Intelligent detection method for surface defects of fine grinding tool

Abstract

The invention relates to the technical field of image processing, in particular to an intelligent detection method for surface defects of a fine grinding tool, which comprises the following steps: acquiring a surface image of a diamond grinding wheel, segmenting the surface image by using an image segmentation model to obtain a working layer image, and establishing a polar coordinate system by taking the central point of the working layer image as an original point, and acquiring a pixel point sequence formed by pixel points on the working layer image through which the ray passes every preset angle in the polar coordinate. According to the invention, the importance of the position of each pixel point in the working layer image, for example, the importance of the pixel point in the middle area or the importance of the area close to the edge, is different, and the diamond grinding wheel mainly utilizes the grinding characteristics of the middle area of the working layer and the diamond particles during working; and the importance degree of each pixel point is obtained and is used in the subsequent process of calculating the defects of the diamond grinding wheel, so that the accuracy of a calculation result is improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and more specifically, to an intelligent detection method for surface defects of fine grinding tools. Background Art

[0002] The hardness of silicon carbide substrate is relatively high, with a Mohs hardness of 9.25, second only to diamond. Silicon carbide substrate has the characteristics of high hardness, high brittleness, good wear resistance, and extremely stable chemical properties, which makes the precision processing of silicon carbide substrate very difficult. Ordinary grinding wheels cannot meet the requirements of silicon carbide substrate chamfering in terms of processing life and processing efficiency. Diamond grinding wheels are circular bonded abrasives with a through hole in the center, made of diamond abrasives as raw materials, metal powder, resin powder, ceramics and electroplated metal as binders. Diamond grinding wheels are different from ordinary abrasive grinding wheels in structure. They are generally composed of diamond abrasive layers, transition layers and substrates. The working layer is also called the diamond layer, which is composed of abrasives, binders and fillers. It is the working part of the grinding wheel, and diamond grinding wheels can be used for rough grinding and fine grinding stages. Diamond grinding wheels have good cutting performance and can effectively remove irregularities and roughness on the surface of silicon carbide.

[0003] At present, when detecting defects in diamond grinding wheels, the main focus is on the working layer of the diamond grinding wheel. The working layer of the diamond grinding wheel is composed of abrasives, binders and fillers. Therefore, when defects appear in the working layer, such as particle shedding and pits, the grayscale value difference between the defective area and the normal area on the working layer will be small, making it difficult to identify, and therefore the defects cannot be accurately detected. Summary of the invention

[0004] The present invention provides an intelligent detection method for surface defects of fine abrasive tools, aiming to solve the problem in the related art that when defects occur in the working layer, such as particle shedding and pits, the grayscale value difference between the defective area and the normal area on the working layer is small, which makes it difficult to identify and therefore the defect cannot be accurately detected.

[0005] The present invention provides an intelligent detection method for surface defects of fine grinding tools, comprising: collecting a surface image of a diamond grinding wheel, segmenting the surface image using an image segmentation model to obtain a working layer image, establishing a polar coordinate system with the center point of the working layer image as the origin, and obtaining a pixel point sequence composed of pixel points on the working layer image passed by rays at every preset angle in the polar coordinates; calculating the importance and defect possibility of the pixel points in the pixel point sequence at each angle in the working layer image, and obtaining the corresponding importance degree sequence and defect possibility sequence at each angle, wherein the importance of the pixel point reflects the distance between the position of the pixel point and the central area and the height of the gray value of the pixel point, and the defect possibility of the pixel point reflects the degree of change of the depth value of the pixel point and the depth values ​​of the surrounding pixels; calculating the difference between the defect possibility sequences at any two angles as the defect degree of the working layer image, and performing defect detection on the surface of the grinding tool based on the size of the defect degree, wherein the difference is also related to the importance of the pixel point. By using the image segmentation model to extract the working layer image and processing the pixel points in combination with the polar coordinate system, the entire working surface can be effectively decomposed into sequences at multiple angles, and the difference in the defect possibility sequence between angles can be calculated. This difference can be used to accurately evaluate the defect degree of the diamond grinding wheel.

