A visual rapid quality inspection method, device and system for automobile die-casting molds

By calculating the main spread coefficient and environmental impact factor of the crack area in the surface image of the automobile die-casting mold, and determining the burst coefficient, the problem of difficult to judge the source of the crack in the prior art is solved, and the optimized design and production process of automobile die-casting molds are realized. Effective improvement.

CN118758949BActive Publication Date: 2024-11-12HUIZHOU CAMEL DIE LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411203817.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-12
Estimated Expiration
2044-08-30

AI Technical Summary

Technical Problem

The prior art is difficult to effectively judge the source and cause of surface cracks in automobile die-casting molds, resulting in great limitations in production process optimization.

Method used

By obtaining the image of the surface of the automotive die-casting mold, detecting the crack area, calculating the main spread coefficient and environmental impact factor of each pixel point, and determining the source position of the crack in combination with the burst coefficient, thereby analyzing the cause of the crack.

Benefits of technology

The positioning of the source position of the crack is achieved, helping engineers analyze the causes of cracks, improve optimized design effects and efficiency, and reduce the limitations of the production process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118758949B_ABST
    Figure CN118758949B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data processing technology, and in particular to a visual rapid quality inspection method, device and system for automobile die-casting molds, the visual rapid quality inspection method for automobile die-casting molds comprising: acquiring an image of the surface of the automobile die-casting mold; detecting a crack area in the image; determining the main propagation coefficient of each pixel point in the crack area under stress, determining the environmental impact factor of each pixel point in the crack area according to the surrounding edges of the crack area and the surrounding area of ​​the crack area, the environmental impact factor indicating the degree of similarity between the crack area and the surrounding edge; determining the burst coefficient of each pixel point in the crack area using the main propagation coefficient and the environmental impact factor; determining the source position of the crack in the crack area from each pixel point according to the burst coefficient. Targeted optimization design of automobile die-casting molds improves the optimization effect and optimization efficiency, and further reduces the limitations of optimizing the production process of automobile die-casting molds.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a visual rapid quality inspection method, device and system for automobile die-casting molds. Background Art

[0002] As an efficient metal forming technology in modern manufacturing, die casting is widely used in the automobile manufacturing industry to produce a variety of key components including engine blocks, structural parts, and transmission parts. Die casting is a process of injecting molten metal into a pre-made automobile die-casting mold to solidify and form various parts. However, the design and manufacturing quality of automobile die-casting molds directly affect the quality and production stability of the final product. If the automobile die-casting mold is not designed reasonably or the structural strength is insufficient, it may cause excessive local stress on the surface, thereby causing cracks. Cracks will not only reduce the service life of automobile die-casting molds, but may also cause product defects, increase scrap rate, and affect production efficiency and cost control.

[0003] In some cases, in order to ensure the performance and accuracy of automobile die-casting molds, it is necessary to detect cracks on the surface of automobile die-casting molds, such as using machine vision technology to obtain surface images of automobile die-casting molds and automatically identify cracks on the surface of automobile die-casting molds, such as identifying and classifying the appearance characteristics of cracks such as length, width and shape. Although cracks on the surface of automobile die-casting molds can be identified, such an identification process often only stays on the surface and cannot effectively judge the source and cause of the cracks. There are great limitations in the optimization of production processes. Summary of the invention

[0004] In order to solve the technical problem of effectively judging the source and cause of cracks to reduce the limitations of optimizing the production process, the purpose of the present invention is to provide a visual rapid quality inspection method, device and system for automotive die-casting molds. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a visual rapid quality inspection method for an automobile die-casting mold, comprising: acquiring an image of the surface of the automobile die-casting mold; detecting a crack area in the image; determining a main propagation coefficient of each pixel point in the crack area under stress, wherein the main propagation coefficient is proportional to the degree of fluctuation of the crack in the crack area, and the main propagation coefficient is inversely proportional to the degree of bifurcation of the crack in the crack area; determining an environmental impact factor of each pixel point in the crack area according to the crack area and the surrounding edges of the surrounding area of ​​the crack area, wherein the environmental impact factor indicates the degree of proximity between the crack area and the surrounding edges, and the environmental impact factor is inversely proportional to the degree of proximity; determining an outbreak coefficient of each pixel point in the crack area using the main propagation coefficient and the environmental impact factor; and determining the source position of the crack in the crack area from each of the pixel points according to the outbreak coefficient.

[0006] Optionally, after determining the source position of the crack in the crack area from each of the pixel points according to the explosion coefficient, the method further comprises: determining the cause of the crack according to the region where the source position is located in the automobile die-casting mold.

[0007] Optionally, determining the cause of the crack according to the area of ​​the automobile die-casting mold where the source position is located includes: if the source position is in a flat area of ​​the automobile die-casting mold, determining that hard particle impurities in the production process of the automobile die-casting mold are the cause of the crack; if the source position is in a concave and convex area of ​​the automobile die-casting mold, determining that design defects and / or die-casting process parameters of the automobile die-casting mold are the cause of the crack; if the source position is in the same area of ​​the automobile die-casting mold for multiple times, determining that the degree of solidification of the same area of ​​the automobile die-casting mold is the cause of the crack.

[0008] Optionally, detecting the crack area in the image includes: detecting the crack edge in the image by an edge detection method; taking each pixel point of the crack edge as the center, selecting the crack area for each pixel point, and the crack area includes a current pixel point of the crack edge and multiple pixel points around the current pixel point.

[0009] Optionally, determining the main propagation coefficient of each pixel in the crack area under stress includes: determining the area of ​​the minimum circumscribed rectangle of the crack area where the current pixel is located, and a first number of pixel points on the edge of the crack area where the current pixel is located; determining the angle of the pixel points on the edge and a second number of pixel points in the crack area where the current pixel is located; determining the main extension coefficient of the current pixel based on the area, the first number, the second number and the angle, the main extension coefficient indicating the degree of fluctuation of the crack in the crack area, and the main extension coefficient is proportional to the degree of fluctuation of the crack in the crack area; extracting the morphological skeleton of the crack area where the current pixel is located, the morphological skeleton including multiple endpoints and intersections; determining the diffusion coefficient of the current pixel based on the number of endpoints and the number of intersections of the morphological skeleton, the diffusion coefficient indicating the degree of bifurcation of the crack in the crack area, and the diffusion coefficient is proportional to the degree of bifurcation of the crack in the crack area; determining the main propagation coefficient of the current pixel based on the main extension coefficient and the diffusion coefficient.

