A method of welding multiple work sites
By setting reference points in the welding area and establishing a coordinate system using a camera, a welding queue was constructed, solving the problems of insufficient precision and high labor costs in the welding of steel reinforcement cages, and realizing efficient and automated welding.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHAOXING CITY INVESTMENT & CONSTRUCTION IND MANUFACTURING CO LTD
- Filing Date
- 2023-12-21
- Publication Date
- 2026-05-29
AI Technical Summary
In existing technologies, the welding of steel reinforcement cages suffers from insufficient precision, high labor costs, and the need for frequent position adjustments in mechanical welding, making it particularly difficult to achieve efficient automated welding in complex environments.
By setting reference points in the welding area, using a first camera to acquire a wide background image and establish a planar coordinate system, and combining this with a second camera to acquire local site images, a welding queue is constructed to achieve automated welding at multiple working sites.
It improves welding precision, reduces labor costs, lowers calculation workload, ensures welding quality, and enables efficient automated operation in complex environments.
Smart Images
Figure CN117733451B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the technical field of general image data processing or generation, and particularly to a welding method with multiple working sites. Background Technology
[0002] Welding is a manufacturing process and technology that joins metals or other thermoplastic materials such as plastics by heating, high temperature, or high pressure. During welding, the workpiece and the solder melt to form a molten zone. Welding can pose dangers to operators, regardless of location, including burns, electric shock, vision impairment, inhalation of toxic fumes, and excessive ultraviolet radiation. Therefore, welding operations are increasingly becoming automated.
[0003] Rebar welding is a common welding operation, which involves using electric welding equipment to extend or cross-connect rebars along the axial direction. With the development of construction, the requirements for the fabrication of rebar cages are becoming increasingly stringent, and horizontal wide rebar cages to be welded are now a regular requirement.
[0004] In existing technologies, the processing methods for such wide-width steel reinforcement cages include manual and mechanical operation. The problems with manual operation are obvious, including but not limited to insufficient precision, high labor costs, and high product scrap rates. Mechanical operation requires pre-entering the coordinate information of the welding points before welding with a mechanical welding torch. However, as is well known, the processing environment is often complex, and the materials to be welded may shift during transportation, and their positions may also differ when placed on the jig. This results in the pre-entered welding points not being the actual welding points. Furthermore, an error in one data point may not necessarily affect the data of other points, and the error in one data point cannot be used to simultaneously optimize other data. This leads to the need for frequent adjustments to the welding position during mechanical welding, resulting in a large computational load and a lack of basic guarantee of welding quality. Summary of the Invention
[0005] This invention solves the problems existing in the prior art and provides a welding method with multiple working sites.
[0006] The technical solution adopted in this invention is a welding method with multiple working sites, the method comprising the following steps:
[0007] Step 1: Set up several reference points in the actual welding area;
[0008] Step 2: Use the first camera to acquire a background width image of the actual welding area, adjust the background width image, and record the adjustment parameters between the background width image and the actual welding area;
[0009] Step 3: Define an origin in the background wide-angle image, and establish a planar coordinate system based on the origin and the adjusted background wide-angle image;
[0010] Step 4: Input the material to be welded into the actual welding area, obtain a new wide-angle image, and adjust the new wide-angle image based on the adjustment parameters;
[0011] Step 5: Obtain the estimated working position in the adjusted new wide-angle image, and construct a welding queue based on the adjustment parameters corresponding to the actual welding area;
[0012] Step 6: Start the welding equipment. The welding equipment acquires local site images in the actual welding area according to the estimated working sites in the welding queue, determines the welding position in the local site images, and performs multi-working site welding.
[0013] Preferably, the actual welding area includes a wide welding plane, which includes an array of welding work points arranged on a horizontal plane; the reference points include at least three markers arranged in conjunction with the wide welding plane.
