Intelligent diagnosis system and method for micro-defects of porcelain insulators based on industrial CT detection

Abstract

The present invention belongs to the technical field of insulator detection and relates to an intelligent diagnostic system and method for micro-defects in porcelain insulators based on industrial CT inspection. This method collects industrial CT inspection photographs of porcelain insulators and extracts the cement layer and porcelain layer in the cross-section and longitudinal section. The method then uses the Yolov11 instance segmentation algorithm to identify defect regions contained in the industrial CT inspection photographs. Based on the cement layer and porcelain layer, the defects are classified into cement layer porosity defects and porcelain layer crack defects. This invention solves the problem of low manual defect identification efficiency in industrial CT inspection photographs of porcelain insulators, achieving fully automated defect identification and significantly reducing the workload of manual defect review.

Description

Technical Field

[0001] The present invention belongs to the technical field of nondestructive testing of insulators, and relates to an intelligent diagnosis system and method for micro-defects of porcelain insulators based on industrial CT detection. Background Art

[0002] Current industrial CT-based micro-defect diagnosis of porcelain insulators relies primarily on manual interpretation of CT scan results, which is labor-intensive and inefficient. Currently, the primary indicators of micro-defects in porcelain insulators are cracks in the porcelain and porosity in the cement layer. Cracks in the porcelain indicate a significant internal defect in the insulator.

[0003] When manual identification is used to carry out defect detection, the number of industrial CT inspection photos generated after a single sample is inspected by industrial CT is extremely large. Manually identifying defects one by one is a huge workload and inefficient. Summary of the Invention

[0004] The present invention provides an intelligent diagnosis system and method for micro-defects of porcelain insulators based on industrial CT detection, which realizes fully automated detection of cracks and pores in cement layers in industrial CT detection photos of porcelain insulators.

[0005] The present invention is implemented by the following technical solution: An intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection, comprising:

[0006] Step 1: Collect industrial CT inspection photos of porcelain insulators and extract the cement layer and porcelain layer in the cross section and longitudinal section;

[0007] Step 2: Use the YOLOv11 instance segmentation algorithm to identify the defect areas contained in the industrial CT inspection photos, and classify the defects into cement layer pore defects and porcelain layer crack defects based on the cement layer and porcelain layer.

[0008] The process of extracting the cement layer and the porcelain layer in the cross section and the longitudinal section comprises the following sub-steps:

[0009] Step S1: extracting parameter information from each industrial CT inspection photo, the parameter information including scene depth and pixel size information;

[0010] Step S2: extracting the size features of the maximum longitudinal and transverse sections;

[0011] Step S2-1: Find two maximum longitudinal sections and one maximum transverse section from the industrial CT inspection photos;

[0012] Step S2-2: Extracting the non-steel cap area, the porcelain layer area, and the steel foot area;

[0013] Step S2-3: Extracting the size features of the largest cross-section photo and the largest longitudinal section;

[0014] Step S2-3-1: Calculate the distance between the steel foot and the steel cap on the X-axis and Y-axis in the maximum cross-sectional photograph;

[0015] Step S2-3-2: Calculate the distance between the widest position of the steel foot and the top of the photo, as well as the horizontal distance between the widest position and the left and right sides of the steel cap in the largest longitudinal section;

[0016] Step S3: extracting the cement layer and the porcelain layer of any longitudinal and transverse sections;

[0017] Step S3-1: Calculate the position of the horizontal projection line of any cross section on the two largest longitudinal sections;

[0018] Step S3-2: Calculation of cross-section porcelain layers;

[0019] Step S3-3: Calculation of cross-section cement layer;

[0020] Step S3-4: Calculation of the longitudinal section porcelain layer;

[0021] Step S3-5: Calculation of the longitudinal section cement layer.

[0022] Further preferably, the industrial CT detection photos include 2 groups of longitudinal section photos and one group of cross-section photos, the 2 groups of longitudinal section photos are respectively called longitudinal section group 1 photos and longitudinal section group 2 photos, the cross-section photo with the largest steel foot area in the same group of photos is called the largest cross-section; the longitudinal section photo with the largest steel foot area in the same group of photos is called the largest longitudinal section.

[0023] Further preferably, the yolov11 instance segmentation algorithm is trained using a calibrated data set to obtain a non-steel cap area recognition model, a porcelain layer area recognition model, and a steel foot area recognition model, respectively; by calling the non-steel cap area recognition model, the porcelain layer area recognition model, and the steel foot area recognition model, the industrial CT detection photos are subjected to instance segmentation, and the non-steel cap area, the porcelain layer area, and the steel foot area are extracted in turn.

[0024] Further preferably, the calculation of the cross-section porcelain layer includes:

[0025] Step S3-2-1: Calculate the distance between the porcelain layer and the steel foot at the position of the horizontal projection line of any cross section on the two largest longitudinal sections;

[0026] Step S3-2-2: Calculate the distance between the porcelain layer and the steel foot on the X-axis and Y-axis in any cross section;

[0027] Step S3-2-3: Draw the porcelain layer area in any cross section.

[0028] Further preferably, the longitudinal section porcelain layer calculation includes:

[0029] Step S3-4-1: Calculate the distance between the inner and outer boundaries of the porcelain layer at any height in any longitudinal section and the central axis of the photograph;

[0030] Step S3-4-2: Calculation of the porcelain layer area of any longitudinal section.

[0031] The present invention solves the problem of low efficiency in manual defect identification of porcelain insulator industrial CT inspection photos, realizes fully automated defect identification, and greatly reduces the workload of manual defect review. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a diagram of the architecture of the present invention.

[0033] Figure 2 Flowchart for extracting cement and porcelain layers in cross-sections and longitudinal sections.

[0034] Figure 3 Schematic diagram of steel foot extraction using yolov11 instance segmentation algorithm.

[0035] Figure 4 The mask map output by the yolov11 instance segmentation algorithm.

[0036] Figure 5 Schematic diagram of the distance between the steel foot and the steel cap in the X-axis and Y-axis directions at the maximum cross-section.

[0037] Figure 6 Schematic diagram of the distance between the widest position of the steel foot and the top of the photo, as well as the distance between the widest position and the steel cap in the largest longitudinal section.

[0038] Figure 7 Schematic diagram of the intersection of the horizontal line of any Y-axis value and the masks of the porcelain layer area and the steel foot area.

[0039] Figure 8 Schematic diagram of the distance between the porcelain piece and the steel foot in the X-axis and Y-axis directions on the cross section.

[0040] Figure 9 Schematic diagram of 8 points on the boundary of the porcelain layer.

[0041] Figure 10 It is the mask map of the porcelain layer area.

[0042] Figure 11 For any longitudinal section Z i and the maximum longitudinal section Z max Projection to M j The projection relationship on .

[0043] Figure 12Schematic diagram of the intersection of P0 and any longitudinal section projection line with the inner and outer boundaries of the porcelain layer.

[0044] Figure 13 is the cross section M j At the maximum longitudinal section Z max Schematic diagram of the position of the upper projection line and its intersection with the boundary of the porcelain layer.

[0045] Figure 14 Schematic diagram of the position relationship of coordinate points on the boundary. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of this application.

