Visual inspection method and system for weld contour

Through the preprocessing and feature data calculation of weld profile image data, the problem of failure to detect weld forming coefficient, fusion ratio and concaveness in the prior art is solved, the weld detection accuracy is improved, the weld quality is ensured, and the quality of welds is improved. Overall product performance.

CN118918082APending Publication Date: 2024-11-08ANHUI HONGLU STEEL CONSTR (GROUP) CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410968341.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-18
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The prior art fails to effectively detect the forming coefficient, fusion ratio, and concaveness of the weld, resulting in low detection accuracy of the weld.

Method used

By obtaining the image data of the weld profile, pre-processing is performed to obtain feature data, calculate the weld forming coefficient and quality evaluation coefficient, and judge it with the preset range and threshold to determine whether the weld quality is qualified.

Benefits of technology

It improves the accuracy of weld inspection, ensures that the weld has good mechanical properties and low crack tendency, and improves product quality and reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918082A_ABST
    Figure CN118918082A_ABST
Patent Text Reader

Abstract

The invention discloses a visual detection method and system for a weld joint contour, relates to the technical field of weld joint contour detection, and solves the technical problem that the weld joint detection precision is not high due to single detection data since the fusion ratio of a weld joint and the concavity and surplus height of the weld joint are not considered in the prior art. The method comprises the steps of obtaining image data of a weld contour; preprocessing the image data to obtain feature data of the weld contour; calculating a weld forming coefficient based on the feature data, and judging whether the weld quality is qualified or not based on the weld forming coefficient and a preset forming coefficient range; and calculating to obtain a weld quality evaluation coefficient based on the characteristic data, and judging whether the weld quality is qualified or not based on the weld quality evaluation coefficient and a preset quality evaluation threshold, so that the technical problem that the weld detection precision is not high due to the fact that the fusion ratio of the weld and the concavity and surplus height of the weld are not detected in the prior art is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of weld profile detection, and in particular relates to a weld profile visual detection method and system. Background Art

[0002] During the welding process, the quality of the weld directly affects the strength and stability of the entire welded structure. Therefore, accurate detection of the weld profile is an important part of ensuring welding quality.

[0003] The prior art (invention patent with application number 2020104749433) discloses a weld contour visual detection method and system, including, first obtaining weld cross-sectional contour data; determining whether there is no weld in the cross-section, if there is a weld, using a horizontal straight line to divide the weld contour area into a feature point area and two weld toe point areas; calculating the feature point coordinates of the feature point area, and according to the calculated weld feature point position, the distance of the weld feature point relative to the center of the scanning device can be obtained, which is the distance of the next movement of the scanning device; using the farthest distance method to determine the two weld toe points in the two weld toe point areas, and according to the weld toe point position, the weld contour height can be obtained; repeating the above process until the tracking scan of the entire weld is completed to obtain the entire weld contour and the weld contour height; the detection method of the invention can accurately identify the characteristics of the weld, with a high accuracy recognition rate, and the identifiable weld width can reach 50 mm. The generated weld contour three-dimensional graphics can truly reflect the morphology of the weld. A fusion ratio that is too low may result in insufficient weld strength, while a fusion ratio that is too high may introduce too much weld metal, affecting the performance of the welded structure; the concavity and reinforcement of the weld directly affect the stress distribution and fatigue life of the weld, and excessive concavity may cause stress concentration, while excessive reinforcement may lead to accelerated corrosion in the weld area; however, the prior art does not take into account the detection of the fusion ratio, concavity and reinforcement of the weld, and since the detection data is single, the detection accuracy of the weld is not high.

[0004] Therefore, the present invention solves the above-mentioned problem by proposing a weld profile visual detection method and system. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a weld profile visual detection method and system, which are used to solve the technical problem that the prior art does not detect the weld forming coefficient, weld fusion ratio, weld concavity and weld excess height, and the detection data is single, resulting in low weld detection accuracy.

