Porcelain insulator insulating magnet defect detection method and system

Abstract

The invention provides a porcelain insulator insulating magnet defect detection method and system. The method comprises the following steps: acquiring a plurality of groups of insulator magnet images in a rotating state through a high-speed camera; wherein the plurality of groups of images are continuous, and two adjacent groups of images have an overlapping region; performing initial defect identification on the plurality of groups of images, and fusing the plurality of groups of images after the initial defect identification to obtain an initial position of the insulator magnet defect; according to the method, the defect area of the insulator can be accurately determined, and the defect detection result of the insulator magnet is determined based on the feature type of the edge curve of the insulator magnet when the defect area is on the edge of the insulator magnet. And the defect detection precision of the insulator magnet can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of appearance defect detection, and in particular to a method and system for detecting defects of insulating magnets of porcelain insulators. Background Art

[0002] Butterfly insulators and pin insulators among porcelain insulator products are mainly used in high and low voltage transmission lines. They need to be able to withstand high temperatures and mechanical external forces. Porcelain insulators are required to have high mechanical strength, good wear resistance, corrosion resistance, good thermal stability and other properties. Defect detection of porcelain insulators is the most critical factor affecting their performance.

[0003] Publication No. CN202211132257.3 discloses a porcelain insulator quality detection method and system, which collects images of porcelain insulators through an image acquisition device to obtain a multi-angle porcelain insulator image information set; obtains a surface predetermined convolution kernel according to the application standard of the porcelain insulator; performs traversal convolution calculation on the multi-angle porcelain insulator image information set according to the surface predetermined convolution kernel to obtain an image convolution calculation result; obtains surface defect feature information of the porcelain insulator according to the image convolution calculation result; performs ultrasonic measurement on the porcelain insulator through an ultrasonic flaw detector to obtain an ultrasonic flaw detection curve; performs waveform analysis on the ultrasonic flaw detection curve to obtain internal defect feature information of the porcelain insulator; inputs the surface defect feature information of the porcelain insulator and the internal defect feature information of the porcelain insulator into a porcelain insulator quality analysis model to obtain a porcelain insulator quality detection result; based on the porcelain insulator quality detection result, the porcelain insulator is operated and maintained; solves the technical problem of how to intelligently realize real-time detection of porcelain insulator quality and improve quality detection efficiency and accuracy, so as to achieve the technical effect that the porcelain insulator can operate stably and reduce economic losses.

[0004] At present, the surface defect detection method based on image processing can often only perform defect detection on the appearance, and cannot accurately obtain the impact of magnetic defects on the magnet, resulting in incomplete defect detection. Summary of the invention

[0005] 1. Technical issues to be resolved In view of the deficiencies in the prior art, the present invention provides a method and system for detecting defects in insulating magnets of porcelain insulators to solve the problems raised in the above background technology.

[0006] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for detecting defects of insulating magnets of porcelain insulators, comprising the following steps: Acquire several groups of images of insulator magnets in a rotating state by using a high-speed camera; Wherein, the plurality of groups of images are continuous, and there are overlapping areas between two adjacent groups of images; Performing preliminary defect recognition on several groups of images respectively, fusing several groups of images after preliminary defect recognition, and obtaining preliminary positions of defects of the insulator magnet; Based on the preliminary location of the insulator magnet defects, the inspection is carried out to determine all defects of the insulator magnet.

[0007] As a preferred embodiment of this embodiment, the acquiring of a plurality of groups of images of insulators in a rotating state by a high-speed camera comprises: The insulator magnet and the high-speed camera are fixed in position, wherein the axis of the high-speed camera is perpendicular to the axis of the insulator magnet; The insulator magnet image collection surface is illuminated by a light source with a fixed brightness, the insulator magnet rotates at a constant speed for one circle, and a high-speed camera is used to collect several groups of insulator magnet images during the rotation process; Among them, the first group of pictures and the last group of pictures in the several groups of insulator magnet pictures are exactly the same.

