A method and system for controlling sheet metal stamping using a heat-treated punch

Abstract

This application relates to the field of metal stamping technology, and particularly to a method and system for controlling sheet metal stamping using a heat-treated punch. The method includes: collecting the surface grayscale image of the stamped part after each stamping during the sheet metal stamping process, and determining the morphological approximation degree of each edge contour in the surface grayscale image after each stamping; determining the morphological evaluation value of each edge contour in the surface grayscale image after each stamping; determining the wrinkling intersection coefficient of each edge contour in the surface grayscale image after each stamping; determining the wrinkling change coefficient of each edge contour in the surface grayscale image after each stamping, and determining the wrinkling interference coefficient of each edge contour in the surface grayscale image after each stamping; determining the wrinkling discrimination coefficient of the surface grayscale image after each stamping, and controlling and adjusting the stamping machine during the stamping process. This application improves the accuracy of detecting the edge contours formed on the surface of the stamped part due to wrinkling, and further enhances the precise control of the stamping control system over the stamping machine.

Description

Technical Field

[0001] This application relates to the technical field of metal stamping, and particularly to a method and system for controlling sheet metal stamping using a heat-treated punch. Background Art

[0002] Stamping relies on a stamping machine and a die to apply external force to a sheet metal, causing it to undergo plastic deformation or separation, thereby obtaining a stamped workpiece with the required shape and size. The hot stamping forming technology utilizes the principle of metal hot plastic forming and adopts a stamping process with a special heat-treated punch, which can achieve quenching heat treatment of the sheet metal during forming, improve the forming performance of the material, and greatly expand the application range of high-strength and ultra-high-strength steel.

[0003] Since it is usually necessary to preset stamping parameters for the control system of the stamping machine during the stamping process, but there are many materials used for stamping, and there are certain differences in the strength, thickness, and processing environment among the materials. The preset stamping parameters may not match the actual stamping situation, resulting in defects such as cracking, wrinkling, and springback in some materials during the stamping process, affecting the quality of the stamped parts. Therefore, the quality of the stamped parts is detected by visual means, and then the stamping parameters of the stamping control system are adjusted to ensure the overall stamping quality of the stamped parts. However, during the production and processing of stamped parts, there will be many stripes on their surfaces, and when wrinkling occurs on the surface, there will also be many stripe patterns on the surface of the stamped parts. It is difficult to detect whether wrinkling occurs on the surface of the stamped parts during the visual inspection process, resulting in errors in the quality inspection of the stamped parts and affecting the production quality. Summary of the Invention

[0004] To solve the above technical problems, a method and system for controlling sheet metal stamping using a heat-treated punch are provided to solve the existing problems.

[0005] The solution of this application to solve the technical problems is to provide a method and system for controlling sheet metal stamping using a heat-treated punch, including the following steps:

[0006] In a first aspect, an embodiment of this application provides a method for controlling sheet metal stamping using a heat-treated punch, and the method includes the following steps:

[0007] Collect the surface grayscale image of the stamped part after each stamping during the sheet metal stamping process, and extract each edge contour in the surface grayscale image after each stamping;

[0008] According to the main extension direction of any edge contour in the surface grayscale image after each stamping, determine the direction representation line of the any edge contour; analyze the differences in the directions and lengths of the direction representation lines of each edge contour and the remaining edge contours, and determine the morphological approximation degree of each edge contour in the surface grayscale image after each stamping; according to the density distribution between different edge contours and the fluctuation of the grayscale values of the pixel points around each edge contour, determine the density fluctuation coefficient of each edge contour in the surface grayscale image after each stamping, and combine the morphological approximation degree and the density fluctuation coefficient to determine the morphological evaluation value of each edge contour in the surface grayscale image after each stamping;

[0009] According to the discreteness of the distance between the upper endpoint of the skeleton extracted from each edge contour and the intersection point, and the number of endpoints and intersection points, determine the wrinkling intersection coefficient of each edge contour in the surface grayscale image after each stamping; analyze the difference in the number of edge contours in the surface grayscale image after two adjacent stampings and the difference in the number of edge pixels on the edge contours at the same position, determine the wrinkling change coefficient of each edge contour in the surface grayscale image after each stamping, and combine the wrinkling intersection coefficient to determine the wrinkling interference coefficient of each edge contour in the surface grayscale image after each stamping;

[0010] Based on the morphological evaluation value and the wrinkling interference coefficient of all edge contours in the surface grayscale image after each stamping, determine the wrinkling discrimination coefficient of the surface grayscale image after each stamping, and based on the wrinkling discrimination coefficient of the surface grayscale images after all previous stampings before the current stamping, control and adjust the stamping machine during the stamping process.

[0011] Preferably, the determination of the direction representation line of the any edge contour includes:

[0012] Denote the connection line between the edge pixel point corresponding to the minimum abscissa and the edge pixel point corresponding to the maximum abscissa on the any edge contour as the horizontal connection line;

[0013] Denote the connection line between the edge pixel point corresponding to the minimum ordinate and the edge pixel point corresponding to the maximum ordinate on the any edge contour as the vertical connection line;

[0014] Count the number of all pixel points on the horizontal connection line, denoted as the horizontal length; count the number of all pixel points on the vertical connection line, denoted as the vertical length;

[0015] If the horizontal length is greater than or equal to the vertical length, use the horizontal connection line as the direction representation line of the any edge contour; otherwise, use the vertical connection line as the direction representation line of the any edge contour.

