Computer-aided metal wire welding quality detection method and system

Abstract

The present application relates to the technical field of metal wire welding quality detection, and specifically to a computer-aided metal wire welding quality detection method and system, the method comprising: obtaining metal wire welding images and infrared thermal imaging images at each acquisition moment during the metal wire welding process; obtaining each welding area in each metal wire welding image; calculating the cross angle of each welding area; and then calculating the welding shape abnormal value of each welding area in each metal wire welding image; calculating the temperature abnormality coefficient of any metal wire where each welding area is located in each metal wire welding image; and then obtaining the welding temperature abnormality factor and the cooling temperature abnormality factor; obtaining the welding thermal effect abnormality parameters of each welding area in each metal wire welding image; and then obtaining the welding abnormality score of each welding area in each metal wire welding image, and detecting the metal wire welding quality. The present application improves the accuracy of metal wire welding quality detection.

Description

Technical Field

[0001] The present application relates to the technical field of metal wire welding quality detection, and in particular to a computer-aided metal wire welding quality detection method and system. Background Art

[0002] In the welding process, the position, shape and accuracy of the welding node directly affect the mechanical properties and safety of the structural parts. Resistance welding is a process that generates heat at the metal contact point through a combination of current and pressure to achieve welding. It has the advantages of fast welding speed, low energy consumption and high weld strength.

[0003] The welding quality detection methods of fully automatic resistance welding machines in mesh processing have covered a variety of means, from traditional current and voltage monitoring to modern visual and ultrasonic detection. Different methods have their own advantages in improving welding quality and detection accuracy. Using high-power fully automatic resistance welding machines for mesh processing can improve production efficiency and control the consistency of processing. The welds are full and uniform, and the welds are very strong. The fully automatic system reduces dependence on manual labor, which not only saves labor costs, but also improves the continuous production capacity of the production line. Automatic detection and quality analysis of welding nodes through computer-aided technology can improve the monitoring accuracy and reliability of the welding process.

[0004] Currently, welding quality assessment mainly uses visual inspection systems, which use high-resolution cameras or industrial cameras with image processing technology to perform real-time inspection of welding points. Due to the numerous and complex defects in the welding area, such as cracks, pores, and numerous welding nodes, it is difficult to directly use traditional visual inspection to detect and comprehensively evaluate the welding quality. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of this application is to provide a computer-aided metal wire welding quality detection method and system, and the technical solutions adopted are as follows:

[0006] In a first aspect, an embodiment of the present application provides a computer-aided metal wire welding quality detection method, the method comprising the following steps:

[0007] Acquire metal wire welding images and infrared thermal imaging images at each acquisition moment during the metal wire welding process;

[0008] Based on the distance between each corner point in the metal wire welding image, each welding area in the metal wire welding image is obtained; and the intersection angle of each welding area is obtained by using the Hough straight line algorithm;

[0009] Based on the number of pixels in the welding area, the number of pixels at the edge of the welding area and the value of the intersection angle of the welding area, the welding shape abnormal value of each welding area in each metal wire welding image is obtained;

[0010] Based on the average temperature information of the pixel points in the infrared thermal imaging image, the temperature anomaly coefficient of any metal wire in each welding area in each metal wire welding image is obtained;

[0011] Obtaining a welding temperature anomaly factor for each welding area in each metal wire welding image based on a temperature anomaly coefficient;

[0012] Based on the average of the differences between the temperature information of the pixel points, the cooling temperature abnormality factor of each welding area in each metal wire welding image is obtained;

[0013] Acquire welding thermal effect abnormal parameters of each welding area in each metal wire welding image based on welding temperature abnormal factors and cooling temperature abnormal factors;

[0014] The welding anomaly score of each welding area in each metal wire welding image is obtained based on the welding shape anomaly value and the welding thermal effect anomaly parameter. The welding area with a welding anomaly score greater than a preset welding score threshold is regarded as a defective welding area to detect the metal wire welding quality.

[0015] Further, the method for obtaining the welding area is:

[0016] For the wire welding image at each acquisition moment, a corner point detection algorithm is used to obtain all corner points in the wire welding image. For each corner point, the three corner points closest to each corner point are obtained, and the quadrilateral area formed by the lines between each corner point and the three corner points closest to it is taken as each welding area.

