of the steel plate;

[0015] determining the weld between the steel bar and the steel plate according to the straight line segment and the vector V.

[0016] In one of the embodiments, the determining the weld between the steel bar and the steel plate according to the straight line segment and the vector V comprises:

[0017] obtaining a steel bar size, a steel plate thickness, and a welding process requirement of the steel bar and steel plate lap joint workpiece, and determining a weld distance d;

[0018] moving the weld distance d in the positive direction and the negative direction of the vector V on the straight line segment respectively, and determining two welds between the steel bar and the steel plate.

[0019] In one of the embodiments, the obtaining a welding image of the steel bar and steel plate lap joint workpiece, and extracting a steel bar main body point cloud and a steel plate plane main body point cloud corresponding to the welding image comprises:

[0020] obtaining a welding image of the steel bar and steel plate lap joint workpiece;

[0021] extracting a welding workpiece point cloud corresponding to the steel bar and steel plate lap joint workpiece according to the welding image;

[0022] extracting a steel bar point cloud and a steel plate plane point cloud in the welding workpiece point cloud;

[0023] performing clustering on the steel bar point cloud and the steel plate plane point cloud respectively to obtain a steel bar main body point cloud and a steel plate plane main body point cloud.

[0024] In one of the embodiments, the performing clustering on the steel bar point cloud and the steel plate plane point cloud respectively to obtain a steel bar main body point cloud and a steel plate plane main body point cloud comprises:

[0025] performing clustering on the steel bar point cloud through a dbscan algorithm, and taking a cluster with the largest number of point clouds in the clustered point cloud as the steel bar main body point cloud;

[0026] The steel plate plane point cloud is clustered by a dbscan algorithm, a cluster point cloud with the most point clouds and adjacent to the reinforcing bar is obtained from the clustered multi-cluster point cloud, and a steel plate plane main body point cloud is obtained.

[0027] In one of the embodiments, constructing a main plane of the steel plate and an auxiliary plane of the steel plate according to the reinforcing bar main body point cloud and the steel plate plane main body point cloud comprises:

[0028] determining a cylinder C corresponding to the reinforcing bar according to the reinforcing bar main body point cloud, and constructing a main plane of the steel plate according to the steel plate plane main body point cloud;

[0029] determining a steel plate plane normal vector corresponding to the main plane of the steel plate;

[0030] constructing an auxiliary plane of the steel plate based on the cylinder C and the steel plate plane normal vector.

[0031] In one of the embodiments, obtaining the intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate comprises:

[0032] determining an auxiliary plane normal vector of the auxiliary plane of the steel plate;

[0033] cross-multiplying the steel plate plane normal vector and the auxiliary plane normal vector of the steel plate to obtain a target direction vector;

[0034] determining the intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate based on the target direction vector.

[0035] In a second aspect, the application further provides a welding seam identification device in a reinforcing bar and steel plate lap joint workpiece. The device comprises:

[0036] a point cloud extraction module configured to obtain a welding image of the reinforcing bar and steel plate lap joint workpiece, and extract a reinforcing bar main body point cloud and a steel plate plane main body point cloud corresponding to the welding image;

[0037] an intersection line determination module configured to construct a main plane of the steel plate and an auxiliary plane of the steel plate according to the reinforcing bar main body point cloud and the steel plate plane main body point cloud, and obtain an intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate; the auxiliary plane of the steel plate is a plane perpendicular to the steel plate;

[0038] a target point cloud extraction module configured to filter point clouds with a distance less than a reinforcing bar radius from the intersection line from the reinforcing bar main body point cloud, and obtain a target point cloud;

[0039] a segmented calculation module configured to calculate a first projection segment formed by projecting the target point cloud on the intersection line, and a second projection segment formed by projecting the steel plate plane main body point cloud on the intersection line;

[0040] A weld seam confirmation module is configured to determine a weld seam between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment.

[0041] In a third aspect, the present application further provides a computer device. The computer device comprises a memory and a processor. The memory stores a computer program. The processor implements the following steps when executing the computer program:

[0042] obtaining a welding image of a steel bar and steel plate lap joint workpiece, and extracting a steel bar main point cloud and a steel plate plane main point cloud corresponding to the welding image;

[0043] constructing a main plane of the steel plate and an auxiliary plane of the steel plate according to the steel bar main point cloud and the steel plate plane main point cloud, and obtaining an intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate; the auxiliary plane of the steel plate is a plane perpendicular to the steel plate;

[0044] screening point clouds with a distance less than a steel bar radius from the intersection line from the steel bar main point cloud to obtain a target point cloud;

[0045] calculating a first projection segment formed after the target point cloud is projected on the intersection line, and a second projection segment formed after the steel plate plane main point cloud is projected on the intersection line;

[0046] determining a weld seam between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment.

[0047] In a fourth aspect, the present application further provides a computer readable storage medium. The computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the following steps:

[0048] obtaining a welding image of a steel bar and steel plate lap joint workpiece, and extracting a steel bar main point cloud and a steel plate plane main point cloud corresponding to the welding image;

[0049] constructing a main plane of the steel plate and an auxiliary plane of the steel plate according to the steel bar main point cloud and the steel plate plane main point cloud, and obtaining an intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate; the auxiliary plane of the steel plate is a plane perpendicular to the steel plate;

[0050] screening point clouds with a distance less than a steel bar radius from the intersection line from the steel bar main point cloud to obtain a target point cloud;

[0051] calculating a first projection segment formed after the target point cloud is projected on the intersection line, and a second projection segment formed after the steel plate plane main point cloud is projected on the intersection line;

[0052] Determine the weld between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment.