[0016] Preferably, determining the morphological approximation degree of each edge contour in the surface grayscale image after each stamping includes:

[0017] Obtaining the slope of the direction representation line of any one of the edge contours and the number of all pixel points on its direction representation line;

[0018] Taking the mean value of the differences in the slopes between any one of the edge contours in the surface grayscale image after each stamping and all the other edge contours as the direction difference coefficient of any one of the edge contours;

[0019] Taking the mean value of the differences in the numbers between any one of the edge contours in the surface grayscale image after each stamping and all the other edge contours as the length difference coefficient of any one of the edge contours;

[0020] Taking the reciprocal of the sum of the direction difference coefficient and the length difference coefficient as the morphological approximation degree of each edge contour in the surface grayscale image after each stamping.

[0021] Preferably, determining the density fluctuation coefficient of each edge contour in the surface grayscale image after each stamping, and combining the morphological approximation degree and the density fluctuation coefficient to determine the morphological evaluation value of each edge contour in the surface grayscale image after each stamping includes:

[0022] Taking the mean value of the distances between the centroid of each edge contour in the surface grayscale image after each stamping and the centroids of all the other edge contours as the distribution coefficient of each edge contour;

[0023] Calculating the degree of dispersion of the grayscale values of all pixel points within the minimum circumscribed rectangle of each edge contour in the surface grayscale image after each stamping;

[0024] Taking the ratio of the degree of dispersion to the distribution coefficient as the density fluctuation coefficient of each edge contour in the surface grayscale image after each stamping;

[0025] Taking the product of the morphological approximation degree and the density fluctuation coefficient as the morphological evaluation value of each edge contour in the surface grayscale image after each stamping.

[0026] Preferably, determining the wrinkling intersection coefficient of each edge contour in the surface grayscale image after each stamping includes:

[0027] Assigning a grayscale value of 1 to the edge pixel points and a grayscale value of 0 to the non-edge pixel points in the surface grayscale image after each stamping to obtain a binary image, extracting the skeletons of all edge contours in the binary image, and denoting the pixel points on the skeletons as skeleton pixel points;

[0028] If there is only one skeleton pixel in the eight-neighborhood of any skeleton pixel on the skeleton corresponding to each edge contour, record the any skeleton pixel as a skeleton endpoint; if there are multiple skeleton pixels in the eight-neighborhood of the any skeleton pixel on the skeleton corresponding to each edge contour, record the any skeleton pixel as a skeleton intersection point;

[0029] Count the number of all skeleton endpoints and skeleton intersection points on the skeleton corresponding to each edge contour, and record it as the bifurcation number;

[0030] Take the minimum value of the distances from each skeleton endpoint on the skeleton corresponding to each edge contour to all skeleton intersection points as the nearest distance;

[0031] Calculate the dispersion degree of the nearest distances of all skeleton endpoints on the skeleton corresponding to each edge contour, and record it as the distance dispersion degree;

[0032] Take the ratio of the bifurcation number to the distance dispersion degree as the wrinkling crossing coefficient of each edge contour in the surface gray-scale image after each stamping.

[0033] Preferably, the determination of the wrinkling change coefficient of each edge contour in the surface gray-scale image after each stamping includes:

[0034] Record the difference between the number of all edge contours in the surface gray-scale image after each stamping and the number of all edge contours in the surface gray-scale image after the adjacent stamping as the first quantity difference;

[0035] Record the edge contour with the same centroid coordinates in the surface gray-scale image after each stamping and the edge contour with the same centroid coordinates in the surface gray-scale image after the adjacent stamping as the relative edge contour of each edge contour;

[0036] Record the difference between the number of all edge pixels on each edge contour and the number of all edge pixels on the relative edge contour as the second quantity difference;

[0037] Take the sum of the first quantity difference and the second quantity difference as the wrinkling change coefficient of each edge contour in the surface gray-scale image after each stamping.

[0038] Preferably, the wrinkling interference coefficient of each edge contour in the surface gray-scale image after each stamping is the product of the wrinkling crossing coefficient and the wrinkling change coefficient.

[0039] Preferably, the determination of the wrinkling discrimination coefficient of the surface gray-scale image after each stamping includes:

[0040] Take the product of the morphological evaluation value and the wrinkling interference coefficient as the discrimination factor of each edge contour in the surface gray-scale image after each stamping;

[0041] The mean value of the discrimination factors of all edge contours in the surface gray-scale image after each stamping is used as the wrinkling discrimination coefficient of the surface gray-scale image after each stamping.

[0042] Preferably, the control adjustment of the stamping machine in the stamping process includes:

[0043] Performing threshold segmentation on the wrinkling discrimination coefficients of the surface gray-scale images after all previous stampings before the current stamping to obtain a segmentation threshold;

[0044] If the wrinkling discrimination coefficient of the surface gray-scale image of the current stamping is less than the segmentation threshold, the stamping machine is not adjusted; otherwise, the stamping machine is controlled and adjusted through a PLC controller.

[0045] In a second aspect, an embodiment of the present application further provides a sheet metal stamping control system using a heat-treated punch, 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 the sheet metal stamping control method described in any one of the above are implemented.