[0047] Example 1

[0048] like Figure 1 As shown, an intelligent diagnosis system for micro-defects of porcelain insulators based on industrial CT detection includes:

[0049] Porcelain insulator structure analysis module: collects industrial CT inspection photos of porcelain insulators and extracts cement layers and porcelain layers in cross-sections and longitudinal sections;

[0050] Porcelain insulator defect detection module: This module uses the Yolov11 instance segmentation algorithm to identify defect areas in industrial CT inspection photos. It then classifies defects into cement layer porosity defects and porcelain layer crack defects based on the cement layer and porcelain layer.

[0051] Example 2

[0052] This embodiment provides an intelligent diagnosis method for micro-defects in porcelain insulators based on industrial CT detection, including:

[0053] Step 1: Collect industrial CT inspection photos of porcelain insulators and extract the cement layer and porcelain layer in the cross section and longitudinal section;

[0054] Step 2: Use the yolov11 instance segmentation algorithm to identify the defect areas contained in the industrial CT inspection photos, and combine the cement layer and porcelain layer to classify the defects into cement layer pore defects and porcelain layer crack defects.

[0055] Before detecting defects, it is necessary to know which areas in the photo are porcelain layers and which are cement layers. This facilitates determining the defect type after the defects are detected in the industrial CT inspection photos. Because the boundary between the cement layer and the porcelain layer in the industrial CT inspection photos is not very clear, it is difficult to distinguish the cement layer and the porcelain layer in the industrial CT inspection photos simply through image recognition. To solve this problem, the characteristic that the layers of porcelain insulators are distributed around the central axis of the porcelain insulator is utilized. By calculating the size parameters of the porcelain layer and the cement layer, the porcelain layer and the cement layer in the industrial CT inspection photos are calculated.

[0056] The process of extracting the cement and porcelain layers in transverse and longitudinal sections consists of the following sub-steps:

[0057] Step S1: extracting parameter information from each industrial CT inspection photo, the parameter information including scene depth and pixel size information;

[0058] Scene depth refers to the distance between the cross-section and the detector during industrial CT imaging. The scene depth is typically measured in the upper left corner of an industrial CT image. Because the coordinate information area occupies a relatively small portion of the image, it cannot be directly extracted through OCR recognition. Therefore, accurate extraction requires integrating the coordinate arrangement characteristics of the industrial CT image. A 100x100 pixel area in the upper left corner of the industrial CT image is captured as a new image and then recognized using OCR. The text within the image is extracted, and the numerical value within the text is found. This numerical value represents the scene depth of the industrial CT image. Since the pixel size of an industrial CT image is represented by a scale bar located below the image, and this area is relatively small, AI detection algorithms cannot effectively extract the scale directly. Therefore, the bottommost area of the industrial CT image is first captured as a new image and then recognized using OCR.

[0059] Step S2: extracting the size features of the maximum longitudinal and transverse sections;

[0060] Step S3: Extract the cement layer and porcelain layer of any longitudinal and transverse sections.

[0061] Specifically, step S2 of extracting the size features of the maximum longitudinal and transverse sections includes the following sub-steps:

[0062] Step S2-1: Find two maximum longitudinal sections and one maximum transverse section from the industrial CT inspection photos;

[0063] To ensure that all defects are detected during industrial CT inspection, three perpendicular directions are selected for scanning, resulting in two sets of longitudinal section images and one set of transverse section images. These two longitudinal section images are referred to as longitudinal section set 1 and longitudinal section set 2, respectively. The transverse section image with the largest steel foot area in the same set of images is called the largest transverse section. The longitudinal section image with the largest steel foot area in the same set of images is called the largest longitudinal section.

[0064] In order to extract the maximum cross-section photo and the maximum longitudinal section, the yolov11 instance segmentation algorithm in this embodiment is used to train the corresponding recognition model to identify the steel feet in the industrial CT inspection photos; instance segmentation is a complex task in the field of computer vision, which requires the model to be able to identify objects of different categories in the image and classify each individual object at the pixel level. Figure 3 As shown, the steel foot part in the industrial CT inspection photo is extracted using the yolov11 instance segmentation algorithm.

[0065] The construction of instance segmentation detection algorithm is similar to that of object detection, which mainly consists of three steps: sample labeling, model training and model detection. The differences between each step and object detection are as follows:

[0066] 1. The sample annotation stage annotates polygonal areas instead of rectangular areas.

[0067] 2. The model training phase uses a segmentation model rather than an object detection model. The principle behind this is that the detection branch first obtains a bounding box for each object. The segmentation branch then uses this to identify pixels in the photo that belong to that instance. Finally, the bounding box is used to extract the pixels within the region into a mask.

[0068] 3. The output of the model detection stage is a mask image (a binary image, 1 represents the object, 0 represents it is not) instead of a bounding box, such as Figure 4 shown.

[0069] Since the main feature of the largest cross-section photo and the largest longitudinal section is that the area of the steel foot region in this photo is the largest among the same group of photos, the area of the steel foot region in each industrial CT detection photo is calculated based on the steel foot region mask map (the sum of the number of pixels with a brightness value of 1 in the steel foot region mask map); finally, the photo with the largest steel foot area in each group of photos is found.

[0070] The extraction process of the maximum cross-section photo is as follows: traverse each cross-section group photo set (where i is the serial number of the cross-section group photo, is the i-th cross-section group photo, and n is the number of cross-section group photos). The yolov11 instance segmentation algorithm is used to extract Steel foot area mask map ; Calculate the total brightness of each steel foot area mask map ,in This function sums the brightness of the image, and returns the sum of the brightness of the input image; traverse each ,turn up The cross-section group photo number i when the value is the largest is marked as the maximum cross-section photo.

[0071] The extraction process of the maximum longitudinal section of a group of longitudinal sections is as follows: traverse the set of longitudinal section photos ,in is the qth longitudinal section photo in the longitudinal section group 1, k is the number of longitudinal section photos in the longitudinal section group 1, and is extracted using the yolov11 instance segmentation algorithm Steel foot area mask map ; Calculate the sum of the brightness of each steel foot area mask map ; Traverse each ,turn up The serial number of the set of longitudinal section photos when the value is the largest is q, and the corresponding set of longitudinal section photos is recorded as the maximum longitudinal section.

[0072] Similarly, the process of extracting the maximum longitudinal section of the two groups of longitudinal sections is as follows: traverse the two groups of longitudinal section photo sets ,in is the e-th longitudinal section photo in the 2 longitudinal sections, r is the number of longitudinal section photos in the 2 longitudinal sections, and is extracted using the yolov11 instance segmentation algorithm. Steel foot area mask map ; Calculate the sum of the brightness of each steel foot area mask map ; Traverse each , find the sum of the brightness The serial number e of the two sets of longitudinal section photos when the value is the largest is taken as the maximum longitudinal section of the two sets of longitudinal section photos.