[0014] Preferably, step 2 includes the following steps:
[0015] Step 2.1: Use the first camera to acquire a wide background image of the actual welding area;
[0016] Step 2.2: Perform trapezoidal correction on the background wide-format image;
[0017] Step 2.3: Obtain the relative position of the reference point in the corrected background wide-angle image;
[0018] Step 2.4: Adjust the background wide-angle image based on the relative position of the reference point so that the position of the reference point relative to the background wide-angle image is consistent with the actual welding area;
[0019] Step 2.5: Record the adjustment parameters from Step 2.2 to Step 2.4.
[0020] Preferably, in step 3, if the background wide image is rectangular, then any corner point of the actual welding area corresponding to the background wide image is taken as the origin; otherwise, the center of the background wide image is defined as the origin.
[0021] Preferably, step 5 includes the following steps:
[0022] Step 5.1: Subtract the adjusted new wide-angle image from the corrected background wide-angle image to obtain the foreground image;
[0023] Step 5.2: Binarize the foreground image, traverse the image, and obtain the feature points to be welded;
[0024] Step 5.3: If the feature point to be welded meets the preset requirements, then take the feature point to be welded as the estimated working point and proceed to step 5.5; otherwise, proceed to the next step.
[0025] Step 5.4: Repeat the theoretical positions of the feature points to be welded in the preset requirements. If there are feature points to be welded, they are counted as estimated working positions. Otherwise, an alarm is triggered and the next step is performed.
[0026] Step 5.5: Based on the coordinates and adjustment parameters of all the estimated working positions, construct the welding queue corresponding to the actual welding area in a preset order.
[0027] Preferably, in step 5.2, isolated points in the foreground image are removed, the line width of the binarized foreground image is obtained, the average value is taken, and a rectangular border is constructed with the average value as the side length. The rectangular border is used to traverse the binarized foreground image. For any pixel, if its neighborhood is not empty and there is a pixel with a perpendicular side to the current neighborhood edge that is not less than α times the side length, it is recorded as a feature point to be welded.
[0028] Preferably, the welding equipment includes a welding torch, and a second camera is provided on the side of the welding torch.
[0029] Preferably, step 6 includes the following steps:
[0030] Step 6.1: Start the welding equipment. The welding equipment will start working according to the estimated working position in the welding queue that corresponds to the actual welding area.
[0031] Step 6.2: Obtain the first estimated working position. Move the welding equipment to this estimated working position and use the second camera to acquire a local position image.
[0032] Step 6.3: Define the ROI region. If there are no welding feature points in the ROI region, proceed to the next step; otherwise, weld directly according to the preset estimated working position and proceed to step 6.5.
[0033] Step 6.4: Magnify the ROI region proportionally according to the preset requirements. If there are still no welding feature points even when the ROI region is equal to the local site image, report an error and stop welding. Otherwise, perform welding and record the actual deviation of the welding feature points and the estimated working site corresponding to the real welding area, and include it in the next round of welding operation.
[0034] Step 6.5: After welding is completed, delete the currently estimated working position from the welding queue. If the welding queue is not empty, return to step 6.2; otherwise, end.
[0035] Preferably, a fill light is provided in conjunction with the second camera; if the error rate is higher than a preset value, the output power of the fill light is adjusted.
[0036] Preferably, the welding queue has the origin as the first element of the queue.
[0037] This invention relates to a multi-working-site welding method, which involves setting several reference points in a real welding area; acquiring a background wide-angle image of the real welding area using a first camera; adjusting the background wide-angle image and recording the adjustment parameters between the background wide-angle image and the real welding area; defining an origin in the background wide-angle image and establishing a planar coordinate system based on the origin and the adjusted background wide-angle image; inputting the material to be welded into the real welding area and acquiring a new wide-angle image, which is then adjusted based on the adjustment parameters; acquiring estimated working sites in the adjusted new wide-angle image and mapping them to the real welding area based on the adjustment parameters to construct a welding queue; starting the welding equipment, which acquires local site images in the real welding area according to the estimated working sites in the welding queue, determines the welding position in the local site images, and performs multi-working-site welding.