[0046] The present application has at least the following beneficial effects:

[0047] Based on the main extension direction of any edge contour in the surface grayscale image after each stamping, the direction characterization line of the any edge contour is determined; by analyzing the differences in the directions and lengths of the direction characterization lines of each edge contour and the remaining edge contours, the morphological approximation degree of each edge contour in the surface grayscale image after each stamping is determined. The beneficial effect is that it considers the approximation of the morphologies between different edge contours, and further reflects the possibility of the wrinkle contour formed by the wrinkling phenomenon caused by stamping for the corresponding edge contour; according to the density of the distribution between different edge contours and the fluctuation of the grayscale values of the pixel points around each edge contour, combined with the morphological approximation degree, the morphological evaluation value of each edge contour in the surface grayscale image after each stamping is determined. The beneficial effect is that it considers the distribution of different edge contours and the change of the pixel points in the area around the edge contour to reflect the possibility of the wrinkle contour formed by the stamping part being wrinkled due to stamping for the corresponding edge contour; according to the number of the upper endpoints and intersection points of the skeleton extracted from each edge contour and the discrete situation of their distances, the wrinkling intersection coefficient of each edge contour in the surface grayscale image after each stamping is determined. The beneficial effect is that it considers the degree of intersection of the edge contour when the wrinkling phenomenon occurs, so as to reflect the possible situation of wrinkling intersection of the edge contour; by analyzing the difference in the number of edge contours in the surface grayscale images after two adjacent stampings and the difference in the number of upper edge pixel points on the edge contours at the same position, the wrinkling change coefficient of each edge contour in the surface grayscale image after each stamping is determined. Combining the wrinkling intersection coefficient, the wrinkling interference coefficient of each edge contour in the surface grayscale image after each stamping is determined. The beneficial effect is that it considers the change in the number of edge contours in the surface grayscale images after two adjacent stampings and the change in the number of pixel points of the same edge contour in the two images, and reflects the possibility of the corresponding edge contour having a wrinkling phenomenon; based on the morphological evaluation value and the wrinkling interference coefficient of all edge contours in the surface grayscale image after each stamping, the wrinkling discrimination coefficient of the surface grayscale image after each stamping is determined. The beneficial effect is that it further reflects the possibility of the edge contour being the wrinkle edge generated by the surface wrinkling of the stamping part; based on the wrinkling discrimination coefficients of the surface grayscale images after all the previous stampings before the current stamping, the stamping machine in the stamping process is controlled and adjusted. The beneficial effect is that it improves the accuracy of detecting the edge contour formed by the wrinkling phenomenon on the surface of the stamping part, and further improves the precise control of the stamping machine by the stamping control system, and can improve the production quality and production efficiency of the stamping part. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The following further elaborates in detail a method for controlling the stamping of a sheet using a heat-treated punch according to the present application with reference to the accompanying drawings.

[0049] Figure 1 It is a flowchart of the steps of a method for controlling the stamping of a sheet using a heat-treated punch provided by an embodiment of the present application;

[0050] Figure 2 It is a flowchart of the steps for obtaining the morphological approximation degree of each edge contour in the surface grayscale image after each stamping provided by the embodiment of the present application;

[0051] Figure 3 It is a flowchart of the steps for obtaining the wrinkling intersection coefficient of each edge contour in the surface grayscale image after each stamping provided by the embodiment of the present application. Detailed implementation manners

[0052] In order to make the objectives, technical solutions and advantages of the present application clearer, the following further elaborates in detail a method and system for controlling sheet metal stamping using a heat-treated punch in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs.

[0054] Please refer to Figure 1 , which shows a flowchart of the steps of a method for controlling sheet metal stamping using a heat-treated punch provided by an embodiment of the present application. The method includes the following steps:

[0055] Step 1, collect the surface grayscale image of the stamped part after each stamping during the sheet metal stamping process, and extract each edge contour in the surface grayscale image after each stamping.

[0056] Place a CMOS high-definition industrial camera above the workbench of the stamping equipment. After each stamping is completed, collect the surface image of the stamped part, perform gray-scale processing on the collected image to obtain the surface grayscale image of each stamping, perform edge detection processing on the surface grayscale image, and extract each edge contour in the surface grayscale image through the erosion and dilation algorithm.

[0057] Preferably, in this embodiment, the canny edge detection algorithm is used for edge detection. Among them, the canny edge detection algorithm and the erosion and dilation algorithm are well-known technologies and will not be elaborated here. As other implementation manners, implementers can use other methods of the existing technology, such as the sobel operator, etc. This embodiment does not make special restrictions on this.

[0058] So far, each edge contour in the surface grayscale image after each stamping is obtained.

[0059] Step 2, according to the main extension direction of any edge contour in the surface grayscale image after each stamping, determine the direction characterization line of any edge contour; analyze the difference in direction and length of the direction characterization line of each edge contour and the other edge contours, and determine the morphological similarity of each edge contour in the surface grayscale image after each stamping; according to the density of distribution between different edge contours and the fluctuation of the grayscale values ​​of the pixels around each edge contour, combined with the morphological similarity, determine the morphological evaluation value of each edge contour in the surface grayscale image after each stamping.

[0060] During the stamping process, the sheet undergoes physical deformation and becomes a stamped part with a specific shape and function. Under normal circumstances, when no wrinkling occurs after the sheet is stamped, the texture formed on the surface of the stamped part is mainly affected by the mold setting, causing the sheet to bend during stamping, thereby forming texture in the surface grayscale image of the stamped part. Such texture formed on the surface of the stamped part is usually wider and longer in overall length, and the direction of the edge contour corresponding to such texture in the surface grayscale image is inconsistent, and there is no obvious regularity. When wrinkling occurs after the sheet is stamped, wrinkles will appear in local areas of the surface of the stamped part, which appear as edge contours with similar morphology in the collected surface grayscale image, and these edge contours are usually along the same direction with obvious regular characteristics.