[0053] In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the following steps:

[0054] Obtain a welding image of a steel bar and steel plate lap joint workpiece, and extract a steel bar main body point cloud and a steel plate plane main body point cloud corresponding to the welding image;

[0055] Construct a steel plate main plane and a steel plate auxiliary plane according to the steel bar main body point cloud and the steel plate plane main body point cloud, and obtain an intersection line of the steel plate auxiliary plane and the steel plate main plane; the steel plate auxiliary plane is a plane perpendicular to the steel plate;

[0056] Screen point clouds with a distance to the intersection line less than a steel bar radius from the steel bar main body point cloud to obtain a target point cloud;

[0057] Calculate a first projection segment formed after the target point cloud is projected on the intersection line, and a second projection segment formed after the steel plate plane main body point cloud is projected on the intersection line;

[0058] Determine the weld between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment.

[0059] The steel bar and steel plate lap joint workpiece weld identification method, device, computer equipment, storage medium and computer program product described above obtain a welding image of a steel bar and steel plate lap joint workpiece, and extract a steel bar main body point cloud and a steel plate plane main body point cloud corresponding to the welding image; construct a steel plate main plane and a steel plate auxiliary plane according to the steel bar main body point cloud and the steel plate plane main body point cloud, and obtain an intersection line of the steel plate auxiliary plane and the steel plate main plane; the steel plate auxiliary plane is a plane perpendicular to the steel plate; screen point clouds with a distance to the intersection line less than a steel bar radius from the steel bar main body point cloud to obtain a target point cloud; calculate a first projection segment formed after the target point cloud is projected on the intersection line, and a second projection segment formed after the steel plate plane main body point cloud is projected on the intersection line; and determine the weld between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment. In the entire process, the weld is identified based on three-dimensional point clouds, and errors are reduced through main body point cloud extraction, intersection line extraction based on the steel plate auxiliary plane, and intersection line projection, so that the weld between the steel bar and the steel plate is finally determined automatically, and efficient and accurate weld identification in the steel bar and steel plate lap joint workpiece can be achieved. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 It is an application environment diagram of the steel bar and steel plate lap joint workpiece weld identification method in one embodiment.

[0061] Figure 2 Flowchart of the method for identifying welds in a steel bar and steel plate lap joint workpiece in one embodiment;

[0062] Figure 3 Flowchart of the method for identifying welds in a steel bar and steel plate lap joint workpiece in another embodiment;

[0063] Figure 4 Schematic diagram of a weld identification structure in a specific application example;

[0064] Figure 5 Schematic diagram of a sub-process of S100 in one embodiment;

[0065] Figure 6 Block diagram of the structure of a device for identifying welds in a steel bar and steel plate lap joint workpiece in one embodiment;

[0066] Figure 7 Internal structure diagram of a computer device in one embodiment. DETAILED DESCRIPTION

[0067] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0068] The method for identifying welds in a steel bar and steel plate lap joint workpiece provided by the embodiments of the present application can be applied to, for example, Figure 1The application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data required by the server 104 to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers. The terminal 102 obtains the welding image of the steel bar and steel plate lap joint workpiece, generates a weld seam identification request, and sends the weld seam identification request to the server 104. The server 104 responds to the request, obtains the welding image, extracts the steel bar main point cloud and the steel plate plane main point cloud corresponding to the welding image; according to the steel bar main point cloud and the steel plate plane main point cloud, the main plane of the steel plate and the auxiliary plane of the steel plate are constructed, and the intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate is obtained; The auxiliary plane of the steel plate is a plane perpendicular to the steel plate; From the steel bar main point cloud, the point cloud with a distance less than the radius of the steel bar from the intersection line is selected to obtain the target point cloud; Calculate the first projection segment formed after the target point cloud is projected on the intersection line, and the second projection segment formed after the steel plate plane main point cloud is projected on the intersection line; Based on the first projection segment and the second projection segment, determine the weld seam between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece. Further, the server 104 can mark the identified weld seam with a highlighted color in the welding image, and the marked image to the terminal 102. Among them, the terminal 102 can be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices, Internet of Things devices can be smart speakers, smart televisions, smart air conditioners, smart car devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers.

[0069] In one embodiment, as shown in Figure 2 , a steel bar and steel plate lap joint workpiece weld seam identification method is provided. The method is applied to the server 104 in Figure 1 for example, including the following steps:

[0070] S100: Obtain the welding image of the steel bar and steel plate lap joint workpiece, and extract the steel bar main point cloud and the steel plate plane main point cloud corresponding to the welding image.

[0071] The welding image can be obtained by receiving a three-dimensional camera of a specified model to capture the steel bar and steel plate lap joint workpiece. Further, in actual application, the three-dimensional camera needs to be directed to the to-be-welded position to be positioned, and the angle deviation cannot exceed 15° to ensure the accuracy of the shooting angle, so as to obtain more complete and accurate image information. The camera distance to the to-be-positioned workpiece is 500 mm, and the positive and negative deviation does not exceed 100 mm. Within this distance range, the camera can clearly capture the details of the workpiece. Moreover, the to-be-welded workpiece must be within the camera field of view and as close to the center of the camera field of view as possible during shooting, so as to ensure that the obtained image quality is high and reduce the problems of missing or distortion of image information caused by deviation of the workpiece position.

[0072] After obtaining the welding image, the welding workpiece point cloud can be obtained. Further, the point cloud can be preprocessed to improve the data quality. Specifically, first, according to the specified camera-to-to-be-welded workpiece shooting distance, a band-pass filter is used to obtain the point cloud with a Z-direction value in the range of 400 mm-600 mm to remove the influence of the cluttered background; then the filtered point cloud is subjected to voxel downsampling filtering to improve the consistency of the point cloud density; and then radius density filtering is used to remove the discrete points in the point cloud and reduce the point cloud noise. After the above preprocessing steps, the overall point cloud of the welding workpiece with high quality can be obtained.