[0006] To achieve the above object, a first aspect of the present invention provides a weld profile visual detection method, comprising:

[0007] Step 1: Obtain image data of weld profile;

[0008] Step 2: Obtain characteristic data of the weld profile by preprocessing the image data;

[0009] Step 3: Calculate the weld forming coefficient based on the characteristic data; judge whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range; if yes, proceed to step 4; if no, send the unqualified information to the user end; wherein the client end includes: a mobile phone or a computer;

[0010] Step 4: Calculate the weld quality assessment coefficient based on the characteristic data, and judge whether the weld quality is qualified based on the weld quality assessment coefficient and the preset quality assessment threshold; if yes, the inspection passes; if not, the unqualified information is sent to the user end.

[0011] Preferably, the step of acquiring image data of the weld profile includes:

[0012] The weld contour image between the parent materials is collected by an image collection device to obtain image data; wherein the image collection device includes: a camera or a video camera.

[0013] Preferably, the preprocessing of the image data to obtain characteristic data of the weld profile includes:

[0014] The image data is converted into a black-and-white binary image by a binarization method; wherein the white in the black-and-white binary image represents the weld area and the black represents the base material area; the binarization method includes: a global threshold or an adaptive threshold;

[0015] The boundary of the weld is identified by an edge detection algorithm; wherein the edge detection algorithm includes: a Canny algorithm or a Sobel algorithm;

[0016] The characteristic data of the weld contour are calculated based on the black-and-white binary image and the weld boundary; wherein the characteristic data include: the calculated thickness of the weld, the weld width, the area of ​​the parent material in the weld, the weld area, the concavity and the excess height.

[0017] It should be noted that the calculated thickness of the weld refers to the expected weld thickness under ideal conditions based on given welding parameters (such as current, voltage, wire diameter, welding speed, etc.) and material properties (such as melting coefficient, thermal conductivity, etc.).

[0018] Preferably, the characteristic data of the weld contour is obtained by calculating based on the black-and-white binary image and the weld boundary, including:

[0019] Connect the opposite weld toes of the weld boundary and obtain the concavity by obtaining the maximum distance between the connecting line and the weld surface;

[0020] By obtaining the distance between the relative weld toe lines, the weld width is obtained;

[0021] The excess height is obtained by obtaining the vertical distance from the weld vertex to the weld toe line;

[0022] The weld area is obtained by obtaining the number of pixels of the entire weld area on the weld cross section in the black-and-white binary image and converting it into the actual area according to the pixel size; the area of ​​the parent material in the weld is obtained by obtaining the number of pixels of the parent material in the weld on the weld cross section and converting it into the actual area according to the pixel size.

[0023] Preferably, the calculated thickness of the weld includes:

[0024] Calculated thickness of butt welds and calculated thickness of fillet welds;

[0025] Calculated thickness of butt weld: The calculated thickness of the weld is obtained by obtaining the distance from the weld root to the line connecting the two weld toes;

[0026] Calculated thickness of fillet weld: The calculated thickness of the weld is obtained by obtaining the distance from the intersection of the inner surfaces of the two parts in the weld cross section to the line connecting the two weld toes.

[0027] Preferably, the calculation of the weld formation coefficient based on the characteristic data includes:

[0028] The calculated thickness of the weld is marked as H, and the weld width is marked as K;

[0029] The weld formation coefficient is calculated by the weld formation coefficient calculation formula: φ=K / H; wherein φ is the weld formation coefficient.

[0030] Preferably, judging whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range includes:

[0031] Determine whether the weld forming coefficient exceeds the preset forming coefficient range; if yes, the weld quality is judged to be unqualified; if no, the weld quality is judged to be qualified; wherein, the preset forming coefficient range is set according to actual welding requirements and application scenarios.

[0032] It should be noted that the weld forming coefficient is too small, the weld will be narrow and deep, this weld may cause serious segregation in the central area of ​​the weld, increased impurity accumulation, decreased thermal crack resistance and other problems; a moderate weld forming coefficient can ensure that the weld has sufficient width and appropriate depth, thereby providing sufficient load-bearing capacity and good welding quality; if the weld forming coefficient is too large, the weld will be wide and shallow, although this weld may have better crack resistance, but a weld that is too wide will increase the consumption of welding materials and welding costs. At the same time, if the weld is too wide and the depth is insufficient, it may also affect the load-bearing capacity and sealing performance of the weld.

