Sealing cover mold wear detection method based on machine vision

Abstract

The invention belongs to the technical field of image processing, and particularly relates to a sealing cover mold wear detection method based on machine vision, which comprises the following steps: shooting a plurality of images of a standard sealing cover mold during horizontal rotation, and calculating the wear of the standard sealing cover mold according to the difference between gray values of the same mold at different spatial positions; determining the gray significance of the mold position at different spatial positions, and determining the gray influence degree of the light of the spatial position on the different mold positions according to the difference between the gray significance of the different mold positions at the same spatial position; obtaining a corrected gray value of the mold position according to the gray influence degree of light rays at different spatial positions on the same mold position, so as to reconstruct a standard image without illumination influence; and comparing the standard image with the image without illumination influence of the to-be-detected sealing cover mold to realize wear detection of the to-be-detected sealing cover mold. Illumination interference is eliminated, and abrasion detection of the sealing cover mold is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing. More specifically, the present invention relates to a method for detecting wear of a sealing cover mold based on machine vision. Background Art

[0002] With the continuous development of modern manufacturing, molds play an important role in product production. Especially for complex-shaped parts such as sealing covers, the use accuracy of the mold directly affects production efficiency and product quality. During long-term use, due to repeated high-intensity machining, impact, and friction, molds are prone to wear. Wear not only leads to a decrease in the dimensional accuracy of products but may also cause problems such as surface cracks and defects in the mold, ultimately affecting production stability and economy. Therefore, the detection of mold wear becomes particularly important.

[0003] With the improvement of industrial automation, detection technologies based on machine vision have gradually become an important direction for mold wear monitoring. For example, Chinese patent documents with the authorization announcement numbers CN118196081B and CN116912261B both use machine vision means to detect defects on the mold surface.

[0004] The method for detecting mold wear based on machine vision mainly relies on the analysis of the mold surface image. By using image processing algorithms to extract the characteristics of the worn area, the wear degree is then judged. Since the sealing cover mold is a metal mold, when detecting the wear of the sealing cover mold surface using machine vision, there is reflection or shadow on the mold surface in the captured image of the sealing cover mold, which affects the accuracy of detecting the wear of the sealing cover mold.

[0005] Therefore, there is an urgent need for an efficient and accurate method for detecting wear of a sealing cover mold based on machine vision to improve the accuracy of detecting the wear of the sealing cover mold. Summary of the Invention

[0006] To solve the technical problem that the reflection or shadow on the mold surface in the sealing cover mold image affects the accuracy of detecting the wear of the sealing cover mold, the present invention provides a method for detecting wear of a sealing cover mold based on machine vision, including: Take multiple images when the standard seal cover mold rotates horizontally. Reconstruct a standard image without the influence of light according to the multiple images of the standard seal cover mold, including: the same mold position of the standard seal cover mold is located at different spatial positions in different images. Determine the gray significance of the mold position at different spatial positions according to the difference between the gray values of the same mold position at different spatial positions; determine the influence degree of the light at this spatial position on the gray values of different mold positions according to the difference between the gray significances of different mold positions at the same spatial position; determine the corrected gray value of this mold position according to the influence degree of the light at different spatial positions on the gray value of the same mold position and the gray value of this mold position at different spatial positions; reconstruct an image according to the corrected gray values of all mold positions to obtain a standard image; take multiple images when the seal cover mold to be detected rotates horizontally, and reconstruct an image to be detected without the influence of light according to the multiple images of the seal cover mold to be detected; compare the standard image with the image to be detected to identify the wear condition of the seal cover mold to be detected.

[0007] The effects are as follows: By reconstructing the standard image and the image to be detected, the present invention eliminates the interference of light changes, makes the image analysis more accurate, reduces the influence of environmental factors on the wear detection of the seal cover, and improves the reliability of the detection; The present invention analyzes based on the gray significance of different mold positions at the same spatial position, can accurately capture the subtle differences on the mold surface, provides high-precision image data for subsequent wear detection, and ensures that even small changes in wear can be detected; The present invention obtains the corrected gray value of this mold position according to the influence degree of the light at different spatial positions on the gray value of the same mold position, can remove the influence of light on this mold position, makes the reconstructed standard image and the image to be detected more accurate, and through the analysis of the standard image and the image to be detected, can accurately reflect the wear conditions of different positions of the seal cover mold to be detected, ensuring the accuracy of wear detection. Especially when dealing with complex mold structures, it can effectively identify early wear and minor defects.