[0010] Optionally, determining the environmental impact factor of each pixel in the crack area according to the surrounding edges of the crack area and the surrounding areas of the crack area includes: extracting the morphological skeleton of the crack area where the pixel is located, and determining the two endpoints with the largest distance in the morphological skeleton; determining the shortest path between the two endpoints in the morphological skeleton as the trend line of the crack area; encoding the surrounding edges using edge chain codes to obtain chain codes; determining the difference between the chain codes of the surrounding edges adjacent to the chain code of the current surrounding edge, normalizing the difference to obtain the roughness of the surrounding edge; determining the distance between the trend line and the surrounding edge; and determining the environmental impact factor of the pixel in the crack area according to the number of the surrounding edges, the roughness, the trend line and the distance.

[0011] Optionally, determining the explosion coefficient of each pixel point in the crack area using the main propagation coefficient and the environmental influence factor includes: determining the product of the main propagation coefficient and the environmental influence factor as the explosion coefficient of the pixel point.

[0012] Optionally, determining the source position of cracks in the crack area from each pixel point according to the explosion coefficient includes: partitioning the image to obtain multiple areas; detecting the maximum explosion coefficient of the explosion coefficients in each area; respectively calculating the difference between the maximum explosion coefficient in each area and the explosion coefficients of pixel points around the pixel point corresponding to the maximum explosion coefficient in the area; taking the pixel point corresponding to the maximum explosion coefficient as the center, determining the change trend of the difference in the explosion coefficients from the center to the pixel points around the center; selecting the position of the pixel point corresponding to the maximum explosion coefficient in the area with the largest change trend as the source position.

[0013] In a second aspect, an embodiment of the present application provides a visual rapid quality inspection device for an automobile die-casting mold, comprising: an acquisition module for acquiring an image of the surface of the automobile die-casting mold; a detection module for detecting a crack area in the image; a determination module for determining a main propagation coefficient of each pixel point in the crack area under stress, wherein the main propagation coefficient is proportional to the degree of fluctuation of the crack in the crack area, and the main propagation coefficient is inversely proportional to the degree of bifurcation of the crack in the crack area; the determination module is also used to determine an environmental impact factor of each pixel point in the crack area according to the crack area and the surrounding edges of the surrounding area of ​​the crack area, wherein the environmental impact factor indicates the degree of proximity between the crack area and the surrounding edge, and the environmental impact factor is inversely proportional to the degree of proximity; the determination module is also used to determine an outbreak coefficient of each pixel point in the crack area using the main propagation coefficient and the environmental impact factor; the determination module is also used to determine the source position of the crack in the crack area from each of the pixels according to the outbreak coefficient.

[0014] In a third aspect, an embodiment of the present application provides a visual rapid quality inspection system for automotive die-casting molds, the visual rapid quality inspection system for automotive die-casting molds comprising: a processor and a memory; wherein the memory is used to store computer programs that can be run on the processor; the processor is used to execute the program stored in the memory to implement the steps of the visual rapid quality inspection method for automotive die-casting molds mentioned in the first aspect above.

[0015] The present invention has the following beneficial effects: by acquiring an image of the surface of an automobile die-casting mold, then detecting the crack area in the image, and then determining the main propagation coefficient of each pixel in the crack area under stress, the main propagation coefficient is proportional to the degree of fluctuation of the crack in the crack area, and the main propagation coefficient is inversely proportional to the degree of bifurcation of the crack in the crack area; according to the first crack edge of the crack area and the surrounding edge of the surrounding area of ​​the crack area, the environmental impact factor of each pixel in the crack area is determined, the environmental impact factor indicates the degree of proximity between the first crack edge and the surrounding edge, and the environmental impact factor is inversely proportional to the degree of proximity; the main propagation coefficient and the environmental impact factor are used to determine the explosion coefficient of each pixel in the crack area; according to the explosion coefficient, the source position of the crack in the crack area is determined from each pixel. In this way, the embodiment of the present application can locate the source position of the crack, thereby helping engineers analyze the cause of the crack, and optimize the design of the automobile die-casting mold in a targeted manner, thereby improving the optimization effect and optimization efficiency, and further reducing the limitations of optimizing the production process of the automobile die-casting mold. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A flowchart of a visual rapid quality inspection method for automotive die-casting molds provided by one embodiment of the present invention;

[0018] Figure 2 A schematic diagram of the module composition of a visual rapid quality inspection device for automobile die-casting molds provided by one embodiment of the present invention;

[0019] Figure 3 A schematic structural diagram of a visual rapid quality inspection system for automotive die-casting molds provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the visual rapid quality inspection method, device and system for automotive die-casting molds proposed by the present invention, 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.

[0021] 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 invention belongs.

[0022] The specific scheme of a visual rapid quality inspection method, device and system for automobile die-casting molds provided by the present invention is described in detail below in conjunction with the accompanying drawings.

[0023] Embodiment 1

[0024] See also Figure 1 , which shows a flow chart of a visual rapid quality inspection method for automotive die-casting molds provided by one embodiment of the present invention, comprising:

[0025] Step S101, acquiring an image of the surface of the automobile die-casting mold.

[0026] Specifically, the automotive die-casting molds used to manufacture automotive parts are usually composed of multiple mold parts, one part of which has the shape of the part and is called a mold, while the other part of the mold has the cavity shape of the part and is called a mold; in order to inspect the automotive die-casting mold, first fix the automotive die-casting mold to be inspected above the die-casting machine to ensure that it is aligned and securely fixed, put the pre-heated die-casting material into the injection cavity of the die-casting machine, use the hydraulic system to close the upper and lower molds, start the die-casting machine and apply sufficient pressure, keep the automotive die-casting mold closed after the die-casting is completed, and let the die-casting material cool and solidify; open the die-casting machine to separate the upper and lower molds and take out the finished product.