[0033] The present invention obtains the weld forming coefficient by calculation based on characteristic data, and judges whether the weld quality is qualified based on the weld forming coefficient and a preset forming coefficient range; it ensures that the weld has good mechanical properties and a low crack tendency, thereby improving the overall quality and reliability of the product, and solves the problem that the inconsistency of the weld forming coefficient directly affects the consistency of product quality.

[0034] Preferably, the weld quality assessment coefficient is calculated based on the characteristic data, including:

[0035] The area of ​​the parent material in the weld is marked as Sn, the weld area is marked as S, the concavity is marked as D and the reinforcement is marked as Y;

[0036] The fusion ratio R of the weld is calculated by the fusion ratio calculation formula: R = Sn / S;

[0037] The weld quality assessment coefficient is calculated using the formula: P = A×e^|R-ZR|+B×e^[(D-ZD)^2 / (2σ^2)]+C×(Y-ZY)^2; where A, B, and C are proportional coefficients, σ ​​is the standard deviation, ZR is the optimum fusion ratio, ZD is the optimum concavity, and ZY is the optimum excess height.

[0038] Preferably, the step of setting a threshold based on the weld quality assessment coefficient and the pre-quality assessment to determine whether the weld quality is qualified includes:

[0039] Determine whether the weld quality assessment coefficient is greater than the preset quality assessment threshold; if yes, the weld quality is judged to be qualified; if not, the weld quality is judged to be unqualified; wherein, the preset quality assessment threshold is set according to actual welding requirements and application scenarios.

[0040] The present invention obtains the weld quality assessment coefficient by calculating the characteristic data, and judges whether the weld quality is qualified based on the weld quality assessment coefficient and the preset quality assessment threshold; it solves the problem that the prior art does not detect the fusion ratio of the weld and the concavity and excess height of the weld, resulting in low weld detection accuracy.

[0041] A second aspect of the present invention provides a weld profile visual inspection system, comprising: a data analysis module, and a data acquisition module and a weld inspection module connected to the data analysis module;

[0042] Data acquisition module: used to obtain image data of weld profile;

[0043] Data analysis module: used to pre-process image data and obtain characteristic data of weld profile;

[0044] Weld detection module: calculate the weld forming coefficient based on the characteristic data; judge whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range; if yes, proceed to step 4; if no, send the unqualified information to the user end; and,

[0045] The weld quality assessment coefficient is calculated based on the characteristic data, and whether the weld quality is qualified is judged based on the weld quality assessment coefficient and the preset quality assessment threshold; if yes, the detection passes; if not, the unqualified information is sent to the user end.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The prior art does not consider the detection of the fusion ratio, concavity and excess height of the weld, and the detection accuracy of the weld is not high due to the single detection data; the present invention obtains image data of the weld contour; obtains feature data of the weld contour by preprocessing the image data; calculates the weld quality assessment coefficient based on the feature data, and judges whether the weld quality is qualified based on the weld quality assessment coefficient and a preset quality assessment threshold; solves the problem that the prior art does not detect the fusion ratio, concavity and excess height of the weld, resulting in low weld detection accuracy.

[0048] 2. The present invention obtains the weld forming coefficient by calculation based on the characteristic data, and judges whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range; it ensures that the weld has good mechanical properties and low crack tendency, and solves the problem that the prior art does not detect the weld forming coefficient and the detection accuracy of the weld is low due to the single detection data. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A schematic diagram of the method steps of an embodiment of the present invention;

[0051] Figure 2 A schematic diagram of a system module according to an embodiment of the present invention;

[0052] Figure 3 The figure is a schematic diagram of a specific process of an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] See also Figure 1 The first aspect of the present invention provides a method for visually detecting a weld profile, comprising:

[0055] Step 1: Obtain image data of weld profile;

[0056] Step 2: Obtain characteristic data of the weld profile by preprocessing the image data;

[0057] Step 3: Calculate the weld forming coefficient based on the characteristic data; judge whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range; if yes, proceed to step 4; if no, send the unqualified information to the user end; wherein the client end includes: a mobile phone or a computer;

[0058] Acquire image data of weld profile, including:

[0059] The weld contour image between the parent materials is collected by an image collection device to obtain image data; wherein the image collection device includes: a camera or a video camera.