[0017] Furthermore, the method for obtaining the intersection angle is:

[0018] For the metal wire welding image at each acquisition moment, a skeleton extraction algorithm is used to obtain each skeleton in the metal wire welding image, and the Hough straight line algorithm is used to process all the skeletons in the metal wire welding image to obtain the skeleton straight lines of all the skeletons, and the welding area where the intersection of any two intersecting skeleton straight lines is located is obtained as the welding area of ​​each intersection;

[0019] For the intersection point of any two intersecting skeleton straight lines, the minimum value of all angles formed by the intersection of the two skeleton straight lines is taken as the intersection angle of each welding area where the intersection point is located.

[0020] Furthermore, the calculation formula of the welding shape abnormal value is: ; In the formula, represents the welding shape abnormal value of the i-th welding area in each wire welding image; represents the number of pixels in the i-th welding area in each wire welding image, Represents the number of pixels at the edge of the i-th welding area, represents the intersection angle of the i-th welding area, and c is the preset parameter adjustment factor; is the normalization function.

[0021] Furthermore, the method for obtaining the temperature anomaly coefficient is:

[0022] Use image feature matching algorithm to match the temperature information of each pixel in the infrared thermal imaging image to each pixel in the wire welding image;

[0023] For each welding area in the wire welding image at each acquisition moment, the mean value of the temperature information of all pixels in the welding area is calculated as the average temperature information of each welding area;

[0024] The calculation formula of the temperature anomaly coefficient is: ;in, represents the temperature anomaly coefficient of any metal wire in the ith welding area in each metal wire welding image; It represents the average temperature information of the vth welding area on any metal wire where the i-th welding area is located in each metal wire welding image, except the i-th welding area. represents the mean value of the average temperature information of all welding areas on any metal wire where the i-th welding area is located in each metal wire welding image; represents the number of welding areas on any metal wire where the i-th welding machine area is located in each metal wire welding image; Represents the absolute value of the difference between the average temperature information of the i-th welding area and the preset expected temperature.

[0025] Furthermore, the method for obtaining the abnormal welding temperature factor is:

[0026] For each welding area in the wire welding image at each acquisition moment, the product of the temperature anomaly coefficients of all the wires in the welding area is calculated as the welding temperature anomaly factor of each welding area.

[0027] Furthermore, the method for obtaining the cooling temperature abnormality factor is:

[0028] The pixel point at the intersection of any two intersecting skeleton lines in the metal wire welding image at each acquisition moment is taken as the initial pixel point, and along the directions on both sides of any skeleton line where the initial pixel point is located, starting from the initial pixel point, the absolute values ​​of the differences between the temperature information of a preset number of consecutive adjacent pixel points are calculated in sequence as the first absolute value of the difference, and the average of all the first absolute values ​​of the difference in the directions on both sides of any skeleton line where the initial pixel point is located is calculated as the temperature change rate difference of the welding area where the initial pixel point is located in the directions on both sides of any skeleton line;

[0029] For each welding area in the wire welding image at each acquisition moment, the absolute value of the difference between the temperature change rate difference of the welding area in the direction of both sides of any skeleton straight line and the temperature change rate difference in the direction of both sides of another skeleton straight line is calculated as the cooling temperature anomaly factor of each welding area in each wire welding image.

[0030] Furthermore, the method for obtaining the abnormal parameters of welding thermal effect is:

[0031] For each welding area in each metal wire welding image, the sum of the welding temperature abnormality factor and the cooling temperature abnormality factor is calculated as the welding thermal effect abnormality parameter of each welding area in each metal wire welding image.

[0032] Furthermore, the calculation formula of the welding abnormality score is: ; In the formula, Score the welding anomaly of the i-th welding area in each wire welding image; is the normalization function, is the welding shape anomaly value of the i-th welding area in each wire welding image, is the abnormal parameter of welding thermal effect of the i-th welding area in each metal wire welding image.