[0027] Among them, the surface of the automobile die-casting mold can be the inner surface of the automobile die-casting mold. When collecting the image of the surface of the automobile die-casting mold, the inner surface of the automobile die-casting mold can be cleaned first, and then a high-resolution camera is installed on the inner surface of the automobile die-casting mold to take a picture of the inner surface of the automobile die-casting mold to be inspected, thereby obtaining an image of the surface of the automobile die-casting mold.

[0028] Furthermore, for the image of the surface of the automobile die-casting mold obtained, since the analysis of cracks on the surface of the automobile die-casting mold is irrelevant to the color space, the collected image of the surface of the automobile die-casting mold is grayed to remove redundant information in the color space, thereby improving the calculation efficiency in subsequent calculations.

[0029] Step S103, detecting crack areas in the image.

[0030] Specifically, for automobile die-casting molds, cracks may appear on the internal surface of the automobile die-casting mold due to improper use or problems with the structure or production materials of the automobile die-casting mold itself. Therefore, it is necessary to detect and obtain the cracks on the surface of the automobile die-casting mold and obtain the causes of the cracks based on the crack characteristics so as to adjust the production or the mold material.

[0031] Furthermore, the cracks generated in automobile die-casting molds include tensile cracks, lightning cracks and tree-branch cracks. Tensile cracks refer to tensile cracks in automobile die-casting molds due to improper pressure in the die-casting process, which are caused by stress. Lightning cracks or tree-branch cracks refer to fracture-like cracks at the jagged grain boundaries caused by material problems in the automobile die-casting mold itself, which present lightning-shaped or tree-branch-shaped diffusion cracks.

[0032] The crack area refers to the area containing the above-mentioned various types of cracks generated on the inner surface of the automobile die-casting mold. The crack area can be determined based on the pixel points of the crack part. For example, an appropriate box is selected according to the image and the crack size, and the area within the box is the crack area.

[0033] In one possible implementation, detecting a crack area in an image includes: detecting the crack edge in the image by an edge detection method, taking each pixel point of the crack edge as the center, selecting a crack area for each pixel point, the crack area including a current pixel point of the crack edge and multiple pixel points around the current pixel point.

[0034] Specifically, for the pixel points on the crack, they can be screened according to the characteristics of the crack where the pixel points are located. For the pixel points on each crack, an appropriate box can be selected according to the image and the crack size, such as selecting a 9×9 pixel point box centered on the pixel points on each crack, and recording the crack within the 9×9 pixel point box as the crack area where the pixel points on each crack are located.

[0035] Step S105, determining the main propagation coefficient of each pixel point in the crack area under the action of stress.

[0036] Among them, the main propagation coefficient is proportional to the fluctuation degree of the crack in the crack area, and the main propagation coefficient is inversely proportional to the bifurcation degree of the crack in the crack area.

[0037] Specifically, the main propagation coefficient can show the propagation direction and degree of cracks, and the main propagation coefficient can be determined by the fluctuation and diffusion of cracks. For each pixel in the crack area, the crack will extend under the action of stress. The strong principal stress will lead to a higher crack roughness in the extension direction of the principal stress and the crack will fluctuate greatly. As the crack spreads, the higher the main propagation coefficient, the greater the degree of crack fluctuation. In addition, due to different crack generation conditions, the propagation and diffusion of the crack part are also different. The cracks generated by stress spread along the direction of stress propagation. The crack area generated by the principal stress will not bifurcate due to the strong stress, but will bifurcate as the stress propagates, forming a lightning-like crack. Therefore, due to the strong stress, the crack will have fewer bifurcations. Therefore, the main propagation coefficient is inversely proportional to the degree of crack bifurcation. Based on this, the extension and diffusion of the cracks at each pixel under stress can be used to determine the main propagation coefficient.

[0038] In one possible implementation, the main spreading coefficient can be determined as follows:

[0039] First, determine the area of ​​the minimum circumscribed rectangle of the crack region where the current pixel point is located, and the first number of pixel points on the edge of the crack region where the current pixel point is located; determine the angle of the pixel points on the edge and the second number of pixel points in the crack region where the current pixel point is located; determine the main extension coefficient of the current pixel point according to the area, the first number, the second number and the angle, the main extension coefficient indicates the degree of fluctuation of the crack in the crack region, and the main extension coefficient is proportional to the degree of fluctuation of the crack in the crack region.

[0040] Specifically, the edge caused by grain boundary fracture due to stress extension is a lightning-like edge. Since stress extension has a main extension direction and bifurcates due to material reasons, the crack in the main extension part is relatively rough due to the strong stress in the main extension part, while the bifurcated crack has a lower roughness due to the weak stress. Therefore, the main extension part of the stress will cause the overall crack to fluctuate greatly due to stress. Therefore, the minimum circumscribed rectangle is constructed for the crack area where each pixel point is located. Since the crack in the main stress part will have a large turning point and its edge is divergent outward, the smaller the pixel point of the crack area is compared with the crack; and for the pixel point on the edge of each crack area, the angle between the two fitting straight lines of the pixel point adjacent to it on its edge can be recorded as the angle of the pixel point on the edge of the crack. Since the pixel point of the crack in the main stress part has more turning points, the angle of the pixel point on its edge is smaller. The number of pixel points in the crack area where the current pixel point is located and the number of pixel points on the edge of the crack area where the current pixel point is located are counted. Therefore, the main extension coefficient of each pixel point in the crack is It can be calculated by the following formula:

[0041] ;

[0042] In the above formula, is any pixel point on the crack area, is any point on the edge of the crack area where the pixel of any crack is located. The crack area The main stretch coefficient of each pixel, The crack area The pixel point on the crack edge The angle of the pixel, The crack area The area of ​​the smallest circumscribed rectangle of the crack area where the pixel point is located, The crack area The number of pixels in the crack area where the pixel is located (the second number), is the hyperbolic tangent function, is the first number of pixels at the edge of the crack area where the current pixel is located, It is the area of ​​the minimum circumscribed rectangle of the crack area where the current pixel is located.