[0038] The beneficial effects of this invention are as follows:
[0039] (1) Avoid the problems of manual operation, ensure that the accuracy meets the requirements, and reduce labor costs;
[0040] (2) The width is adjusted by adjusting the first camera, and the working area of the welding equipment is divided. It does not rely entirely on the input coordinates and makes up for the image error caused by the first camera.
[0041] (3) Fine-area adjustments are made by using the adjustment of the second camera, so that the welding equipment can search for the working position in a limited area with a high probability, thus ensuring the controllability of mechanical control;
[0042] (4) No special environmental control is required for the processing area; it is only necessary to ensure that there is no significant deviation at the start of the welding operation.
[0043] (5) The overall workload is small and the welding quality is guaranteed. Attached Figure Description
[0044] Figure 1 This is a flowchart of the method of the present invention;
[0045] Figure 2 This is a flowchart illustrating the process of adjusting the background width image in this invention;
[0046] Figure 3 This is a flowchart of constructing the welding queue in this invention;
[0047] Figure 4 This is a flowchart of the welding process in this invention;
[0048] Figure 5 This is a schematic diagram of the application device structure of the present invention. Detailed Implementation
[0049] The present invention will be further described in detail below with reference to embodiments, but the scope of protection of the present invention is not limited thereto.
[0050] This invention relates to a welding method with multiple working sites, the method comprising the following steps:
[0051] Step 1: Set up several reference points in the actual welding area;
[0052] Step 2: Use the first camera 2 to acquire a background width image of the actual welding area, adjust the background width image, and record the adjustment parameters between the background width image and the actual welding area;
[0053] Step 3: Define an origin in the background wide-angle image, and establish a planar coordinate system based on the origin and the adjusted background wide-angle image;
[0054] Step 4: Input the material to be welded into the actual welding area, obtain a new wide-angle image, and adjust the new wide-angle image based on the adjustment parameters;
[0055] Step 5: Obtain the estimated working position in the adjusted new wide-angle image, and construct a welding queue based on the adjustment parameters corresponding to the actual welding area;
[0056] Step 6: Start the welding equipment. The welding equipment acquires local site images in the actual welding area according to the estimated working sites in the welding queue, determines the welding position in the local site images, and performs multi-working site welding.
[0057] In this invention, step 1 sets several reference points in the actual welding area. The actual welding area here refers to a wide welding plane, which generally includes more than 100 working points to be welded. These working points to be welded are generally arranged in an array on the horizontal plane.
[0058] In this invention, for subsequent image calibration, at least three markers set in conjunction with the wide welding plane are used as reference points; it should be noted that there is no third reference point on the line segment between any two reference points or on the extension of the line segment, such as setting pointers at the corner positions of the welding jig 1.
[0059] In this invention, a wide background image of the actual welding area is acquired using a first camera 2. The first camera 2 is generally positioned directly above the center of the actual welding area. Therefore, the wide background image at this location is usually distorted and needs to be adjusted. After adjustment, the adjustment parameters between the wide background image and the actual welding area need to be recorded for adjustment of subsequent newly acquired images. Of course, the first camera 2 can be set as a line scan camera to minimize the deformation.
[0060] In this invention, after constructing a planar coordinate system for the background wide-width image that has been adjusted, the material to be welded is input into the actual welding area. At this time, a new wide-width image can be acquired, and the new wide-width image can be adjusted based on the aforementioned adjustment parameters to shorten the adjustment time. After obtaining the estimated working position, due to the existence of the reference point, the estimated working position can be mapped to the actual welding area and a welding queue can be constructed.
[0061] In this invention, the welding equipment arrives at the estimated working position in the order of the welding queue. By acquiring local position images, only local adjustments are needed to obtain the accurate welding position, thus achieving rapid welding of multiple working positions.