[0061] Based on the above analysis, the position coordinates of the edge pixels on each edge contour are analyzed to determine the direction characterization line of each edge contour to reflect the overall extension direction of the corresponding edge contour, specifically:

[0062] The connecting line between the edge pixel point corresponding to the minimum horizontal coordinate and the edge pixel point corresponding to the maximum horizontal coordinate on each edge contour in the surface grayscale image after each punching is recorded as a horizontal connecting line;

[0063] The connecting line between the edge pixel point corresponding to the minimum ordinate and the edge pixel point corresponding to the maximum ordinate on each edge contour in the surface grayscale image after each punching is recorded as the longitudinal connecting line;

[0064] Counting the number of all pixels on the horizontal line, and recording it as the horizontal length; counting the number of all pixels on the vertical line, and recording it as the vertical length;

[0065] If the horizontal length is greater than or equal to the longitudinal length, the horizontal connection line is used as the direction characterizing line of each edge contour in the surface grayscale image after each punching; if the horizontal length is less than the longitudinal length, the longitudinal connection line is used as the direction characterizing line of each edge contour in the surface grayscale image after each punching;

[0066] Further, the flowchart of the method for obtaining the morphological approximation degree of each edge contour in the surface grayscale image after each stamping provided in this embodiment is as Figure 2 shown.

[0067] First, based on the difference in the corresponding slopes of the direction representation lines of different edge contours and the difference in the lengths of the direction representation lines, determine the morphological approximation degree to reflect the similarity of the extension directions between different edge contours, and further reflect the possibility of the wrinkle contour caused by the wrinkling phenomenon during stamping. Specifically:

[0068] Obtain the slope of the direction representation line of any edge contour in the surface grayscale image after each stamping, and the number of all pixel points on the direction representation line;

[0069] Record the difference in the slopes between any edge contour in the surface grayscale image after each stamping and the rest of the edge contours as the first difference;

[0070] Preferably, in this embodiment, record the difference in the slopes between any edge contour in the surface grayscale image after each stamping and the rest of the edge contours as the first difference.

[0071] Take the mean value of the first differences between any edge contour in the surface grayscale image after each stamping and all the rest of the edge contours as the direction difference coefficient of the any edge contour;

[0072] Record the difference in the number between any edge contour in the surface grayscale image after each stamping and the rest of the edge contours as the second difference;

[0073] Preferably, in this embodiment, record the difference in the number between any edge contour in the surface grayscale image after each stamping and the rest of the edge contours as the second difference.

[0074] Take the mean value of the second differences between any edge contour in the surface grayscale image after each stamping and all the rest of the edge contours as the length difference coefficient of the any edge contour;

[0075] Take the reciprocal of the sum of the direction difference coefficient and the length difference coefficient as the morphological approximation degree of each edge contour in the surface grayscale image after each stamping;

[0076] Preferably, in this embodiment, the calculation method of the morphological approximation degree of each edge contour in the surface grayscale image after each stamping is: , where is the th morphological approximation degree of the is the direction difference coefficient of the th edge contour in the surface gray-scale image after the th stamping, is the length difference coefficient of the th edge contour in the surface gray-scale image after the th stamping, is a preset value greater than 0 to avoid a zero denominator, and its value range is in , in this embodiment, the value is 0.1. As other implementation manners, the implementer can set it according to the actual situation.

[0077] It should be noted that the smaller the direction difference coefficient, the more consistent the corresponding edge contour is with the other edge contours in direction; the smaller the length difference coefficient, the closer the corresponding edge contour is to the other edge contours in length, and the greater the obtained morphological approximation degree, indicating that the corresponding edge contour is more similar to the other edge contours in morphology, and the greater the possibility that the corresponding edge is the wrinkle contour formed by the wrinkling phenomenon caused by stamping.

[0078] Furthermore, as the punching force continuously increases during the stamping process, axial stress is generated during the movement of the punch driving the sheet metal. The axial stress will generate axial strain and radial strain, resulting in deformation of this part of the sheet metal. When wrinkling occurs, the generated wrinkles will mainly concentrate at the punch position of the die, showing the characteristic of locally dense distribution of the corresponding edge contour in the surface gray-scale image. And because the wrinkled area will increase the undulation degree of the stamping part surface, it will cause large fluctuations in the gray-scale changes of the pixel points in the corresponding area of the image.

[0079] Based on the above analysis, based on the dense situation of different edge contour distributions and the gray-scale changes of the pixel points around the edge contours, a dense fluctuation coefficient is determined, specifically:

[0080] The mean value of the distances between the centroid of each edge contour in the surface gray-scale image after each stamping and the centroids of all the other edge contours is used as the distribution coefficient of each edge contour;

[0081] Preferably, in this embodiment, the mean value of the Euclidean distances between the centroid of each edge contour in the surface gray-scale image after each stamping and the centroids of all the other edge contours is used as the distribution coefficient of each edge contour. Among them, the calculation of the Euclidean distance is a well-known technology and will not be elaborated here. As other implementation manners, the implementer can adopt other methods of the existing technology, such as the Manhattan distance, etc. This embodiment does not make special restrictions on this.

[0082] Calculate the dispersion degree of the gray-scale values of all pixel points within the minimum circumscribed rectangle of each edge contour in the surface gray-scale image after each stamping;

[0083] Preferably, in this embodiment, the variance of the grayscale values ​​of all pixels within the minimum circumscribed rectangle of each edge contour in the surface grayscale image after each stamping is calculated. As other implementation methods, the implementer may adopt other methods of the prior art to measure the degree of discreteness, such as standard deviation, coefficient of variation, etc. This embodiment does not impose any special restrictions on this.

