Diamond wire coil quality evaluation method and device, electronic equipment and storage medium

Abstract

The invention provides a diamond wire coil quality evaluation method and device, electronic equipment and a storage medium, and the method comprises the steps: for each diamond wire image in a diamond wire image set, carrying out the preprocessing of the diamond wire image, and obtaining a preprocessed image; determining an agglomeration parameter and a sand crushing parameter of the preprocessed image according to diamond pixel points in the preprocessed image; determining an agglomeration judgment result of whether the diamond wire image is an agglomeration image or not according to the agglomeration parameters; according to the sand crushing parameters, determining whether the diamond wire image is a sand crushing judgment result of a sand crushing image; according to the agglomeration judgment result and the broken sand judgment result of each diamond wire image, the proportion of the number of images determined as agglomeration images and broken sand images in the diamond wire image set is obtained; judging whether the proportion of the number of the images exceeds a set qualified proportion or not, and obtaining a proportion judgment result; and determining the quality state of the diamond wire coil to be detected according to the proportion judgment result.

Description

Technical Field

[0001] The present application relates to the field of device detection, and in particular, to a quality evaluation method, device, electronic device and storage medium for a diamond wire roll. Background Art

[0002] Diamond wire has the advantages of high processing efficiency and small cutting gap. It is widely used in the slicing processing of hard and brittle materials such as single crystal silicon, sapphire, single crystal silicon carbide, and ceramics.

[0003] When the diamond wire is sand-plated, the agglomeration and crushing of the sand successfully attached to the diamond wire will directly affect the quality of the diamond wire. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a quality evaluation method, device, electronic device and storage medium for a diamond wire reel, so as to achieve quality evaluation of the diamond wire reel.

[0005] An embodiment of the present application provides a quality evaluation method for a diamond wire roll, comprising: obtaining a diamond wire image set corresponding to a diamond wire roll to be tested; wherein the diamond wire image set includes a plurality of diamond wire images; for each diamond wire image in the diamond wire image set: preprocessing the diamond wire image to obtain a preprocessed image; wherein the preprocessed image includes a plurality of background pixels and a plurality of diamond pixels; determining agglomeration parameters and crushed sand parameters of the preprocessed image according to the diamond pixels in the preprocessed image; determining agglomeration judgment results of whether the diamond wire image is agglomeration image according to the agglomeration parameters; and determining a crushed sand judgment results of whether the diamond wire image is a crushed sand image according to the crushed sand parameters; obtaining the proportion of images determined as agglomeration images and crushed sand images in the diamond wire image set according to the agglomeration judgment results and crushed sand judgment results of each diamond wire image; judging whether the proportion of the number of images exceeds a set qualified proportion, and obtaining a proportion judgment result; and determining the quality status of the diamond wire roll to be tested according to the proportion judgment result.

[0006] In the above implementation process, by obtaining each diamond wire image corresponding to the diamond wire roll to be tested, and determining the agglomeration parameters and crushed sand parameters of the diamond wire image, it is determined whether the diamond wire image is agglomerated image or a crushed sand image based on the agglomeration parameters and crushed sand parameters, respectively, and then the quality status of the diamond wire roll to be tested is determined based on whether the proportion of the number of images determined as agglomerated images and crushed sand images in the diamond wire image set exceeds the set qualified proportion. In this way, the quality detection and evaluation of the diamond wire roll is realized, and automatic detection can be realized on the machine based on the image, with high detection efficiency and low cost.

[0007] Further, according to the diamond pixel points in the preprocessed image, the agglomeration parameters of the preprocessed image are determined, including: obtaining the target diamond pixel density of the preprocessed image according to the sum of the diamond pixel density of each column of pixels in the preprocessed image under M vertical sliding windows; wherein M is an integer greater than 1; obtaining the target diamond sand density of the preprocessed image according to the sum of the sand density of each row of pixels in the preprocessed image under N horizontal windows, wherein N is an integer greater than 1, and the agglomeration parameters include the target diamond pixel density and the target diamond sand density.