[0060] By preprocessing the image data, the characteristic data of the weld profile is obtained, including:

[0061] The image data is converted into a black-and-white binary image by a binarization method; wherein the white in the black-and-white binary image represents the weld area and the black represents the base material area; the binarization method includes: a global threshold or an adaptive threshold;

[0062] The boundary of the weld is identified by an edge detection algorithm; wherein the edge detection algorithm includes: a Canny algorithm or a Sobel algorithm;

[0063] The characteristic data of the weld contour are calculated based on the black-and-white binary image and the weld boundary; wherein the characteristic data include: the calculated thickness of the weld, the weld width, the area of ​​the parent material in the weld, the weld area, the concavity and the excess height.

[0064] The characteristic data of the weld contour is obtained based on the black and white binary image and the weld boundary calculation, including:

[0065] Connect the opposite weld toes of the weld boundary and obtain the concavity by obtaining the maximum distance between the connecting line and the weld surface;

[0066] By obtaining the distance between the relative weld toe lines, the weld width is obtained;

[0067] The excess height is obtained by obtaining the vertical distance from the weld vertex to the weld toe line;

[0068] The weld area is obtained by obtaining the number of pixels of the entire weld area on the weld cross section in the black-and-white binary image and converting it into the actual area according to the pixel size; the area of ​​the parent material in the weld is obtained by obtaining the number of pixels of the parent material in the weld on the weld cross section and converting it into the actual area according to the pixel size.

[0069] Calculated thickness of welds, including:

[0070] Calculated thickness of butt welds and calculated thickness of fillet welds;

[0071] Calculated thickness of butt weld: The calculated thickness of the weld is obtained by obtaining the distance from the weld root to the line connecting the two weld toes;

[0072] Calculated thickness of fillet weld: The calculated thickness of the weld is obtained by obtaining the distance from the intersection of the inner surfaces of the two parts in the weld cross section to the line connecting the two weld toes.

[0073] The weld formation factor is calculated based on the characteristic data, including:

[0074] The calculated thickness of the weld is marked as H, and the weld width is marked as K;

[0075] The weld formation coefficient is calculated by the weld formation coefficient calculation formula: φ=K / H; wherein φ is the weld formation coefficient.

[0076] The quality of the weld is judged based on the weld forming coefficient and the preset forming coefficient range, including:

[0077] Determine whether the weld forming coefficient exceeds the preset forming coefficient range; if yes, the weld quality is judged to be unqualified; if no, the weld quality is judged to be qualified; wherein, the preset forming coefficient range is set according to actual welding requirements and application scenarios.

[0078] For example, suppose that a shipbuilding factory is conducting a quality inspection on a ship hull butt weld in a welding workshop. The image acquisition device used is a high-precision industrial camera that can capture clear images of the weld contour. The specific steps are as follows:

[0079] Step 1: Obtain image data of weld profile;

[0080] Image acquisition: Use an industrial camera to capture images of the base material weld from several directions to ensure that the image is clear and contains the complete weld contour.

[0081] Step 2: Image data preprocessing;

[0082] Binarization processing:

[0083] The color image is converted into a black and white binary image using the global threshold method. A suitable threshold (such as 128) is set so that the weld area (higher brightness) appears white and the base material area (lower brightness) appears black.

[0084] Edge Detection:

[0085] Use the Canny edge detection algorithm to identify the weld boundary. Adjust the low and high thresholds of the Canny algorithm to ensure that the weld boundary is accurately identified.

[0086] Step 3: Calculate the characteristic data of the weld profile;

[0087] Calculated thickness (H): For butt welds, measure the distance from the weld root to the line connecting the two weld toes to obtain the calculated thickness (H) of the weld. Assume H = 3mm.

[0088] Weld width (K): Measure the distance from the weld toe line to get the weld width (K). Assume K = 10mm.

[0089] Concavity: Draw a line between the relative weld toes of the weld boundary and measure the maximum distance between the line and the weld surface to obtain the concavity (D). Assume D = 2 mm.