[0008] Preferably, the gray significance satisfies the expression: ; where represents the gray significance of the target mold position at the th spatial position, and the target mold position is any position on the standard seal cover mold; represents the gray value of the pixel point corresponding to the target mold position at the th spatial position; represents the th gray value in the gray sequence of the target mold position; , respectively represent the maximum gray value and the minimum gray value in the gray sequence of the target mold position; is the length of the gray - level sequence at the target mold position; is the absolute - value symbol; the gray - level sequence is a sequence composed of gray - level values at the target mold position at different spatial positions.

[0009] Its effect is that by calculating the gray - level saliency of the target mold position at different spatial positions, it can effectively measure the gray - level changes of the target mold position at different angles and positions, and can provide more stable and reliable gray - level data for subsequent image reconstruction and wear detection, improving the detection accuracy and robustness.

[0010] Preferably, the gray - level influence degree satisfies the expression: ; where represents the gray - level influence degree of the light at the th spatial position on the target mold position; represents the gray - level saliency of the target mold position at the th spatial position; represents the th gray - level saliency in the gray - level saliency sequence at the th spatial position; and respectively represent the maximum and minimum gray - level saliencies in the gray - level saliency sequence at the th spatial position; represents the length of the gray - level saliency sequence at the th spatial position; the gray - level saliency sequence is a sequence composed of the gray - level saliencies of all mold positions passing through the th spatial position at the th spatial position.

[0011] Its effect is that by calculating the gray - level influence degree of the light at different spatial positions on the target mold position, the influence of illumination on the gray - level value of the mold image is effectively considered. By comparing the gray - level saliencies of different mold positions at different spatial positions, the influence of illumination on different structures of the mold is quantified, so as to further adjust the gray - level data of each mold position subsequently, making the final image still maintain consistency under illumination changes, thereby improving the accuracy of image reconstruction and the reliability of detection results, especially providing a more stable and accurate detection basis in a complex illumination environment.

[0012] Preferably, the corrected gray - level value satisfies the expression: ; where represents the corrected gray - level value of the target mold position; represents the gray - level influence degree of the light at the th spatial position on the target mold position; Indicates the grayscale value of the pixel corresponding to the target mold position at the th spatial position; Indicates the length of the grayscale sequence of the target mold position, and the grayscale sequence is a sequence composed of grayscale values of the target mold position at different spatial positions; Indicates the exponential function with the natural constant as the base.

[0013] The effect is that: the present invention fully considers the degree of light influence, weights the grayscale values of the target mold position at different spatial positions, and the weight is determined by the degree of influence of the light at different spatial positions on the grayscale of the target mold position, so that the finally obtained corrected grayscale value can better reflect the true image features without light influence, helps to reduce the interference of light changes on the image quality, improves the stability and accuracy of the image, and enhances the reliability of subsequent wear detection of the sealing cover mold.

[0014] Preferably, reconstructing an image according to the corrected grayscale values of all mold positions to obtain a standard image includes: arranging the corrected grayscale values of all mold positions according to the positions corresponding to all mold positions in the first captured image to form a new image as the standard image.

[0015] Preferably, comparing the standard image with the image to be detected to identify the wear condition of the sealing cover mold to be detected includes: performing template matching on the standard image and the image to be detected; determining the defect possibility of each pixel point in the mold area of the image to be detected according to the difference between each pixel point in the mold area of the image to be detected and the corresponding matching pixel point in the standard image; screening the worn area according to the size of the defect possibility.

[0016] Preferably, the defect possibility satisfies the expression: ; where represents the defect possibility of the target pixel point, and the target pixel point is any pixel point in the mold area of the image to be detected; represents the grayscale value of the target pixel point; represents the grayscale value of the matching pixel point of the target pixel point in the standard image; represents the range of grayscale values in the mold area of the standard image; represents the grayscale value of the th pixel point within the neighborhood range of the target pixel point; represents the grayscale value of the matching pixel point of the th pixel point within the neighborhood range of the target pixel point in the standard image; represents the size of the neighborhood range of the target pixel point; is a hyperparameter; is the natural constant An exponential function with a base of...