[0008] The diamond pixel density can intuitively reflect the number of diamond pixels per unit area on the diamond wire, and the diamond grit density can intuitively reflect the number of diamonds per unit area on the diamond wire, so the diamond pixel density and the diamond grit density can reflect the agglomeration of the grit on the diamond wire. Therefore, in the above implementation, by obtaining the target diamond pixel density and the target diamond grit density, the detection of whether the diamond wire image is an agglomerated image can be made more accurate. The target diamond pixel density is determined by the sum of the diamond pixel densities under multiple windows, and the target diamond grit density is determined by the sum of the grit densities under multiple windows, so that the target diamond pixel density and the target diamond grit density can be obtained by combining the situations of multiple windows, so as to be more accurate, thereby making the detection of whether the diamond wire image is an agglomerated image more accurate.

[0009] Further, according to the sum of diamond pixel densities of each column of pixels in the preprocessed image under M vertical sliding windows, a target diamond pixel density of the preprocessed image is obtained, including: for the i-th column of pixels in the preprocessed image, M diamond pixel densities of the i-th column of pixels under the M vertical sliding windows are obtained, and the sum of the M diamond pixel densities corresponding to the i-th column of pixels is used as the sum of diamond pixel densities of the i-th column of pixels, where i is sequentially taken from 1 to M; according to the sum of diamond pixel densities corresponding to all columns of pixels in the preprocessed image, the target diamond pixel density is obtained; wherein the target diamond pixel density is not less than the average value of the sum of diamond pixel densities corresponding to all columns of pixels in the preprocessed image.

[0010] In the above implementation, the target diamond pixel density is determined based on the sum of the M diamond pixel densities corresponding to each column of pixels by obtaining the sum of the M diamond pixel densities corresponding to each column of pixels, and the target diamond pixel density is not less than the average value of the sum of the diamond pixel densities corresponding to all columns of pixels in the preprocessed image. The target diamond pixel density obtained in this way is more representative, thereby making the detection of whether the diamond wire image is an agglomerated image more accurate.

[0011] Furthermore, for the i-th column of pixels in the preprocessed image, M diamond pixel densities of the i-th column of pixels under the M vertical sliding windows are obtained, including: for each vertical sliding window in the i-th column of pixels, if i is less than half the window width of the vertical sliding window, the number of image columns in the vertical sliding window is expanded by mirror flipping; based on the image of the expanded vertical sliding window, the diamond pixel density of the i-th column of pixels under the vertical sliding window is obtained; if i is not less than half the window width of the vertical sliding window, the diamond pixel density of the i-th column of pixels under the vertical sliding window is obtained according to the image under the vertical sliding window.

[0012] In the above implementation, the number of image columns in the vertical sliding window is expanded by mirror flipping, so that each vertical sliding window has complete pixels, so that the diamond pixel density of each vertical sliding window is more reliable, and the obtained target diamond pixel density is more representative.

[0013] Further, according to the sum of the sand density of each row of pixels in the preprocessed image under N horizontal windows, the target diamond sand density of the preprocessed image is obtained, including: for the k-th row of pixels in the preprocessed image, the N sand densities of the k-th row of pixels in the N horizontal sliding windows are obtained, and the sum of the N sand densities corresponding to the k-th row of pixels is taken as the sum of the sand density of the k-th row of pixels, and k is taken from 1 to N in sequence; according to the sum of the sand density corresponding to all rows of pixels in the preprocessed image, the target diamond sand density is obtained; wherein, the target diamond sand density is not less than the average value of the sum of the sand density corresponding to all rows of pixels in the preprocessed image.

[0014] By obtaining the sum of N sand densities corresponding to each column of pixels, the target diamond sand density is determined based on the sum of N sand densities corresponding to each column of pixels, and the target diamond sand density is not less than the average value of the sum of sand densities corresponding to all columns of pixels in the preprocessed image. The target diamond sand density obtained in this way is more representative, thereby making the detection of whether the diamond wire image is a clustered image more accurate.