[0073] Specifically, after obtaining the three-dimensional point cloud information corresponding to the welding image, the filtered point cloud is recorded and reserved as Cloud_1. From Cloud_1, it can be seen that the overall point cloud is composed of an incomplete cylindrical steel bar point cloud and a flat steel plate point cloud. In order to accurately obtain the lap weld of the steel bar and the steel plate, the main point cloud of the steel bar part and the main point cloud of the steel plate part need to be extracted from the overall point cloud. Here, a clustering algorithm can be directly used to extract the steel bar main point cloud and the steel plate flat main point cloud from the overall point cloud.

[0074] S200: According to the steel bar main point cloud and the steel plate flat main point cloud, a steel plate main plane and a steel plate auxiliary plane are constructed, and an intersection line of the steel plate auxiliary plane and the steel plate main plane is obtained. The steel plate auxiliary plane is a plane perpendicular to the steel plate.

[0075] In theory, if the reinforcing bar and the steel plate are completely attached, the intersection line of the reinforcing bar and the steel plate can be obtained by directly introducing the reinforcing bar cylindrical expression and the steel plate plane expression. However, in the actual working condition, there may be a certain angle difference between the reinforcing bar and the steel plate, and the intersection line of the reinforcing bar and the steel plate cannot be directly analyzed through the parameter expression. Therefore, it is necessary to construct a steel plate auxiliary plane to assist in calculation. Specifically, the steel plate auxiliary plane is perpendicular to the steel plate plane. In actual application, the steel plate auxiliary plane can be constructed based on the steel plate main plane and the reinforcing bar axis. After the steel plate auxiliary plane is constructed, a reference plane related to the position relationship between the reinforcing bar and the steel plate can be obtained, and the intersection line is calculated based on the reinforcing bar auxiliary plane and the steel plate main plane.

[0076] S300: Selecting, from the reinforcing bar main point cloud, point clouds with a distance to the intersection line less than the reinforcing bar radius, to obtain a target point cloud.

[0077] In order to further determine the point cloud data related to the weld, it is necessary to select the part of the point cloud on the steel plate main point cloud Cloud_P1 close to the intersection line L2. The specific operation is to calculate the distance of all points of the steel plate main point cloud Cloud_P1 to the intersection line L2, and select the part of the point cloud with a distance less than the reinforcing bar radius R, denoted as Cloud_P2. The role of this step is to remove the point cloud far from the theoretical intersection line and not likely to be the weld related area, so as to narrow the data range, improve the efficiency and accuracy of subsequent calculation, and make the obtained Cloud_P2 more accurately reflect the part of the steel plate main close to the intersection line of the reinforcing bar and the steel plate.

[0078] S400: Calculating a first projection segment formed after the target point cloud is projected on the intersection line, and a second projection segment formed after the steel plate plane main point cloud is projected on the intersection line.

[0079] The first projection segment formed after all points of Cloud_P2 are projected on the intersection line L2 is denoted as Sp. This segment represents the part of the steel plate main point cloud Cloud_P1 close to the intersection line L2, corresponding to the range on the intersection line L2 of the reinforcing bar and the steel plate. By calculating this segment, the projection range of the area on the steel plate main that may produce weld with the reinforcing bar on the intersection line can be determined. At the same time, the second projection segment formed after all points of the steel plate main point cloud Cloud_C1 are projected on L2 is denoted as Sc. This segment represents the projection of the steel plate main point cloud Cloud_C1 on the intersection line L2 of the reinforcing bar and the steel plate. This calculation can determine the projection range of the area on the steel plate main that may produce weld with the reinforcing bar on the intersection line. The calculation of these two segments provides key data for subsequent determination of the specific position of the weld on the intersection line.

[0080] S500: Determining the weld between the reinforcing bar and the steel plate in the reinforcing bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment.

[0081] Here, the intersection of the projection segments Sp and Sc can be calculated first, and the intersection is used to intercept a straight line segment on the intersection line L2 to obtain the straight line segment S1. This straight line segment S1 reflects the area of the common projection of the steel bar and the steel plate on the intersection line L2, that is, the representation of the area where the steel bar and the steel plate can actually produce a weld on the intersection line. Further, the weld to be welded in the steel bar and steel plate lap joint is located on both sides of the intersection line L2 of the steel bar and the steel plate, so the weld between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece can be determined within a reasonable range based on the straight line segment S1.

[0082] The above steel bar and steel plate lap joint workpiece weld identification method obtains a welding image of the steel bar and steel plate lap joint workpiece, extracts steel bar main body point cloud and steel plate plane main body point cloud corresponding to the welding image; constructs a steel plate main plane and a steel plate auxiliary plane according to the steel bar main body point cloud and the steel plate plane main body point cloud, and obtains an intersection line of the steel plate auxiliary plane and the steel plate main plane; the steel plate auxiliary plane is a plane perpendicular to the steel plate; the target point cloud is obtained by screening the point cloud with a distance less than the steel bar radius from the intersection line from the steel bar main body point cloud; the first projection segment formed after the target point cloud is projected on the intersection line, and the second projection segment formed after the steel plate plane main body point cloud is projected on the intersection line are calculated; based on the first projection segment and the second projection segment, the weld between the steel bar and the steel plate in the steel bar and steel plate lap joint workpiece is determined. In the whole process, the weld is identified based on the three-dimensional point cloud, and the error is reduced through the main body point cloud extraction, the intersection line extraction based on the steel plate auxiliary plane, and the intersection line projection, and finally the weld between the steel bar and the steel plate is automatically determined, which can realize efficient and accurate weld identification in the steel bar and steel plate lap joint workpiece.

[0083] In one embodiment, as shown in FIG. 5, Figure 3 S500 includes:

[0084] S520: Calculate the intersection between the first projection segment and the second projection segment, and intercept a straight line segment on the intersection line based on the intersection.