[0006] Furthermore, defect detection is performed on the surface of the grinding tool based on the magnitude of the defect, including: if the defect degree of the working layer image is greater than or equal to a preset error, it is determined that a defect occurs on the surface of the diamond grinding wheel, and an alarm is issued.

[0007] Furthermore, the calculation formula of the difference is: ; In the formula, For the The defect possibility sequence under the angle and the The difference between the defect probability sequences is For the The first in the importance sequence under each angle The importance of each pixel, Indicates The first in the importance sequence under each angle The importance of each pixel; is the number of pixels that the ray passes through at any angle, For the The first in the defect possibility sequence under the angle The probability of pixel defects is For the The first in the defect possibility sequence under the angle The probability of pixel defects is Represents a normalized function. This difference calculation formula refines the combination of defect probability and importance, making it more sensitive in the detection of local and minor defects. Considering the importance increases the sensitivity to key areas and can more keenly reveal surface problems.

[0008] Furthermore, collecting the surface image of the diamond grinding wheel includes: photographing the surface of the diamond grinding wheel using a depth camera to obtain the surface image of the diamond grinding wheel.

[0009] Furthermore, the surface image is segmented using the image segmentation model, including: inputting the surface image into the image segmentation model and outputting the working layer image. By inputting the working layer image, the segmentation model can automatically distinguish the working layer from other irrelevant parts and extract the area related to defect detection. In this way, the subsequent analysis can be focused on the working layer without processing all the information in the entire image, thereby greatly improving the efficiency and accuracy of subsequent processing.

[0010] Furthermore, the training process of the image segmentation model includes: marking the pixel points belonging to the working layer in the surface image of the diamond grinding wheel as 1, and the pixel points belonging to the non-working layer as 0, to obtain a 0-1 marked image; inputting the 0-1 marked image into the semantic segmentation network model for training; and completing the training in response to the loss function being less than a preset value or reaching a preset number of training times to obtain an image segmentation model.

[0011] Furthermore, obtaining the pixel point sequence includes: establishing a ray in polar coordinates at every preset angle, and forming a pixel point sequence by forming all pixel points of the working layer image through which the ray passes, thereby obtaining a pixel point sequence corresponding to each angle.

[0012] Further, the importance of each pixel in the working layer image is obtained, including: performing edge detection on the working layer image to obtain two circular edges, the two edges being a first circular edge and a second circular edge; the importance of each pixel in the working layer image is related to the difference in distance between the pixel and the first circular edge and the second circular edge, respectively, and is also related to the difference between the gray value of the pixel and the average gray value of all pixels in the working layer image. By comprehensively considering the difference in distance between the pixel and the edge and the difference in gray value, the importance of each pixel in the working layer image can be more comprehensively evaluated, thereby enhancing the accuracy and reliability of detection.

[0013] Furthermore, the defect possibility of each pixel is obtained, including: establishing a sliding window in the working layer image, calculating the average value of the difference between the depth value of the pixel and all the pixel points in the sliding window, and taking the product of the average value and the variance of the depth value of all the pixel points in the sliding window as the defect possibility of the pixel point. The difference and variance of the local depth value reflect the surface morphology and possible damage (such as wear, cracks or depressions) of the area, which can reveal subtle defects in the mold surface that are not easy to capture through global information, thereby improving the sensitivity of detection.

[0014] Furthermore, the preset angle is 1 degree.

[0015] Beneficial effects: (I) The importance of the location of each pixel in the working layer image is determined, for example, the pixel is in the middle area or near the edge area. Based on the fact that the diamond wheel mainly uses the grinding characteristics of the middle area of ​​the working layer and the diamond particles during operation, the importance of each pixel is obtained and used in the subsequent steps of calculating the defects of the diamond wheel, thereby improving the accuracy of the calculation results.