[0015] Furthermore, for the k-th row of pixels in the preprocessed image, N sand densities of the k-th row of pixels in the N horizontal sliding windows are obtained, including: for each horizontal sliding window in the k-th row of pixels in the preprocessed image, if k is less than half the window width of the horizontal sliding window, the number of image rows in the horizontal sliding window is expanded by mirror flipping; based on the image of the expanded horizontal sliding window, the sand density of the k-th row of pixels under the horizontal sliding window is obtained; if k is not less than half the window width of the horizontal sliding window, the sand density of the k-th row of pixels under the horizontal sliding window is obtained according to the image under the horizontal sliding window.

[0016] In the above implementation, the number of image columns in the horizontal sliding window is expanded by mirror flipping, so that each horizontal sliding window has complete pixels, so that the diamond pixel density of each horizontal sliding window is more reliable, and the obtained target diamond pixel density is more representative.

[0017] Furthermore, according to the diamond pixel points in the preprocessed image, the crushed sand parameters of the preprocessed image are determined, including: performing a connected domain analysis on the preprocessed image to obtain a diamond connected domain set corresponding to the diamond pixel points; according to the area of ​​a single diamond and the area of ​​each connected domain in the diamond connected domain set, obtaining the crushed sand ratio of the preprocessed image, and using the crushed sand ratio as the crushed sand parameter of the preprocessed image.

[0018] In the above implementation, the broken sand ratio reflects the ratio of broken sand to total diamond sand. Therefore, based on the broken sand ratio, it is possible to more accurately determine whether the diamond wire image is a broken sand image, thereby improving the detection accuracy of the broken sand image.

[0019] Furthermore, the broken sand ratio of the preprocessed image is obtained according to the area of ​​a single diamond and the area of ​​each connected domain in the diamond connected domain set, including: comparing the area of ​​a single diamond with the area of ​​each connected domain in the diamond connected domain set to obtain the number of broken sand connected domains whose areas are smaller than the area of ​​a single diamond; and determining the broken sand ratio of the preprocessed image according to the number of broken sand connected domains and the total amount of diamonds in the preprocessed image.

[0020] In the above implementation, through the area of ​​a single diamond and the area of ​​each connected domain, it is easy to determine which connected domains are broken sand connected domains, and then determine the broken sand ratio based on the number of broken sand connected domains, thereby realizing the rapid determination of broken sand parameters.

[0021] Further, according to the number of connected domains of broken sand and the total amount of diamonds in the preprocessed image, the proportion of broken sand in the preprocessed image is determined, including: according to the set pixel amount, selecting a large connected domain subset from the diamond connected domain set, wherein the number of pixels contained in each connected domain in the large connected domain subset is greater than the set pixel amount; using the number of single pixels corresponding to a single diamond, splitting each connected domain in the large connected domain subset to obtain the number of split connected domains corresponding to the large connected subset; determining the total amount of diamonds in the preprocessed image according to the number of remaining connected domains and the number of split connected domains; wherein the remaining connected domain data is the total amount of all connected domains in the diamond connected domain set except the large connected domain subset; determining the proportion of broken sand in the preprocessed image according to the number of connected domains of broken sand and the total amount of diamonds.

[0022] It can be understood that the number of pixels in a connected domain can reflect the area of ​​each connected domain. Considering that there may be bonding between diamonds in the image, for the case where multiple diamonds are bonded together, these multiple diamonds correspond to one connected domain. In addition, the number of single pixels corresponding to a single diamond is smaller than that of multiple diamonds bonded together. Therefore, according to the set pixel number, it can be judged whether each connected domain in the upper connected domain set is a connected domain corresponding to multiple diamonds bonded together (i.e., a connected domain in a large connected domain subset), and the connected domain in the large connected domain subset is split using the number of single pixels corresponding to a single diamond. The split connected domain after splitting can be considered as a connected domain with only a single diamond, so that the total amount of diamonds determined according to the number of remaining connected domains and the number of split connected domains is more accurate, making the determination result of the proportion of broken sand more reliable.