[0084] The ratio of the discrete degree to the distribution coefficient is used as the intensive fluctuation coefficient of each edge contour in the surface grayscale image after each punching;

[0085] Preferably, in this embodiment, the calculation method of the dense fluctuation coefficient of each edge contour in the surface grayscale image after each punching is: ,in, For the The grayscale image of the surface after the first punching The intensive fluctuation coefficient of the strip edge profile, For the The grayscale image of the surface after the first punching The discreteness of the edge contour of the strip, For the The grayscale image of the surface after the first punching The distribution coefficient of the strip edge profile, To preset a value greater than 0, in order to avoid the denominator being 0, The value range is , in this embodiment, The value is 0.1. As other implementation methods, the implementer can set it according to the actual situation.

[0086] It should be noted that, if the distribution coefficient is smaller, it means that the distance between the corresponding edge contour and the other edge contours is closer, which means that the edge contour distribution is denser; the greater the degree of discreteness, the more drastic the fluctuation of the grayscale value of the pixel points in the surrounding area of ​​the corresponding edge contour; the larger the obtained intensive fluctuation coefficient, the more likely the corresponding edge contour is to be a wrinkle contour caused by wrinkling due to deformation of the stamping part during the stamping process.

[0087] Further, based on the morphology similarity and the intensive fluctuation coefficient, a morphology evaluation value is determined, specifically:

[0088] The product of the morphological similarity and the intensive fluctuation coefficient is used as the morphological evaluation value of each edge contour in the surface grayscale image after each punching;

[0089] It should be noted that, the larger the morphological evaluation value is, the more likely the corresponding edge profile is a wrinkle profile formed by wrinkling of the stamped part due to stamping.

[0090] So far, the morphological evaluation value of each edge contour in the surface grayscale image after each stamping is obtained.

[0091] Step 3: According to the discrete situation of the distance between the upper endpoint and the intersection point of the skeleton extracted from each edge contour, and the number of endpoints and intersection points, determine the wrinkling intersection coefficient of each edge contour in the surface grayscale image after each stamping; analyze the difference in the number of edge contours in the surface grayscale images after two adjacent stampings, and the difference in the number of edge pixels on the edge contours at the same position, determine the wrinkling change coefficient of each edge contour in the surface grayscale image after each stamping, and combine the wrinkling intersection coefficient to determine the wrinkling interference coefficient of each edge contour in the surface grayscale image after each stamping.

[0092] Furthermore, due to the different shapes of the sheet metal molds, the sheet metal may be stamped into different shapes according to actual production needs. Therefore, large edge contours may be formed on the surface of the stamped parts. These large edge contours may intersect with the edge contours formed during other stamping processes, that is, the phenomenon of edge contour intersection occurs. When the wrinkling phenomenon occurs, the wrinkle contours may also appear in the large edge contours and intersect with the large edge contours. Since these large edge contours are usually long and thick, when the wrinkle contours intersect with them, it causes the large edge contours to bifurcate. By the edge detection method, the wrinkle contours and the large edge contours may be recognized as one edge contour, so that they cannot be distinguished according to the characteristics of the wrinkle contours in the image, resulting in a smaller morphological evaluation value calculated above, and further causing deviation in the detection of the wrinkle contours.

[0093] When there are large edge contours on the surface of the stamped parts, under normal circumstances, the bifurcation phenomenon in the large edge contours is relatively light, specifically manifested as fewer branches on the edge and a greater difference in length between the branches. When the wrinkling phenomenon occurs, the intersection of the wrinkled edge and the large edge contour will generate a large number of branches on the large edge contour, and since the approximation degree between the wrinkled edges is high, the lengths of the branches generated on this large edge contour are approximately the same.

[0094] Furthermore, the step flow chart of the method for obtaining the wrinkling intersection coefficient of each edge contour in the surface grayscale image after each stamping provided in this embodiment is as Figure 3 shown.

[0095] Based on the above analysis, analyze the intersection situation of different edge contours, and determine the wrinkling intersection coefficient to reflect the degree of intersection phenomenon of the corresponding edge contour. Specifically:

[0096] Assign the grayscale value of the edge pixels in the surface grayscale image after each stamping to 1, and assign the grayscale value of the non-edge pixels to 0 to obtain a binary image;

[0097] Adopt a skeleton extraction algorithm to extract the skeletons of all edge contours in the binary image, and denote the pixel points on the skeletons as skeleton pixel points;

[0098] It should be noted that the skeleton extraction algorithm and the binarization process are well-known technologies and will not be elaborated here.

[0099] If there is only one skeleton pixel point in the eight-neighborhood of any skeleton pixel point on the skeleton corresponding to each edge contour, denote the any skeleton pixel point as a skeleton end point; if there are more than 2 skeleton pixel points in the eight-neighborhood of the any skeleton pixel point on the skeleton corresponding to each edge contour, denote the any skeleton pixel point as a skeleton intersection point;

[0100] Count the number of all skeleton end points and skeleton intersection points on the skeleton corresponding to each edge contour, and denote it as the bifurcation number;

[0101] Denote the minimum value of the distances between each skeleton end point and all skeleton intersection points on the skeleton corresponding to each edge contour as the nearest distance;

[0102] Preferably, in this embodiment, denote the minimum value of the Euclidean distances between each skeleton end point and all skeleton intersection points on the skeleton corresponding to each edge contour as the nearest distance. As other implementation manners, implementers can use other methods in the prior art to measure distances, for example, Manhattan distance, etc. This embodiment does not make special restrictions on this.