[0016] (ii) By calculating the defect possibility of each pixel point through the depth change of each pixel point in the working layer image, and using the importance of the pixel point as the weight to weight the defect possibility of the pixel point, the accurate defect degree can be obtained, thereby improving the accuracy of defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic diagram schematically showing a top view of a diamond grinding wheel according to an embodiment of the present invention; Figure 2 is a flowchart schematically illustrating a method for calculating a defect degree of a working layer image according to an embodiment of the present invention; DETAILED DESCRIPTION The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0018] like Figure 1 and Figure 2 As shown, S101: collecting grinding wheel images.

[0019] Specifically, a diamond grinding wheel is photographed using a camera device, which is a depth camera, to obtain a surface image of the diamond grinding wheel. The surface image of the diamond grinding wheel is grayscaled, and the grayscale processing reduces the complexity of calculations in subsequent steps and better extracts and analyzes features in the image.

[0020] In one embodiment, when a diamond grinding wheel is used to grind silicon carbide, the working layer grinds the silicon carbide. Figure 1When defects occur in the surface image of the diamond wheel (shaded area in the middle), it will directly affect the grinding effect of silicon carbide. Therefore, in this embodiment, the working layer needs to be segmented to detect the defects of the working layer. Specifically, the surface image of the diamond wheel is input into a trained image segmentation model, and the working layer image and the non-working layer image are output. At this point, the working layer image of the diamond wheel can be obtained. Among them, the process of training the image segmentation model is as follows: the pixels belonging to the working layer in the surface image of the diamond wheel are marked as 1, and the pixels belonging to the non-working layer are marked as 0, thereby obtaining a 0-1 marked image; then the 0-1 marked image is input into the semantic segmentation network model for training; in response to the loss function being less than a preset value or reaching a preset number of training times, the training is completed to obtain a trained image segmentation model.

[0021] In one embodiment, a polar coordinate system is established with the midpoint of the working layer image as the origin. In the polar coordinate system, a ray is constructed in the working layer image at every preset angle from 0 to 360 degrees, wherein the preset angle is 1 degree. In other embodiments, the preset angle may be 2 degrees or 5 degrees, etc., which may be manually adjusted according to specific implementation conditions. A pixel point sequence is constructed based on the pixel points of the working surface passed by the ray at each angle.

[0022] S102: Calculate the importance of each pixel.

[0023] In one embodiment, during the use of a diamond grinding wheel, different areas on the working layer of the diamond grinding wheel play different roles in actual use, so their importance is different. For example, the central area of ​​the grinding wheel usually bears the greatest cutting pressure, so it is more important; while the edge area may bear less pressure, so it is less important. The reason is that the central area: usually the grinding effect is the most important. If there are defects in the central area, it will affect the stability and accuracy of the entire grinding process. Therefore, the importance of the pixels in this area is greater. Edge area: Although edge defects may affect the stability of the grinding wheel, they usually have little effect on the grinding effect, so the importance of the pixels in this area is less. It is further explained that the working layer of the diamond grinding wheel is composed of abrasives, binders and fillers. Abrasives (especially diamond abrasives) are the most important components in diamond grinding wheels. Its hardness and cutting ability are usually very strong. Therefore, in the working layer image, abrasives usually correspond to higher grayscale value areas. Therefore, the higher the grayscale value of a pixel, the more likely it is that it belongs to abrasives, and the greater the importance of the pixel. The lower the grayscale value of a pixel, the less likely it is that it belongs to abrasives, and the less important the pixel. Therefore, the importance of a pixel can be calculated based on the distance between the pixel and the central area and the grayscale value of each pixel.

[0024] In one embodiment, to determine whether each pixel in the working layer image is close to an edge, it is first necessary to perform edge detection on the working layer image to obtain two edges. In order to facilitate subsequent calculations, the two edges need to be distinguished by size, and divided into a first circular edge and a second circular edge, and the first circular edge is the largest edge. Among them, the canny edge detection algorithm can be used to perform edge detection on the working layer image. It should be noted that the geometric center point of the first circular edge can be used as the midpoint of the working layer image.

[0025] In one embodiment, a calculation method is provided to calculate the importance of each pixel, and the calculation formula is: .