[0033] In a second aspect, an embodiment of the present application also provides a computer-aided metal wire welding quality detection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0034] This application has at least the following beneficial effects:

[0035] The present application uses a dual-sensor camera to simultaneously acquire metal wire welding images and infrared thermal imaging images for analysis based on the traditional visual inspection system. Based on the division of welding areas in the metal wire welding image, the welding shape anomaly value is calculated to characterize the shape characteristics of each welding area. In order to reflect the degree of uneven temperature on each metal wire where each welding area is located, a temperature anomaly coefficient is constructed, and then a welding temperature anomaly factor and a cooling temperature anomaly factor are constructed to reflect the abnormal welding temperature during the welding process of the welding area at each acquisition moment and the abnormal temperature cooling after welding. The welding thermal effect anomaly parameters are obtained according to the welding temperature anomaly factor and the cooling temperature anomaly factor, and the welding anomaly score is constructed in combination with the welding shape anomaly value to detect the metal wire welding quality, thereby avoiding the problem of detection errors caused by single variable analysis and improving the accuracy of metal wire welding quality detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 A flowchart of a computer-aided wire welding quality detection method according to an embodiment of the present application;

[0038] Figure 2 A schematic diagram of a welding area provided for one embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the computer-aided wire welding quality detection method and system proposed in the present application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0041] The specific scheme of the computer-aided metal wire welding quality detection method and system provided in the present application is described in detail below with reference to the accompanying drawings.

[0042] See also Figure 1 , which shows a flowchart of a method for computer-aided metal wire welding quality detection provided by an embodiment of the present application, the method comprising the following steps:

[0043] Step S1, acquiring a metal wire welding image and an infrared thermal imaging image at each acquisition moment during a metal wire welding process.

[0044] A dual-sensor camera is used, including an RGB camera and an infrared thermal imager, where the RGB camera is used to capture visible light images and the infrared thermal imager is used to capture infrared thermal imaging images. After the equipment is calibrated, the dual-sensor camera is installed directly above the welding area to ensure that the instrument field of view can cover the entire welding area. After welding begins, the dual-sensor camera starts working at the same time with the welding frequency of the welding joint as the image acquisition frequency to obtain the metal wire welding image and infrared thermal imaging image of the welding area at each acquisition time. Among them, the metal wire welding image is an RGB image.

[0045] Wavelet transform is used to remove noise from the collected wire welding image and infrared thermal imaging image, so as to maintain the edge and detail information of the image. Wavelet transform is a well-known technology and will not be described in detail in this embodiment.

[0046] Step S2, based on the distance between each corner point in the metal wire welding image, obtain each welding area in the metal wire welding image; use the Hough line algorithm to obtain the intersection angle of each welding area; based on the number of pixels in the welding area, the number of edge pixels in the welding area and the value of the intersection angle of the welding area, obtain the welding shape abnormal value of each welding area in each metal wire welding image.

[0047] In the welding quality inspection process of welding process, heat input not only affects the formation of welding shape, but also has an important impact on the problems of thermal cracking, deformation and residual stress of welding joints. Furthermore, the geometric shape of welding joints will also affect the distribution of thermal effects and heat conduction during welding.

[0048] There are many metal wires in the metal mesh, and the welding area formed is complex. The shape and positioning of the welding area are crucial to evaluating whether the welding is qualified, and positioning analysis of the welding area is required.

[0049] The welding area is the intersection area of ​​multiple welds. Specifically, the metal transverse wires and longitudinal wires in the metal mesh are distributed vertically, where the metal transverse wires and metal transverse wires are collectively referred to as metal wires. However, there may be certain deviations in the arrangement of the metal wires during actual operation, resulting in changes in the crossing angles, which directly affects the stress intensity of the weld. Moreover, since the spot welding process is a preset program, the geometric shape of the welding result in the welding area changes, affecting the quality of the weld.

[0050] According to the above analysis, in order to obtain the welding area where two metal wires intersect, for the metal wire welding image at each acquisition time, a corner point detection algorithm is used to obtain all corner points in the metal wire welding image. For each corner point, the three corner points closest to each corner point are obtained, and the quadrilateral area formed by the lines between each corner point and the three corner points closest to it is taken as each welding area. The schematic diagram of the welding area is shown in the figure below. Figure 2 shown.

[0051] For the wire welding image at each acquisition moment, a skeleton extraction algorithm is used to obtain each skeleton in the wire welding image, and the Hough line algorithm is used to process all the skeletons in the wire welding image to obtain the skeleton straight lines of all skeletons, and the welding area where the intersection of any two intersecting skeleton straight lines is located is obtained as the welding area of ​​each intersection.