[0017] Preferably, screening the worn area according to the magnitude of the defect possibility includes: taking the area composed of all pixel points with a defect possibility greater than a preset wear threshold as the worn area.

[0018] The beneficial effects of the present invention are as follows: By reconstructing the standard image and the image to be detected, the present invention eliminates the interference of illumination changes, making image analysis more accurate, reducing the influence of environmental factors on the wear detection of the sealing cover, and improving the reliability of detection; The present invention analyzes based on the gray-scale significance at the same spatial position for different mold positions, can accurately capture the subtle differences on the mold surface, provides high-precision image data for subsequent wear detection, and ensures that even small changes in wear can be detected; The present invention obtains the corrected gray-scale value of the mold position according to the degree of influence of the light at different spatial positions on the gray scale of the same mold position, can remove the influence of illumination on this mold position, makes the reconstructed standard image and the image to be detected more accurate, and through the analysis of the standard image and the image to be detected, can accurately reflect the wear conditions of different positions of the sealing cover mold to be detected, ensuring the accuracy of wear detection. Especially when dealing with complex mold structures, it can effectively identify early wear and minor defects. Description of the Drawings

[0019] By reading the following detailed description with reference to the drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 is a flowchart schematically showing a method for detecting wear of a sealing cover mold based on machine vision in the present invention; Figure 2 is a first schematic diagram schematically showing an image of a standard sealing cover mold taken during rotation; Figure 3 is a second schematic diagram schematically showing an image of a standard sealing cover mold taken during rotation; Figure 4 is a third schematic diagram schematically showing an image of a standard sealing cover mold taken during rotation; Figure 5 is a flowchart schematically showing step S2 of a method for detecting wear of a sealing cover mold based on machine vision in the present invention. Detailed Embodiments

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0021] The following will describe in detail the specific implementation manners of the present invention in conjunction with the accompanying drawings.

[0022] An embodiment of the present invention discloses a method for detecting wear of a sealing cap mold based on machine vision. Referring to Figure 1 , it includes steps S1 to S4: S1. Take multiple images of the standard sealing cap mold during horizontal rotation.

[0023] Place the standard sealing cap mold without wear horizontally on the detection table, and the detection table can rotate horizontally. During the process of the detection table rotating one week horizontally, use the camera above the detection table to take pictures of the standard sealing cap mold on the detection table. The implementer can set the angular velocity of the detection table rotation and the shooting frequency of the camera according to the actual implementation situation. For example, the angular velocity of the detection table rotation is , and the shooting frequency is once every 0.02 seconds. Figure 2 , Figure 3 and Figure 4 are respectively the first schematic diagram, the second schematic diagram, and the third schematic diagram of the images taken during the rotation process.

[0024] Perform semantic segmentation on the taken images to obtain the mold area in the images.

[0025] S2. Reconstruct a standard image without the influence of light according to multiple images of the standard sealing cap mold.

[0026] It should be noted that the same position on the standard sealing cap mold will pass through different spatial positions during the rotation process and correspond to different pixel points in different images. For example, Figure 2 the midpoint is located at the position of Figure 3 the midpoint after rotation, and Figure 4 the midpoint . Different positions on the mold may pass through the same spatial position during the rotation process. For example, Figure 2 the midpoint is located at the position of Figure 3 the midpoint after a certain rotation, Figure 2 the midpoint is located at the position of Figure 4 the midpoint after a certain rotation,Figure 2 midpoint the position of, Figure 3 midpoint the position of and Figure 4 midpoint the position is at Figure 2 , Figure 3 , Figure 4 in the same row and the same column, that is Figure 2 midpoint , Figure 3 midpoint and Figure 4 midpoint the pixel positions in the image are the same. Therefore Figure 2 midpoint the position of, Figure 3 midpoint the position of and Figure 4 midpoint the position of are the same spatial position. The present invention eliminates the interference of light according to the difference in gray values of the same die position on the standard seal cover die at different spatial positions, and performs image reconstruction, so as to obtain a standard image without the influence of light.

[0027] Specifically, the flowchart of step S2 is referred to Figure 5 , including steps S201 to S204: S201. Determine the gray significance of the die position at different spatial positions according to the difference between the gray values of the same die position at different spatial positions.

[0028] For any position on the standard seal cover die, according to the angular velocity of rotation of the test bench and the shooting frequency, obtain the gray values of the corresponding pixel points of the die position in different captured images, and form a gray sequence of the die position in the order of shooting time.