[0085] In the previous steps, the first projection segment Sp formed after the target point cloud is projected on the intersection line and the second projection segment Sc formed after the steel plate plane main body point cloud is projected on the intersection line have been calculated respectively. The intersection of the two projection segments is calculated by using the method of set operation. Specifically, the first projection segment and the second projection segment are regarded as two point sets on the intersection line, and by traversing the two point sets, the points belonging to both point sets are found, and the set of these points is the intersection of the two point sets.

[0086] The intersection represents the overlapping part of the steel bar and the steel plate projection on the intersection line during the lap process. In practical terms, this part of the area is most likely the projection position of the weld on the intersection line. Based on this intersection, a straight line segment S1 is obtained by intercepting a straight line segment on the intersection line. The purpose of this step is to further narrow the range where the weld may exist, providing more accurate basic data for subsequent accurate determination of the weld position. By intercepting the straight line segment S1, the areas on the intersection line where the weld cannot exist are removed, reducing the amount of data for subsequent processing and improving the calculation efficiency and accuracy.

[0087] S540: Solve the vector V perpendicular to the intersection line on the main plane of the steel plate.

[0088] After determining the straight line segment S1, in order to further determine the specific position of the weld on the main plane of the steel plate, the vector V perpendicular to the intersection line needs to be solved on the main plane PL1 of the steel plate. Let the normal of the main plane of the steel plate be P n (X pn , Y pn , Z pn ), and the direction vector of the intersection line be L v (X lv , Y lv , Z lv ). According to the properties of vector cross product, the cross product of two vectors is a vector perpendicular to the two vectors. Therefore, the vector V is calculated by the formula V(x, y, z) = P n (X pn , Y pn , Z pn ) x L v (X lv , Y lv , Z lv ). The vector V is located on the main plane PL1 of the steel plate and is perpendicular to the intersection line L2. In practical terms, it represents the offset direction on both sides of the intersection line along the direction perpendicular to the intersection line on the main plane of the steel plate. The purpose of this step is to provide directional information for subsequent determination of the specific position of the weld on both sides of the intersection line. Since the weld is usually formed on both sides of the intersection line of the steel bar and the steel plate, solving the vector V perpendicular to the intersection line and located on the main plane of the steel plate is a key step for accurately determining the position of the weld.

[0089] S560: Determine the weld between the steel bar and the steel plate according to the straight line segment and the vector V.

[0090] According to the straight line segment S1 and the vector V, a certain weld spacing is selected to determine the weld between the steel bar and the steel plate. Specifically, the straight line segment S1 can be taken as a reference to move a selected weld spacing in the positive and negative directions of the vector V to determine two welds between the steel bar and the steel plate. The specific welds identified are as Figure 4As shown, Figure 4 The middle green part is the steel bar main point cloud, the gray part is the steel plate main point cloud, and the blue line segment is the intersection of the steel bar and the steel plate on the intersection line. The two red line segments are the two welds to be welded positioned by the algorithm after offset according to the process requirements. The two welds are located on the steel plate plane and distributed on both sides of the intersection line of the steel bar and the steel plate.

[0091] In one of the embodiments, determining the weld between the steel bar and the steel plate according to the straight line segment and the vector V includes: obtaining the steel bar size, the steel plate thickness, and the welding process requirement of the steel bar and steel plate overlap workpiece, determining the weld distance d; moving the weld distance d along the positive and negative directions of the vector V on the straight line segment respectively, to determine two welds between the steel bar and the steel plate.

[0092] In the case of the steel bar and the steel plate overlap, the weld to be welded is located on both sides of the intersection line of the steel bar and the steel plate, and the distance to the intersection line needs to be determined according to the steel bar size, the steel plate thickness, the process requirement and other data. Specifically, a corresponding relationship table can be constructed based on historical data, and when the weld distance d needs to be determined, the corresponding accurate weld distance d can be obtained based on the corresponding relationship table. According to the straight line segment S1 and the vector V obtained in the foregoing and the determined offset distance d, the positions of the two welds to be welded are calculated by the formulas S2(x, y, z) = S1(x, y, z) + dV(x, y, z); S3(x, y, z) = S1(x, y, z) - dV(x, y, z). The formula S2(x, y, z) = S1(x, y, z) + dV(x, y, z) represents moving the straight line segment S1 along the positive direction of the vector V by a distance d to obtain the position of one of the welds; the formula S3(x, y, z) = S1(x, y, z) - dV(x, y, z) represents moving the straight line segment S1 along the negative direction of the vector V by a distance d to obtain the position of the other weld.

[0093] In this embodiment, the specific position of the weld in the three-dimensional space is accurately determined in combination with the actual welding process requirement. By considering the steel bar size, the steel plate thickness and the process requirement, the offset distance d is determined, and the vector V and the straight line segment S1 calculated in the foregoing are used, so that the two welds to be welded can be accurately positioned, providing clear guidance for subsequent welding operation, and ensuring the welding quality and the accuracy of the welding process.

[0094] In one of the embodiments, as shown in Figure 5 S100 includes:

[0095] S120: Obtain the welding image of the steel bar and steel plate overlap workpiece.

[0096] The welding image can be specifically acquired by a three-dimensional camera. The server acquires the welding image acquired by the three-dimensional camera. Further, the three-dimensional camera can be directed towards the to-be-positioned welding position, an angle deviation can not exceed 15°, the camera is 500 mm away from the to-be-positioned workpiece, a positive or negative deviation is not more than 100 mm, the camera photographs the workpiece, and the to-be-welded workpiece must be in the field of view of the camera and as close to the center of the field of view of the camera as possible during photographing.

[0097] S140: According to the welding image, a welding workpiece point cloud corresponding to the steel bar and steel plate lap joint workpiece is extracted.