[0062] Step 2 includes the following steps:
[0063] Step 2.1: Use the first camera 2 to acquire a wide background image of the actual welding area;
[0064] Step 2.2: Perform trapezoidal correction on the background wide-format image;
[0065] Step 2.3: Obtain the relative position of the reference point in the corrected background wide-angle image;
[0066] Step 2.4: Adjust the background wide-angle image based on the relative position of the reference point so that the position of the reference point relative to the background wide-angle image is consistent with the actual welding area;
[0067] Step 2.5: Record the adjustment parameters from Step 2.2 to Step 2.4.
[0068] In this invention, after trapezoidal correction, the adjustment parameters of each pixel in the background wide-angle image may differ. Furthermore, the pixel position is adjusted by referring to a reference point to obtain the adjusted adjustment parameters, so that each pixel can be mapped to the actual position of the real welding area based on the adjustment parameters. Here, the set of adjustment parameters can be regarded as the transformation mapping relationship between camera coordinates and world coordinates.
[0069] In step 3, if the background wide image is rectangular, then any corner point of the actual welding area corresponding to the background wide image is taken as the origin; otherwise, the center of the background wide image is defined as the origin.
[0070] In this invention, the background wide image is generally rectangular. Here, "rectangular" refers to the actual working area being rectangular. The origin is taken as a corner point, usually the upper left corner. Otherwise, the center area is defined as the origin for ease of operation.
[0071] Step 5 includes the following steps:
[0072] Step 5.1: Subtract the adjusted new wide-angle image from the corrected background wide-angle image to obtain the foreground image;
[0073] Step 5.2: Binarize the foreground image, traverse the image, and obtain the feature points to be welded;
[0074] In step 5.2, isolated points in the foreground image are removed, the line width of the binarized foreground image is obtained, the average value is taken, and a rectangular border is constructed with the average value as the side length. The rectangular border is used to traverse the binarized foreground image. For any pixel, if its neighborhood is not empty and there is a pixel with a perpendicular side to the current neighborhood edge that is not less than α times the side length, it is recorded as a feature point to be welded.
[0075] Step 5.3: If the feature point to be welded meets the preset requirements, then take the feature point to be welded as the estimated working point and proceed to step 5.5; otherwise, proceed to the next step.
[0076] Step 5.4: Repeat the theoretical positions of the feature points to be welded in the preset requirements. If there are feature points to be welded, they are counted as estimated working positions. Otherwise, an alarm is triggered and the next step is performed.
[0077] Step 5.5: Based on the coordinates and adjustment parameters of all the estimated working positions, construct the welding queue corresponding to the actual welding area in a preset order.
[0078] In this invention, the method of acquiring the foreground image is conventional. The acquired foreground image is binarized. Considering the environment of the welding site, the binarized image generally contains noise, i.e., isolated points, which need to be removed. For the binarized foreground image after noise removal, it is generally a line or a network of intersecting lines. The lines represent the steel bars to be welded, which generally have a certain outer diameter. Therefore, a rectangular frame is constructed with the average value as the side length. By traversing the foreground image, the actual feature points to be welded can be obtained. The feature points to be welded include the intersection of the intersecting steel bars and the contact area between the two ends of the steel bars. Therefore, when the neighborhood of a pixel is not empty and there are pixels with a perpendicular side of not less than α times the side length of the current neighborhood edge, it indicates that this is the location of a "knot", which is obviously wider than the average width of the lines, so welding is required. Here, α is generally 0.12, but it can be adjusted based on the actual working conditions.
[0079] In this invention, after obtaining the feature points to be welded in step 5.2, if the layout of these feature points to be welded conforms to the theoretical value, then it is the estimated working position. Otherwise, it is necessary to re-traverse the theoretical positions (theoretical locations) of the feature points to be welded in the preset requirements. When there are no feature points to be welded at the theoretical location, an alarm needs to be issued, but it does not affect the subsequent welding operation. The work can be completed by re-welding.