[0073] Step S2-2: Extracting the non-steel cap area, the porcelain layer area, and the steel foot area;

[0074] By organizing and calibrating the non-steel cap area, porcelain layer area, and steel foot area in industrial CT inspection photos of various types of porcelain insulators, a data set is formed. Then, the calibrated data set is used to train the yolov11 instance segmentation algorithm to obtain the non-steel cap area recognition model, porcelain layer area recognition model, and steel foot area recognition model respectively. By calling the non-steel cap area recognition model, porcelain layer area recognition model, and steel foot area recognition model to perform instance segmentation on the industrial CT inspection photos, the non-steel cap area, porcelain layer area, and steel foot area are extracted in turn; and the corresponding maximum cross-section mask map sets are obtained respectively. . A set of maximum longitudinal section mask maps . 2 sets of maximum longitudinal section mask maps ,in They are the mask images of the non-steel cap area of the maximum cross section, the maximum longitudinal section of the longitudinal section 1 group, and the maximum longitudinal section of the longitudinal section 2 group. They are the mask images of the porcelain layer area of the largest cross section, the largest longitudinal section of the longitudinal section 1 group, and the largest longitudinal section of the longitudinal section 2 group. They are the steel foot area mask images of the maximum cross section, the maximum longitudinal section of the longitudinal section 1 group, and the maximum longitudinal section of the longitudinal section 2 group.

[0075] Step S2-3: Extracting the size features of the largest cross-section photo and the largest longitudinal section;

[0076] Step S2-3-1: Calculate the distance between the steel foot and the steel cap on the X and Y axes in the maximum cross-sectional photograph:

[0077] Let the origin O of the plane coordinate system A in the maximum cross-section photo be the upper left corner of the photo, the positive direction of the X axis be the horizontal direction from left to right, the positive direction of the Y axis be the vertical direction from top to bottom, and the length unit in the X and Y axes be mm. The distance between the steel foot and the steel cap on the maximum cross-section on the X and Y axes includes four distances (such as Figure 5 shown), which are the distances between the steel foot and the steel cap in the negative direction of the X axis. , the distance between the steel foot and the steel cap in the positive direction of the X axis , the distance between the steel foot and the steel cap in the negative direction of the Y axis , the distance between the steel foot and the steel cap in the positive direction of the Y axis To calculate these four distances, we must first calculate the steel foot area's bounding box in the plane coordinate system A of the maximum cross-sectional photo. , and the bounding box of the non-steel cap area in the plane coordinate system A in the maximum cross-sectional view .in The leftmost X coordinate of the steel foot bounding box, The rightmost X coordinate of the steel foot bounding box, is the top Y coordinate of the steel foot bounding box, The Y coordinate of the bottom side of the steel foot bounding box. is the leftmost X coordinate of the bounding box of the non-steel cap area, is the rightmost X coordinate of the bounding box of the non-steel cap area, is the top Y coordinate of the bounding box of the non-steel cap area, It is the bottom Y coordinate of the bounding box of the non-steel cap area.

[0078] Then by the formula 、 、 、 Calculate the four distances between the steel foot and the steel cap on the X-axis and Y-axis at the maximum cross section.

[0079] Mask image of the steel foot area through the largest cross section Calculate the bounding box of the steel foot area:

[0080] Mask map of the steel foot area traversing the maximum cross section For each pixel, record the pixel coordinates of each pixel whose brightness value is not 0 to form a pixel coordinate set: .in is the pixel coordinate of the steel foot area mask map, is the pixel horizontal coordinate of the steel foot area mask map, is the pixel ordinate of the steel foot area mask image, is the number of pixels in the horizontal direction, is the number of pixels in the vertical direction; traverse Set, find the maximum value of the pixel horizontal coordinate of the steel foot area mask map and minimum value , the maximum value of the pixel vertical coordinate of the steel foot area mask and minimum value ;but 、 、 、 , where F is the pixel size of the image.

[0081] Mask image of the non-steel cap area through the maximum cross section Calculate the bounding box of the non-steel cap area:

[0082] Traverse the non-steel cap area mask map of the maximum cross section For each pixel, record the pixel coordinates of each pixel whose brightness value is not 0 to form a pixel coordinate set: ,in is the pixel coordinate of the non-steel cap area mask, is the pixel horizontal coordinate of the non-steel cap area mask map, is the pixel ordinate of the non-steel cap area mask map, is the number of pixels in the horizontal direction, is the number of pixels in the vertical direction; traverse Set, find the maximum value of the pixel horizontal coordinate of the non-steel cap area mask map and minimum value , the maximum value of the pixel vertical coordinate of the non-steel cap area mask and minimum value ,but 、 、 、 , where F is the pixel size of the image.

[0083] Step S2-3-2: Calculate the distance between the widest point of the steel foot and the top of the photo, as well as the horizontal distance between the widest point and the left and right sides of the steel cap in the largest longitudinal section:

[0084] Assume that there is a two-dimensional plane coordinate system in the maximum longitudinal section, the origin O of the coordinate system is the upper left corner of the photo, the positive direction of the X axis is the horizontal direction from left to right, and the positive direction of the Y axis is the vertical direction from top to bottom; the distance between the widest position of the steel foot in the maximum longitudinal section and the top of the photo is , The distance between the widest position of the steel foot and the left side of the steel cap And the distance between the steel foot and the right side of the steel cap at the widest position of the steel foot ,like Figure 6 As shown. Since there are two longitudinal section groups of photos, 、 、 There are also two sets of data. Let the distance between the widest position of the steel foot in the largest longitudinal section of the longitudinal section 1 and the top of the photo be , The horizontal distances between the steel foot and the left and right sides of the steel cap at the widest position of the steel foot in the largest longitudinal section of group 1 are 、 .

[0085] Similarly, let the distance between the widest position of the steel foot in the largest longitudinal section of the two groups of longitudinal sections and the top of the photo be The horizontal distances between the widest position of the steel foot and the left and right sides of the steel cap in the largest longitudinal section of the two groups are 、 .

[0086] The distance between the widest position of the steel foot and the top of the photo in the largest longitudinal section of the first group The calculation process is as follows: Count the steel foot area mask of the largest longitudinal section of a group of longitudinal sections by row The sum of the brightness of the steel foot area mask image is obtained by ,in Mask map of the steel foot area The line number of any line, Mask map of the steel foot area No. The sum of the brightness of all pixels in the row, Mask map of the steel foot area Pixel height; find the set The row number with the highest brightness value in is , .

[0087] Similarly, the distance between the widest position of the steel foot and the top of the photo in the largest longitudinal section of the 2nd group of longitudinal sections is The calculation process is as follows: Count the steel foot area mask of the two groups of maximum longitudinal sections by row The sum of the brightness of the steel foot area mask image is obtained by .in Mask map of the steel foot area The line number of any line, Mask map of the steel foot area No. The sum of the brightness of all pixels in the row, Mask map of the steel foot area Pixel height; find the set The row number with the highest brightness value in is ,but .

[0088] The horizontal distances between the steel foot and the left and right sides of the steel cap at the widest position of the steel foot in the largest longitudinal section of the longitudinal section 1 are 、 The calculation process is as follows: Read the steel foot area mask map of the largest longitudinal section of the longitudinal section 1 group , find the Y-axis coordinate value of the widest position of the steel foot in the pixel coordinate system ; traverse No. For all pixels in the row, find the pixel coordinates of the first pixel whose brightness value is not 0 ; Find the pixel coordinates of the last pixel whose brightness value is not 0 ; Traverse the non-steel cap area mask map of the largest longitudinal section of the longitudinal section 1 group No. For all pixels in the row, find the pixel coordinates of the first pixel whose brightness value is not 0 ; Find the pixel coordinates of the last pixel whose brightness value is not 0 ;but ; .

