A method for adjusting parameters of a weld seam sensor
By adopting a multi-level, step-by-step automatic debugging method, the problem of existing weld sensor parameter debugging relying on operator experience has been solved, and an efficient and stable parameter debugging process has been achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 伯朗特机器人股份有限公司
- Filing Date
- 2026-03-18
- Publication Date
- 2026-06-19
AI Technical Summary
Existing methods for adjusting weld sensor parameters rely on the operator's personal experience, resulting in low efficiency and poor stability of the adjustment results.
A multi-layered, step-by-step automatic debugging method is adopted. Through steps such as image preprocessing, special point extraction, and weld seam recognition, a real-time scoring model is established, and the weld seam sensor parameters are automatically adjusted to form a standardized debugging process.
This reduces reliance on operator experience and improves the stability and efficiency of weld sensor parameter tuning.
Smart Images
Figure CN122244409A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of automated welding, and in particular to a method for adjusting weld seam sensor parameters. Background Technology
[0002] With the advancement of automation technology, the use of vision sensor-based automated welding is becoming increasingly common in factory production involving welding. However, before automated welding can begin, the vision sensors need to be configured to ensure accurate extraction of weld features. Parameter configuration typically involves setting numerous parameters, and current methods mostly rely on manual adjustments by engineers. These engineers adjust each parameter individually on the sensor's software interface based on their personal experience, considering the weld type and site conditions. This method is inefficient, and the quality and stability of the results are highly dependent on the operator's skill and experience. Summary of the Invention
[0003] Based on this, the purpose of the present invention is to provide a method for adjusting the parameters of a weld seam sensor, which has the advantages of not relying on the user's adjustment experience and having high adjustment stability.
[0004] A method for adjusting weld seam sensor parameters specifically includes the following steps:
[0005] S1 generates initial parameters and obtains the parameters to be debugged and the image to be identified for debugging; S2 adjusts the exposure time, binarization threshold, and filter intensity parameters in the parameters to be adjusted through image preprocessing; S3 uses a specially designed point extraction and debugging method to adjust the breakpoint distance, window size, and mutation threshold parameters in the parameters to be debugged. S4 uses weld seam identification and adjustment to adjust the maximum gap, maximum drop and angle deviation parameters among the parameters to be adjusted. S5 saves the parameter configuration.
[0006] The weld sensor parameter debugging method described in this invention forms a stable automatic debugging process through multi-level step-by-step automatic debugging, which greatly facilitates the operator's debugging of the weld sensor, eliminates the need to rely on the operator's personal experience for debugging, and has high stability.
[0007] Furthermore, step S2 specifically includes the following steps: S201 Acquire point cloud data of the debugging image to be identified and extract feature values of the debugging image to be identified; S202 Select partial feature values of the image to be identified and debugged, including the average intensity of the strip edges. Main strip pixels Total strip pixels and noise area proportion A clarity score is obtained by scoring the data. Connectivity score Ratio to noise score Compared with clarity-based scoring Connectivity score Ratio to noise score Image preprocessing score ; S203 Determine the image preprocessing score If the preset conditions for preprocessing and debugging are met, save the parameters to be debugged and continue with subsequent steps; otherwise, score based on sharpness. Connectivity score Ratio to noise score Adjust the parameters of exposure time, binarization threshold, and filter intensity to determine if the preset conditions for preprocessing debugging are met. After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S201.
[0008] Further, in step S202,
[0009]
[0010]
[0011] .
[0012] Further, in step S203, when the following conditions are met... When the time is up, save the parameters to be debugged and continue with the next steps; When not satisfied At that time, if Then increase the exposure time and / or increase the binarization threshold; if Then reduce the breakpoint distance and / or enable adaptive binarization; if If so, then median filtering will be enabled.