[0043] For the pixel points in each crack area, since the crack in the main stress extension part will be rougher and higher due to the strong main stress, and the crack will have greater fluctuations, the angle of the pixel points on the edge of the crack area where each pixel point is located is 180° The average of the difference The larger the value, the smaller the angle of the pixel points on the edge of the crack area, so the rougher the crack edge; and the ratio of the area of ​​the minimum circumscribed rectangle of the crack area to the number of pixels in the crack area is used. The larger the value, the larger the minimum circumscribed rectangle is compared to the crack area, which indicates that the crack corresponding to the circumscribed rectangle has greater tortuosity, so the greater the main elongation coefficient, and the degree of tortuosity of the crack is There is a positive correlation.

[0044] Then, the morphological skeleton of the crack area where the current pixel is located is extracted, and the morphological skeleton includes multiple endpoints and intersections. The diffusion coefficient of the current pixel is determined according to the number of endpoints and the number of intersections of the morphological skeleton. The diffusion coefficient indicates the degree of bifurcation of the crack in the crack area, and the diffusion coefficient is proportional to the degree of bifurcation of the crack in the crack area.

[0045] Specifically, due to different crack generation conditions, the cracks in the crack area spread differently. The cracks generated by stress spread in the direction of stress propagation. The main stress crack part will not bifurcate due to strong stress, but will bifurcate as the stress propagates, forming a lightning-like crack. Therefore, the diffusion coefficient of the crack in the main stress part is relatively small and there are fewer bifurcations. Therefore, in order to obtain the diffusion coefficient of the crack area where the pixel is located, the morphological skeleton of the crack area is first extracted using morphology, and the diffusion coefficient of each pixel in the crack area is solved according to the extracted morphological skeleton; the obtained morphological skeleton contains several endpoints and several intersections. For the lightning crack formed by grain boundary fracture, there will be more bifurcations, so its morphological skeleton will have more endpoints. The fewer the intersections, the more endpoints diverge from one intersection. Therefore, the diffusion coefficient of each pixel can be determined by the following formula:

[0046] ;

[0047] In the above formula, The crack region The diffusion coefficient of each pixel is The crack region The number of endpoints on the morphological skeleton of the crack area where the pixel is located, The crack region The number of intersection points on the morphological skeleton of the crack area where the pixel is located, is the hyperbolic tangent function.

[0048] For the morphological skeleton of the crack region where each pixel is located, the ratio of the number of endpoints to the number of intersections on the morphological skeleton of the crack region where the pixel is located is calculated. For endpoints, the larger the value, the more endpoints there are on the surface, and there are more bifurcations in this part of the crack, so the diffusion coefficient is higher, and for intersections, the larger the ratio, the more endpoints there are compared to the number of intersections, so there are more bifurcations and the higher the diffusion coefficient. Therefore, the diffusion coefficient is inversely proportional to the number of intersections and directly proportional to the number of endpoints.

[0049] Finally, the main spreading coefficient of the current pixel is determined according to the main extension coefficient and the diffusion coefficient.

[0050] Specifically, after determining the main extension coefficient and the diffusion coefficient, the product of the two can be used as the main spread coefficient, which is specifically expressed by the following formula:

[0051] ;

[0052] For the pixels in each crack region, since the main extension coefficient of the main stress part is larger and there are fewer bifurcations in the main extension part due to the strong stress, the diffusion coefficient The main spreading coefficient Negative correlation, main extension coefficient The main spreading coefficient There is a positive correlation.

[0053] Step S107, determining the environmental impact factor of each pixel in the crack area according to the crack area and the surrounding edges of the area surrounding the crack area.

[0054] Among them, the environmental impact factor indicates the proximity between the crack area and the surrounding edge, and the environmental impact factor is inversely proportional to the proximity.

[0055] Specifically, for the crack edges in the image, cracks caused by stress are usually located in key parts of the automobile die-casting mold, such as holes, bosses, inner corners, etc., and the cracks caused by stress due to stress propagation are located next to the key parts of the automobile die-casting mold, and their propagation direction is along the edge of the automobile die-casting mold; but due to material problems of the bicycle die-casting mold, stress will cause grain boundary fracture-like cracks when propagating, and will branch out from the main stress direction and propagate in random directions. Therefore, the environmental impact factor of each pixel point in the image is determined based on whether there is a key edge of the automobile die-casting mold with the same propagation direction around the crack area where each pixel point is located.

[0056] In a possible implementation, when determining the environmental impact factor, the following method may be used:

[0057] Extract the morphological skeleton of the crack area where the pixel is located, and determine the two endpoints with the largest distance in the morphological skeleton; determine the shortest path between the two endpoints in the morphological skeleton as the trend line of the crack area; use the edge chain code to encode the surrounding edges to obtain the chain code; determine the difference between the chain codes of the surrounding edges adjacent to the current chain code of the surrounding edges, normalize the difference and obtain the roughness of the surrounding edges; determine the distance between the trend line and the surrounding edges; determine the environmental impact factor of the pixel in the crack area according to the number, roughness, trend line and distance of the surrounding edges.