[0089] Similarly, the horizontal distance between the widest position of the steel foot and the left and right sides of the steel cap in the largest longitudinal section of the 2nd group of longitudinal sections is 、 The calculation process is as follows: Read the steel foot area mask map of the two largest longitudinal sections , find the Y-axis coordinate value of the widest position of the steel foot in the pixel coordinate system ; traverse No. For all pixels in the row, find the pixel coordinates of the first pixel whose brightness value is not 0 and the pixel coordinates of the last pixel whose brightness value is not 0 ; Traverse the non-steel cap area mask map of the two largest longitudinal sections of the longitudinal section No. Line, find the pixel coordinates of the first pixel whose brightness value is not 0 ; Find the pixel coordinates of the last pixel whose brightness value is not 0 ,but: ; .

[0090] Specifically, extracting the cement layer and porcelain layer of any longitudinal and transverse sections includes the following sub-steps:

[0091] Step S3-1: Calculation of the horizontal projection line position of any cross section on the two largest longitudinal sections:

[0092] Step S3-2: Calculation of cross-section porcelain layers;

[0093] Step S3-3: Calculation of cross-section cement layer;

[0094] Step S3-4: Calculation of the longitudinal section porcelain layer;

[0095] Step S3-5: Calculation of the longitudinal section cement layer.

[0096] Among them, the horizontal projection line position of any cross section on the two largest longitudinal sections is calculated as follows:

[0097] The scene depth value of any known cross section , find the Y-axis coordinate value of the horizontal projection line of the cross section on the two largest longitudinal sections under the plane coordinate system A. The calculation process is as follows:

[0098] For any cross section , the calculation formula for the Y coordinate value of the horizontal projection line in the maximum longitudinal section of the longitudinal section 1 group (the Y axis coordinate under the plane coordinate system A) is: ,in is the y-axis coordinate of the horizontal projection line of the maximum cross section in the maximum longitudinal section of the longitudinal section 1 group, which is numerically the same as Equal, g is any cross section With the maximum cross section The scene depth distance.

[0099] For any cross section , the calculation formula for the Y coordinate value of the horizontal projection line in the maximum longitudinal section of the two groups of longitudinal sections (the Y axis coordinate under the plane coordinate system A) is: ,in is the y-axis coordinate of the horizontal projection line of the maximum transverse section in the maximum longitudinal section of the longitudinal section group 2, which is numerically the same as equal.

[0100] The calculation of cross-section porcelain layers includes:

[0101] Step S3-2-1: Calculation of the distance between the porcelain layer and the steel foot at the horizontal projection line of any cross section on the two largest longitudinal sections:

[0102] It is known that the Y coordinate value of the horizontal projection line of the cross section in the two largest longitudinal sections is 、 , calculate the distance between the inner and outer boundaries of the porcelain layer and the steel foot at the horizontal projection line position.

[0103] In the largest longitudinal section, after superimposing the porcelain layer area mask map and the steel foot area mask map, any given Y-axis coordinate value , and make a horizontal projection line, we can get The horizontal projection line of the 4 intersections with the mask boundary of the porcelain layer and the 2 intersections with the mask boundary of the steel foot, such as Figure 7 As shown, the intersection points from left to right are The intersection of the horizontal projection line and the left outer edge of the tile layer mask 、 The intersection of the horizontal projection line and the left inner boundary of the tile layer mask 、 The intersection of the horizontal projection line and the left boundary of the steel foot mask boundary 、 The intersection of the horizontal projection line and the right boundary of the steel foot mask boundary 、 The intersection of the horizontal projection line and the right inner boundary of the tile layer mask 、 The intersection of the horizontal projection line and the right outer edge of the tile layer mask .

[0104] Since the industrial CT inspection image contains two groups of longitudinal sections, corresponding to the two largest longitudinal sections, there will also be two groups of the above 6 intersection points. For a group of maximum longitudinal section photos of the longitudinal section, any given Y-axis coordinate value , let the intersection points be the largest longitudinal section of the longitudinal section group 1 The intersection of the horizontal projection line and the left outer edge of the tile layer mask , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the left inner boundary of the tile layer mask , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the left boundary of the steel foot mask boundary , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the right boundary of the steel foot mask boundary , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the right inner boundary of the tile layer mask , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the right outer edge of the tile layer mask ; Correspondingly, in the largest longitudinal section of the longitudinal section 1 group, the distance from the outside of the porcelain layer to the left side of the steel foot is , the distance from the inner side of the porcelain layer to the left side of the steel foot is , the distance from the outside of the porcelain layer to the right side of the steel foot is , the distance from the inner side of the porcelain layer to the right side of the steel foot is , where Dist is the distance calculation formula , for point The X-axis coordinate, for point The Y-axis coordinate, for point The X-axis coordinate, for point The Y-axis coordinate of .

[0105] For the largest longitudinal section photo in the two longitudinal sections, any given Y-axis coordinate value , let the intersection be the maximum longitudinal section of the two groups of longitudinal sections The intersection of the horizontal projection line and the left outer edge of the tile layer mask , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the left inner boundary of the tile layer mask , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the left boundary of the steel foot mask boundary , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the right boundary of the steel foot mask boundary , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the right inner boundary of the tile layer mask , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the right outer edge of the tile layer mask Correspondingly, in the largest longitudinal section of the 2nd group of longitudinal sections, the distance from the outside of the porcelain layer to the left side of the steel foot is , the distance from the inner side of the porcelain layer to the left side of the steel foot is , the distance from the outside of the porcelain layer to the right side of the steel foot is , the distance from the inner side of the porcelain layer to the right side of the steel foot is .

[0106] Step S3-2-2: Calculate the distance between the ceramic layer and the steel foot on the X and Y axes in any cross section:

[0107] It is known that the distances between the porcelain layer and the steel foot at the horizontal projection line of any cross section on the two largest longitudinal sections are: 、 It is necessary to calculate the distance between the inner and outer boundaries of the cross-section porcelain layer and the steel foot on the X and Y axes. Figure 8 As shown, there are 8 distance parameters in total, which are: the distance from the outside of the porcelain layer to the left side of the steel foot in the X-axis direction ; Distance from the inner side of the porcelain layer to the left side of the steel foot in the X-axis direction ; Distance from the outside of the porcelain layer to the right side of the steel foot in the X-axis direction ; Distance from the inner side of the porcelain layer to the right side of the steel foot in the X-axis direction ; Distance from the outside of the porcelain layer to the upper side of the steel foot in the Y-axis direction ; Distance from the inner side of the porcelain layer to the upper side of the steel foot in the Y-axis direction ; Distance from the outside of the porcelain layer to the lower side of the steel foot in the Y-axis direction ; Distance from the inner side of the porcelain layer to the lower side of the steel foot in the Y-axis direction .

[0108] Since the widest position of the steel foot in the largest longitudinal section is the horizontal projection position of the largest cross section, 、 、 、 and 、 、 、 The values of are pairwise equal, so by matching 、 、 、 and 、 、 、 The value of can be used to find the horizontal distance between the steel foot and the left and right sides of the steel cap at the widest position of the steel foot in the two largest longitudinal sections, which is the projection of the steel foot and the steel cap in the X-axis direction or the Y-axis direction of the largest cross section.

[0109] In addition, since the projection direction of any cross section on the maximum longitudinal section is the same as the projection direction of the maximum cross section on the maximum longitudinal section, the projection direction of any cross section on the maximum longitudinal section can be obtained based on the projection direction of the maximum cross section on the maximum longitudinal section.