[0008] As a preferred embodiment of the present invention, after acquiring the plurality of sets of images of the insulator magnet in the rotating state, the method further comprises: acquiring edge lines of a first set of images among the plurality of sets of images of the insulator magnet, wherein the edge lines include a top edge line and a bottom edge line; The contour of the rear insulator magnet image is compared with the edge lines of the first set of images to complete the pre-identification of magnet defects.

[0009] As a preferred embodiment of this embodiment, preliminary defect recognition is performed on several groups of images respectively, and several groups of images after preliminary defect recognition are fused to obtain preliminary positions of defects of the insulator magnet; Divide the insulator magnet region in the image into a number of independent pixel grids, and determine the grayscale value of each independent pixel grid; If the grayscale value of the independent pixel grid is less than the preset grayscale value, the position of the independent pixel grid is the preliminary position of the insulator magnet defect; The regional shape of the surface defects of the insulator magnet is determined based on the independent pixel grids whose grayscale values ​​are less than the preset grayscale values, and the proportion of the area where the grayscale values ​​of the independent pixel grids are less than the preset grayscale values ​​is calculated to obtain the preliminary position of the insulator magnet defects.

[0010] As a preferred embodiment of this invention, the detection based on the preliminary position of the insulator magnet defect to determine all defects of the insulator magnet includes: Based on the determined preliminary position, an image whose preliminary position is located in the middle area of ​​the insulator magnet is selected to obtain a first detection image; Extracting image feature information of the preliminary position based on the first detection image; Acquiring defect data corresponding to the insulator magnet, and determining the type of the defect based on the defect data and image feature information; Acquire the image of the preliminary position divided by the edge of the insulator magnet to obtain a second detection image; According to the insulator magnet profile, an edge image of the insulator magnet edge corresponding to the preliminary position in the second detection image is obtained; The defect detection algorithm is used to calculate the edge image to complete the defect detection of the preliminary position and obtain the insulator magnet defect result.

[0011] As a preferred embodiment of this embodiment, the step of calculating the edge image using a defect detection algorithm to complete the defect detection at the preliminary position and obtaining the insulator magnet defect result includes: Calculate the inner boundary threshold and the outer boundary threshold according to the grayscale values ​​of the independent pixel grids in the boundary curve in the edge image; Extracting the position of the edge curve of the insulator magnet corresponding to the preliminary position, and extracting the pixel points of the position of the edge curve of the insulator magnet corresponding to the preliminary position that meet the preset conditions according to the inner boundary threshold and the outer boundary threshold; The feature type of the edge curve position of the insulator magnet corresponding to the preliminary position is determined to obtain the defect result of the insulator magnet.

[0012] As a preferred embodiment of this embodiment, a porcelain insulator insulating magnet defect detection system is used to implement the above-mentioned porcelain insulator insulating magnet defect detection method, including: An acquisition module, comprising a plurality of groups of images of the insulator magnet in a rotating state, wherein the plurality of groups of images are continuous and two adjacent groups of images have overlapping areas; A first recognition module, which obtains edge lines of a first group of images among a plurality of groups of insulator magnet images, wherein the edge lines include a top edge line and a bottom edge line, and compares the contour of the rear insulator magnet image with the edge lines of the first group of images to complete pre-recognition of magnet defects; A second recognition module is used to perform preliminary defect recognition on the plurality of groups of images pre-recognized by the first recognition module, and fuse the plurality of groups of images after preliminary defect recognition to obtain preliminary positions of the insulator magnet defects; The defect detection module is used to detect the preliminary position obtained by the first identification module and determine all defects of the insulator magnet.