[0058] Specifically, morphology is used to obtain the morphological skeleton of the crack area where each pixel point in the automobile die-casting mold image is located, and the two endpoints with the largest distance in the morphological skeleton are obtained, and the shortest path of the two endpoints with the largest distance in the morphological skeleton is obtained and recorded as the trend line of the crack area. Since the trend of the crack caused by stress is similar to the edge of the surrounding part of the automobile die-casting mold, it is necessary to judge based on the edge of the surrounding area of ​​the crack area; the surrounding area of ​​the crack area is selected according to the size of the image and the size of the crack. In the embodiment of the present application, the part of 50×50 around the pixel point except the crack area is selected as the surrounding area of ​​the crack area where the pixel point is located, and the edge in the surrounding area is recorded as the surrounding edge of the crack area. The surrounding edge is encoded using an edge chain code to obtain a chain code, and the difference between adjacent chain codes is normalized to obtain the roughness of each surrounding edge. ; From this, the environmental impact factor coefficient of each pixel in the crack area can be obtained. Specifically, the environmental impact factor coefficient can be calculated using the following formula: :

[0059] ;

[0060] In the formula, is the surrounding edge of the crack area where any pixel point in the crack area is located, The crack region The trend line of the crack area where the pixel point is located, The crack region The area around the crack area where the pixel is located Around the edge of the strip, is the number of edges around the crack region where any pixel point is located. is the environmental impact factor coefficient.

[0061] For the surrounding edge of the crack area where each pixel is located, the roughness of the pixel in the surrounding area The lower the roughness, the smoother the surrounding edge. Since the image contains not only the edge of the crack area, but also the texture on the automobile die-casting mold and the edge of the mold line, the lower the roughness, the more the surrounding edge belongs to the line edge of the automobile die-casting mold. Since the extension direction of the principal stress is the same as the line on the automobile die-casting mold, the DTW value between the surrounding edge and the trend line of the crack area obtained by the dynamic time warping (DTW) algorithm is The smaller the value, the closer the trend of the surrounding edge is to the crack area, and the distance between the trend line of the crack area and the surrounding edge is The smaller the value, the closer the surrounding edge is to the crack area, so the higher the environmental impact factor is. is a natural exponential function.

[0062] Step S109, determining the explosion coefficient of each pixel in the crack area using the main propagation coefficient and the environmental influence factor.

[0063] Specifically, for the environmental impact factor and the main spreading coefficient of each pixel point obtained by the above operation, since the main spreading coefficient and the environmental impact factor in the main stress part are relatively high, the source point of the stress can be obtained through the burst coefficient.

[0064] Specifically, the product of the main spreading coefficient and the environmental impact factor can be determined as the burst coefficient of the pixel point, which is calculated using the following formula:

[0065] ;

[0066] The explosion coefficient of each pixel in the crack area is calculated by the above formula .

[0067] Step S1011, determining the source position of the crack in the crack area from each pixel point according to the explosion coefficient.

[0068] Specifically, the source position refers to the position where the stress explosion coefficient of the stress action point is the highest, or the position where the stress explosion coefficient of the stress action point is the highest and the explosion coefficients of surrounding pixel points gradually decrease.

[0069] When determining the source position, the following method can be used: determining the source position of the crack in the crack area from each pixel point according to the explosion coefficient includes: partitioning the image to obtain multiple areas; detecting the maximum explosion coefficient of the explosion coefficient in each area; respectively calculating the difference between the maximum explosion coefficient in each area and the explosion coefficient of the pixel points around the pixel point corresponding to the maximum explosion coefficient in the area; taking the pixel point corresponding to the maximum explosion coefficient as the center, determining the change trend of the difference in the explosion coefficient from the center to the pixel points around the center; selecting the position of the pixel point corresponding to the maximum explosion coefficient in the area with the largest change trend as the source position.

[0070] Specifically, since the stress explosion coefficient at the stress action point is the highest and the explosion coefficient of the surrounding stress will decay, the stress explosion coefficient map of each area in the image is obtained according to the stress explosion coefficient, and the maximum point (maximum explosion coefficient) of the local stress explosion coefficient in each area is obtained according to the change of the explosion coefficient. The difference between the explosion coefficient of each maximum point and the adjacent pixel point is calculated, and then a point with the highest explosion coefficient and the largest attenuation of the difference in the explosion coefficient of the surrounding pixel points is determined, and marked as the stress explosion origin, that is, the explosion origin is located, that is, the source position of the crack generation and propagation, and it is marked.

[0071] According to the technical solution disclosed in the embodiment of the present application, by acquiring an image of the surface of the automobile die-casting mold, and then detecting the crack area in the image, the main propagation coefficient of each pixel in the crack area under stress is determined, and the main propagation coefficient is proportional to the degree of fluctuation of the crack in the crack area, and the main propagation coefficient is inversely proportional to the degree of bifurcation of the crack in the crack area; the environmental impact factor of each pixel in the crack area is determined according to the first crack edge of the crack area and the surrounding edge of the surrounding area of ​​the crack area, and the environmental impact factor indicates the degree of proximity between the first crack edge and the surrounding edge, and the environmental impact factor is inversely proportional to the degree of proximity; the main propagation coefficient and the environmental impact factor are used to determine the burst coefficient of each pixel in the crack area; the source position of the crack in the crack area is determined from each pixel according to the burst coefficient. In this way, the embodiment of the present application can locate the source position of the crack, thereby helping engineers analyze the causes of the cracks, and optimize the design of the automobile die-casting mold in a targeted manner, thereby improving the optimization effect and optimization efficiency, and further reducing the limitations of the production process of optimizing the automobile die-casting mold.

[0072] After determining the source position of the crack in the crack area from each pixel point according to the explosion coefficient, the method further includes: determining the cause of the crack according to the region where the source position is located in the automobile die-casting mold.

[0073] Specifically, after determining the source location, the cause of the crack is analyzed based on the area of ​​the source location of the crack in the automobile die-casting mold, that is, the cause of the crack will be different if the source location is in different areas of the automobile die-casting mold. Among them, if the source location is in the flat area of ​​the automobile die-casting mold, the hard particle impurities in the production process of the automobile die-casting mold are determined to be the cause of the crack; if the source location is in the concave and convex area of ​​the automobile die-casting mold, the design defects and / or die-casting process parameters of the automobile die-casting mold are determined to be the cause of the crack; if the source location is in the same area of ​​the automobile die-casting mold for multiple times, the solidification degree in the same area of ​​the automobile die-casting mold is determined to be the cause of the crack.