[0052] Furthermore, for the intersection of any two intersecting skeleton lines, the minimum value of all angles formed by the intersection of the two skeleton lines is used as the intersection angle of each welding area where the intersection is located. The skeleton extraction algorithm and the Hough line algorithm are well-known technologies and are not described in detail in this embodiment.

[0053] Furthermore, in order to characterize the shape characteristics of each welding area, based on the number of pixels in the welding area, the number of pixels at the edge of the welding area and the value of the intersection angle of the welding area, the welding shape abnormal value of each welding area in the wire welding image at each acquisition time is calculated. The calculation formula is: ; In the formula, represents the welding shape abnormal value of the i-th welding area in each wire welding image; represents the number of pixels in the i-th welding area in each wire welding image, Represents the number of pixels at the edge of the i-th welding area, Indicates the intersection angle of the i-th welding area, c is a preset parameter adjustment factor, which is used to prevent the value in the absolute value from being 0, thereby causing invalid calculation. In this application, the value of the preset parameter adjustment factor is 1, and the implementer can select other values ​​according to actual conditions; is the normalization function.

[0054] It should be noted that It represents the ratio of the number of pixels in the welding area to the number of pixels at the edge of the welding area, reflecting the ratio of the area of ​​the welding area to the perimeter of the contour, reflecting the shape characteristics of the welding area. The larger the value of is, the more irregular the shape of the welding area is, and the larger the value of the welding shape abnormality is; It is the difference between the crossing angle of the two metal wires in the welding area and the crossing angle in the vertical state. The greater the angle difference, the more non-standard the arrangement of the two metal wires constituting the welding area. At this time, the larger the value of the obtained welding shape anomaly value; conversely, the smaller the value of the obtained welding shape anomaly value.

[0055] Step S3, based on the difference between the average temperature information of the welding area and the average situation of the average temperature information of the welding area on each metal wire, and the difference between the average temperature information of the welding area and the preset expected temperature, obtain the temperature anomaly coefficient of any metal wire in each welding area in each metal wire welding image; based on the temperature anomaly coefficient, obtain the welding temperature anomaly factor of each welding area in each metal wire welding image.

[0056] The images obtained by the dual sensors belong to two different dimensional types of information. By spatially registering the wire welding image and the infrared thermal imaging image, the two images collected at the same time can be accurately aligned, and the specific positions of several welding areas in the wire welding image are mapped in the infrared thermal imaging image, so that the thermal effect of the welding process can be analyzed. The thermal effect of the welding area is closely related to the temperature distribution. The thermal information provided by the infrared thermal imaging image can provide an in-depth understanding of the temperature changes in the welding area, the range of the heat-affected zone, and the heat distribution, which is crucial for detecting welding quality.

[0057] According to the temperature information of each pixel in the infrared thermal imaging image at each acquisition time, the metal wire welding image and the infrared thermal imaging image at each acquisition time are matched using an image feature matching algorithm, and the temperature information of each pixel in the infrared thermal imaging image is matched to each pixel in the metal wire welding image. The image feature matching algorithm selected in this embodiment is the SIFT algorithm, and the implementer can select other image feature matching algorithms according to actual conditions; among them, the SIFT algorithm is a well-known technology and will not be described in detail in this embodiment.

[0058] At this point, the temperature information of each pixel in the metal wire welding image at each acquisition moment is obtained.

[0059] Furthermore, during the welding process of the welding machine, the temperature distribution of the welding area on each metal wire is mainly caused by the production heat effect. The spot welding technology is used to weld the welding areas of each metal wire on the mesh simultaneously. The temperature distribution at the welding area position on each metal wire in the infrared thermal imaging image should be uniform. However, when there are problems such as poor welds, uneven molten pools, offsets, or unstable welding current during the welding process, it will cause uneven heat distribution in the welding area, showing obvious abnormal hot spots or temperature gradients.

[0060] For each welding area in the wire welding image at each acquisition time, the average value of the temperature information of all pixels in the welding area is calculated as the average temperature information of each welding area.