[0110] After obtaining the projection direction of any cross section on the two largest longitudinal sections, combined with the horizontal projection position of the cross section on the two largest longitudinal sections 、 As well as the two largest longitudinal sections of the porcelain layer area mask map and the steel foot area mask map, and then get the horizontal projection position 、 The distance between the porcelain layer and the steel foot 、 、 、 、 、 、 、 .

[0111] Finally, combining the projection directions of the maximum longitudinal sections of the longitudinal section 1 group and the maximum longitudinal sections of the longitudinal section 2 group on any cross section, we can find 、 、 、 、 、 、 、 and 、 、 、 、 、 、 、 relationship.

[0112] The specific matching rules are as follows:

[0113] when , When , the maximum steel foot position of the largest longitudinal section of group 1 is the projection of the cross section in the X-axis direction; the maximum steel foot position of the largest longitudinal section of group 2 is the projection of the cross section in the Y-axis direction. The X-axis direction of the largest longitudinal section of group 1 is the same as the X-axis direction of the cross section, and the X-axis direction of the largest longitudinal section of group 2 is the same as the Y-axis direction of the cross section. Then the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

[0114] when , When , the maximum steel foot position of the largest longitudinal section of the first group of longitudinal sections is the projection of the cross section in the X-axis direction. The maximum steel foot position of the largest longitudinal section of the second group of longitudinal sections is the projection of the cross section in the Y-axis direction. The X-axis direction of the largest longitudinal section of the first group of longitudinal sections is opposite to the X-axis direction of the cross section, and the X-axis direction of the largest longitudinal section of the second group of longitudinal sections is the same as the Y-axis direction of the cross section. Then the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

[0115] when , When , the maximum steel foot position of the largest longitudinal section of the longitudinal section 1 is the projection of the X-axis direction of the cross section. The maximum steel foot position of the largest longitudinal section of the longitudinal section 1 is the projection of the Y-axis direction of the cross section. The X-axis direction of the largest longitudinal section of the longitudinal section 1 is the same as the X-axis direction of the cross section, and the X-axis direction of the largest longitudinal section of the longitudinal section 2 is opposite to the Y-axis direction of the cross section. Then the distance parameter values of the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

[0116] when , When , the maximum steel foot position of the largest longitudinal section of the first group of longitudinal sections is the projection of the cross section in the X-axis direction. The maximum steel foot position of the largest longitudinal section of the second group of longitudinal sections is the projection of the cross section in the Y-axis direction. The X-axis direction of the largest longitudinal section of the first group of longitudinal sections is opposite to the X-axis direction of the cross section, and the X-axis direction of the largest longitudinal section of the second group of longitudinal sections is opposite to the Y-axis direction of the cross section. Then the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

[0117] when , When , the maximum steel foot position of the largest longitudinal section of the first group of longitudinal sections is the projection of the Y-axis direction of the cross section. The maximum steel foot position of the largest longitudinal section of the second group of longitudinal sections is the projection of the X-axis direction of the cross section. The X-axis direction of the largest longitudinal section of the first group of longitudinal sections is the same as the Y-axis direction of the cross section, and the X-axis direction of the largest longitudinal section of the second group of longitudinal sections is the same as the X-axis direction of the cross section. Then the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

[0118] when , When , the maximum steel foot position of the largest longitudinal section of the first group of longitudinal sections is the projection of the Y-axis direction of the cross section. The maximum steel foot position of the largest longitudinal section of the second group of longitudinal sections is the projection of the X-axis direction of the cross section. The X-axis direction of the largest longitudinal section of the first group of longitudinal sections is opposite to the Y-axis direction of the cross section, and the X-axis direction of the largest longitudinal section of the second group of longitudinal sections is the same as the X-axis direction of the cross section. Then the distance parameter values of the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

[0119] when , When , the maximum steel foot position of the largest longitudinal section of the first group of longitudinal sections is the projection of the Y-axis direction of the cross section. The maximum steel foot position of the largest longitudinal section of the second group of longitudinal sections is the projection of the X-axis direction of the cross section. The X-axis direction of the largest longitudinal section of the first group of longitudinal sections is the same as the Y-axis direction of the cross section, and the X-axis direction of the largest longitudinal section of the second group of longitudinal sections is opposite to the X-axis direction of the cross section. Then the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

[0120] when , When , the maximum steel foot position of the largest longitudinal section of the first group of longitudinal sections is the projection of the Y-axis direction of the cross section. The maximum steel foot position of the largest longitudinal section of the second group of longitudinal sections is the projection of the X-axis direction of the cross section. The X-axis direction of the largest longitudinal section of the first group of longitudinal sections is opposite to the Y-axis direction of the cross section, and the X-axis direction of the largest longitudinal section of the second group of longitudinal sections is opposite to the X-axis direction of the cross section. Then the distance parameter values of the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

[0121] Step S3-2-3: Drawing of the porcelain layer area in any cross section:

[0122] When calculating the distance between the porcelain layer and the steel foot in the X and Y axis directions of any cross section 、 、 、 、 、 、 、 After that, 8 points on the inner and outer boundaries of the porcelain layer can be determined on the cross section, of which the upper, lower, left and right points on the outer boundary are , the four points on the inner boundary are: ,like Figure 9 As shown. Create a mask map with the same size as the cross section, with all default brightness values set to 0. On this mask map, use the four points on the outer boundary: , 4 points on the inner boundary: The coordinate values of are used to draw the inner and outer rings of the ceramic part respectively. The brightness between the inner and outer rings is filled to 1. Finally, a mask image of the ceramic layer area matching the cross section is obtained, as shown in the figure below. Figure 10 shown.

[0123] The calculation process of the cross-section cement layer is as follows: after obtaining the mask map of the porcelain layer area in the cross section Then, the non-steel cap area mask of the cross section is extracted by calling the segmentation algorithm. Subtract the brightness of the corresponding coordinate positions of the two porcelain layer area masks and the non-steel cap area mask map, and the resulting cement layer mask map of the cross section is obtained. .

[0124] The calculation process of the longitudinal section porcelain layer is as follows:

[0125] Step S3-4-1, calculate the distance between the inner and outer boundaries of the porcelain layer at any longitudinal section at any height and the central axis of the photograph.

[0126] Take a set of longitudinal section photos as an example, and take any cross section M j As the projection surface, any longitudinal section Z i and the maximum longitudinal section Z max Projection to M j On top; then its projection relationship is as follows Figure 11 As shown, there is the following relationship:

[0127] 1. The two projection lines are parallel, and the distance between the two projection lines is the difference between the scene depth values of any longitudinal section and the maximum longitudinal section.

[0128] 2. When shooting any longitudinal section, the range of the detector is the same, so Z i and Z max In M j The lengths of the projection lines on are the same.

[0129] 3. Based on the above two points, it can be inferred that any longitudinal section Z i In M j The center position P of the projection line on i With the maximum longitudinal section at M j The line connecting the center position P0 of the projection line on the cross section M j The X-axis is perpendicular to the line segment P i P0 is perpendicular to the X-axis.