[0029] It should be noted that due to the complex shooting environment and different light conditions at different spatial positions, there will be reflections or shadows on the surface of the mold in the captured images, resulting in different colors presented by the same position on the mold when passing through different spatial positions during rotation, that is, there are differences in the gray values in the gray scale sequence of the same mold position. When the light at a certain spatial position has a greater impact on the color presented by the mold position in the image, the difference between the gray value of the pixel corresponding to the mold position at this spatial position and the gray value of the pixel corresponding to the mold position at other spatial positions is greater, that is, the gray value of the pixel corresponding to the mold position at this spatial position is more prominent in the gray scale sequence of the mold position. Therefore, for any mold position, the present invention obtains the gray significance of the mold position at different spatial positions according to the gray scale sequence of the mold position, so as to subsequently determine the degree of influence of the light at different spatial positions on the gray scale of the target mold position according to the gray significance.

[0030] Specifically, the gray significance satisfies the expression: ; Among them, any position on the standard sealing cover mold is denoted as the target mold position, represents the gray significance of the target mold position at the th spatial position; represents the gray value of the pixel corresponding to the target mold position at the th spatial position; represents the th gray value in the gray scale sequence of the target mold position; represents the maximum gray value in the gray scale sequence of the target mold position; represents the minimum gray value in the gray scale sequence of the target mold position; represents the length of the gray scale sequence of the target mold position; is the absolute value symbol; reflects the average difference between the gray value of the target mold position at the th spatial position and the gray values of the target mold position at other spatial positions, is used to normalize the average difference . When the average difference is greater, the gray value of the pixel corresponding to the target mold position at the th spatial position is more prominent in the gray scale sequence of the target mold position, and the light at the th spatial position has a greater impact on the color presented by the target mold position in the image.

[0031] S202. Determine the degree of influence of the light at this spatial position on the gray scales of different mold positions according to the differences between the gray significances of different mold positions at the same spatial position.

[0032] It should be noted that there are different structures on the mold surface, and the influence of light at the same spatial position on the colors presented by different structures on the mold surface in the image is different. Different positions on the mold may pass through the same spatial position during rotation. Therefore, in the present invention, the gray-scale significance of all mold positions passing through the same spatial position at this spatial position is analyzed to obtain the degree of influence of the light at this spatial position on the gray scale of the target mold position.

[0033] Specifically, for any spatial position, the gray-scale significance of all mold positions passing through this spatial position during the rotation of the detection table at this spatial position is sorted in the time order of the mold positions passing through this spatial position during the rotation of the detection table to form the gray-scale significance sequence of this spatial position.

[0034] According to the gray-scale significance sequence of the th spatial position, obtain the degree of influence of the light at the th spatial position on the gray scale of the target mold position: wherein, represents the degree of influence of the light at the th spatial position on the gray scale of the target mold position; represents the gray-scale significance of the target mold position at the th spatial position; represents the th gray-scale significance in the gray-scale significance sequence of the th spatial position; represents the maximum gray-scale significance in the gray-scale significance sequence of the th spatial position; represents the minimum gray-scale significance in the gray-scale significance sequence of the th spatial position; represents the length of the gray-scale significance sequence of the th spatial position. When the gray-scale significance of the target mold position at the th spatial position is greater, and the gray-scale significance of the target mold position at the th spatial position is more greater than the gray-scale significance of the remaining mold positions at the th spatial position, the degree of influence of the light at the th spatial position on the gray scale of the target mold position is greater.

[0035] S203. Determine the corrected gray-scale value of the mold position according to the degree of influence of the light at different spatial positions on the gray scale of the same mold position and the gray-scale value of the mold position at different spatial positions.