[0061] Based on the above analysis, in order to reflect the degree of uneven temperature on each metal wire where each welding area is located, based on the difference between the average temperature information of the welding area and the average situation of the average temperature information of the welding area on each metal wire, and the difference between the average temperature information of the welding area and the preset expected temperature, the temperature anomaly coefficient of any metal longitudinal wire in each welding area in the metal wire welding image at each acquisition moment is calculated, and the calculation formula is: ;in, represents the temperature anomaly coefficient of any metal wire in the ith welding area in each metal wire welding image; It represents the average temperature information of the vth welding area on any metal wire where the i-th welding area is located in each metal wire welding image, except for the i-th welding area. represents the mean value of the average temperature information of all welding areas on any metal wire where the i-th welding area is located in each metal wire welding image; represents the number of welding areas on any metal wire where the i-th welding machine area is located in each metal wire welding image; Indicates the absolute value of the difference between the average temperature information of the ith welding area and the preset expected temperature. The output current of the welding machine and the resistance of the metal wire are fixed values, and the preset expected temperature of each welding area is calculated according to Joule's law. The Joule's law is a well-known technology and will not be described in detail in this embodiment.

[0062] It should be noted that It reflects the temperature distribution difference in the welding area of ​​the metal wire. The larger the value, the greater the possibility of abnormal welding temperature in the same metal longitudinal wire. It reflects the deviation between the average temperature information of the welding area and the preset expected temperature. The larger the value is, the greater the deviation is, the more abnormal the welding temperature in the welding area is, and the larger the temperature anomaly coefficient obtained at this time is; conversely, the smaller the temperature anomaly coefficient obtained is.

[0063] Furthermore, based on the above analysis, for each welding area in the metal wire welding image at each acquisition moment, the product of the temperature anomaly coefficients of all the metal wires in the welding area is calculated as the welding temperature anomaly factor of each welding area, which comprehensively reflects the degree of abnormality of the welding temperature on all the metal wires in the welding area.

[0064] Step S4, based on the average of the differences between the temperature information of the pixel points, obtain the cooling temperature abnormality factor of each welding area in each metal wire welding image; based on the welding temperature abnormality factor and the cooling temperature abnormality factor, obtain the welding thermal effect abnormality parameter of each welding area in each metal wire welding image.

[0065] The cooling process after welding is a heat conduction process. The temperature of the welding area gradually diffuses to the surrounding area over a period of time. When the welding condition of the welding area is good, the degree of temperature conduction to the surrounding areas should be similar.

[0066] Further, according to the above analysis, for the pixel point at the intersection of any two intersecting skeleton lines in the metal wire welding image at each acquisition moment, take the initial pixel point, and along the directions on both sides of any skeleton line where the initial pixel point is located, starting from the initial pixel point, the absolute values ​​of the difference between the temperature information of a preset number of consecutive adjacent pixel points are calculated in sequence as the first absolute value of the difference, and the average of all the first absolute values ​​of the difference on both sides of any skeleton line where the initial pixel point is located is calculated as the difference in the temperature change rate of the welding area where the initial pixel point is located in the directions on both sides of any skeleton line.

[0067] For each welding area in the wire welding image at each acquisition moment, the absolute value of the difference between the temperature change rate difference of the welding area in the direction on both sides of any skeleton straight line and the temperature change rate difference in the direction on both sides of another skeleton straight line is calculated as the cooling temperature anomaly factor of each welding area in the wire welding image at each acquisition moment.

[0068] It should be noted that the temperature of the welding area gradually diffuses to the metal wire over a period of time. When the temperature change rates on both sides of the welding area in the same direction of the metal wire are more consistent, and the change rate of the difference in the temperature change rates on both sides in different directions is more consistent, it means that the welding effect of the current welding area is better. At this time, the value of the cooling temperature anomaly factor is smaller; conversely, the value of the cooling temperature anomaly factor is larger.

[0069] Furthermore, in order to reflect the abnormal conditions of welding temperature during welding and the abnormal conditions of temperature cooling after welding in the welding area at each acquisition moment, the welding thermal effect abnormality parameters of each welding area in the metal wire welding image at each acquisition moment are obtained based on the welding temperature abnormality factor and the cooling temperature abnormality factor. The calculation method is as follows: for each welding area in the metal wire welding image at each acquisition moment, the sum of the welding temperature abnormality factor and the cooling temperature abnormality factor is calculated as the welding thermal effect abnormality parameter of each welding area in the metal wire welding image at each acquisition moment.