[0013] Furthermore, step S3 specifically includes the following steps: S301 acquires the point cloud data of the debugging image to be identified and extracts the feature values of the debugging image to be identified; S302 Select a subset of feature values from the image to be identified, including the average gradient magnitude of feature points. Area of marked region ROI area ratio outlier ratio The significance score is obtained by scoring. Coverage score Outlier score ; S303 Significance Score Coverage score Outlier score Does the preset conditions for special point extraction and debugging meet? If so, save the parameters to be debugged and continue with subsequent steps; otherwise, score based on saliency. Coverage score Outlier score Adjust parameters such as breakpoint distance, window size, and mutation threshold to determine if the preset conditions for special point extraction and debugging are met. After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S301.
[0014] Further, in step S302,
[0015]
[0016] .
[0017] Further, in step S303, when the condition is satisfied... , , If the time is right, save the parameters to be debugged and continue with the subsequent steps; Otherwise, if If so, then lower the mutation threshold; if If so, reduce the distance between the breakpoints; If so, the window size will be increased.
[0018] Furthermore, step S4 specifically includes the following steps: S401 acquires the point cloud data of the debugging image to be identified and extracts the feature values of the debugging image to be identified; S402 Select partial feature values of the image to be identified and debugged, including the fitting residual. Compliance Points Total points The geometric fit score is obtained by scoring. Constraint satisfaction score ;
[0019]
[0020] S403 Geometric Fit Score Constraint satisfaction score If the preset conditions for weld seam identification and debugging are met, save the parameters to be debugged and continue to the next step; otherwise, score according to geometric fitting. Constraint satisfaction score Adjust the parameters of breakpoint distance, window size, and abrupt change threshold to determine whether the preset conditions for weld seam identification and debugging are met. After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S401.
[0021] Further, in step S402,
[0022] .
[0023] Further, in step S403, when the condition is satisfied... , If the time is right, save the parameters to be debugged and continue with the subsequent steps; Otherwise, if Then reduce the breakpoint distance and / or increase the window size; if This will increase the maximum gap and / or drop.
[0024] To better understand and implement this invention, the following detailed description is provided in conjunction with the accompanying drawings. Attached Figure Description
[0025] Figure 1 This is a schematic diagram of the weld sensor parameter adjustment method according to an embodiment of the present invention; Figure 2 This is a flowchart of step S2 of the weld sensor parameter debugging method according to an embodiment of the present invention. Detailed Implementation
[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0027] In the description of this invention, it should be noted that the terms "first", "second", "third" and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0028] Embodiments Please see Figure 1This invention provides a method for adjusting weld seam sensor parameters, comprising the following steps: S1: Generate initial parameters and obtain the parameters to be debugged and the image to be identified for debugging; S2: Adjust the exposure time, binarization threshold, and filter intensity parameters in the parameters to be adjusted through image preprocessing; S3: Debug by extracting special points, and adjust the breakpoint distance, window size and mutation threshold parameters in the parameters to be debugged; S4: Through weld seam identification and debugging, adjust the maximum gap, maximum drop and angle deviation parameters in the parameters to be debugged; S5: Save parameter configuration.
[0029] Step S1 specifically includes the following steps: S101: Based on the preset recommended parameter values, set the initial parameters and use the initial parameters as the parameters to be debugged; S102: Acquire at least one initial image of the target workpiece under initial parameters as the image to be identified and debugged.
[0030] In some embodiments, recommended parameter values are preset by manually setting them in advance.
[0031] Please see Figure 2 Furthermore, step S2 includes the following steps: S201: Acquire point cloud data of the image to be identified and extract feature values of the image to be identified; In some embodiments, the feature values of the debugging image to be identified include, but are not limited to, the brightness features, strip quality features, and surface features of the debugging image to be identified; the brightness features are obtained by calculating the average brightness of the debugging image to be identified, the strip quality features are obtained by calculating the mean edge gradient of the debugging image to be identified, and the surface features are obtained by calculating the point cloud height variance of the debugging image to be identified.