[0080] In this invention, the coordinates of the estimated working position in the plane coordinate system are obtained, mapped to the actual welding area through the aforementioned adjustment parameters, and a welding queue of the actual welding area is constructed in sequence.
[0081] The welding equipment includes a welding torch 4, and a second camera 5 is provided on the side of the welding torch 4.
[0082] Step 6 includes the following steps:
[0083] Step 6.1: Start the welding equipment. The welding equipment will start working according to the estimated working position in the welding queue that corresponds to the actual welding area.
[0084] Step 6.2: Obtain the first estimated working position. Move the welding equipment to this estimated working position and use the second camera 5 to acquire local position images.
[0085] Step 6.3: Define the ROI area. If there are no welding feature points in the ROI area, proceed to the next step; otherwise, weld directly according to the preset estimated working position and proceed to step 6.5.
[0086] Step 6.4: Magnify the ROI region proportionally according to the preset requirements. If there are still no welding feature points even when the ROI region is equal to the local site image, report an error and stop welding. Otherwise, perform welding and record the actual deviation of the welding feature points and the estimated working site corresponding to the real welding area, and include it in the next round of welding operation.
[0087] Step 6.5: After welding is completed, delete the currently estimated working position from the welding queue. If the welding queue is not empty, return to step 6.2; otherwise, end.
[0088] The second camera 5 is equipped with a fill light; if the error rate is higher than the preset value, the output power of the fill light is adjusted.
[0089] In this invention, a second camera 5 is provided on the side of the welding torch 4 to acquire local images. The second camera 5 should be equipped with a protective cover or other structure to prevent the high temperature from affecting its performance. When the error rate is too large, the brightness, i.e. the output power, should be adjusted manually or by an electric controller.
[0090] In this invention, for each estimated working position, the welding equipment uses the second camera 5 to acquire local position images and delineate the ROI region. Generally, the ROI is delineated at the center of the local position image. If there are no welding feature points in the ROI region, the ROI is enlarged proportionally. If the ROI region is equal to the local position image and there are still no welding feature points, an error is reported, welding is stopped, and on-site troubleshooting is carried out. Otherwise, welding proceeds normally and deviations (if any) are recorded. In the next round of welding operations, the deviation direction is the priority search direction.
[0091] The welding queue has the origin as its first element.
[0092] In this invention, the origin is taken as the first element of the queue. Welding is not performed here, but rather used for calibration. When the position is accurate, welding is performed on the subsequent positions in the welding queue.
[0093] In the application of this invention, welding equipment will be set up, including a welding jig 1, a first camera 2 set above the welding jig 1, and welding equipment set up in conjunction with the welding jig 1. The welding equipment includes a welding robot 3, and the welding robot 3 is equipped with a welding torch 4 and a second camera 5.
[0094] In this invention, a welding jig 1 is used to receive the material to be welded, a first camera 2 is used to acquire a global background wide-angle image and a new wide-angle image, a second camera 5 is used to acquire local site images, and the welding gun 4 of the welding robot 3 is used to perform welding and achieve fine-tuning of the position.
[0095] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0096] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0097] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0098] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0099] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.
[0100] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.
Claims
1. A welding method with multiple working sites, characterized in that: The method includes the following steps: Step 1: Set up several reference points in the actual welding area; Step 2: Use the first camera to acquire a background width image of the actual welding area, adjust the background width image, and record the adjustment parameters between the background width image and the actual welding area; Step 3: Define an origin in the background wide-angle image, and establish a planar coordinate system based on the origin and the adjusted background wide-angle image; Step 4: Input the material to be welded into the actual welding area, obtain a new wide-angle image, and adjust the new wide-angle image based on the adjustment parameters; Step 5: Obtain the estimated working position in the adjusted new wide-angle image, and construct a welding queue based on the adjustment parameters corresponding to the actual welding area; Step 6: Start the welding equipment. The welding equipment acquires local site images in the actual welding area according to the estimated working sites in the welding queue, determines the welding position in the local site images, and performs multi-working site welding.