[0090] Reinforcement: Measure the vertical distance from the weld vertex to the weld toe line to get the reinforcement (Y). Assume Y = 4mm.

[0091] Area calculation: Count the number of pixels in the weld area on the cross section of the weld in the black and white binary image, convert the counted number of pixels into actual area units, and obtain the weld area (S) and the area of ​​the parent material in the weld (Sn). Assume S = 500mm 2 , Sn=300mm 2 .

[0092] Step 4: Calculate the weld formation coefficient;

[0093] Calculate the weld formation coefficient by the weld formation coefficient calculation formula: φ = K / H;

[0094] Substitute the known data: φ=10mm / 3mm≈3.33;

[0095] Step 5: Determine the quality of the weld;

[0096] Assume that the preset forming coefficient range is (2.0, 4.0); since 2.0<φ≈3.33<4.0, the weld quality is judged to be qualified within the preset range.

[0097] If the weld quality is qualified, the system can automatically record the data and continue to inspect the next weld.

[0098] If the weld quality is unqualified (for example, the forming coefficient exceeds the preset range), the system will send the unqualified information to the user end (such as a mobile phone or computer) to notify relevant personnel to handle it.

[0099] The present invention monitors the weld forming factor to ensure that the weld has good mechanical properties and low crack tendency, thereby improving the overall quality and reliability of the product. This process not only improves the detection efficiency, but also ensures the consistency of product quality, which is of great significance for improving the competitiveness of the manufacturing industry.

[0100] Step 4: Calculate the weld quality assessment coefficient based on the characteristic data, and judge whether the weld quality is qualified based on the weld quality assessment coefficient and the preset quality assessment threshold; if yes, the inspection passes; if not, the unqualified information is sent to the user end.

[0101] The weld quality assessment coefficient is calculated based on the characteristic data, including:

[0102] The area of ​​the parent material in the weld is marked as Sn, the weld area is marked as S, the concavity is marked as D and the reinforcement is marked as Y;

[0103] The fusion ratio R of the weld is calculated by the fusion ratio calculation formula: R = Sn / S;

[0104] The weld quality assessment coefficient is calculated using the formula: P = A×e^|R-ZR|+B×e^[(D-ZD)^2 / (2σ^2)]+C×(Y-ZY)^2; where A, B, and C are proportional coefficients, σ ​​is the standard deviation, ZR is the optimum fusion ratio, ZD is the optimum concavity, and ZY is the optimum excess height.

[0105] Based on the weld quality assessment coefficient and the pre-quality assessment threshold, the weld quality is judged to be qualified, including:

[0106] Determine whether the weld quality assessment coefficient is greater than the preset quality assessment threshold; if yes, the weld quality is judged to be qualified; if not, the weld quality is judged to be unqualified; wherein, the preset quality assessment threshold is set according to actual welding requirements and application scenarios.

[0107] For example, suppose that in a welding workshop of an automobile manufacturer, the quality of a critical weld of a car chassis is evaluated. The specific process is as follows:

[0108] 1. Weld characteristic data;

[0109] Assume that the following weld feature data is obtained from image processing:

[0110] Area of ​​parent material in weld (Sn): 300mm 2 ;

[0111] Welding area (S): 500mm 2 ;

[0112] Concavity (D): 2mm;

[0113] Residual height (Y): 4mm;

[0114] 2. Preset parameters;

[0115] Optimum fusion ratio (ZR): 0.6;

[0116] Optimum concavity (ZD): 1mm;

[0117] Optimum residual height (ZY): 3mm;

[0118] Standard deviation (σ): 0.5;

[0119] Proportional coefficients (A, B, C): assumed to be 0.1, 0.5, 0.3 respectively;

[0120] 3. Calculate the fusion ratio (R) of the weld;

[0121] According to the fusion ratio calculation formula: R = Sn / S, substituting the characteristic data into the result: R = 300 / 500 = 0.6;

[0122] 4. Calculate the weld quality assessment coefficient;

[0123] The weld quality assessment coefficient is calculated by the formula: P = A × e^|R-ZR|+B × e^[(D-ZD)^2 / (2σ^2)]+C × (Y-ZY)^2;

[0124] P=0.1×e^|0.6-0.6|+0.5×e^[(2-1)^2 / (2×0.5^2)]+0.3×(4-3)^2;

[0125] We get P≈4.1.