[0098] According to the acquired welding image, an existing three-dimensional reconstruction technology (such as a structured light three-dimensional reconstruction, a laser scanning three-dimensional reconstruction, etc.) is used to convert the two-dimensional welding image into a three-dimensional welding workpiece point cloud. In this process, first, according to the specified photographing distance of the camera to the to-be-welded workpiece, a band-pass filter is used to process the acquired point cloud. The function of the band-pass filter is to filter out the point cloud with a Z direction value in the range of 400 mm-600 mm, and remove the influence of the cluttered background. Because in the actual scene, in addition to the steel bar and steel plate lap joint workpiece, there can be other objects or backgrounds around, the point cloud generated by these backgrounds will interfere with the subsequent analysis of the steel bar and steel plate point cloud, and the irrelevant point cloud can be effectively removed by the band-pass filter, improving the quality of the point cloud data. Next, the filtered point cloud is subjected to voxel downsampling filtering. Voxel downsampling filtering is to divide the three-dimensional space into small voxels (cubic units), and then select a representative point (such as the center of gravity point) in each voxel to represent all points in the voxel. The purpose of this step is to improve the consistency of the point cloud density. Since the point cloud density may vary in different areas when collecting the point cloud, excessive density difference will affect the processing effect of the subsequent algorithm, and the voxel downsampling filtering can make the point cloud density more uniform, providing more stable point cloud data for subsequent processing. Finally, radius density filtering is used to remove discrete points in the point cloud and reduce point cloud noise. Radius density filtering defines a radius range around each point, counts the number of points within the radius range, and if the number of points is too small, the point is considered to be a discrete point or noise point and is removed. This step can further purify the point cloud data and remove isolated points caused by measurement errors or other interference factors, thereby obtaining a high-quality to-be-welded workpiece whole point cloud, and the filtered point cloud is recorded as Cloud_1.

[0099] S160: Extract the steel bar point cloud and the steel plate plane point cloud in the welding workpiece point cloud.

[0100] As can be seen from the Cloud_1, the overall point cloud is composed of an incomplete cylindrical reinforcement point cloud and a planar steel plate point cloud. In order to accurately obtain the lap weld between the reinforcement and the steel plate, it is necessary to extract the point cloud of the reinforcement part and the point cloud of the steel plate part from the overall point cloud, and to perform parameterized expression. The server can specifically use a fitting algorithm to extract the reinforcement point cloud and the steel plate planar point cloud in the welding workpiece point cloud. Further, the RANSAC (Random Sample Consensus) fitting plane algorithm and the RANSAC fitting cylinder algorithm can be used to extract the point cloud to obtain the reinforcement point cloud and the steel plate planar point cloud.

[0101] Extracting the steel plate planar point cloud: First, use the RANSAC algorithm to fit the steel plate planar point cloud from the welding workpiece point cloud Cloud_1. The RANSAC fitting algorithm is a robust method for extracting a target model from a three-dimensional point cloud containing noise and outliers. When fitting the steel plate planar point cloud, a distance threshold (the maximum allowed distance from the specified model, considering that the general industrial steel plate is relatively consistent, the distance threshold is set to 2mm), the number of iterations (set to one-fifth of the total number of point clouds) and the proportion of inliers threshold (since the steel plate body is large, the proportion of inliers when fitting the steel plate can be 70%). Through multiple iterations, the algorithm will find a best plane model that maximizes the number of inliers that satisfy the distance threshold. The parameterized expression obtained after fitting is the general equation of the plane Ax+By+Cz+D=0, which is converted into the point formula, where n=(A, B, C), and the point (x_0, y_0, z_0) on the plane. Denote the conversion to the point formula, the normal vector of the plane (PlaneNormal) is P n (X pn , Y pn , Z pn ), and the point on the plane is P p (X pp , Y pp , Z pp ). Denote the planar point cloud extracted from the welding workpiece point cloud Cloud_1 as Cloud_P0, and the remaining point cloud as Cloud_2. By using the RANSAC algorithm to extract the steel plate planar point cloud, the steel plate part can be accurately separated from the overall point cloud, providing a basis for subsequent analysis of the lap relationship between the steel plate and the reinforcement.

[0102] Extracting the steel bar cylindrical point cloud: using the RANSAC algorithm to fit the steel bar cylindrical point cloud from Cloud_2. Similarly, when fitting the cylinder, the corresponding parameters also need to be set, such as the distance threshold, the number of iterations, and the inner point proportion threshold (the inner point proportion when fitting the cylinder can be 10%). The parameterized expression obtained after fitting contains 7 parameters, which are the cylindrical radius R, the cylindrical axis vector (CylindricalAxis) Ca(Xca, Yca, Zca), and a point Cp(Xcp, Ycp, Zcp) on the cylindrical axis. Denote the extracted cylindrical point cloud as Cloud_C0. By extracting the steel bar cylindrical point cloud through the RANSAC algorithm, the shape and position information of the steel bar can be accurately obtained, providing key data for subsequent analysis of the lap weld between the steel bar and the steel plate.

[0103] S180: clustering the steel bar point cloud and the steel plate plane point cloud respectively to obtain the steel bar main body point cloud and the steel plate plane main body point cloud.

[0104] Due to the complex field environment, the plane point cloud Cloud_P0 and the cylindrical point cloud Cloud_C0 may contain interference data. For example, if there is a plane object with the same height as the steel plate in the camera field of view, the point cloud formed by this plane will also be included in Cloud_C0, forming interference data. In order to avoid the errors that interference data may bring to subsequent calculations, the server uses clustering algorithms to cluster the steel bar point cloud and the steel plate plane point cloud respectively to obtain the steel bar main body point cloud and the steel plate plane main body point cloud. Specifically, the dbscan (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm can be used to cluster the plane point cloud Cloud_P0 and the cylindrical point cloud Cloud_C0 respectively, and the steel plate plane main body point cloud and the steel bar main body point cloud.

[0105] In one embodiment, clustering the steel bar point cloud and the steel plate plane point cloud respectively to obtain the steel bar main body point cloud and the steel plate plane main body point cloud includes:

[0106] Step 1: clustering the steel bar point cloud by the dbscan algorithm, and taking the cluster with the most point clouds in the clustered point cloud as the steel bar main body point cloud.