[0013] As a preferred embodiment of the present invention, the specific process of the second identification module obtaining the preliminary position of the insulator magnetic defect includes: Divide the insulator magnet region in the image into a number of independent pixel grids, and determine the grayscale value of each independent pixel grid; If the grayscale value of the independent pixel grid is less than the preset grayscale value, the position of the independent pixel grid is the preliminary position of the insulator magnet defect; The regional shape of the surface defects of the insulator magnet is determined based on the independent pixel grids whose grayscale values ​​are less than the preset grayscale values, and the proportion of the area where the grayscale values ​​of the independent pixel grids are less than the preset grayscale values ​​is calculated to obtain the preliminary position of the insulator magnet defects.

[0014] As a preferred embodiment of this embodiment, the process of the defect detection module determining all defects of the insulator magnet includes: Based on the determined preliminary position, an image whose preliminary position is located in the middle area of ​​the insulator magnet is selected to obtain a first detection image; Extracting image feature information of the preliminary position based on the first detection image; Acquiring defect data corresponding to the insulator magnet, and determining the type of the defect based on the defect data and image feature information; Acquire the image of the preliminary position divided by the edge of the insulator magnet to obtain a second detection image; According to the insulator magnet profile, an edge image of the insulator magnet edge corresponding to the preliminary position in the second detection image is obtained; The defect detection algorithm is used to calculate the edge image to complete the defect detection of the preliminary position and obtain the insulator magnet defect result.

[0015] As a preferred embodiment of this embodiment, the step of calculating the edge image using a defect detection algorithm to complete the defect detection at the preliminary position and obtaining the insulator magnet defect result includes: Calculate the inner boundary threshold and the outer boundary threshold according to the grayscale values ​​of the independent pixel grids in the boundary curve in the edge image; Extracting the position of the edge curve of the insulator magnet corresponding to the preliminary position, and extracting the pixel points of the position of the edge curve of the insulator magnet corresponding to the preliminary position that meet the preset conditions according to the inner boundary threshold and the outer boundary threshold; The feature type of the edge curve position of the insulator magnet corresponding to the preliminary position is determined to obtain the defect result of the insulator magnet.

[0016] (III) Beneficial effects The present invention provides a method and system for detecting defects of porcelain insulator insulating magnets, which have the following beneficial effects: by obtaining a plurality of groups of images of insulator magnets in a rotating state, determining the preliminary position of the surface defect of the magnet by the grayscale value of the independent pixel grid of the insulator magnet area, by the feature information of the preliminary position on the image, and by obtaining the image of the preliminary position divided by the edge of the insulator magnet, according to the contour of the insulator magnet, obtaining the edge image of the edge of the insulator magnet corresponding to the preliminary position in the second detection image, using a defect detection algorithm to calculate the edge image to complete the defect detection of the preliminary position, and obtaining the defect result of the insulator magnet. The present invention can accurately determine the defect area of ​​the insulator, and based on the feature type of the edge curve of the insulator magnet when the defect area is at the edge of the insulator magnet, determine the depth of influence of the insulator magnet defect on the magnet, which is conducive to improving the accuracy of insulator magnet defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of the method for detecting defects of insulating magnets of porcelain insulators of the present invention; Figure 2 The block diagram of the porcelain insulator insulating magnet defect detection system of the present invention. DETAILED DESCRIPTION

[0018] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0019] The disclosure below provides many different embodiments or examples to realize different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or reference letters in different examples, and this repetition is for the purpose of simplicity and clarity, which itself does not indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides various specific examples of processes and materials, but those of ordinary skill in the art can be aware of the application of other processes and / or the use of other materials.

[0020] like Figure 1 As shown, an embodiment of the present invention provides a method for detecting defects of an insulating magnet of a porcelain insulator, characterized in that it includes the following steps: S1: Acquire several sets of images of insulator magnets in rotating states through a high-speed camera; Among them, several groups of images are continuous, and there are overlapping areas between two adjacent groups of images; Each set of images contains half of the area of ​​the insulator magnet.