[0126] 5. Determine the quality of the weld;

[0127] Assume that the preset quality assessment threshold is set to 5. According to the comparison between the weld quality assessment coefficient and the preset threshold:

[0128] Since the weld quality assessment coefficient (4.1) is less than the preset quality assessment threshold (5);

[0129] Therefore, the weld quality was judged to be unqualified.

[0130] The system sends the unqualified information to the user end (such as a mobile phone or computer), notifying relevant personnel that the weld needs further inspection or repair.

[0131] At the same time, the system can record detailed test data and evaluation results for subsequent analysis and improvement.

[0132] In this specific example, although the fusion ratio of the weld reaches the optimum fusion ratio, the concavity and the residual height deviate from the optimum value, resulting in the weld quality evaluation coefficient not reaching the preset threshold, thereby determining that the weld quality is unqualified. The present invention comprehensively considers multiple weld characteristic parameters, quantifies the weld quality through a mathematical model, and provides a scientific basis for the optimization and quality control of the welding process.

[0133] See also Figure 2-Figure 3 , a second aspect of the present invention provides a weld profile visual inspection system, comprising: a data analysis module, and a data acquisition module and a weld inspection module connected to the data analysis module;

[0134] Data acquisition module: used to obtain image data of weld profile;

[0135] Data analysis module: used to pre-process image data and obtain characteristic data of weld profile;

[0136] Weld detection module: calculate the weld forming coefficient based on the characteristic data; judge whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range; if yes, proceed to step 4; if no, send the unqualified information to the user end; and,

[0137] The weld quality assessment coefficient is calculated based on the characteristic data, and whether the weld quality is qualified is judged based on the weld quality assessment coefficient and the preset quality assessment threshold; if yes, the detection passes; if not, the unqualified information is sent to the user end.

[0138] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0139] Working principle of the present invention:

[0140] The present invention obtains image data of the weld contour; pre-processes the image data to obtain characteristic data of the weld contour; calculates a weld forming coefficient based on the characteristic data; judges whether the weld quality is qualified based on the weld forming coefficient and a preset forming coefficient range; if yes, proceeds to step four; if no, sends unqualified information to a user end; and, calculates a weld quality assessment coefficient based on the characteristic data, and judges whether the weld quality is qualified based on the weld quality assessment coefficient and a preset quality assessment threshold; if yes, the detection passes; if no, sends unqualified information to the user end; the technical problem that the prior art does not detect the fusion ratio of the weld and the concavity and excess height of the weld, resulting in low weld detection accuracy is solved.

[0141] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A weld profile visual detection method, characterized in that: include: Step 1: Obtain image data of weld profile; Step 2: Obtain characteristic data of the weld profile by preprocessing the image data; Step 3: Calculate the weld forming coefficient based on the characteristic data; determine whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range; If yes, proceed to step 4; If not, the unqualified information is sent to the user end; wherein the client end includes: a mobile phone or a computer; Step 4: Calculate the weld quality assessment coefficient based on the characteristic data, and judge whether the weld quality is qualified based on the weld quality assessment coefficient and the preset quality assessment threshold; if yes, the inspection passes; if not, the unqualified information is sent to the user end.

2. A weld profile visual detection method according to claim 1, characterized in that: The step of obtaining image data of the weld contour comprises: The weld contour image between the parent materials is collected by an image collection device to obtain image data; wherein the image collection device includes: a camera or a video camera.

3. A weld profile visual detection method according to claim 1, characterized in that: The method of obtaining characteristic data of the weld profile by preprocessing the image data includes: The image data is converted into a black-and-white binary image by a binarization method; wherein the white in the black-and-white binary image represents the weld area and the black represents the base material area; the binarization method includes: a global threshold or an adaptive threshold; The boundary of the weld is identified by an edge detection algorithm; wherein the edge detection algorithm includes: a Canny algorithm or a Sobel algorithm; The characteristic data of the weld contour are calculated based on the black-and-white binary image and the weld boundary; wherein the characteristic data include: the calculated thickness of the weld, the weld width, the area of ​​the parent material in the weld, the weld area, the concavity and the excess height.