[0074] That is to say, when the source position is in the flat area of ​​the automobile die-casting mold (there is no bump), it can be judged that the crack is most likely caused by hard particle impurities in the quenching process or the die-casting process. If the source position is in the bumpy area of ​​the automobile die-casting mold, it can be judged that the crack is most likely caused by design defects of the automobile die-casting mold or improper die-casting process parameters. If the source position of the crack often appears in a certain position (the same area), it is necessary to reinforce or optimize the design of the specific position on the automobile die-casting mold. In addition, according to the location where the cracks often appear, the attention and quality inspection standards of the specific position can be increased to ensure that the key control points in the production process are effectively monitored. After determining the cause of the crack, the automobile die-casting mold can be optimized and designed in a targeted manner, which improves the optimization effect and efficiency, and further reduces the limitations of optimizing the production process of automobile die-casting molds.

[0075] Embodiment 2

[0076] Corresponding to the visual rapid quality inspection method for automobile die-casting molds provided in the above embodiment, based on the same technical concept, the embodiment of the present application also provides a visual rapid quality inspection device for automobile die-casting molds, Figure 2 Schematic diagram of the module composition of the visual rapid quality inspection device for automobile die-casting molds provided in the embodiment of the present application, which is used to perform Figure 1 The automotive die casting mold described uses a visual rapid quality inspection method, such as Figure 2 As shown, the visual rapid quality inspection device 200 for automobile die-casting molds includes: an acquisition module 201, which is used to acquire an image of the surface of the automobile die-casting mold; a detection module 202, which is used to detect the crack area in the image; a determination module 203, which is used to determine the main propagation coefficient of each pixel point in the crack area under stress, the main propagation coefficient is proportional to the fluctuation degree of the crack in the crack area, and the main propagation coefficient is inversely proportional to the bifurcation degree of the crack in the crack area; the determination module 203 is also used to determine the environmental impact factor of each pixel point in the crack area according to the surrounding edges of the crack area and the surrounding area of ​​the crack area, the environmental impact factor indicates the degree of proximity between the crack area and the surrounding edge, and the environmental impact factor is inversely proportional to the degree of proximity; the determination module 203 is also used to determine the outbreak coefficient of each pixel point in the crack area using the main propagation coefficient and the environmental impact factor; the determination module 203 is also used to determine the source position of the crack in the crack area from each pixel point according to the outbreak coefficient.

[0077] According to the technical solution disclosed in the embodiment of the present application, by acquiring an image of the surface of the automobile die-casting mold, and then detecting the crack area in the image, the main propagation coefficient of each pixel in the crack area under stress is determined, and the main propagation coefficient is proportional to the degree of fluctuation of the crack in the crack area, and the main propagation coefficient is inversely proportional to the degree of bifurcation of the crack in the crack area; the environmental impact factor of each pixel in the crack area is determined according to the first crack edge of the crack area and the surrounding edge of the surrounding area of ​​the crack area, and the environmental impact factor indicates the degree of proximity between the first crack edge and the surrounding edge, and the environmental impact factor is inversely proportional to the degree of proximity; the main propagation coefficient and the environmental impact factor are used to determine the burst coefficient of each pixel in the crack area; the source position of the crack in the crack area is determined from each pixel according to the burst coefficient. In this way, the embodiment of the present application can locate the source position of the crack, thereby helping engineers analyze the causes of the cracks, and optimize the design of the automobile die-casting mold in a targeted manner, thereby improving the optimization effect and optimization efficiency, and further reducing the limitations of the production process of optimizing the automobile die-casting mold.

[0078] It should be noted that the visual rapid quality inspection device for automobile die-casting molds provided in the embodiment of the present application and the visual rapid quality inspection method for automobile die-casting molds provided in the embodiment of the present application are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned visual rapid quality inspection method for automobile die-casting molds, and has the same or similar beneficial effects, and the repeated parts will not be repeated.

[0079] Embodiment 3

[0080] Corresponding to the visual rapid quality inspection method for automobile die-casting molds provided in the above embodiment, based on the same technical concept, the embodiment of the present application also provides a visual rapid quality inspection system for automobile die-casting molds, which is used to execute the visual rapid quality inspection method for automobile die-casting molds. Figure 3 A schematic diagram of a structure of a visual rapid quality inspection system for automotive die-casting molds for implementing each embodiment of the present application is shown as follows: Figure 3 The visual rapid quality inspection system for automobile die casting molds may have relatively large differences due to different configurations or performances, and may include one or more processors 301 and memory 302, the memory 302 is used to store computer programs that can be run on the processor, and the processor is used to execute the program stored in the memory to achieve the above Figure 1 The memory 302 may be a temporary storage or a permanent storage. The application stored in the memory 302 may include one or more modules (not shown in the figure), each of which may include a series of computer executable instructions in the visual rapid quality inspection system for automotive die casting molds.

[0081] Furthermore, the processor 301 can be configured to communicate with the memory 302, and execute a series of computer executable instructions in the memory 302 on the visual rapid quality inspection system for automotive die casting molds. The visual rapid quality inspection system for automotive die casting molds can also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input and output interfaces 305, and one or more keyboards 306.

[0082] Specifically in this embodiment, the visual rapid quality inspection system for automobile die-casting molds includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to achieve the above Figure 1 The various steps in the method embodiment are similar to those in the method embodiment, and have the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present application will not be described again here.

[0083] It should be noted that the visual rapid quality inspection system for automobile die-casting molds provided in the embodiment of the present application and the visual rapid quality inspection method for automobile die-casting molds provided in the embodiment of the present application are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned visual rapid quality inspection method for automobile die-casting molds, and has the same or similar beneficial effects, and the repeated parts will not be repeated.

[0084] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. 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.

[0085] The various embodiments in this specification 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.