[0070] It should be noted that the abnormal welding thermal effect parameters comprehensively reflect the abnormal welding temperature during the welding process and the abnormal temperature cooling after welding. When the abnormal conditions of the welding temperature and the temperature cooling after welding are more significant, the value of the abnormal welding thermal effect parameters obtained at this time is larger; conversely, the value of the abnormal welding thermal effect parameters obtained is smaller.

[0071] Step S5, obtaining a welding anomaly score for each welding area in each metal wire welding image based on the welding shape anomaly value and the welding thermal effect anomaly parameter, and detecting the metal wire welding quality.

[0072] The abnormal value of welding shape reflects the abnormal degree of the geometric dimensions of the weld, which directly affects the mechanical properties of the weld; the abnormal parameters of welding thermal effect reflect the abnormal conditions of the metal structure and welding residual stress, and indirectly reflect the performance and quality of the welded parts.

[0073] Further, according to the above analysis, in order to reflect the abnormal degree of the welding area during the welding process, the welding abnormality score of each welding area in the wire welding image at each acquisition moment is calculated, and the calculation formula is: ; In the formula, Score the welding anomaly of the i-th welding area in each wire welding image; is the normalization function, is the welding shape anomaly value of the i-th welding area in each wire welding image, is the abnormal parameter of welding thermal effect of the i-th welding area in each metal wire welding image.

[0074] It should be noted that the welding anomaly score reflects the degree of abnormality of the welding area during the welding process. When the shape of the metal wire in the welding area is relatively abnormal during the welding process and the welding temperature is relatively abnormal, the values ​​of the welding shape abnormality value and the welding thermal effect abnormality parameter are large, and the obtained welding anomaly score is large; otherwise, the obtained welding anomaly score is small.

[0075] At this point, the welding anomaly score of each welding area in each metal wire welding image at each acquisition time is obtained.

[0076] The welding quality of the metal wire is detected based on the welding anomaly score, and a preset welding score threshold is set. In this embodiment, the value of the preset welding score threshold is 0.8. The implementer can select other values ​​according to actual conditions. When the value of the welding anomaly score of the welding area is greater than the preset welding score threshold, the welding area is regarded as a defective welding area.

[0077] The detection results are processed by computers. When welding defects occur, the welding defect areas are marked and counted. When the frequency of welding defects is low, only the welding defect areas are identified and recorded. When the frequency of welding defects is high, production needs to be stopped immediately and the production equipment needs to be checked and repaired.

[0078] Based on the same inventive concept as the above method, an embodiment of the present application also provides a computer-aided metal wire welding quality detection system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above-mentioned computer-aided metal wire welding quality detection methods are implemented.

[0079] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0080] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0081] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A computer-aided metal wire welding quality detection method, characterized in that: The method comprises the following steps: Acquire metal wire welding images and infrared thermal imaging images at each acquisition moment during the metal wire welding process; Based on the distance between each corner point in the metal wire welding image, each welding area in the metal wire welding image is obtained; and the intersection angle of each welding area is obtained by using the Hough straight line algorithm; Based on the number of pixels in the welding area, the number of pixels at the edge of the welding area and the value of the intersection angle of the welding area, the welding shape abnormal value of each welding area in each metal wire welding image is obtained; Based on the average temperature information of the pixel points in the infrared thermal imaging image, the temperature anomaly coefficient of any metal wire in each welding area in each metal wire welding image is obtained; Obtaining a welding temperature anomaly factor for each welding area in each metal wire welding image based on a temperature anomaly coefficient; Based on the average of the differences between the temperature information of the pixel points, the cooling temperature abnormality factor of each welding area in each metal wire welding image is obtained; Acquire welding thermal effect abnormal parameters of each welding area in each metal wire welding image based on welding temperature abnormal factors and cooling temperature abnormal factors; The welding anomaly score of each welding area in each metal wire welding image is obtained based on the welding shape anomaly value and the welding thermal effect anomaly parameter. The welding area with a welding anomaly score greater than a preset welding score threshold is regarded as a defective welding area to detect the metal wire welding quality.

2. The computer-aided wire welding quality detection method according to claim 1, characterized in that: The method for obtaining the welding area is: For the wire welding image at each acquisition moment, a corner point detection algorithm is used to obtain all corner points in the wire welding image. For each corner point, the three corner points closest to each corner point are obtained, and the quadrilateral area formed by the lines between each corner point and the three corner points closest to it is taken as each welding area.