[0130] like Figure 12 As shown, based on Figure 11 On the basis of the above, the cross-section of the porcelain layer area is drawn out, and P0 is connected with any longitudinal section Z i The intersection of the projection line and the inner and outer boundaries of the porcelain layer, Figure 12 The meanings of the intersection points are: P a is the cross section M j The inner side of the porcelain layer and the longitudinal section Z i The left intersection point of the projection line, P b is the cross section M j The outer side of the porcelain layer and the longitudinal section Z i The right intersection point of the projection line, P c is the cross section M j The inner side of the porcelain layer and the longitudinal section Z max The left intersection point of the projection line, P d is the cross section M j The outer side of the porcelain layer and the longitudinal section Z max The right intersection point of the projection lines.

[0131] based on Figure 12 , we have the following conclusions:

[0132] 1. Line segment The length meets ,in is the scene depth value of the maximum longitudinal section, is the scene depth value of any longitudinal section.

[0133] 2. Line Segment The length satisfies the formula ,in For line segments Length, For line segments length.

[0134] 3. Since the porcelain layer is annular, ,so .

[0135] 4. Similarly, line segments The length satisfies the formula ,in For line segments Length, For line segments length.

[0136] 5. Similarly, since the porcelain layer is annular, ,therefore .

[0137] 6. During the above process, The value is equal to the cross section M j At the maximum longitudinal section Z max At the position of the upper projection line, the center point of the projection line P cj To the right outer boundary point P of the porcelain layer fj distance, The value is equal to the cross section M j At the maximum longitudinal section Z max At the position of the upper projection line, the center point of the projection line P cj To the left inner boundary point P of the porcelain layer bj The distance, such as Figure 13 shown.

[0138] The distance between the projection line and the top of the photo , is the y-axis coordinate of the horizontal projection line of the maximum cross section in the maximum longitudinal section of the longitudinal section 1 group, which is numerically the same as equal.

[0139] therefore Figure 13 middle, , .

[0140] Since the height of the photos in the same set of longitudinal sections is the same, the cross section M j At the projection position of any longitudinal section Zi, the right outer boundary of the porcelain layer of the longitudinal section Zi and the cross section M j The distance R from the center point of the projection position mj Numerically equal to The distance E between the left inner boundary of the porcelain layer and the center point of the horizontal position mj Numerically equal to .

[0141] In the same way, the distance between the longitudinal section Zi and the top of the photo can be calculated. At the horizontal position, the distance between the left outer boundary of the porcelain layer and the center point of the horizontal position is assumed to be B mj , and the distance between the right inner boundary of the porcelain layer and the center point of the horizontal position, assuming it is Q mj .

[0142] Through the above process, any longitudinal section Z i At any height The distance between the inner and outer boundaries of the porcelain layer and the center point of the photo is calculated, and finally a photo Z is formed. i At any height, the set of distances between the inner and outer boundaries of the ceramic layer and the center point of the photo ,in For at height The distance between the left outer edge of the porcelain layer and the center point of the horizontal position at that height, For at height The distance between the left inner boundary of the porcelain layer and the center point of the horizontal position at that height, For at height The distance between the right inner boundary of the porcelain layer and the center point of the horizontal position at that height, For at height The distance between the right outer edge of the porcelain layer and the center point of the horizontal position at that height.

[0143] Step S3-4-2: Calculation of the porcelain layer area of any longitudinal section:

[0144] Step S3-4-2-1: Create a vertical section Z i Mask images of the same size , its brightness defaults to 0;

[0145] Step S3-4-2-2: Assume that the longitudinal section is in plane coordinate system A and its width is , the height is , the pixel size is ;

[0146] Step S3-4-2-3: Create an empty collection , used to record the coordinates of points on the boundary of the porcelain layer;

[0147] Step S3-4-2-4: Based on the plane coordinate system A, define a variable to record the y-axis coordinate value , the initial value is 0;

[0148] Step S3-4-2-5: Calculate the height value The set of distances between the inner and outer boundaries of the lower porcelain layer and the central axis of the photo ;

[0149] Step S3-4-2-6: When If it is not empty, the coordinate point 、 、 、 Add to collection In the figure, the position relationship of the four coordinate points on the boundary is as follows: Figure 14 As shown;

[0150] Step S3-4-2-7: The value increases the size of the pixel by one , repeat steps S3-3-2-5 to S3-4-2-6 until Finish;

[0151] Step S3-4-2-8: Gather The coordinates of the points in the mask map Draw a closed polygon on it, fill the brightness of the polygon area with 1, and finally the mask map It is the longitudinal section Z i The mask map of the porcelain layer area.

[0152] The longitudinal section cement layer calculation referred to in step S3-5 involves drawing a cement layer region mask for all longitudinal section images. For each longitudinal section image, the brightness of each pixel in the non-steel cap region mask is subtracted from the brightness of the corresponding pixel in the steel foot region mask and the porcelain layer region mask to obtain the cement layer region mask for the longitudinal section image.

[0153] In this embodiment, the specific process of step 2 is as follows:

[0154] Step A1: For any industrial CT inspection photo, use the yolov11 instance segmentation algorithm to perform defect detection. Suppose each defect mask in the industrial CT inspection photo is The mask image of the ceramic layer area of the industrial CT inspection photo is ;

[0155] Step A2: Traverse , find the pixels whose brightness value is not 0, and set the pixel coordinates Record in a collection , is the horizontal coordinate of the i-th pixel of the s-th defect mask image, is the horizontal coordinate of the i-th pixel of the s-th defect mask image;

[0156] Step A3: Iterate over the collection For each pixel coordinate in the image, read the brightness value of the pixel at the same position in the porcelain layer area mask image; if the brightness value of the pixel at the corresponding position in the porcelain layer area mask image is not 0, the defect is a porcelain layer crack, otherwise it is a cement layer pore.

[0157] The above disclosures are merely some preferred embodiments of the present invention, and certainly cannot be used to limit the scope of the present invention. A person skilled in the art will understand that all or part of the processes of the above embodiments and equivalent changes made in accordance with the claims of the present invention still fall within the scope of the invention.