2. The multi-working-site welding method according to claim 1, characterized in that: The actual welding area includes a wide welding plane, which includes an array of welding work points arranged on a horizontal plane; the reference points include at least three markers arranged in conjunction with the wide welding plane.
3. The multi-working-site welding method according to claim 1, characterized in that: Step 2 includes the following steps: Step 2.1: Use the first camera to acquire a wide background image of the actual welding area; Step 2.2: Perform trapezoidal correction on the background wide-format image; Step 2.3: Obtain the relative position of the reference point in the corrected background wide-angle image; Step 2.4: Adjust the background wide-angle image based on the relative position of the reference point so that the position of the reference point relative to the background wide-angle image is consistent with the actual welding area; Step 2.5: Record the adjustment parameters from Step 2.2 to Step 2.
4.
4. The multi-working-site welding method according to claim 1, characterized in that: In step 3, if the background wide image is rectangular, then any corner point of the actual welding area corresponding to the background wide image is taken as the origin; otherwise, the center of the background wide image is defined as the origin.
5. The multi-working-site welding method according to claim 1, characterized in that: Step 5 includes the following steps: Step 5.1: Subtract the adjusted new wide-angle image from the corrected background wide-angle image to obtain the foreground image; Step 5.2: Binarize the foreground image, traverse the image, and obtain the feature points to be welded; Step 5.3: If the feature point to be welded meets the preset requirements, then take the feature point to be welded as the estimated working point and proceed to step 5.5; otherwise, proceed to the next step. Step 5.4: Repeat the theoretical positions of the feature points to be welded in the preset requirements. If there are feature points to be welded, they are counted as estimated working positions. Otherwise, an alarm is triggered and the next step is performed. Step 5.5: Based on the coordinates and adjustment parameters of all the estimated working positions, construct the welding queue corresponding to the actual welding area in a preset order.
6. The multi-working-site welding method according to claim 5, characterized in that: In step 5.2, isolated points in the foreground image are removed, the line width of the binarized foreground image is obtained, the average value is taken, and a rectangular border is constructed with the average value as the side length. The rectangular border is used to traverse the binarized foreground image. For any pixel, if its neighborhood is not empty and there is a pixel with a perpendicular side to the current neighborhood edge that is not less than α times the side length, it is recorded as a feature point to be welded.
7. The multi-working-site welding method according to claim 1, characterized in that: The welding equipment includes a welding torch, and a second camera is fitted to the side of the welding torch.
8. A multi-working-site welding method according to claim 7, characterized in that: Step 6 includes the following steps: Step 6.1: Start the welding equipment. The welding equipment will start working according to the estimated working position in the welding queue that corresponds to the actual welding area. Step 6.2: Obtain the first estimated working position. Move the welding equipment to this estimated working position and use the second camera to acquire a local position image. Step 6.3: Define the ROI region. If there are no welding feature points in the ROI region, proceed to the next step; otherwise, weld directly according to the preset estimated working position and proceed to step 6.
5. Step 6.4: Magnify the ROI region proportionally according to the preset requirements. If there are still no welding feature points even when the ROI region is equal to the local site image, report an error and stop welding. Otherwise, perform welding and record the actual deviation of the welding feature points and the estimated working site corresponding to the real welding area, and include it in the next round of welding operation. Step 6.5: After welding is completed, delete the currently estimated working position from the welding queue. If the welding queue is not empty, return to step 6.2; otherwise, end.
9. A multi-working-site welding method according to claim 8, characterized in that: The second camera is equipped with a supplementary light; if the error rate is higher than the preset value, the output power of the supplementary light is adjusted.
10. A multi-working-site welding method according to claim 1, characterized in that: The welding queue has the origin as its first element.