Claims

1. A visual rapid quality inspection method for automobile die-casting molds, characterized in that: The visual rapid quality inspection method for automobile die-casting molds comprises: Acquire images of automotive die-casting mold surfaces; detecting crack regions in the image; Determine a main propagation coefficient of each pixel point in the crack region under stress, wherein the main propagation coefficient is proportional to the degree of fluctuation of the crack in the crack region, and the main propagation coefficient is inversely proportional to the degree of bifurcation of the crack in the crack region; Determining an environmental impact factor of each pixel in the crack region according to the crack region and the surrounding edge of the surrounding region of the crack region, wherein the environmental impact factor indicates the degree of similarity between the crack region and the surrounding edge, and the environmental impact factor is inversely proportional to the degree of similarity; Determine the explosion coefficient of each pixel point in the crack area by using the main propagation coefficient and the environmental influence factor; Determine the source position of the crack in the crack area from each of the pixel points according to the explosion coefficient; After determining the source position of the crack in the crack area from each of the pixel points according to the explosion coefficient, the method further includes: Determining the cause of the crack according to the source location being in the area of ​​the automobile die-casting mold; The detecting a crack region in the image comprises: detecting crack edges in the image by an edge detection method; Taking each pixel point of the crack edge as the center, selecting the crack region for each pixel point, wherein the crack region includes a current pixel point of the crack edge and a plurality of pixel points around the current pixel point; Determining the main propagation coefficient of each pixel point in the crack region under stress includes: Determine the area of ​​the minimum circumscribed rectangle of the crack region where the current pixel point is located, and a first number of pixel points at the edge of the crack region where the current pixel point is located; Determine the angle of the pixel point of the edge and the second number of pixel points in the crack area where the current pixel point is located, and record the angle between two fitted straight lines on the edge of the pixel point of the edge and the pixel points adjacent to it as the angle of the pixel point of the crack edge; The main extension coefficient of the current pixel is determined according to the area, the first number, the second number and the angle, the main extension coefficient indicating the degree of fluctuation of the crack in the crack region, and the main extension coefficient is proportional to the degree of fluctuation of the crack in the crack region, and the main extension coefficient ZS i Calculate using the following formula: ; In the above formula, i is any pixel point on the crack area, j is any point on the edge of the crack area where the pixel point of any crack is located, and ZS i is the principal extension coefficient of the i-th pixel point on the crack area, is the angle of the jth pixel on the crack edge where the i-th pixel in the crack area is located, S i is the area of ​​the minimum circumscribed rectangle of the crack area where the i-th pixel point is located, is the second number of pixels in the crack area where the i-th pixel is located, is a hyperbolic tangent function, and n is the first number of pixel points at the edge of the crack region where the current pixel point is located; Extracting a morphological skeleton of the crack region where the current pixel point is located, wherein the morphological skeleton includes a plurality of endpoints and intersections; The diffusion coefficient of the current pixel is determined according to the number of endpoints and the number of intersections of the morphological skeleton, and the diffusion coefficient indicates the degree of bifurcation of the crack in the crack region, and the diffusion coefficient is proportional to the degree of bifurcation of the crack in the crack region. i Calculate using the following formula: ; In the above formula, MY i is the diffusion coefficient of the i-th pixel in the crack region, d i is the number of endpoints on the morphological skeleton of the crack region where the i-th pixel point is located, h i is the number of intersection points on the morphological skeleton of the crack region where the i-th pixel point is located, is the hyperbolic tangent function; The main spreading coefficient of the current pixel is determined according to the main extension coefficient and the diffusion coefficient. i The calculation is done by the following formula: ; The step of determining the environmental impact factor of each pixel in the crack area according to the crack area and the surrounding edges of the surrounding area of ​​the crack area comprises: Extracting a morphological skeleton of the crack region where the pixel point is located, and determining two endpoints with the largest distance in the morphological skeleton; Determining the shortest path between the two endpoints in the morphological skeleton as a trend line of the crack region; Encoding the surrounding edges using edge chain codes to obtain chain codes; Determine the difference between the chain codes of the surrounding edges adjacent to the current chain code of the surrounding edges, normalize the difference, and obtain the roughness c of the surrounding edges. i,x ; determining a distance between the trend line and the surrounding edge; The environmental impact factor of the pixel point in the crack area is determined according to the number of the surrounding edges, the roughness, the trend line and the distance. The environmental impact factor WZ i The following formula is used for calculation: ; Where x is the edge of the crack region where any pixel point is located, and q is i is the trend line of the crack area where the i-th pixel point is located, b i,x is the xth surrounding edge of the crack region where the i-th pixel point in the crack region is located, m is the number of surrounding edges of the crack region where any pixel point in the crack region is located, is the DTW value between the surrounding edge and the trend line of the crack area obtained according to the dynamic time warping algorithm; The method of determining the explosion coefficient of each pixel point in the crack area by using the main propagation coefficient and the environmental influence factor includes: Determine the product of the main spreading coefficient and the environmental impact factor as the burst coefficient of the pixel point; Determining the source position of the crack in the crack area from each of the pixel points according to the explosion coefficient includes: Partitioning the image to obtain multiple regions; Detecting the maximum burst coefficient of the burst coefficients in each of the regions; Calculating the difference between the maximum burst coefficient in each of the regions and the burst coefficients of the pixel points around the pixel point corresponding to the maximum burst coefficient in the region; Taking the pixel point corresponding to the maximum burst coefficient as the center, determining the change trend of the difference of the burst coefficients of the pixel points starting from the center to the surrounding pixel points of the center; The position of the pixel point corresponding to the maximum burst coefficient of the area with the largest change trend is selected as the source position.

2. The visual rapid quality inspection method for automobile die-casting molds according to claim 1 is characterized in that: Determining the cause of the crack according to the source location being in the area of ​​the automobile die-casting mold includes: If the source position is in a flat area of ​​the automobile die-casting mold, it is determined that hard particle impurities in the production process of the automobile die-casting mold are the cause of the crack; If the source position is in the concave-convex area of ​​the automobile die-casting mold, it is determined that the design defects of the automobile die-casting mold and / or the die-casting process parameters are the cause of the crack; If the source position is located in the same area of ​​the automobile die-casting mold for multiple times, it is determined that the solidification degree in the same area of ​​the automobile die-casting mold is the cause of the crack.