[0107] The cylindrical point cloud Cloud_C0 is clustered by the dbscan algorithm. The dbscan is a density-based unsupervised clustering method, and its core idea is to divide the sufficiently dense region into clusters and identify noise points. It has the advantages of being able to process non-convex clusters, strong adaptability, automatic identification of noise points, enhanced robustness of results, no need for manual setting of the number of clusters, and reduced dependence on prior knowledge. After clustering Cloud_C0, the cluster with the most point clouds is taken as the main body of the reinforcement point cloud, denoted as Cloud_C1, which has the same parameterized expression as Cloud_C0, including the cylindrical radius R, the cylindrical axis vector (CylindricalAxis) Ca(Xca, Yca, Zca), and a point Cp(Xcp, Ycp, Zcp) on the cylindrical axis, denoted as C1. Through the dbscan clustering algorithm, the interference data in the reinforcement point cloud can be effectively removed, and a more accurate reinforcement main body point cloud is obtained.

[0108] Step 2: The steel plate plane point cloud is clustered by the dbscan algorithm, and the cluster with the most point clouds and adjacent to the reinforcement is obtained from the clustered multi-cluster point cloud, to obtain the main body point cloud of the steel plate plane.

[0109] The steel plate plane point cloud Cloud_P0 is clustered by the dbscan algorithm, and the cluster with the most point clouds and adjacent to the reinforcement is obtained from the clustered multi-cluster point cloud, to obtain the main body point cloud of the steel plate plane. n (Xpn, Ypn, Zpn), and a point P p (Xpp, Ypp, Zpp) on the plane is denoted as PL1. In this way, the main body point cloud of the steel plate overlapping with the reinforcement can be accurately extracted from the plane point cloud, and other possible interference plane point clouds are excluded, providing reliable data support for accurately determining the overlapping weld between the reinforcement and the steel plate. Further, the cluster of reinforcement vectors can be obtained by traversing the reinforcement main body point cloud Cloud_C1 and each point in each cluster after clustering, calculating the distance between two points, and determining which cluster is adjacent to the reinforcement point cloud according to the nearest distance.

[0110] In one embodiment, the main plane of the steel plate and the auxiliary plane of the steel plate are constructed according to the reinforcement main body point cloud and the steel plate plane main body point cloud, including:

[0111] Step 1: Determine the cylindrical C corresponding to the reinforcement according to the reinforcement main body point cloud, and construct the main plane of the steel plate according to the steel plate plane main body point cloud.

[0112] Since the steel bar can be generally approximated as a cylinder in macroscopic view, a cylinder fitting algorithm (e.g. least square method to fit a cylinder) is used to determine the cylinder C corresponding to the steel bar based on the point cloud of the steel bar body. The cylinder C can accurately describe the geometry of the steel bar in space, and its parameters include the equation of the cylinder axis, the radius, etc. Based on the point cloud of the main plane of the steel plate, a plane fitting algorithm (e.g. least square method to fit a plane) is used to construct the main plane of the steel plate. The plane represents the main position and attitude of the steel plate in space.

[0113] Step 2: Determine the steel plate plane normal vector corresponding to the main plane of the steel plate.

[0114] The steel plate plane normal vector is a vector perpendicular to the main plane of the steel plate, which describes the direction of the main plane of the steel plate.

[0115] Step 3: Based on the cylinder C and the steel plate plane normal vector, construct the auxiliary plane of the steel plate.

[0116] In theory, if the steel bar and the steel plate are completely attached, the intersection line of the steel bar and the steel plate can be obtained by directly introducing the steel bar cylinder expression and the steel plate plane expression. However, in actual working conditions, there may be a certain angle difference between the steel bar and the steel plate, which cannot be directly analyzed by the parameter expression to obtain the intersection line of the steel bar and the steel plate. Therefore, the auxiliary plane PL2 is constructed by passing through the axis of the steel bar and being perpendicular to the steel plate. The vector N3 is calculated by combining the normal vector Pn perpendicular to the main plane PL1 of the steel plate and the axis vector Ca of the cylinder C1. The auxiliary plane PL2 is obtained by combining the cylinder axis point Cp.

[0117] In one embodiment, obtaining the intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate comprises:

[0118] Step 1: Determine the steel plate auxiliary plane normal vector corresponding to the auxiliary plane of the steel plate.

[0119] The steel plate auxiliary plane normal vector is a vector perpendicular to the auxiliary plane of the steel plate, which determines the directional characteristics of the auxiliary plane of the steel plate.

[0120] Step 2: Cross multiply the steel plate plane normal vector and the steel plate auxiliary plane normal vector to obtain the target direction vector.

[0121] The steel plate plane normal vector is a vector perpendicular to the main plane of the steel plate. In this step, the steel plate plane normal vector (A pl1 , B pl1 , C pl1 ) is cross multiplied with the steel plate auxiliary plane normal vector (A pl2 , B pl2 , C pl2 ) to obtain the target direction vector L v (X lv , Y lv , Z lv) while being perpendicular to the normal vector of the steel plate main plane PL1 and the normal vector of the steel plate auxiliary plane PL2.

[0122] Step 3: Based on the target direction vector, determine the intersection line of the steel plate auxiliary plane and the steel plate main plane.

[0123] The intersection line L2 is both the intersection line of the steel bar and the steel plate in theory, and the parametric expression of the intersection line L2 is composed of the direction vector L v (X lv , Y lv , Z lv ) of the straight line and a point L p (X lp , Y lp , Z lp ) on the straight line.

[0124] In practical applications, the entire intersection line calculation process includes the following two steps:

[0125] Step 1: First, the normal vector of the steel plate auxiliary plane PL2 and the normal vector of the steel plate main plane PL1 are crossed to obtain the direction vector L v of the intersection line L2.