Claims

1. An intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection, characterized in that: include: Step 1: Collect industrial CT inspection photos of porcelain insulators. The industrial CT inspection photos include two groups of longitudinal section photos and one group of cross-section photos. The two groups of longitudinal section photos are respectively called longitudinal section group 1 and longitudinal section group 2. The cross-section photo with the largest steel foot area in the same group of photos is called the largest cross-section; the longitudinal section photo with the largest steel foot area in the same group of photos is called the largest longitudinal section. Extract the cement layer and porcelain layer in the cross-section and longitudinal sections; including: S1: Extract scene depth and pixel size information of each industrial CT inspection photo; S2: Extract the size features of the maximum longitudinal and transverse sections; S2-1: Find two maximum longitudinal sections and one maximum transverse section from the industrial CT inspection photos; S2-2: Use the calibrated dataset to train the YOLOv11 instance segmentation algorithm, and obtain the non-steel cap area recognition model, the porcelain layer area recognition model, and the steel foot area recognition model. Perform instance segmentation on the industrial CT inspection photos and extract the non-steel cap area, porcelain layer area, and steel foot area in turn; Step S2-3: Extracting the size features of the largest cross-section photo and the largest longitudinal section; Step S2-3-1: Let the origin O of the plane coordinate system A in the maximum cross-section photo be the upper left corner of the photo, the positive direction of the X axis be the horizontal direction from left to right, and the positive direction of the Y axis be the vertical direction from top to bottom. The distance between the steel foot and the steel cap on the maximum cross-section on the X axis and Y axis includes four distances, namely the distance between the steel foot and the steel cap on the negative direction of the X axis , the distance between the steel foot and the steel cap in the positive direction of the X axis , the distance between the steel foot and the steel cap in the negative direction of the Y axis , the distance between the steel foot and the steel cap in the positive direction of the Y axis ; The steel foot area is bounded by the steel foot in the plane coordinate system A of the maximum cross-sectional photo , and the bounding box of the non-steel cap area in the plane coordinate system A in the maximum cross-sectional view ;in The leftmost X coordinate of the steel foot bounding box, The rightmost X coordinate of the steel foot bounding box, is the top Y coordinate of the steel foot bounding box, The Y coordinate of the bottom side of the steel foot bounding box; is the leftmost X coordinate of the bounding box of the non-steel cap area, is the rightmost X coordinate of the bounding box of the non-steel cap area, is the top Y coordinate of the bounding box of the non-steel cap area, is the bottom Y coordinate of the bounding box of the non-steel cap area; through the formula 、 、 、 Calculate the four distances between the steel foot and the steel cap on the X and Y axes at the maximum cross section; Step S2-3-2: Calculate the distance between the widest position of the steel foot and the top of the photo, as well as the horizontal distance between the widest position and the left and right sides of the steel cap in the largest longitudinal section; Step S3: extracting the cement layer and porcelain layer of any longitudinal and transverse sections based on the projection relationship; Step S3-1: Calculate the position of the horizontal projection line of any cross section on the two largest longitudinal sections; Step S3-2: Calculation of cross-section porcelain layers; Step S3-3: Calculation of cross-section cement layer; Step S3-4: Calculation of the longitudinal section porcelain layer; Step S3-5: Calculation of longitudinal section cement layer; Step 2: Use the yolov11 instance segmentation algorithm to identify the defect areas contained in the industrial CT inspection photos, and combine the cement layer and porcelain layer to classify the defects into cement layer pore defects and porcelain layer crack defects.

2. The intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to claim 1 is characterized in that: The corresponding maximum cross-section mask map set is obtained in S2-2 ; A set of maximum longitudinal section mask images for the longitudinal section ; 2 sets of maximum longitudinal section mask maps ,in They are the mask images of the non-steel cap area of the maximum cross section, the maximum longitudinal section of the longitudinal section group 1, and the maximum longitudinal section of the longitudinal section group 2; These are the mask images of the porcelain layer area of the largest cross section, the largest longitudinal section of group 1 of longitudinal sections, and the largest longitudinal section of group 2 of longitudinal sections; They are the steel foot area mask images of the maximum cross section, the maximum longitudinal section of the longitudinal section 1 group, and the maximum longitudinal section of the longitudinal section 2 group.

3. The intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to claim 2 is characterized in that: Mask image of the steel foot area through the largest cross section Calculate the bounding box of the steel foot area: Mask map of the steel foot area traversing the maximum cross section For each pixel, record the pixel coordinates of each pixel whose brightness value is not 0 to form a pixel coordinate set: ;in is the pixel coordinate of the steel foot area mask map, is the pixel horizontal coordinate of the steel foot area mask map, is the pixel ordinate of the steel foot area mask image, is the number of pixels in the horizontal direction, is the number of pixels in the vertical direction; traverse Set, find the maximum value of the pixel horizontal coordinate of the steel foot area mask map and minimum value , the maximum value of the pixel vertical coordinate of the steel foot area mask and minimum value ;but 、 、 、 , where F is the pixel size of the image; Mask image of the non-steel cap area through the maximum cross section Calculate the bounding box of the non-steel cap area: Traverse the non-steel cap area mask map of the maximum cross section For each pixel, record the pixel coordinates of each pixel whose brightness value is not 0 to form a pixel coordinate set: ,in is the pixel coordinate of the non-steel cap area mask, is the pixel horizontal coordinate of the non-steel cap area mask map, is the pixel ordinate of the non-steel cap area mask image, is the number of pixels in the horizontal direction, is the number of pixels in the vertical direction; traverse Set, find the maximum value of the pixel horizontal coordinate of the non-steel cap area mask map and minimum value , the maximum value of the pixel vertical coordinate of the non-steel cap area mask and minimum value ,but 、 、 、 , where F is the pixel size of the image.

4. The intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to claim 3 is characterized in that: Step S2-3-2 specifically includes: The distance between the widest position of the steel foot and the top of the photo in the largest longitudinal section of the first group The calculation process is as follows: Count the steel foot area mask of the largest longitudinal section of a group of longitudinal sections by row The sum of the brightness of the steel foot area mask image is obtained by ,in Mask map of the steel foot area The line number of any line, Mask map of the steel foot area No. The sum of the brightness of all pixels in the row, Mask map of the steel foot area Pixel height; find the set The row number with the highest brightness value in is , ; The distance between the widest part of the steel foot and the top of the photo in the largest longitudinal section of the two groups of longitudinal sections The calculation process is as follows: Count the steel foot area mask of the two groups of maximum longitudinal sections by row The sum of the brightness of the steel foot area mask image is obtained by ;in Mask map of the steel foot area The line number of any line, Mask map of the steel foot area No. The sum of the brightness of all pixels in the row, Mask map of the steel foot area Pixel height; find the set The row number with the highest brightness value in is ,but ; The horizontal distances between the steel foot and the left and right sides of the steel cap at the widest position of the steel foot in the largest longitudinal section of the longitudinal section 1 are 、 The calculation process is as follows: Read the steel foot area mask map of the largest longitudinal section of the longitudinal section 1 group , find the Y-axis coordinate value of the widest position of the steel foot in the pixel coordinate system ; traverse No. For all pixels in the row, find the pixel coordinates of the first pixel whose brightness value is not 0 ; Find the pixel coordinates of the last pixel whose brightness value is not 0 ; Traverse the non-steel cap area mask map of the largest longitudinal section of the longitudinal section 1 group No. For all pixels in the row, find the pixel coordinates of the first pixel whose brightness value is not 0 ; Find the pixel coordinates of the last pixel whose brightness value is not 0 ;but ; ; The horizontal distance between the widest position of the steel foot and the left and right sides of the steel cap in the largest longitudinal section of the 2nd group of longitudinal sections 、 The calculation process is as follows: Read the steel foot area mask map of the maximum longitudinal section of the 2 groups of longitudinal sections , find the Y-axis coordinate value of the widest position of the steel foot in the pixel coordinate system ; traverse No. For all pixels in the row, find the pixel coordinates of the first pixel whose brightness value is not 0 and the pixel coordinates of the last pixel whose brightness value is not 0 ; Traverse the non-steel cap area mask map of the two largest longitudinal sections of the longitudinal section No. Line, find the pixel coordinates of the first pixel whose brightness value is not 0 ; Find the pixel coordinates of the last pixel whose brightness value is not 0 ,but: ; .

5. The intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to claim 4 is characterized in that: The calculation of the cross-section porcelain layer includes: Step S3-2-1: Calculate the distance between the porcelain layer and the steel foot at the position of the horizontal projection line of any cross section on the two largest longitudinal sections; Step S3-2-2: Calculate the distance between the porcelain layer and the steel foot on the X-axis and Y-axis in any cross section; Step S3-2-3: Draw the porcelain layer area in any cross section.