[0026] In the formula, Indicates the first The importance of each pixel, Indicates the first pixels, Indicates the distance from the first circular edge to the first circular edge in the working layer image. The nearest pixel , Indicates the distance from the second circular edge to the first The nearest pixel , Indicates Pixels and The distance between pixels, Indicates Pixels and The distance between pixels. Indicates Pixels and The distance between pixels, Indicates the first The gray value of a pixel, Represents the average gray value of all pixels in the working layer image. The natural constant The exponential function with base . The distance between any two pixels can be calculated using the Euclidean distance algorithm.

[0027] in, The value of The difference between the shortest distances of the pixels to the first circular edge and the second circular edge. The larger the difference, the greater the difference. The closer the pixel is to the two edges, the The smaller the difference, the smaller the pixel importance; The more the pixel point is biased towards the middle area between the two edges, the The greater the importance of a pixel, the When the value of is equal to 0, The value of is 1, indicating that The pixel point is located in the central area of ​​the first circular edge and the second circular edge, and the pixel point is most important at this time. The smaller the value, the The smaller the gray value of a pixel, the smaller the possibility that the pixel belongs to abrasive. The smaller the importance of each pixel, the smaller the importance of each pixel. At this point, the importance of each pixel in the pixel sequence can be obtained, and then the importance sequence can be obtained.

[0028] Through the above steps, the distance difference and grayscale difference of pixels are combined to evaluate the importance of pixels, which can comprehensively consider two important visual features: the area where the pixels are located and the grayscale value difference. The distance difference of the edge reflects the change in physical form, while the grayscale difference reflects the change in the brightness of the image or the surface material. Combining the two can more accurately evaluate the importance of each area in the image, especially for those defective areas that are both located in the central area and have significant grayscale changes, which can gain higher attention and thus improve the accuracy of defect detection.

[0029] S103: Calculate the defect probability of each pixel.

[0030] In one embodiment, since the working surface of the diamond grinding wheel is composed of a mixture of abrasives, binders and fillers, when the working surface of the diamond grinding wheel is damaged, the texture of the damaged area is similar to that of the normal area, so it is difficult to identify the defect area on the working surface of the diamond grinding wheel. However, the depth of the defect area is different from that of the normal area, so it is necessary to calculate the possibility of each pixel input defect based on the depth information of each pixel. The depth information can be obtained from the depth image of the surface of the diamond grinding wheel.

[0031] In one embodiment, obtaining the defect possibility of each pixel includes: establishing a sliding window in the working layer image, calculating the average value of the difference between the pixel and the depth values ​​of all the pixels in the sliding window, and taking the product of the average value and the variance of the depth values ​​of all the pixels in the sliding window as the defect possibility of the pixel. The sliding window method can focus on each local area in the image, thereby capturing local details and small surface changes.

[0032] Specifically, a sliding window is set with any pixel point as the center in the working layer image, and the size of the sliding window is In other embodiments, the size of the sliding window can also be The defect probability of a pixel is positively correlated with the difference between the depth value of any pixel in the sliding window and the mean depth value of all pixels in the sliding window, and is also positively correlated with the variance of the depth values ​​of all pixels in the sliding window.

[0033] In one embodiment, the calculation formula for the defect probability of each pixel is: In the formula, For the working layer image The probability of pixel defects is For the working layer image The depth value of each pixel, is the mean depth value of all pixels in the window, Indicates the number of all pixels in the window. is the standard normalization function.

[0034] in, Indicates the first The difference between the pixel point and the mean depth value of all pixels in the sliding window, the greater the difference, the The more pixels there are, the greater the possibility of defects. Represents the variance of the depth values ​​of all pixels in the window. The larger the value, the greater the fluctuation of the depth value in the window. The greater the defect probability of each pixel, the greater the defect probability of each pixel in the pixel sequence.

[0035] For example, when the angle is 1 degree, the sequence of pixels formed by the ray passing through the working surface is , is the number of pixels that the ray passes through the working surface. According to the above method, the importance and defect possibility of each pixel are calculated, then the importance sequence of each pixel at an angle of 1 degree can be obtained as follows: , the defect possibility sequence composed of each pixel point at an angle of 1 degree is obtained as follows: .