[0032] S202: Select partial feature values of the image to be identified and debugged, including the average intensity of the strip edges. Main strip pixels Total strip pixels and noise area proportion A clarity score is obtained by scoring the data. Connectivity score Ratio to noise score Image preprocessing scoring ;
[0033]
[0034]
[0035]
[0036] S203: Determine the image preprocessing score If the preset conditions for preprocessing and debugging are met, save the parameters to be debugged and continue with subsequent steps; otherwise, score based on sharpness. Connectivity score Ratio to noise score Adjust the parameters of exposure time, binarization threshold, and filter intensity to determine if the preset conditions for preprocessing debugging are met. Specifically, when the following conditions are met When the time is up, save the parameters to be debugged and continue with the next steps; When not satisfied At that time, if Then increase the exposure time and / or increase the binarization threshold; if Then reduce the breakpoint distance and / or enable adaptive binarization; if If the median filter is activated, the filter intensity will be changed. Through the above adjustments, the exposure time, binarization threshold, and filter intensity parameters will reach the target state.
[0037] After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S201.
[0038] Furthermore, step S3 includes the following steps: S301: Acquire point cloud data of the debugging image to be identified and extract feature values of the debugging image to be identified; S302: Select a subset of feature values from the image to be identified, including the average gradient magnitude of feature points. Area of marked region ROI area ratio outlier ratio The significance score is obtained by scoring. Coverage score Outlier score ;
[0039]
[0040]
[0041] S303: Determine the significance score Coverage score Outlier score Does the preset conditions for special point extraction and debugging meet? If so, save the parameters to be debugged and continue with subsequent steps; otherwise, score based on saliency. Coverage score Outlier score Adjust parameters such as breakpoint distance, window size, and mutation threshold to determine if the preset conditions for special point extraction and debugging are met. Specifically, when satisfying , , If the time is right, save the parameters to be debugged and continue with the subsequent steps; Otherwise, if If so, then lower the mutation threshold; if If so, reduce the distance between the breakpoints; If the window size is increased, the window size is increased; in some embodiments, the window size is increased by increasing the minimum number of points in the window; through the above adjustments, the breakpoint distance, window size, and mutation threshold parameters are brought to the target state.
[0042] After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S301.
[0043] Further, step S4 includes the following steps: S401: Acquire point cloud data of the image to be identified and extract feature values of the image to be identified; S402: Select partial feature values of the image to be identified and debugged, including the fitting residual. Compliance Points Total points The geometric fit score is obtained by scoring. Constraint satisfaction score ;
[0044]
[0045] S403: Geometric Fit Score Constraint satisfaction score If the preset conditions for weld seam identification and debugging are met, save the parameters to be debugged and continue to the next step; otherwise, score according to geometric fitting. Constraint satisfaction score Adjust the parameters of breakpoint distance, window size, and abrupt change threshold to determine whether the preset conditions for weld seam identification and debugging are met. Specifically, when satisfying , If the time is right, save the parameters to be debugged and continue with the subsequent steps; Otherwise, if Then reduce the breakpoint distance and / or increase the window size; if If the maximum gap and / or drop are increased, the maximum gap, maximum drop, and angle deviation parameters can be adjusted to reach the target state.
[0046] After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S401.
[0047] The weld sensor parameter debugging method described in this invention establishes a real-time scoring model and performs multi-level step-by-step automatic debugging, forming a standardized debugging process. This reduces the reliance on operator experience during the debugging process and has high stability.
[0048] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and the present invention also intends to include these modifications and variations.
Claims
1. A method for adjusting parameters of a weld seam sensor, characterized in that, Specifically, the following steps are included: S1 generates initial parameters and obtains the parameters to be debugged and the image to be identified for debugging; S2 adjusts the exposure time, binarization threshold, and filter intensity parameters in the parameters to be adjusted through image preprocessing; S3 uses a specially designed point extraction and debugging method to adjust the breakpoint distance, window size, and mutation threshold parameters in the parameters to be debugged. S4 uses weld seam identification and adjustment to adjust the maximum gap, maximum drop and angle deviation parameters among the parameters to be adjusted. S5 saves the parameter configuration.