[0036] The gray value of the target mold position is corrected according to the influence degree of the light at each spatial position on the gray level of the target mold position, and the corrected gray value of the target mold position is obtained: ; Among them, represents the corrected gray value of the target mold position; represents the influence degree of the light at the th spatial position on the gray level of the target mold position; represents the gray value of the pixel corresponding to the target mold position at the th spatial position; represents the length of the gray level sequence of the target mold position; represents the exponential function with the natural constant as the base; in the present invention, is used as the weight of the gray value of the pixel corresponding to the target mold position at the th spatial position. When the influence degree of the gray level is larger, the influence of the light at the th spatial position on the color presentation of the target mold position in the image is greater, and the difference between the actual color of the target mold position and the gray value of the pixel corresponding to the target mold position at the th spatial position is larger. At this time, the weight is smaller, and when obtaining the corrected gray value of the target mold position, the gray value of the pixel corresponding to the target mold position at the th spatial position is less referenced; on the contrary, when the influence degree of the gray level is smaller, the influence of the light at the th spatial position on the color presentation of the target mold position in the image is smaller, and the actual color of the target mold position is closer to the gray value of the pixel corresponding to the target mold position at the th spatial position. At this time, the weight is larger, and when obtaining the corrected gray value of the target mold position, the gray value of the pixel corresponding to the target mold position at the th spatial position is more referenced, so that the corrected gray value can better remove the influence of the light and approach the actual color of the target mold position.

[0037] Similarly, the corrected gray values of all mold positions are obtained.

[0038] S204. Reconstruct an image according to the corrected gray values of all mold positions to obtain a standard image.

[0039] Arrange the corrected gray values of all mold positions according to the corresponding positions of all mold positions in the first captured image to form a new image as the standard image.

[0040] So far, a standard image of the standard sealing cover mold without the influence of light has been obtained.

[0041] S3. Take multiple images when the sealing cover mold to be detected rotates horizontally, and reconstruct the image to be detected without the influence of light based on the multiple images of the sealing cover mold to be detected.

[0042] For the sealing cover mold to be detected, use the methods in steps S1 and S2 to generate an image as the image to be detected.

[0043] S4. Compare the standard image with the image to be detected to identify the wear condition of the sealing cover mold to be detected.

[0044] Perform template matching on the standard image and the image to be detected, and determine the defect possibility of each pixel point in the mold area of the image to be detected according to the difference between each pixel point in the mold area of the image to be detected and the corresponding matching pixel point in the standard image: ; Among them, any pixel point in the mold area of the image to be detected is denoted as the target pixel point, represents the defect possibility of the target pixel point; represents the gray value of the target pixel point; represents the gray value of the matching pixel point of the target pixel point in the standard image; represents the range of gray values in the mold area of the standard image; represents the gray value of the th pixel point within the neighborhood range of the target pixel point; represents the gray value of the th pixel point within the neighborhood range of the target pixel point in the standard image; represents the size of the neighborhood range of the target pixel point, which is an eight-neighborhood in this embodiment, that is , and in other embodiments, the implementer can set the size of the neighborhood range according to the actual implementation situation; is a hyperparameter used to prevent from being too large, resulting in constantly approaching 0, and the empirical value of is 5, and the implementer can also set it according to the actual implementation situation; is the exponential function with the natural constant as the base, which is used for negative correlation mapping of

[0045] When the difference between the gray value of the target pixel point and the gray value of the matching pixel point of the target pixel point in the standard image The larger it is, the more likely the target pixel is a worn pixel or a noise pixel. Since there is a certain area for the worn area, multiple worn pixels appear adjacent to each other. At this time, if the difference is closer to the difference between the neighboring pixel and the corresponding matching pixel, the smaller it is, the more likely the target pixel is located in the worn area, and the greater the defect possibility of the target pixel.

[0046] Similarly, obtain the defect possibility of each pixel in the mold area of the image to be detected, and use the area composed of all pixels with a defect possibility greater than the preset wear threshold as the worn area. Among them is set by the implementer according to the actual implementation situation, and is not specifically limited. For example .

[0047] So far, the wear detection of the sealing cover mold has been realized.

[0048] In the description of this specification, the meanings of "a plurality" and "several" are at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0049] Although this specification has shown and described multiple embodiments of the present invention, it is obvious to those skilled in the art that such embodiments are provided only by way of example. Those skilled in the art will think of many changes, alterations and alternative ways without departing from the spirit and concept of the present invention. It should be understood that various alternative solutions to the embodiments of the present invention described herein can be adopted in the process of practicing the present invention.