[0103] Calculate the dispersion degree of the nearest distances of all skeleton end points on the skeleton corresponding to each edge contour, and denote it as the distance dispersion degree;

[0104] Preferably, in this embodiment, calculate the variance of the nearest distances of all skeleton end points on the skeleton corresponding to each edge contour. As other implementation manners, implementers can use other methods in the prior art, for example, standard deviation, coefficient of variation, etc. This embodiment does not make special restrictions on this.

[0105] Denote the ratio of the bifurcation number to the distance dispersion degree as the wrinkling intersection coefficient of each edge contour in the surface gray image after each stamping;

[0106] It should be noted that when calculating the ratio of the number of bifurcations to the distance dispersion, to avoid a zero denominator, a preset value greater than 0 is added to the denominator. In this embodiment, the preset value greater than 0 is taken as 0.1. As other implementation manners, the implementer can set it according to the actual situation. Secondly, when there is an intersection between a certain edge contour and the edge contour formed by wrinkling, the bifurcation situation on this edge contour is more serious. The larger the number of endpoints on the edge contour and the smaller the dispersion degree, it indicates that the bifurcation lengths of the corresponding edge contour are more consistent, and the obtained wrinkling intersection coefficient is larger, indicating that this edge contour is more likely to have a wrinkling intersection situation.

[0107] Furthermore, usually the same batch of stamped parts uses the same mold, so the edges formed by each stamping are relatively similar. When wrinkling occurs on the surface of the stamped part, the number of edge contours in the image will increase and may intersect with the original edge contours in the image, resulting in an increase in the number of pixel points included in the detected edge contours. Secondly, the positions of the edge contours on the surface of the stamped part are the same after each stamping. When the centroid of each edge contour in the surface grayscale image after each stamping coincides with the centroid of the edge contour in the surface grayscale image after the previous stamping, it indicates that these two edge contours are the same edge contour.

[0108] Based on the above analysis, analyze the change in the number of edge contours on the surface grayscale image after two consecutive stampings, as well as the change in the number of pixel points on the edge contours, and determine the wrinkling change coefficient, specifically:

[0109] Record the difference between the number of all edge contours in the surface grayscale image after each stamping and the number of all edge contours in the surface grayscale image after the previous stamping as the first quantity difference;

[0110] Preferably, in this embodiment, record the absolute value of the difference between the number of all edge contours in the surface grayscale image after each stamping and the number of all edge contours in the surface grayscale image after the previous stamping as the first quantity difference.

[0111] Record the edge contour with the same centroid coordinates in the surface grayscale image after each stamping as the relative edge contour of each edge contour compared with the edge contour with the same centroid coordinates in the surface grayscale image after the previous stamping;

[0112] Record the difference between the number of all edge pixels on each edge contour in the surface grayscale image after each stamping and the number of all edge pixels on the relative edge contour as the second quantity difference;

[0113] Preferably, in this embodiment, the absolute value of the difference between the number of all edge pixels on each edge contour in the surface grayscale image after each stamping and the number of all edge pixels on the relative edge contour is denoted as the second quantity difference.

[0114] The sum of the first quantity difference and the second quantity difference is used as the wrinkling change coefficient of each edge contour in the surface grayscale image after each stamping;

[0115] It should be noted that if there is no relative edge contour in the surface grayscale image of the previous stamping for a certain edge contour, the number of all edge pixels of the relative edge contour is 0, that is, the second quantity difference of each edge contour is the number of all edge pixels on the corresponding edge contour.

[0116] It should be noted that the larger the first quantity difference and the larger the second quantity difference, that is, the larger the wrinkling change coefficient, the more likely it is that the edge contour intersects with the wrinkle edge due to the wrinkling phenomenon on the surface of the stamped part, resulting in a large change, indicating that the corresponding edge contour is more likely to be the wrinkling edge generated by the wrinkling on the surface of the stamped part.

[0117] Further, based on the wrinkling intersection coefficient and the wrinkling change coefficient, a wrinkling interference coefficient is determined, specifically:

[0118] The product of the wrinkling intersection coefficient and the wrinkling change coefficient is used as the wrinkling interference coefficient of each edge contour in the surface grayscale image after each stamping;

[0119] It should be noted that the larger the obtained wrinkling interference coefficient, the more likely it is that the corresponding edge contour changes due to the intersection of the wrinkling edge generated when the wrinkling phenomenon appears on the surface of the stamped part, and the more likely the edge contour is to be the wrinkling edge.

[0120] Thus, the wrinkling interference coefficient of each edge contour in the surface grayscale image after each stamping is obtained.

[0121] Step 4: Based on the shape evaluation value and the wrinkling interference coefficient of all edge contours in the surface grayscale image after each stamping, determine the wrinkling discrimination coefficient of the surface grayscale image after each stamping. Based on the wrinkling discrimination coefficients of the surface grayscale images after all previous stampings before the current stamping, control and adjustment of the stamping machine during the stamping process are performed.

[0122] Further, based on the shape evaluation value and the wrinkling interference coefficient, a discrimination factor is determined, specifically:

[0123] The product of the shape evaluation value and the wrinkling interference coefficient is used as the discrimination factor of each edge contour in the surface grayscale image after each stamping;

[0124] It should be noted that the closer the shape of each edge contour is to the shapes of the other edge contours, the denser the distribution, the greater the gray-scale change of the pixel points in the surrounding area, or the more branches there are in the edge contour, and the greater the change difference in the number of edge contours and the number of their edge pixel points compared with the surface gray-scale image of the previous stamping, the greater the obtained discrimination factor, indicating that the edge contour is more likely to be the wrinkle edge caused by the wrinkling of the stamping part surface.