[0023] Furthermore, based on the agglomeration judgment result and the broken sand judgment result of each diamond wire image, the proportion of the number of images determined as agglomeration images and broken sand images in the diamond wire image set is obtained, including: based on the agglomeration judgment result of each diamond wire image and the broken sand judgment result of each diamond wire image, the number of double-defect images in the diamond wire image set that are both agglomeration images and broken sand images is obtained; based on the number of double-defect images and the total amount of the diamond wire image set, the proportion of the number of images is determined.

[0024] In the above implementation method, the image quantity ratio determined based on the number of dual-defect images that are both agglomerated images and broken sand images and the total number of diamond wire image sets can more objectively reflect the overall quality of the diamond wire roll. Therefore, based on the image quantity ratio, the quality of the diamond wire roll can be evaluated more objectively, thereby improving the reliability of the final evaluation conclusion.

[0025] Further, the image quantity ratio is determined based on the number of dual-defect images and the total number of the diamond wire image set, including: calculating the ratio of the number of dual-defect images to the total number of the diamond wire image set; and using the ratio as the image quantity ratio.

[0026] In the above implementation method, the number of double-defect images represents the number of diamond wire images with both substandard agglomeration and substandard sand crushing. Then the ratio of the number of double-defect images to the total number of diamond wire image sets represents the ratio of diamond wire images with both substandard agglomeration and substandard sand crushing to the total number of images. Based on this ratio, it is possible to more objectively evaluate whether the quality of the diamond wire is qualified.

[0027] Furthermore, the method also includes: obtaining the number of clustered images in the diamond wire image set that are clustered images according to the clustering judgment result of each diamond wire image; obtaining the number of broken sand images in the diamond wire image set that are broken sand images according to the broken sand judgment result of each diamond wire image; correspondingly, determining the image quantity ratio according to the number of double defect images and the total number of the diamond wire image set, including: determining the number of defective images in the diamond wire image set according to the number of clustered images, the number of broken sand images and the number of double defect images; calculating the ratio of the number of defective images to the total number of the diamond wire image set, and using the ratio as the image quantity ratio.

[0028] In the above implementation method, the number of defective images in the diamond wire image set determined based on the number of agglomerated images, the number of broken sand images and the number of double-defective images can represent the number of diamond wire images with at least substandard agglomeration or substandard broken sand conditions, and the ratio of the number of defective images to the total number of diamond wire images represents the ratio of substandard images to the total number of images. Based on this ratio, it is possible to more objectively evaluate whether the quality of the diamond wire is qualified.

[0029] The embodiment of the present application also provides a quality evaluation device for a diamond wire roll, comprising: an acquisition module, used to acquire a diamond wire image set corresponding to a diamond wire roll to be tested; wherein the diamond wire image set includes a plurality of diamond wire images; an image processing module, used to pre-process the diamond wire image for each diamond wire image in the diamond wire image set to obtain a pre-processed image; wherein the pre-processed image includes a plurality of background pixels and a plurality of diamond pixels; according to the diamond pixels in the pre-processed image, determine a clustering parameter and a crushed sand parameter of the pre-processed image; according to the clustering parameter, determine a clustering judgment result of whether the diamond wire image is a clustered image; and according to the crushed sand parameter, determine a crushed sand judgment result of whether the diamond wire image is a crushed sand image; a proportion determination module, used to acquire the proportion of images determined as clustered images and crushed sand images in the diamond wire image set according to the clustering judgment result and the crushed sand judgment result of each diamond wire image; a judgment module, used to judge whether the proportion of the number of images exceeds a set qualified proportion and acquire a proportion judgment result; a quality status determination module, used to determine the quality status of the diamond wire roll to be tested according to the proportion judgment result.