6. The intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to claim 5 is characterized in that: In step S3-2-1, It is known that the Y coordinate value of the horizontal projection line of the cross section in the two largest longitudinal sections is 、 , calculate the distance between the inner and outer boundaries of the porcelain layer and the steel foot at the horizontal projection line position; In the largest longitudinal section, after superimposing the porcelain layer area mask map and the steel foot area mask map, any given Y-axis coordinate value , make a horizontal projection line, and get The four intersection points of the horizontal projection line with the boundary of the porcelain layer mask and the two intersection points of the steel foot mask boundary are as follows from left to right: The intersection of the horizontal projection line and the left outer edge of the tile layer mask 、 The intersection of the horizontal projection line and the left inner boundary of the tile layer mask 、 The intersection of the horizontal projection line and the left boundary of the steel foot mask boundary 、 The intersection of the horizontal projection line and the right boundary of the steel foot mask boundary 、 The intersection of the horizontal projection line and the right inner boundary of the tile layer mask 、 The intersection of the horizontal projection line and the right outer edge of the tile layer mask ; For a set of maximum longitudinal section photos, any given Y-axis coordinate value , let the intersection points be the largest longitudinal section of the longitudinal section group 1 The intersection of the horizontal projection line and the left outer edge of the tile layer mask , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the left inner boundary of the tile layer mask , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the left boundary of the steel foot mask boundary , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the right boundary of the steel foot mask boundary , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the right inner boundary of the tile layer mask , longitudinal section 1 group of the largest longitudinal section The intersection of the horizontal projection line and the right outer edge of the tile layer mask ; Correspondingly, in the largest longitudinal section of the longitudinal section 1 group, the distance from the outside of the porcelain layer to the left side of the steel foot is , the distance from the inner side of the porcelain layer to the left side of the steel foot is , the distance from the outside of the porcelain layer to the right side of the steel foot is , the distance from the inner side of the porcelain layer to the right side of the steel foot is , where Dist is the distance calculation formula , for point The X-axis coordinate, for point The Y-axis coordinate, for point The X-axis coordinate, for point The Y-axis coordinate of ; For the largest longitudinal section photo in the two longitudinal sections, any given Y-axis coordinate value , let the intersection be the maximum longitudinal section of the two groups of longitudinal sections The intersection of the horizontal projection line and the left outer edge of the tile layer mask , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the left inner boundary of the tile layer mask , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the left boundary of the steel foot mask boundary , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the right boundary of the steel foot mask boundary , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the right inner boundary of the tile layer mask , the largest longitudinal section of the two groups The intersection of the horizontal projection line and the right outer edge of the tile layer mask Correspondingly, in the largest longitudinal section of the 2nd group of longitudinal sections, the distance from the outside of the porcelain layer to the left side of the steel foot is , the distance from the inner side of the porcelain layer to the left side of the steel foot is , the distance from the outside of the porcelain layer to the right side of the steel foot is , the distance from the inner side of the porcelain layer to the right side of the steel foot is .

7. The intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to claim 6 is characterized in that: Step S3-2-2: Calculate the distance between the ceramic layer and the steel foot on the X and Y axes in any cross section: It is known that the distances between the porcelain layer and the steel foot at the horizontal projection line of any cross section on the two largest longitudinal sections are: 、 ; It is necessary to calculate the distance between the inner and outer boundaries of the cross-section porcelain layer and the steel foot on the X-axis and Y-axis, a total of 8 distance parameters, which are: the distance from the outer side of the porcelain layer to the left side of the steel foot in the X-axis direction ; Distance from the inner side of the porcelain layer to the left side of the steel foot in the X-axis direction ; Distance from the outside of the porcelain layer to the right side of the steel foot in the X-axis direction ; Distance from the inner side of the porcelain layer to the right side of the steel foot in the X-axis direction ; Distance from the outside of the porcelain layer to the upper side of the steel foot in the Y-axis direction ; Distance from the inner side of the porcelain layer to the upper side of the steel foot in the Y-axis direction ; Distance from the outside of the porcelain layer to the lower side of the steel foot in the Y-axis direction ; Distance from the inner side of the porcelain layer to the lower side of the steel foot in the Y-axis direction ; Since the widest position of the steel foot in the largest longitudinal section is the horizontal projection position of the largest cross section, 、 、 、 and 、 、 、 The values of are pairwise equal, so by matching 、 、 、 and 、 、 、 The value of , and then the horizontal distance between the steel foot and the left and right sides of the steel cap at the widest position of the steel foot in the two largest longitudinal sections can be found, which is the projection of the steel foot and the steel cap in the X-axis direction or the Y-axis direction of the largest cross section; After obtaining the projection direction of any cross section on the two largest longitudinal sections, combined with the horizontal projection position of the cross section on the two largest longitudinal sections 、 As well as the two largest longitudinal sections of the porcelain layer area mask map and the steel foot area mask map, and then get the horizontal projection position 、 The distance between the porcelain layer and the steel foot 、 、 、 、 、 、 、 ; Finally, combining the projection direction of the maximum longitudinal section of the longitudinal section 1 group and the maximum longitudinal section of the longitudinal section 2 group on any cross section, we find 、 、 、 、 、 、 、 and 、 、 、 、 、 、 、 relationship; The specific matching rules are as follows: when , When , the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 ; when , When , the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 ; when , When , the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 ; when , When , the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 ; when , When , the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 ; when , When , the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 ; when , When , the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 ; when , When , the distance parameter values between the porcelain piece and the steel foot on the X-axis and Y-axis in the cross section are: 、 、 、 、 、 、 、 .

8. The intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to claim 7 is characterized in that: The calculation of the longitudinal section porcelain layer includes: Step S3-4-1: Calculate the distance between the inner and outer boundaries of the porcelain layer at any height in any longitudinal section and the central axis of the photograph; Step S3-4-2: Calculation of the porcelain layer area of any longitudinal section.

9. The intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to claim 1 is characterized in that: The specific process of step 2 is as follows: Step A1: For any industrial CT inspection photo, use the yolov11 instance segmentation algorithm to perform defect detection; let each defect mask in the industrial CT inspection photo be The mask image of the ceramic layer area of the industrial CT inspection photo is ; Step A2: Traverse , find the pixels whose brightness value is not 0, and set the pixel coordinates Record in a collection , is the horizontal coordinate of the i-th pixel of the s-th defect mask image, is the horizontal coordinate of the i-th pixel of the s-th defect mask image; Step A3: Iterate over the collection For each pixel coordinate in the image, read the brightness value of the pixel at the same position in the porcelain layer area mask image; if the brightness value of the pixel at the corresponding position in the porcelain layer area mask image is not 0, the defect is a porcelain layer crack, otherwise it is a cement layer pore.

10. A system for implementing the intelligent diagnosis method for micro-defects of porcelain insulators based on industrial CT detection according to any one of claims 1 to 9, characterized in that: include: Porcelain insulator structure analysis module: collects industrial CT inspection photos of porcelain insulators and extracts cement layers and porcelain layers in cross-sections and longitudinal sections; Porcelain insulator defect detection module: This module uses the Yolov11 instance segmentation algorithm to identify defect areas in industrial CT inspection photos. It then classifies defects into cement layer porosity defects and porcelain layer crack defects based on the cement layer and porcelain layer.

Images