[0036] S104: Calculate the defect degree of the working layer image.

[0037] It should be noted that, when there is no defect on the working surface of the diamond grinding wheel, the difference between each defect possibility sequence is basically 0, so the difference between the two defect possibility sequences can be used to characterize the defect degree of the working layer image. If the difference between the two defect possibility sequences is smaller, the possibility of defects on the working surface of the diamond grinding wheel is smaller, and the defect degree of the working layer image is smaller. And considering the importance of the location of the pixel point, the importance sequence of each pixel point is used to weight it, thereby improving the accuracy of the final calculation of the defect degree.

[0038] In one embodiment, the difference between any two defect possibility sequences is calculated as follows: ; In the formula, For the The defect possibility sequence under the angle and the The difference between the defect probability sequences is For the The first in the importance sequence under each angle The importance of each pixel, Indicates The first in the importance sequence under each angle The importance of each pixel; is the number of pixels that the ray passes through at any angle, For the The first in the defect possibility sequence under the angle The probability of pixel defects is For the The first in the defect possibility sequence under the angle The probability of pixel defects is represents the normalization function. Indicates The comprehensive weight of the pixels, Indicates The first in the defect possibility sequence under the angle The probability of a pixel defect is The first in the defect possibility sequence under the angle The greater the difference, the greater the difference between the two defect possibility sequences, and the greater the possibility of defects.

[0039] It should be noted that The importance sequence is: A sequence of the importance of all pixels on the working surface that the ray passes through at an angle. The defect possibility sequence is: At each angle, the defect probability of all pixels on the working surface that the ray passes through is composed of a sequence. So far, the difference between any two defect probability sequences can be calculated, and the difference between any two defect probability sequences can be used as the defect degree of the working layer image to determine whether the diamond grinding wheel has defects.

[0040] According to the above steps, when calculating the defect degree of the working layer image, not only the difference between the defect possibility sequences is considered, but also the importance sequence of the pixel points is introduced, which can make the detection results more refined and comprehensive. For example, pixels with higher importance (pixels close to the center area and with higher grayscale values) have a greater weight in the difference calculation, ensuring that these pixels that are more critical to the performance of the mold receive more attention, thereby improving the accuracy of defect assessment.

[0041] S105: Detect defects based on the degree of defects in the working layer image.

[0042] Specifically, if the degree of defect of the working layer image is greater than or equal to the preset error, it is judged that the diamond grinding wheel has defects, and an alarm is issued at this time to remind the staff to replace or repair the diamond grinding wheel. If the degree of defect of the working layer image is less than the preset error, it is judged that the diamond grinding wheel has no defects and the diamond grinding wheel can continue to be used. The empirical value of the preset error is 2. In other embodiments, the empirical value of the preset error can be 3 or 5, etc., which can be adjusted according to the real-time situation. It should be noted that the difference between the corresponding defect possibility sequences at any two angles is used as the defect degree of the working layer image. When a defect is determined, the defect area can also be located to facilitate inspection by the staff.

[0043] In one embodiment, locating the defect area includes constructing a matrix according to the difference between all defect possibility sequences, and the matrix is ​​shown in the following table.

[0044]

[0045] For example, in the matrix, the difference between the defect possibility sequence corresponding to angle 1 and the defect possibility sequence corresponding to angle 359 is 2, indicating that the defect degree of the working layer image is also equal to 2. At this time, the defect degree of the working layer image is equal to the preset error. At this time, it is judged that the diamond grinding wheel has a defect, and an alarm is issued to remind the staff to replace or repair the diamond grinding wheel. At this point, the defect detection of the diamond grinding wheel is completed.