3. A visual rapid quality inspection device for automobile die-casting molds, characterized in that: The visual rapid quality inspection device for automobile die-casting molds comprises: An acquisition module, used for acquiring an image of the surface of the automobile die-casting mold; A detection module, used for detecting a crack area in the image; A determination module, used to determine a main propagation coefficient of each pixel point in the crack region under stress, wherein the main propagation coefficient is proportional to the degree of fluctuation of the crack in the crack region, and the main propagation coefficient is inversely proportional to the degree of bifurcation of the crack in the crack region; The determination module is further used to determine an environmental impact factor of each pixel in the crack area according to the crack area and the surrounding edge of the surrounding area of ​​the crack area, wherein the environmental impact factor indicates the degree of similarity between the crack area and the surrounding edge, and the environmental impact factor is inversely proportional to the degree of similarity; The determination module is further used to determine the explosion coefficient of each pixel point in the crack area by using the main propagation coefficient and the environmental influence factor; The determination module is further used to determine the source position of the crack in the crack area from each of the pixel points according to the explosion coefficient, and after determining the source position of the crack in the crack area from each of the pixel points according to the explosion coefficient, determine the cause of the crack according to the source position being in the area of ​​the automobile die-casting mold; The detecting a crack region in the image comprises: detecting crack edges in the image by an edge detection method; Taking each pixel point of the crack edge as the center, selecting the crack region for each pixel point, wherein the crack region includes a current pixel point of the crack edge and a plurality of pixel points around the current pixel point; Determining the main propagation coefficient of each pixel point in the crack region under stress includes: Determine the area of ​​the minimum circumscribed rectangle of the crack region where the current pixel point is located, and a first number of pixel points at the edge of the crack region where the current pixel point is located; Determine the angle of the pixel point of the edge and the second number of pixel points in the crack area where the current pixel point is located, and record the angle between two fitted straight lines on the edge of the pixel point of the edge and the pixel points adjacent to it as the angle of the pixel point of the crack edge; The main extension coefficient of the current pixel is determined according to the area, the first number, the second number and the angle, the main extension coefficient indicating the degree of fluctuation of the crack in the crack region, and the main extension coefficient is proportional to the degree of fluctuation of the crack in the crack region, and the main extension coefficient ZS i Calculate using the following formula: ; In the above formula, i is any pixel point on the crack area, j is any point on the edge of the crack area where the pixel point of any crack is located, and ZS i is the principal extension coefficient of the i-th pixel point on the crack area, is the angle of the jth pixel on the crack edge where the i-th pixel in the crack area is located, S i is the area of ​​the minimum circumscribed rectangle of the crack area where the i-th pixel point is located, is the second number of pixels in the crack area where the i-th pixel is located, is a hyperbolic tangent function, and n is the first number of pixel points at the edge of the crack region where the current pixel point is located; Extracting a morphological skeleton of the crack region where the current pixel point is located, wherein the morphological skeleton includes a plurality of endpoints and intersections; The diffusion coefficient of the current pixel is determined according to the number of endpoints and the number of intersections of the morphological skeleton, and the diffusion coefficient indicates the degree of bifurcation of the crack in the crack region, and the diffusion coefficient is proportional to the degree of bifurcation of the crack in the crack region. i Calculate using the following formula: ; In the above formula, MY i is the diffusion coefficient of the i-th pixel in the crack region, d i is the number of endpoints on the morphological skeleton of the crack region where the i-th pixel point is located, h i is the number of intersection points on the morphological skeleton of the crack region where the i-th pixel point is located, is the hyperbolic tangent function; The main spreading coefficient of the current pixel is determined according to the main extension coefficient and the diffusion coefficient. i The calculation is done by the following formula: ; The step of determining the environmental impact factor of each pixel in the crack area according to the crack area and the surrounding edges of the surrounding area of ​​the crack area comprises: Extracting a morphological skeleton of the crack region where the pixel point is located, and determining two endpoints with the largest distance in the morphological skeleton; Determining the shortest path between the two endpoints in the morphological skeleton as a trend line of the crack region; Encoding the surrounding edges using edge chain codes to obtain chain codes; Determine the difference between the chain codes of the surrounding edges adjacent to the current chain code of the surrounding edges, normalize the difference, and obtain the roughness of the surrounding edges ; determining a distance between the trend line and the surrounding edge; The environmental impact factor of the pixel point in the crack area is determined according to the number of the surrounding edges, the roughness, the trend line and the distance. The environmental impact factor WZ i The following formula is used for calculation: ; Where x is the edge of the crack region where any pixel point is located, and q is i is the trend line of the crack area where the i-th pixel point is located, b i,x is the xth surrounding edge of the crack region where the i-th pixel point in the crack region is located, m is the number of surrounding edges of the crack region where any pixel point in the crack region is located, is the DTW value between the surrounding edge and the trend line of the crack area obtained according to the dynamic time warping algorithm; The method of determining the explosion coefficient of each pixel point in the crack area by using the main propagation coefficient and the environmental influence factor includes: Determine the product of the main spreading coefficient and the environmental impact factor as the burst coefficient of the pixel point; Determining the source position of the crack in the crack area from each of the pixel points according to the explosion coefficient includes: Partitioning the image to obtain multiple regions; Detecting the maximum burst coefficient of the burst coefficients in each of the regions; Calculating the difference between the maximum burst coefficient in each of the regions and the burst coefficients of the pixel points around the pixel point corresponding to the maximum burst coefficient in the region; Taking the pixel point corresponding to the maximum burst coefficient as the center, determining the change trend of the difference of the burst coefficients of the pixel points starting from the center to the surrounding pixel points of the center; The position of the pixel point corresponding to the maximum burst coefficient of the area with the largest change trend is selected as the source position.

4. A visual rapid quality inspection system for automotive die-casting molds, characterized in that: include: A processor and a memory; wherein the memory is used to store a computer program that can be run on the processor; The processor is used to execute the program stored in the memory to implement the steps of the visual rapid quality inspection method for automobile die-casting molds as described in any one of claims 1-2.