[0030] An embodiment of the present application also provides an electronic device, comprising: a processor and a memory; the processor is used to execute one or more programs stored in the memory to implement any of the above-mentioned quality evaluation methods for diamond wire rolls.

[0031] A computer-readable storage medium is also provided in an embodiment of the present application. The computer-readable storage medium stores one or more programs. The one or more programs can be executed by one or more processors to implement any of the above-mentioned quality evaluation methods for diamond wire rolls. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 A schematic flow chart of a quality evaluation method for a diamond wire coil provided in an embodiment of the present application;

[0034] Figure 2 A schematic diagram of an agglomeration judgment structure and a flow chart of obtaining a crushed sand judgment result provided in an embodiment of the present application;

[0035] Figure 3 A schematic diagram of an image preprocessing process provided in an embodiment of the present application;

[0036] Figure 4 A schematic diagram of a vertical sliding window extending beyond the edge of an image provided in an embodiment of the present application;

[0037] Figure 5 A method for Figure 4 Schematic diagram after mirror expansion of the situation;

[0038] Figure 6 A schematic diagram of a horizontal sliding window extending beyond the edge of an image provided in an embodiment of the present application;

[0039] Figure 7 A method for Figure 6 Schematic diagram after mirror expansion of the situation;

[0040] Figure 8 A binarized diamond wire image provided in an embodiment of the present application;

[0041] Fig. 9 A schematic structural diagram of a quality evaluation device for a diamond wire roll provided in an embodiment of the present application;

[0042] Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0044] In order to achieve quality evaluation of diamond wire coils, a quality evaluation method for diamond wire coils is provided in an embodiment of the present application. Figure 1 As shown, Figure 1 A schematic flow chart of a quality evaluation method for a diamond wire coil provided in an embodiment of the present application includes:

[0045] S101: Obtain a diamond wire image set corresponding to the diamond wire reel to be tested.

[0046] The diamond wire image set includes a plurality of diamond wire images.

[0047] In an embodiment of the present application, during the production process of the diamond wire roll to be tested or after the production of the diamond wire roll to be tested is completed, an image of at least a portion of the diamond wire of the diamond wire roll to be tested can be captured by a camera or other photographic equipment to obtain a diamond wire image set corresponding to the diamond wire roll to be tested, but this is not a limitation.

[0048] S102: for each diamond wire image in the diamond wire image set, determining a clustering judgment result of whether each diamond wire image is a clustering image, and a crushing sand judgment result of whether each diamond wire image is a crushing sand image.

[0049] In step S102, for each diamond wire image in the diamond wire image set, Figure 2 As shown in the figure, the sand shortage judgment result can be determined according to the following process:

[0050] S1021: pre-processing the diamond wire image to obtain a pre-processed image corresponding to the diamond wire image.

[0051] The pre-processed image includes a plurality of background pixels and a plurality of diamond pixels. Of course, the pre-processed image may also include a plurality of busbar pixels, which is not specifically limited in the present application.

[0052] It can be understood that the diamond wire image mainly includes three parts: diamond, mother wire (i.e., wire for attaching diamond), and background. In most cases, the three parts of diamond, mother wire, and background will have large differences in color. Therefore, in a feasible implementation of the embodiment of the present application, Figure 3 As shown, step S1021 may include:

[0053] S301: Convert the diamond wire image into a grayscale image.

[0054] Exemplarily, the values ​​of the three channels BGR (blue, green and red) of the diamond wire image can be read, and then the BGR values ​​are converted into a single value by, but not limited to, histogram equalization, averaging, etc., to obtain a grayscale image. For example, the three-channel RGB pixel value of each pixel in the colored diamond wire image can be converted into a single-channel grayscale value, for example, the formula Gray = R*0.299+G*0.587+B*0.114 can be used for conversion, where Gray is the grayscale value of the pixel after conversion, R is the color intensity value of the colored pixel in the red channel, G is the color intensity value of the colored pixel in the green channel, and B is the color intensity value of the colored pixel in the blue channel.