[0021] S2: performing preliminary defect recognition on several groups of images respectively, fusing several groups of images after preliminary defect recognition, and obtaining preliminary positions of defects of insulator magnets; Specifically, the insulator magnet region in the image is divided into a number of independent pixel grids, and the grayscale value of each independent pixel grid is determined; If the grayscale value of the independent pixel grid is less than the preset grayscale value, the position of the independent pixel grid is the preliminary position of the insulator magnet defect; The regional shape of the surface defects of the insulator magnet is determined based on the independent pixel grids whose grayscale values ​​are less than the preset grayscale values, and the proportion of the area where the grayscale values ​​of the independent pixel grids are less than the preset grayscale values ​​is calculated to obtain the preliminary position of the insulator magnet defects.

[0022] Among them, since several groups of images are continuous and have overlapping areas, that is, if there is a first defect in one group of pictures, there will also be other pictures. The positional relationship of the first defect in the two groups of pictures can be determined through affine transformations such as rotation, scaling and translation. Therefore, based on the positional relationship of the first defect in several groups of pictures, the first defects in the pictures are fused to obtain a more accurate first appearance defect, so that the defects can be better identified.

[0023] In other embodiments, the quality of the insulator may be classified by level by setting the grayscale value of the independent pixel grid to be less than a preset grayscale value area ratio threshold.

[0024] It can be understood that the size of the independent pixel grid is determined by the resolution of the high-speed camera, and the preset grayscale value is determined by the mode of the grayscale values ​​of all independent pixel grids in the collected magnet image.

[0025] S3: Detection is performed based on the preliminary location of the insulator magnet defects to determine all defects of the insulator magnet.

[0026] Based on the determined preliminary position, an image whose preliminary position is located in the middle area of ​​the insulator magnet is selected to obtain a first detection image; Extracting image feature information of a preliminary position based on the first detection image; The characteristic information includes intuitive information such as the shape, size, and color of the initial position corresponding to the magnet.

[0027] Obtain defect data corresponding to the insulator magnet, and determine the type of the defect based on the defect data and image feature information; The defect data is specifically defect prior knowledge corresponding to the insulator magnet. Therefore, the defect category of the insulator magnet is determined from at least one possible defect category according to the defect prior knowledge and attribute information of the product image.

[0028] Acquire an image segmented by the edge of the insulator magnet at the preliminary position to obtain a second detection image; According to the contour of the insulator magnet, an edge image of the edge of the insulator magnet corresponding to the preliminary position in the second detection image is obtained; The defect detection algorithm is used to calculate the edge image to complete the defect detection at the initial position and obtain the defect results of the insulator magnet.

[0029] Specifically, by analyzing the image data of the same defect in the insulator magnet at different positions, the concave or convex appearance of the magnet defect can be accurately obtained. Combined with the change in grayscale value, the specific type and cause of the magnet defect can be accurately analyzed, thereby facilitating the judgment of whether the insulator will affect subsequent use.

[0030] Among them, the defect detection algorithm is used to calculate the edge image to complete the defect detection of the preliminary position, and the defect results of the insulator magnet are obtained, including: Calculate the inner boundary threshold and the outer boundary threshold according to the grayscale values ​​of the independent pixel grids in the boundary curve in the edge image; Extract the position of the edge curve of the insulator magnet corresponding to the preliminary position, and extract the pixel points that meet the preset conditions of the position of the edge curve of the insulator magnet corresponding to the preliminary position according to the inner boundary threshold and the outer boundary threshold; Determine the feature type of the preliminary position corresponding to the edge curve position of the insulator magnet, and obtain the defect results of the insulator magnet.

[0031] Specifically, the inner boundary threshold and outer boundary threshold calculated by the grayscale values ​​of independent pixels in the boundary curve in the edge image can accurately determine the impact of magnet defects on the outside of the magnet, and then accurately determine the type of magnet defects.

[0032] After acquiring several groups of images of the insulator magnet in a rotating state, the method further includes: acquiring edge lines of a first group of images among the several groups of images of the insulator magnet, wherein the edge lines include a top edge line and a bottom edge line; The contour of the rear insulator magnet image is compared with the edge lines of the first set of images to complete the pre-identification of magnet defects.