2. The method for adjusting weld sensor parameters according to claim 1, characterized in that, Step S2 specifically includes the following steps: S201 Acquire point cloud data of the debugging image to be identified and extract feature values of the debugging image to be identified; S202 Select partial feature values of the image to be identified and debugged, including the average intensity of the strip edges. Main strip pixels Total strip pixels and noise area proportion A clarity score is obtained by scoring the data. Connectivity score Ratio to noise score Compared with clarity-based scoring Connectivity score Ratio to noise score Image preprocessing score ; S203 Determine the image preprocessing score If the preset conditions for preprocessing and debugging are met, save the parameters to be debugged and continue with subsequent steps; otherwise, score based on sharpness. Connectivity score Ratio to noise score Adjust the parameters of exposure time, binarization threshold, and filter intensity to determine if the preset conditions for preprocessing debugging are met. After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S201.
3. The method for adjusting weld sensor parameters according to claim 2, characterized in that: In step S202, 。 4. The method for adjusting weld sensor parameters according to claim 3, characterized in that: In step S203, when the following conditions are met... When the time is up, save the parameters to be debugged and continue with the next steps; When not satisfied At that time, if Then increase the exposure time and / or increase the binarization threshold; if Then reduce the breakpoint distance and / or enable adaptive binarization; if If so, then median filtering will be enabled.
5. The method for adjusting weld sensor parameters according to claim 1, characterized in that, Step S3 specifically includes the following steps: S301 acquires the point cloud data of the debugging image to be identified and extracts the feature values of the debugging image to be identified; S302 Select a subset of feature values from the image to be identified, including the average gradient magnitude of feature points. Area of marked region ROI area ratio outlier ratio The significance score is obtained by scoring. Coverage score Outlier score ; S303 Significance Score Coverage score Outlier score Does the preset conditions for special point extraction and debugging meet? If so, save the parameters to be debugged and continue with subsequent steps; otherwise, score based on saliency. Coverage score Outlier score Adjust parameters such as breakpoint distance, window size, and mutation threshold to determine if the preset conditions for special point extraction and debugging are met. After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S301.
6. The method for adjusting weld sensor parameters according to claim 5, characterized in that: In step S302, 。 7. The method for adjusting weld sensor parameters according to claim 6, characterized in that: In step S303, when the condition is satisfied , , If the time is right, save the parameters to be debugged and continue with the subsequent steps; Otherwise, if If so, then lower the mutation threshold; if If so, reduce the distance between the breakpoints; If so, the window size will be increased.
8. The method for adjusting weld sensor parameters according to claim 1, characterized in that, Step S4 specifically includes the following steps: S401 acquires the point cloud data of the debugging image to be identified and extracts the feature values of the debugging image to be identified; S402 Select partial feature values of the image to be identified and debugged, including the fitting residual. Compliance Points Total points The geometric fit score is obtained by scoring. Constraint satisfaction score ; S403 Geometric Fit Score Constraint satisfaction score If the preset conditions for weld seam identification and debugging are met, save the parameters to be debugged and continue to the next step; otherwise, score according to geometric fitting. Constraint satisfaction score Adjust the parameters of breakpoint distance, window size, and abrupt change threshold to determine whether the preset conditions for weld seam identification and debugging are met. After adjustment, use the adjusted parameters as the parameters to be debugged, re-acquire the image to be identified and debugged, and return to step S401.
9. The method for adjusting weld sensor parameters according to claim 8, characterized in that: In step S402, 。 10. The method for adjusting weld sensor parameters according to claim 8, characterized in that: In step S403, when the condition is satisfied , If the time is right, save the parameters to be debugged and continue with the subsequent steps; Otherwise, if Then reduce the breakpoint distance and / or increase the window size; if This will increase the maximum gap and / or drop.