3. The computer-aided metal wire welding quality detection method according to claim 1, characterized in that: The method for obtaining the intersection angle is: For the metal wire welding image at each acquisition moment, a skeleton extraction algorithm is used to obtain each skeleton in the metal wire welding image, and the Hough straight line algorithm is used to process all the skeletons in the metal wire welding image to obtain the skeleton straight lines of all the skeletons, and the welding area where the intersection of any two intersecting skeleton straight lines is located is obtained as the welding area of ​​each intersection; For the intersection point of any two intersecting skeleton straight lines, the minimum value of all angles formed by the intersection of the two skeleton straight lines is taken as the intersection angle of each welding area where the intersection point is located.

4. The computer-aided wire welding quality detection method according to claim 1, characterized in that: The calculation formula of the welding shape abnormal value is: ; In the formula, represents the welding shape abnormal value of the i-th welding area in each metal wire welding image; represents the number of pixels in the i-th welding area in each wire welding image, Represents the number of pixels at the edge of the i-th welding area, represents the intersection angle of the i-th welding area, and c is the preset parameter adjustment factor; is the normalization function.

5. The computer-aided metal wire welding quality detection method according to claim 1, characterized in that: The method for obtaining the temperature anomaly coefficient is: Use image feature matching algorithm to match the temperature information of each pixel in the infrared thermal imaging image to each pixel in the wire welding image; For each welding area in the wire welding image at each acquisition moment, the mean value of the temperature information of all pixels in the welding area is calculated as the average temperature information of each welding area; The calculation formula of the temperature anomaly coefficient is: ;in, represents the temperature anomaly coefficient of any metal wire in the ith welding area in each metal wire welding image; It represents the average temperature information of the vth welding area on any metal wire where the i-th welding area is located in each metal wire welding image, except the i-th welding area. represents the mean value of the average temperature information of all welding areas on any metal wire where the i-th welding area is located in each metal wire welding image; represents the number of welding areas on any metal wire where the i-th welding machine area is located in each metal wire welding image; Represents the absolute value of the difference between the average temperature information of the i-th welding area and the preset expected temperature.

6. The computer-aided metal wire welding quality detection method according to claim 1, characterized in that: The method for obtaining the abnormal welding temperature factor is: For each welding area in the wire welding image at each acquisition moment, the product of the temperature anomaly coefficients of all the wires in the welding area is calculated as the welding temperature anomaly factor of each welding area.

7. The computer-aided metal wire welding quality detection method according to claim 1, characterized in that: The method for obtaining the cooling temperature abnormality factor is: The pixel point at the intersection of any two intersecting skeleton lines in the metal wire welding image at each acquisition moment is taken as the initial pixel point, and along the directions on both sides of any skeleton line where the initial pixel point is located, starting from the initial pixel point, the absolute values ​​of the differences between the temperature information of a preset number of consecutive adjacent pixel points are calculated in sequence as the first absolute value of the difference, and the average of all the first absolute values ​​of the difference in the directions on both sides of any skeleton line where the initial pixel point is located is calculated as the temperature change rate difference of the welding area where the initial pixel point is located in the directions on both sides of any skeleton line; For each welding area in the wire welding image at each acquisition moment, the absolute value of the difference between the temperature change rate difference of the welding area in the direction of both sides of any skeleton straight line and the temperature change rate difference in the direction of both sides of another skeleton straight line is calculated as the cooling temperature anomaly factor of each welding area in each wire welding image.

8. The computer-aided metal wire welding quality detection method according to claim 1, characterized in that: The method for obtaining the abnormal parameters of welding thermal effect is as follows: For each welding area in each metal wire welding image, the sum of the welding temperature abnormality factor and the cooling temperature abnormality factor is calculated as the welding thermal effect abnormality parameter of each welding area in each metal wire welding image.

9. The computer-aided metal wire welding quality detection method according to claim 1, characterized in that: The calculation formula of the welding abnormality score is: ; In the formula, Score the welding anomaly of the i-th welding area in each wire welding image; is the normalization function, is the welding shape anomaly value of the i-th welding area in each wire welding image, is the abnormal parameter of welding thermal effect of the i-th welding area in each metal wire welding image.

10. A computer-aided metal wire welding quality detection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the computer-aided wire welding quality detection method as described in any one of claims 1 to 9 are implemented.

Images