4. A weld profile visual detection method according to claim 3, characterized in that: The characteristic data of the weld contour is obtained by calculating the black-and-white binary image and the weld boundary, including: Connect the opposite weld toes of the weld boundary and obtain the concavity by obtaining the maximum distance between the connecting line and the weld surface; By obtaining the distance between the relative weld toe lines, the weld width is obtained; The excess height is obtained by obtaining the vertical distance from the weld vertex to the weld toe line; The weld area is obtained by obtaining the number of pixels of the entire weld area on the weld cross section in the black-and-white binary image and converting it into the actual area according to the pixel size; the area of ​​the parent material in the weld is obtained by obtaining the number of pixels of the parent material in the weld on the weld cross section and converting it into the actual area according to the pixel size.

5. A weld profile visual detection method according to claim 3, characterized in that: The calculated thickness of the weld includes: Calculated thickness of butt welds and calculated thickness of fillet welds; Calculated thickness of butt weld: The calculated thickness of the weld is obtained by obtaining the distance from the weld root to the line connecting the two weld toes; Calculated thickness of fillet weld: The calculated thickness of the weld is obtained by obtaining the distance from the intersection of the inner surfaces of the two parts in the weld cross section to the line connecting the two weld toes.

6. A weld profile visual detection method according to claim 1, characterized in that: The method of calculating the weld formation coefficient based on the characteristic data includes: The calculated thickness of the weld is marked as H, and the weld width is marked as K; The weld formation coefficient is calculated by the weld formation coefficient calculation formula: φ=K / H; wherein φ is the weld formation coefficient.

7. A weld profile visual detection method according to claim 1, characterized in that: The method of judging whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range includes: Determine whether the weld forming coefficient exceeds the preset forming coefficient range; if yes, the weld quality is judged to be unqualified; if no, the weld quality is judged to be qualified; wherein, the preset forming coefficient range is set according to actual welding requirements and application scenarios.

8. A weld profile visual detection method according to claim 1, characterized in that: The weld quality assessment coefficient is calculated based on the characteristic data, including: The area of ​​the parent material in the weld is marked as Sn, the weld area is marked as S, the concavity is marked as D and the reinforcement is marked as Y; The fusion ratio R of the weld is calculated by the fusion ratio calculation formula: R = Sn / S; The weld quality assessment coefficient is calculated using the formula: P = A×e^|R-ZR|+B×e^[(D-ZD)^2 / (2σ^2)]+C×(Y-ZY)^2; where A, B, and C are proportional coefficients, σ ​​is the standard deviation, ZR is the optimum fusion ratio, ZD is the optimum concavity, and ZY is the optimum excess height.

9. A weld profile visual detection method according to claim 1, characterized in that: The method of determining whether the weld quality is qualified by setting a threshold based on the weld quality assessment coefficient and the pre-quality assessment includes: Determine whether the weld quality assessment coefficient is greater than the preset quality assessment threshold; if yes, the weld quality is judged to be qualified; if not, the weld quality is judged to be unqualified; wherein, the preset quality assessment threshold is set according to actual welding requirements and application scenarios.

10. A weld profile visual inspection system, used to execute a weld profile visual inspection method according to any one of claims 1 to 8, characterized in that: include: A data analysis module, and a data acquisition module and a weld detection module connected to the data analysis module; Data acquisition module: used to obtain image data of weld profile; Data analysis module: used to pre-process image data and obtain characteristic data of weld profile; Weld detection module: calculate the weld forming coefficient based on the characteristic data; judge whether the weld quality is qualified based on the weld forming coefficient and the preset forming coefficient range; if yes, proceed to step 4; If not, the unqualified information will be sent to the user end; as well as, The weld quality assessment coefficient is calculated based on the characteristic data, and whether the weld quality is qualified is judged based on the weld quality assessment coefficient and the preset quality assessment threshold; if yes, the test is passed; If not, the unqualified information will be sent to the user end.

Citation Information

Cited By

  • Detection method of pre-encapsulated nest box

    CN119845981A

  • Robot vision flexible detection system and method

    CN121235977A