Claims

1. A method for detecting wear of a sealing cover mold based on machine vision, characterized in that, Including: Taking multiple images when the standard seal cover mold rotates horizontally, and reconstructing a standard image without the influence of light according to the multiple images of the standard seal cover mold, including: The same mold position of the standard seal cover mold is located at different spatial positions in different images. According to the difference between the gray values of the same mold position at different spatial positions, the gray significance of the mold position at different spatial positions is determined; according to the difference between the gray significances of different mold positions at the same spatial position, the influence degree of the light at this spatial position on the gray values of different mold positions is determined; according to the influence degree of the light at different spatial positions on the gray value of the same mold position and the gray value of this mold position at different spatial positions, the corrected gray value of this mold position is determined; reconstructing an image according to the corrected gray values of all mold positions to obtain a standard image; Taking multiple images when the seal cover mold to be detected rotates horizontally, and reconstructing an image to be detected without the influence of light according to the multiple images of the seal cover mold to be detected; Comparing the standard image with the image to be detected to identify the wear condition of the seal cover mold to be detected.

2. The method for detecting wear of a sealing cover mold based on machine vision according to claim 1, wherein The gray significance satisfies the expression: ； Among them, represents the gray-scale significance when the target die position is at the th spatial position, and the target die position is any position on the standard seal cover die; represents the gray-scale value of the pixel corresponding to the target die position at the th spatial position; represents the th gray-scale value in the gray-scale sequence of the target die position; , respectively represent the maximum gray-scale value and the minimum gray-scale value in the gray-scale sequence of the target die position; is the length of the gray-scale sequence of the target die position; is the absolute value symbol; the gray-scale sequence is a sequence composed of the gray-scale values of the target die position at different spatial positions.

3. A method for detecting wear of a sealing cover mold based on machine vision according to claim 1, characterized in that, The gray influence degree satisfies the expression: ； Among them, represents the degree of gray-scale influence of the light rays at the th spatial position on the target mold position; represents the gray-scale significance of the target mold position at the th spatial position; represents the th gray-scale significance in the gray-scale significance sequence of the th spatial position; , respectively represent the maximum gray-scale significance and the minimum gray-scale significance in the gray-scale significance sequence of the th spatial position; represents the length of the gray-scale significance sequence of the th spatial position; the gray-scale significance sequence is composed of the gray-scale significances of all mold positions passing through the th spatial position at the th spatial position.

4. A method for detecting wear of a sealing cover mold based on machine vision according to claim 1 or 3, characterized in that, The corrected gray value satisfies the expression: ； Among them, represents the corrected gray value of the target mold position; represents the influence degree of the light at the -th spatial position on the gray value of the target mold position; represents the gray value of the pixel corresponding to the target mold position at the -th spatial position; represents the length of the gray value sequence of the target mold position, and the gray value sequence is a sequence composed of the gray values of the target mold position at different spatial positions; represents the exponential function with the natural constant as the base.

5. A method for detecting wear of a sealing cap mold based on machine vision according to claim 1, characterized in that, The reconstructing an image according to the corrected gray values of all mold positions to obtain a standard image includes: Arranging the corrected gray values of all mold positions according to the positions corresponding to them in the first image obtained by shooting to form a new image as the standard image.

6. The method for detecting the wear of a sealing cover mold based on machine vision according to claim 1, wherein, The comparing the standard image with the image to be detected to identify the wear condition of the seal cover mold to be detected includes: Performing template matching on the standard image and the image to be detected; determining the defect possibility of each pixel point in the mold area of the image to be detected according to the difference between each pixel point in the mold area of the image to be detected and the corresponding matching pixel point in the standard image; screening the wear area according to the size of the defect possibility.

7. A method for detecting wear of a sealing cover mold based on machine vision according to claim 6, characterized in that, The defect possibility satisfies the expression: ； Among them, represents the defect possibility of the target pixel point, and the target pixel point is any pixel point in the mold area of the image to be detected; represents the gray value of the target pixel point; represents the gray value of the matching pixel point of the target pixel point in the standard image; represents the range of gray values in the mold area of the standard image; represents the gray value of the th pixel point within the neighborhood range of the target pixel point; represents the gray value of the matching pixel point of the th pixel point within the neighborhood range of the target pixel point in the standard image; represents the size of the neighborhood range of the target pixel point; is a hyperparameter; is the exponential function with the natural constant as the base.

8. A method for detecting wear of a sealing cover mold based on machine vision according to claim 6, characterized in that, The screening the wear area according to the size of the defect possibility includes: Taking the area composed of all pixel points with a defect possibility greater than a preset wear threshold as the wear area.

Images