[0046] The technical features of the above-mentioned embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. It should be pointed out that for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. An intelligent detection method for surface defects of fine grinding tools, characterized in that: include: The surface image of the diamond grinding wheel is collected, and the surface image is segmented using an image segmentation model to obtain a working layer image, and a polar coordinate system is established with the center point of the working layer image as the origin, and a pixel point sequence composed of pixel points on the working layer image passed by the ray at every preset angle in the polar coordinates is obtained; Calculate the importance and defect probability of the pixel points in the pixel point sequence at each angle in the working layer image, and obtain the corresponding importance degree sequence and defect probability sequence at each angle, wherein the importance of the pixel point reflects the distance between the position of the pixel point and the central area and the gray value of the pixel point, and the defect probability of the pixel point reflects the degree of change of the depth value of the pixel point and the depth values ​​of the surrounding pixels; The difference between the defect possibility sequences at any two angles is calculated as the defect degree of the working layer image, and defect detection is performed on the mold surface based on the magnitude of the defect degree, wherein the difference is also related to the importance of the pixel point.

2. The intelligent detection method for surface defects of fine grinding tools according to claim 1 is characterized in that: Defect detection is performed on the surface of the mold based on the degree of the defect, including: If the defect degree of the working layer image is greater than or equal to the preset error, it is determined that a defect occurs on the surface of the diamond grinding wheel and an alarm is issued.

3. The intelligent detection method for surface defects of fine grinding tools according to claim 1 is characterized in that: The calculation formula of the difference is: ； In the formula, For the The defect possibility sequence under the angle and the The difference between the defect probability sequences is For the The first in the importance sequence under each angle The importance of each pixel, Indicates The first in the importance sequence under each angle The importance of each pixel; is the number of pixels that the ray passes through at any angle, For the The first in the defect possibility sequence under the angle The probability of pixel defects is For the The first in the defect possibility sequence under the angle The probability of pixel defects is Represents the normalization function.

4. The intelligent detection method for surface defects of fine grinding tools according to claim 1 is characterized in that: Acquire surface images of diamond grinding wheels, including: The surface of the diamond grinding wheel is captured by a depth camera to obtain a surface image of the diamond grinding wheel.

5. The intelligent detection method for surface defects of fine grinding tools according to claim 1 is characterized in that: Surface image segmentation using image segmentation models, including: The surface image is input into an image segmentation model, and a working layer image is output.

6. The intelligent detection method for surface defects of fine grinding tools according to claim 5 is characterized in that: The training process of the image segmentation model includes: The pixel points belonging to the working layer in the surface image of the diamond grinding wheel are marked as 1, and the pixel points belonging to the non-working layer are marked as 0, so as to obtain a 0-1 marked image; Inputting the 0-1 labeled image into a semantic segmentation network model for training; In response to the loss function being less than a preset value or reaching a preset number of training times, the training is completed to obtain an image segmentation model.

7. The intelligent detection method for surface defects of fine grinding tools according to claim 1 is characterized in that: Get pixel sequence, including: A ray is established in polar coordinates at every preset angle, and all the pixels of the working layer image through which the ray passes are combined into a pixel point sequence, thereby obtaining a pixel point sequence corresponding to each angle.

8. The intelligent detection method for surface defects of fine grinding tools according to claim 1 is characterized in that: Calculate the importance of each pixel in the working layer image, including: ； In the formula, Indicates the first The importance of each pixel, Indicates the first pixels, Indicates the distance from the first circular edge to the first circular edge in the working layer image. The nearest pixel , Indicates the distance from the second circular edge to the first The nearest pixel , Indicates Pixels and The distance between pixels, Indicates Pixels and The distance between pixels, Indicates Pixels and The distance between pixels, Indicates the first The gray value of a pixel, Represents the average gray value of all pixels in the working layer image. The natural constant An exponential function with base .

9. The intelligent detection method for surface defects of fine grinding tools according to claim 1, characterized in that: Get the defect possibility of each pixel, including: A sliding window is established in the working layer image, the average value of the difference between the pixel point and the depth values ​​of all the pixels in the sliding window is calculated, and the product of the average value and the variance of the depth values ​​of all the pixels in the sliding window is used as the defect possibility of the pixel point.

10. The intelligent detection method for surface defects of fine grinding tools according to claim 1, characterized in that: The preset angle is 1 degree.

Images