[0033] Furthermore, a plurality of images of insulators in a rotating state are obtained by a high-speed camera, including: The insulator magnet and the high-speed camera are fixed in position, wherein the axis of the high-speed camera is perpendicular to the axis of the insulator magnet; The insulator magnet image collection surface is illuminated by a light source with a fixed brightness, the insulator magnet rotates at a constant speed for one circle, and a high-speed camera is used to collect several groups of insulator magnet images during the rotation process; Among them, the first group of pictures and the last group of pictures in several groups of insulator magnet pictures are exactly the same.

[0034] Specifically, by rotating the insulator, it is possible to prevent dirt contaminated on the insulator during the production process from affecting the detection accuracy.

[0035] The defect detection method for porcelain insulator insulating magnets provided by the present invention determines the preliminary position of the magnet surface defect by the grayscale value of the independent pixel grid of the insulator magnet area through obtaining several groups of insulator magnet pictures in a rotating state, and obtains the image of the preliminary position divided by the edge of the insulator magnet according to the feature information of the preliminary position on the image, and obtains the edge image of the edge of the insulator magnet corresponding to the preliminary position in the second detection image according to the contour of the insulator magnet, and uses the defect detection algorithm to calculate the edge image to complete the defect detection of the preliminary position, and obtains the defect result of the insulator magnet. The present invention can accurately determine the defect area of ​​the insulator, and based on the feature type of the edge curve of the insulator magnet when the defect area is at the edge of the insulator magnet, determine the influence depth of the insulator magnet defect on the magnet, which is conducive to improving the accuracy of insulator magnet defect detection.

[0036] like Figure 2 As shown, a porcelain insulator insulating magnet defect detection system is used to implement the above-mentioned porcelain insulator insulating magnet defect detection method, comprising: An acquisition module, comprising several groups of images of insulator magnets in a rotating state, wherein the groups of images are continuous and two adjacent groups of images have overlapping areas; A first recognition module obtains edge lines of a first group of images among a plurality of groups of insulator magnet images, wherein the edge lines include a top edge line and a bottom edge line, and compares the contour of the rear insulator magnet image with the edge lines of the first group of images to complete pre-recognition of magnet defects; The second recognition module is used to perform preliminary defect recognition on the several groups of images pre-recognized by the first recognition module, and fuse the several groups of images after preliminary defect recognition to obtain the preliminary position of the insulator magnet defect; The defect detection module is used to detect the preliminary position obtained by the first identification module and determine all defects of the insulator magnet.

[0037] Furthermore, the specific process of the second identification module obtaining the preliminary position of the insulator magnetic defect includes: Divide the insulator magnet region in the image into a number of independent pixel grids, and determine the grayscale value of each independent pixel grid; If the grayscale value of the independent pixel grid is less than the preset grayscale value, the position of the independent pixel grid is the preliminary position of the insulator magnet defect; The regional shape of the surface defects of the insulator magnet is determined based on the independent pixel grids whose grayscale values ​​are less than the preset grayscale values, and the proportion of the area where the grayscale values ​​of the independent pixel grids are less than the preset grayscale values ​​is calculated to obtain the preliminary position of the insulator magnet defects.

[0038] Furthermore, the defect detection module determines all defects of the insulator magnet through the following process: Based on the determined preliminary position, an image whose preliminary position is located in the middle area of ​​the insulator magnet is selected to obtain a first detection image; Extracting image feature information of a preliminary position based on the first detection image; Obtain defect data corresponding to the insulator magnet, and determine the type of the defect based on the defect data and image feature information; Acquire an image segmented by the edge of the insulator magnet at the preliminary position to obtain a second detection image; According to the contour of the insulator magnet, an edge image of the edge of the insulator magnet corresponding to the preliminary position in the second detection image is obtained; The defect detection algorithm is used to calculate the edge image to complete the defect detection at the initial position and obtain the defect results of the insulator magnet.