[0126] Step 2: Convert the steel plate auxiliary plane PL2 and the steel plate main plane PL1 into general formula (Ax+By+Cz+D=0), and consider that x, y, z are brought into a special value (for example, z=0), then the two infinite solution three-element first-order equation groups are converted into a two-element first-order equation group with a unique solution. Solve the two-element first-order equation group, and the point that falls on both planes is obtained. The point is a point on the straight line Lp, and the point is a point on the intersection line of the two planes.

[0127] It should be understood that although each step in the flowchart involved in each of the above embodiments is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each of the above embodiments can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0128] Based on the same inventive concept, the application further provides a reinforcing steel bar and steel plate lap joint weld seam identification device for implementing the above-mentioned reinforcing steel bar and steel plate lap joint weld seam identification method. The device provides a solution to the implementation scheme as described in the above-mentioned method, and therefore the specific limitations in one or more reinforcing steel bar and steel plate lap joint weld seam identification device embodiments provided below can refer to the limitations of the reinforcing steel bar and steel plate lap joint weld seam identification method described above, which will not be repeated here.

[0129] In one embodiment, as shown in Figure 6 The application also provides a reinforcing steel bar and steel plate lap joint weld seam identification device. The device comprises:

[0130] A point cloud extraction module 100 is configured to obtain a welding image of a reinforcing steel bar and steel plate lap joint, and extract a reinforcing steel bar body point cloud and a steel plate plane body point cloud corresponding to the welding image.

[0131] A intersection line determination module 200 is configured to construct a main steel plate plane and an auxiliary steel plate plane according to the reinforcing steel bar body point cloud and the steel plate plane body point cloud, and obtain an intersection line of the auxiliary steel plate plane and the main steel plate plane; the auxiliary steel plate plane is a plane perpendicular to the steel plate.

[0132] A target point cloud extraction module 300 is configured to filter point clouds with a distance to the intersection line less than a reinforcing steel bar radius from the reinforcing steel bar body point cloud to obtain a target point cloud.

[0133] A segmented calculation module 400 is configured to calculate a first projection segment formed by projecting the target point cloud on the intersection line, and a second projection segment formed by projecting the steel plate plane body point cloud on the intersection line.

[0134] A weld seam confirmation module 500 is configured to determine a weld seam between the reinforcing steel bar and the steel plate in the reinforcing steel bar and steel plate lap joint based on the first projection segment and the second projection segment.

[0135] In one embodiment, the weld seam confirmation module 500 is further configured to calculate an intersection between the first projection segment and the second projection segment, and intercept a straight line segment on the intersection line based on the intersection; solve a vector V perpendicular to the intersection line on the main steel plate plane; determine the weld seam between the reinforcing steel bar and the steel plate according to the straight line segment and the vector V.

[0136] In one embodiment, the weld seam confirmation module 500 is further configured to obtain a reinforcing steel bar size, a steel plate thickness, and a welding process requirement of the reinforcing steel bar and steel plate lap joint, determine a weld seam distance d; move the weld seam distance d in the positive and negative directions of the vector V on the straight line segment respectively, and determine two weld seams between the reinforcing steel bar and the steel plate.

[0137] In one of the embodiments, the point cloud extraction module 100 is further configured to acquire a welding image of the steel bar and steel plate lap joint workpiece; extract a welding workpiece point cloud corresponding to the steel bar and steel plate lap joint workpiece according to the welding image; extract a steel bar point cloud and a steel plate plane point cloud in the welding workpiece point cloud; and respectively cluster the steel bar point cloud and the steel plate plane point cloud to obtain a steel bar main body point cloud and a steel plate plane main body point cloud.

[0138] In one of the embodiments, the point cloud extraction module 100 is further configured to cluster the steel bar point cloud by using a dbscan algorithm, and take a cluster with the largest number of point clouds in the clustered point cloud as the steel bar main body point cloud; cluster the steel plate plane point cloud by using the dbscan algorithm, and obtain a cluster with the largest number of point clouds and adjacent to the steel bar in the clustered point cloud to obtain the steel plate plane main body point cloud.

[0139] In one of the embodiments, the intersection line determination module 200 is further configured to determine a cylinder C corresponding to the steel bar according to the steel bar main body point cloud, and construct a main plane of the steel plate according to the steel plate plane main body point cloud; determine a normal vector of the main plane of the steel plate; and construct an auxiliary plane of the steel plate based on the cylinder C and the normal vector of the main plane of the steel plate.

[0140] In one of the embodiments, the intersection line determination module 200 is further configured to determine a normal vector of the auxiliary plane of the steel plate; cross multiply the normal vector of the main plane of the steel plate and the normal vector of the auxiliary plane of the steel plate to obtain a target direction vector; and determine an intersection line of the auxiliary plane of the steel plate and the main plane of the steel plate based on the target direction vector.

[0141] The above-mentioned various modules in the steel bar and steel plate lap joint workpiece welding seam identification device can be realized by software, hardware and combinations thereof in whole or in part. The above-mentioned various modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to call and execute the operations corresponding to the above-mentioned various modules by the processor.

[0142] In one embodiment, a computer device is provided, which can be a server, and the internal structure diagram thereof can be as shown in Figure 7 The computer device includes a processor, a memory and a network interface connected by a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is configured to store preset data. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a steel bar and steel plate lap joint workpiece welding seam identification method.

[0143] Those skilled in the art can understand that Figure 7 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0144] In one embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the above-mentioned weld seam identification method in a steel bar and steel plate lap joint workpiece.

[0145] In one embodiment, a computer readable storage medium is provided, storing a computer program, and the computer program is executed by a processor to implement the above-mentioned weld seam identification method in a steel bar and steel plate lap joint workpiece.