[0125] Furthermore, based on the discrimination factor, the wrinkling phenomenon on the surface of the stamping part is evaluated to control the stamping process. Specifically:

[0126] The mean value of the discrimination factors of all edge contours in the surface gray-scale image after each stamping is used as the wrinkling discrimination coefficient of the surface gray-scale image after each stamping;

[0127] Threshold segmentation is performed on the wrinkling discrimination coefficients of all surface gray-scale images after stamping before the current stamping to obtain a segmentation threshold;

[0128] Preferably, in this embodiment, cross-validation is performed on the wrinkling discrimination coefficients of all surface gray-scale images after stamping before the current stamping to obtain a segmentation threshold. Among them, the cross-validation method is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt other methods of existing technologies, such as the Otsu threshold segmentation method, etc. This embodiment does not make special restrictions on this.

[0129] If the wrinkling discrimination coefficient of the surface gray-scale image of the current stamping is less than the segmentation threshold, the stamping machine in the stamping process is not adjusted. If the wrinkling discrimination coefficient of the surface gray-scale image of the current stamping is greater than or equal to the segmentation threshold, the stamping control system controls and adjusts the stamping machine through a PLC (Programmable Logic Controller) controller;

[0130] It should be noted that when the wrinkling discrimination coefficient is greater than or equal to the segmentation threshold, the stamping control system outputs a control signal, and the PLC controller controls the motor according to the control signal. The motor adjusts the output power according to the actual control, so that the stamping speed and stamping force of the stamping machine are reduced, thereby reducing the overall force on the sheet during stamping and reducing the possibility of wrinkling.

[0131] Based on the same inventive concept as the above method, the embodiment of the present application also provides a sheet stamping control system using a heat-treated punch, 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, it implements the steps of any one of the above-mentioned sheet stamping control methods using a heat-treated punch.

[0132] It should be understood that although Figure 1 the steps in the flowchart of Figure 1 are shown in sequence according to the indication of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, there is no strict order limit for the execution of these steps, and these steps can be executed in other orders. Moreover,

[0133] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0134] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation to the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made. Therefore, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application all fall within the protection scope of the technical solution of the present application.

Claims

1. A plate stamping control method using a heat-treated punch, characterized in that: The method comprises the following steps: Collect the grayscale image of the stamped part surface after each stamping process in the sheet metal stamping process, and extract each edge contour in the grayscale image of the surface after each stamping process; According to the main extension direction of any edge contour in the surface grayscale image after each stamping, determine the direction characterization line of any edge contour; the direction characterization line reflects the overall extension direction of the corresponding edge contour; analyze the difference in the direction and length of the direction characterization line of each edge contour and the other edge contours, and determine the morphological similarity of each edge contour in the surface grayscale image after each stamping; according to the density of distribution between different edge contours and the fluctuation of the grayscale value of the pixel points around each edge contour, determine the density fluctuation coefficient of each edge contour in the surface grayscale image after each stamping, and combine the morphological similarity and the density fluctuation coefficient to determine the morphological evaluation value of each edge contour in the surface grayscale image after each stamping; the larger the density fluctuation coefficient, the more likely the corresponding edge contour is a wrinkle contour caused by wrinkling due to deformation of the stamped part during the stamping process; According to the discreteness of the distance between the endpoint and the intersection point on the skeleton extracted from each edge contour, and the number of the endpoint and the intersection point, the wrinkle cross coefficient of each edge contour in the surface grayscale image after each stamping is determined; the wrinkle cross coefficient reflects the degree of the cross phenomenon of the corresponding edge contour; Analyze the difference in the number of edge contours in the surface grayscale image after two adjacent stampings, as well as the difference in the number of edge pixels on the edge contour at the same position, determine the wrinkle variation coefficient of each edge contour in the surface grayscale image after each stamping, and determine the wrinkle interference coefficient of each edge contour in the surface grayscale image after each stamping in combination with the wrinkle intersection coefficient; the larger the wrinkle variation coefficient, the more likely the corresponding edge contour is to intersect with the wrinkle edge due to the wrinkling phenomenon generated on the surface of the stamping part; the larger the wrinkle interference coefficient, the more likely the corresponding edge contour is to change due to the intersection of the wrinkle edge generated when the wrinkle phenomenon occurs on the surface of the stamping part, and the more likely the corresponding edge contour is to be a wrinkled edge; Based on the morphological evaluation values ​​and the wrinkle interference coefficients of all edge contours in the surface grayscale image after each stamping, the wrinkle discrimination coefficient of the surface grayscale image after each stamping is determined; based on the wrinkle discrimination coefficient of the surface grayscale images after all stampings before the current stamping, the stamping machine of the stamping process is controlled and adjusted.

2. A plate stamping control method using a heat-treated punch as claimed in claim 1, characterized in that: The step of determining a direction characterizing line of any edge contour comprises: The line between the edge pixel point with the smallest horizontal coordinate and the edge pixel point with the largest horizontal coordinate on any edge contour is recorded as a horizontal line; The line between the edge pixel point with the smallest ordinate and the edge pixel point with the largest ordinate on any edge contour is recorded as a vertical line; Counting the number of all pixels on the horizontal line, and recording it as the horizontal length; counting the number of all pixels on the vertical line, and recording it as the vertical length; If the horizontal length is greater than or equal to the vertical length, the horizontal connection line is used as the direction characterizing line of any edge contour; otherwise, the vertical connection line is used as the direction characterizing line of any edge contour.