[0039] Furthermore, the defect detection algorithm is used to calculate the edge image to complete the defect detection of the initial position, and the defect results of the insulator magnet are obtained, including: Calculate the inner boundary threshold and the outer boundary threshold according to the grayscale values ​​of the independent pixel grids in the boundary curve in the edge image; Extract the position of the edge curve of the insulator magnet corresponding to the preliminary position, and extract the pixel points that meet the preset conditions of the position of the edge curve of the insulator magnet corresponding to the preliminary position according to the inner boundary threshold and the outer boundary threshold; Determine the feature type of the preliminary position corresponding to the edge curve position of the insulator magnet, and obtain the defect results of the insulator magnet.

[0040] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting defects in insulating magnets of porcelain insulators, characterized in that: The following steps are involved: Acquire several groups of images of insulator magnets in a rotating state by using a high-speed camera; Wherein, the plurality of groups of images are continuous, and there are overlapping areas between two adjacent groups of images; Performing preliminary defect recognition on several groups of images respectively, fusing several groups of images after preliminary defect recognition, and obtaining preliminary positions of defects of the insulator magnet; Based on the preliminary location of the insulator magnet defects, the inspection is carried out to determine all defects of the insulator magnet.

2. A method for detecting defects in insulating magnets of porcelain insulators according to claim 1, characterized in that: The method of acquiring a plurality of groups of images of insulators in a rotating state by a high-speed camera comprises: The insulator magnet and the high-speed camera are fixed in position, wherein the axis of the high-speed camera is perpendicular to the axis of the insulator magnet; The insulator magnet image collection surface is illuminated by a light source with a fixed brightness, the insulator magnet rotates at a constant speed for one circle, and a high-speed camera is used to collect several groups of insulator magnet images during the rotation process; Among them, the first group of pictures and the last group of pictures in the several groups of insulator magnet pictures are exactly the same.

3. A method for detecting defects in insulating magnets of porcelain insulators according to claim 1, characterized in that: After acquiring the plurality of sets of images of the insulator magnet in the rotating state, the method further includes: acquiring edge lines of a first set of images among the plurality of sets of images of the insulator magnet, wherein the edge lines include a top edge line and a bottom edge line; The contour of the rear insulator magnet image is compared with the edge lines of the first set of images to complete the pre-identification of magnet defects.

4. A method for detecting defects in insulating magnets of porcelain insulators according to claim 1, characterized in that: The preliminary defect recognition is performed on the plurality of groups of images respectively, and the preliminary defect recognition is performed on the plurality of groups of images to obtain the preliminary position of the insulator magnet defect; Divide the insulator magnet region in the image into a number of independent pixel grids, and determine the grayscale value of each independent pixel grid; If the grayscale value of the independent pixel grid is less than the preset grayscale value, the position of the independent pixel grid is the preliminary position of the insulator magnet defect; The regional shape of the surface defects of the insulator magnet is determined based on the independent pixel grids whose grayscale values ​​are less than the preset grayscale values, and the proportion of the area where the grayscale values ​​of the independent pixel grids are less than the preset grayscale values ​​is calculated to obtain the preliminary position of the insulator magnet defects.

5. A method for detecting defects in insulating magnets of porcelain insulators according to claim 1, characterized in that: The detection is performed based on the preliminary position of the insulator magnet defect to determine all defects of the insulator magnet including: Based on the determined preliminary position, an image whose preliminary position is located in the middle area of ​​the insulator magnet is selected to obtain a first detection image; Extracting image feature information of the preliminary position based on the first detection image; Acquiring defect data corresponding to the insulator magnet, and determining the type of the defect based on the defect data and image feature information; Acquire the image of the preliminary position divided by the edge of the insulator magnet to obtain a second detection image; According to the insulator magnet profile, an edge image of the insulator magnet edge corresponding to the preliminary position in the second detection image is obtained; The defect detection algorithm is used to calculate the edge image to complete the defect detection of the preliminary position and obtain the insulator magnet defect result.