[0146] In one embodiment, a computer program product is provided, comprising a computer program, and the computer program is executed by a processor to implement the above-mentioned weld seam identification method in a steel bar and steel plate lap joint workpiece.

[0147] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. The non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. The volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0148] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0149] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be noted that for those skilled in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for identifying weld seams in workpieces where reinforcing bars and steel plates overlap, characterized in that, The method includes: Acquire welding images of the workpiece where the reinforcing bar and steel plate overlap, and extract the main point cloud of the reinforcing bar and the main point cloud of the steel plate corresponding to the welding image; Based on the main point cloud of the reinforcing bars and the main point cloud of the steel plate, construct the main plane and auxiliary plane of the steel plate, and obtain the intersection line between the auxiliary plane and the main plane of the steel plate; the auxiliary plane of the steel plate is a plane perpendicular to the steel plate. The target point cloud is obtained by filtering the point cloud of the main steel reinforcement body that is less than the radius of the steel reinforcement at the distance from the intersection line. Calculate the first projection segment formed by the projection of the target point cloud onto the intersection line, and the second projection segment formed by the projection of the main point cloud of the steel plate plane onto the intersection line; Based on the first projection segment and the second projection segment, the weld between the reinforcing bar and the steel plate in the reinforcing bar and steel plate lap joint workpiece is determined.

2. The method according to claim 1, characterized in that, The determination of the weld between the reinforcing bar and the steel plate in the lap joint of the reinforcing bar and the steel plate, based on the first projection segment and the second projection segment, includes: Calculate the intersection between the first projection segment and the second projection segment, and extract a straight line segment on the intersection line based on the intersection; Solve for the vector V perpendicular to the intersection line on the main plane of the steel plate; The weld between the reinforcing bar and the steel plate is determined based on the straight line segmentation and the vector V.

3. The method according to claim 2, characterized in that, Based on the linear segmentation and the vector V, the weld between the reinforcing bar and the steel plate is determined as follows: Obtain the dimensions of the reinforcing bars, the thickness of the steel plates, and the welding process requirements for the lap joints between the reinforcing bars and the steel plates, and determine the weld distance d; The weld seam distance d is moved along the positive and negative directions of the vector V on the straight line segment to determine the two weld seams between the reinforcing bar and the steel plate.

4. The method according to claim 1, characterized in that, The step of acquiring a welding image of the lapped joint between the reinforcing bar and the steel plate, and extracting the main point cloud of the reinforcing bar and the main planar point cloud of the steel plate corresponding to the welding image, includes: Obtain welding images of the workpiece where reinforcing bars and steel plates overlap; Based on the welding image, extract the point cloud of the welded workpiece corresponding to the lap joint of the reinforcing bar and the steel plate; Extract the point cloud of reinforcing bars and the planar point cloud of steel plates from the point cloud of the welded workpiece; Clustering is performed on the point cloud of the reinforcing bars and the planar point cloud of the steel plate to obtain the main point cloud of the reinforcing bars and the main point cloud of the planar point cloud of the steel plate.

5. The method according to claim 4, characterized in that, The step of clustering the point cloud of the reinforcing bars and the planar point cloud of the steel plate to obtain the main point cloud of the reinforcing bars and the main point cloud of the planar point cloud of the steel plate includes: The point cloud of the reinforcing bars is clustered using the dbscan algorithm, and the cluster with the most points in the clustered point cloud is taken as the main point cloud of the reinforcing bars. The point cloud of the steel plate is clustered using the dbscan algorithm. The point cloud with the most points and adjacent to the reinforcing bars is obtained from the clustered point cloud, thus obtaining the main point cloud of the steel plate.

6. The method according to claim 1, characterized in that, Based on the main point cloud of the reinforcing bars and the main point cloud of the steel plate, the main plane and auxiliary plane of the steel plate are constructed as follows: The cylindrical C corresponding to the reinforcing bar is determined based on the main point cloud of the reinforcing bar, and the main plane of the steel plate is constructed based on the main point cloud of the steel plate plane. Determine the normal vector of the steel plate plane corresponding to the principal plane of the steel plate; Based on the cylinder C and the plane normal vector of the steel plate, an auxiliary plane of the steel plate is constructed.

7. The method according to claim 6, characterized in that, Obtaining the intersection line between the auxiliary plane of the steel plate and the main plane of the steel plate includes: Determine the normal vector of the auxiliary plane of the steel plate corresponding to the auxiliary plane of the steel plate; The target direction vector is obtained by cross-product of the normal vector of the steel plate plane and the normal vector of the auxiliary plane of the steel plate. Based on the target direction vector, the intersection line between the auxiliary plane of the steel plate and the main plane of the steel plate is determined.

8. A weld identification device for lap joints of reinforcing bars and steel plates, characterized in that, The device includes: The point cloud extraction module is used to acquire welding images of the workpiece where the steel bar and steel plate overlap, and to extract the main point cloud of the steel bar and the main point cloud of the steel plate planar structure corresponding to the welding image. The intersection line determination module is used to construct the main plane of the steel plate and the auxiliary plane of the steel plate based on the main point cloud of the reinforcing bars and the main point cloud of the steel plate plane, and to obtain the intersection line between the auxiliary plane of the steel plate and the main plane of the steel plate; the auxiliary plane of the steel plate is a plane perpendicular to the steel plate; The target point cloud extraction module is used to filter point clouds from the main point cloud of the reinforcing bar that are less than the radius of the reinforcing bar at the distance from the intersection line to the main point cloud of the reinforcing bar, so as to obtain the target point cloud. The segmented calculation module is used to calculate the first projection segment formed by the projection of the target point cloud onto the intersection line, and the second projection segment formed by the projection of the main point cloud of the steel plate plane onto the intersection line. The weld confirmation module is used to determine the weld between the reinforcing bar and the steel plate in the reinforcing bar and steel plate lap joint workpiece based on the first projection segment and the second projection segment.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images