3. A plate stamping control method using a heat-treated punch as claimed in claim 1, characterized in that: Determining the morphological similarity of each edge contour in the surface grayscale image after each punching includes: Obtaining the slope of the direction characterization line of any edge contour and the number of all pixel points on the direction characterization line; The average of the differences in the slopes between any one edge profile and all other edge profiles in the surface grayscale image after each punching is used as the direction difference coefficient of any one edge profile; The average of the differences in the number between any one edge contour and all other edge contours in the surface grayscale image after each punching is used as the length difference coefficient of any one edge contour; The reciprocal of the sum of the direction difference coefficient and the length difference coefficient is used as the morphological approximation of each edge contour in the surface grayscale image after each punching.

4. A plate stamping control method using a heat-treated punch as claimed in claim 1, characterized in that: The step of determining the dense fluctuation coefficient of each edge contour in the surface grayscale image after each punching, and combining the morphological approximation and the dense fluctuation coefficient to determine the morphological evaluation value of each edge contour in the surface grayscale image after each punching comprises: The average of the distances between the centroid of each edge contour and the centroids of all other edge contours in the surface grayscale image after each punching is taken as the distribution coefficient of each edge contour; Calculate the discrete degree of the grayscale values ​​of all pixels within the minimum circumscribed rectangle of each edge contour in the surface grayscale image after each punching; The ratio of the discrete degree to the distribution coefficient is used as the intensive fluctuation coefficient of each edge contour in the surface grayscale image after each punching; The product of the morphological similarity and the dense fluctuation coefficient is used as the morphological evaluation value of each edge contour in the surface grayscale image after each punching.

5. A plate stamping control method using a heat-treated punch as claimed in claim 1, characterized in that: Determining the wrinkle cross coefficient of each edge contour in the surface grayscale image after each punching includes: The grayscale values ​​of edge pixels in the surface grayscale image after each punching are assigned to 1, and the grayscale values ​​of non-edge pixels are assigned to 0, to obtain a binary image, and the skeletons of all edge contours in the binary image are extracted, and the pixels on the skeletons are recorded as skeleton pixels; If there is only one skeleton pixel point in the eight-neighborhood of any skeleton pixel point on the skeleton corresponding to each edge contour, the any skeleton pixel point is recorded as a skeleton endpoint; if there are multiple skeleton pixels in the eight-neighborhood of any skeleton pixel point on the skeleton corresponding to each edge contour, the any skeleton pixel point is recorded as a skeleton intersection point; Count the number of all skeleton endpoints and skeleton intersections on the skeleton corresponding to each edge contour, and record it as the number of bifurcations; The minimum value of the distance between each skeleton endpoint and all skeleton intersection points on the skeleton corresponding to each edge contour is recorded as the closest distance; Calculate the discrete degree of the nearest distances of all skeleton endpoints on the skeleton corresponding to each edge contour, recorded as distance discreteness; The ratio of the number of bifurcations to the distance dispersion is used as the wrinkle cross coefficient of each edge contour in the surface grayscale image after each punching.

6. A plate stamping control method using a heat-treated punch as claimed in claim 1, characterized in that: The method of determining the wrinkle variation coefficient of each edge contour in the surface grayscale image after each punching comprises: The difference between the number of all edge contours in the grayscale image of the surface after each punching and the number of all edge contours in the grayscale image of the surface after the adjacent punching is recorded as a first quantity difference; The centroid coordinates of each edge contour in the grayscale image of the surface after each punching and the edge contour with the same centroid coordinates in the grayscale image of the surface after the adjacent punching are recorded as the relative edge contour of each edge contour; Recording the difference between the number of all edge pixels on each edge contour and the number of all edge pixels on the relative edge contour as a second quantity difference; The sum of the first quantity difference and the second quantity difference is used as the wrinkle variation coefficient of each edge contour in the surface grayscale image after each punching.

7. A plate stamping control method using a heat-treated punch as claimed in claim 1, characterized in that: The wrinkle interference coefficient of each edge contour in the surface grayscale image after each punching is the product of the wrinkle cross coefficient and the wrinkle variation coefficient.

8. A plate stamping control method using a heat-treated punch as claimed in claim 1, characterized in that: Determining the wrinkle discrimination coefficient of the surface grayscale image after each punching includes: The product of the morphological evaluation value and the wrinkling interference coefficient is used as a discriminant factor of each edge contour in the surface grayscale image after each punching; The average value of the discrimination factors of all edge contours in the surface grayscale image after each punching is used as the wrinkle discrimination coefficient of the surface grayscale image after each punching.

9. A plate stamping control method using a heat-treated punch as claimed in claim 1, characterized in that: The control and adjustment of the punching machine in the punching process includes: Performing threshold segmentation on the wrinkle discrimination coefficient of the surface grayscale image after all punchings before the current punching to obtain a segmentation threshold; If the wrinkle discrimination coefficient of the current stamping surface grayscale image is less than the segmentation threshold, the stamping machine is not adjusted. Otherwise, the stamping machine is controlled and adjusted through the PLC controller.

10. A sheet metal stamping control system using a heat-treated punch, 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 a sheet metal stamping control method using a heat-treated punch as described in any one of claims 1 to 9 are implemented.

Images