6. A method for detecting defects in insulating magnets of porcelain insulators according to claim 5, characterized in that: The step of calculating the edge image using a defect detection algorithm to complete the defect detection at the preliminary position and obtaining the insulator magnet defect result includes: Calculate the inner boundary threshold and the outer boundary threshold according to the grayscale values ​​of the independent pixel grids in the boundary curve in the edge image; Extracting the position of the edge curve of the insulator magnet corresponding to the preliminary position, and extracting the pixel points of the position of the edge curve of the insulator magnet corresponding to the preliminary position that meet the preset conditions according to the inner boundary threshold and the outer boundary threshold; The feature type of the edge curve position of the insulator magnet corresponding to the preliminary position is determined to obtain the defect result of the insulator magnet.

7. A porcelain insulator insulating magnet defect detection system, used to implement the porcelain insulator insulating magnet defect detection method according to any one of claims 1 to 6, characterized in that: include: An acquisition module, comprising a plurality of groups of images of the insulator magnet in a rotating state, wherein the plurality of groups of images are continuous and two adjacent groups of images have overlapping areas; A first recognition module, which obtains edge lines of a first group of images among a plurality of groups of insulator magnet images, wherein the edge lines include a top edge line and a bottom edge line, and compares the contour of the rear insulator magnet image with the edge lines of the first group of images to complete pre-recognition of magnet defects; A second recognition module is used to perform preliminary defect recognition on the plurality of groups of images pre-recognized by the first recognition module, and fuse the plurality of groups of images after preliminary defect recognition to obtain preliminary positions of the insulator magnet defects; The defect detection module is used to detect the preliminary position obtained by the first identification module and determine all defects of the insulator magnet.

8. A porcelain insulator insulating magnet defect detection system according to claim 7, characterized in that: The specific process of the second identification module obtaining the preliminary position of the insulator magnetic defect includes: Divide the insulator magnet region in the image into a number of independent pixel grids, and determine the grayscale value of each independent pixel grid; If the grayscale value of the independent pixel grid is less than the preset grayscale value, the position of the independent pixel grid is the preliminary position of the insulator magnet defect; The regional shape of the surface defects of the insulator magnet is determined based on the independent pixel grids whose grayscale values ​​are less than the preset grayscale values, and the proportion of the area where the grayscale values ​​of the independent pixel grids are less than the preset grayscale values ​​is calculated to obtain the preliminary position of the insulator magnet defects.

9. A porcelain insulator insulating magnet defect detection system according to claim 7, characterized in that: The process of the defect detection module determining all defects of the insulator magnet includes: Based on the determined preliminary position, an image whose preliminary position is located in the middle area of ​​the insulator magnet is selected to obtain a first detection image; Extracting image feature information of the preliminary position based on the first detection image; Acquiring defect data corresponding to the insulator magnet, and determining the type of the defect based on the defect data and image feature information; Acquire the image of the preliminary position divided by the edge of the insulator magnet to obtain a second detection image; According to the insulator magnet profile, an edge image of the insulator magnet edge corresponding to the preliminary position in the second detection image is obtained; The defect detection algorithm is used to calculate the edge image to complete the defect detection of the preliminary position and obtain the insulator magnet defect result.

10. A porcelain insulator insulating magnet defect detection system according to claim 9, characterized in that: The step of calculating the edge image using a defect detection algorithm to complete the defect detection at the preliminary position and obtaining the insulator magnet defect result includes: Calculate the inner boundary threshold and the outer boundary threshold according to the grayscale values ​​of the independent pixel grids in the boundary curve in the edge image; Extracting the position of the edge curve of the insulator magnet corresponding to the preliminary position, and extracting the pixel points of the position of the edge curve of the insulator magnet corresponding to the preliminary position that meet the preset conditions according to the inner boundary threshold and the outer boundary threshold; The feature type of the edge curve position of the insulator magnet corresponding to the preliminary position is determined to obtain the defect result of the insulator magnet.

Images