[0055] S302: clustering the pixels in the grayscale image using three preset cluster centers to obtain a ternary image.

[0056] The ternary image includes a plurality of background pixels, a plurality of diamond pixels and a plurality of busbar pixels. In one embodiment, the ternary image may be used as a pre-processed image.

[0057] In an embodiment of the present application, different clustering algorithms can be used to perform clustering operations on the grayscale values ​​of multiple pixels. For example, clustering algorithms such as kmeans clustering algorithm, Mean-Shift clustering algorithm, and composite clustering algorithm can be used to perform clustering operations on the grayscale values ​​of multiple pixels.

[0058] It can be understood that by clustering the pixels in the grayscale image through three cluster centers, multiple cluster regions can be obtained, each cluster region is composed of the same type of pixels, and there are three types of pixels: background pixels, diamond pixels and busbar pixels, corresponding to the three cluster centers respectively.

[0059] The three initial cluster centers include a first initial cluster center corresponding to the background, a second initial cluster center corresponding to the busbar, and a third initial cluster center corresponding to the diamond. Based on the three initial cluster centers, the pixel values ​​of multiple pixels in the grayscale image are clustered using the K-means clustering algorithm. In the clustering process, if the clustering algorithm has converged, the clustering result with the largest grayscale value at this time can be used as the diamond pixel, the clustering result with the smallest grayscale value can be used as the background pixel, and the clustering result with the grayscale value in the middle can be used as the busbar pixel; conversely, if the clustering algorithm has not converged, the cluster center can be recalculated according to the grayscale value of the pixel points belonging to the same cluster center until the clustering algorithm converges, and three clusters can be obtained, each of which includes multiple pixels.

[0060] In this application, three different gray levels correspond to multiple background pixels, multiple diamond pixels, and multiple busbar pixels. Separating the background pixels, busbar pixels, and diamond pixels of the diamond wire in the ternary image can make the accuracy of the determined diamond pixel points higher and improve the overall detection accuracy of the sand volume detection.

[0061] After clustering is completed, a corresponding ternary image is generated for the grayscale image based on the clustering result. The electronic device can unify the pixel values ​​of each cluster of pixels. In other words, the pixels in the same cluster have the same pixel value, and the pixels of the three clusters have three pixel values ​​in the image, thereby converting the grayscale image into a ternary image. Here, the pixel value set for the pixels of each cluster can be the actual cluster center of the cluster after clustering is completed.

[0062] In the embodiment of the present application, after obtaining the grayscale image and before clustering, a morphological processing method (such as an erosion and dilation algorithm) can be used to eliminate at least part of the noise points in the grayscale image to make the boundaries between diamonds clearer.

[0063] S303: Convert the ternary image into a binary image.

[0064] Among them, the binary image is used as the preprocessed image.

[0065] In some embodiments, after obtaining the ternary image, the pixels in the ternary image can be classified to obtain the category information of the pixels corresponding to the three pixel values. In the grayscale image, the diamond pixel has the highest brightness and the highest pixel value, so the category information of the pixel to which the maximum pixel value among the three pixel values ​​belongs can be determined as the diamond pixel. At this time, the pixels corresponding to the remaining two pixel values ​​can be determined to be background pixels and busbar pixels, and these two types of pixels are classified as non-diamond pixels.

[0066] In some embodiments, the pixel values ​​of non-diamond pixels can be set to zero, or the pixel values ​​of non-diamond pixels can be set to the pixel values ​​of background pixels or busbar pixels; similarly, the pixel values ​​of diamond pixels can be set to specified positive integers or remain unchanged, thereby converting the ternary image into a binary image, and the binary image is used as a preprocessed image.

[0067] Through the above measures, after the diamond wire image is converted into a grayscale image, the pixels of the grayscale image are clustered based on the initial cluster center, so that the grayscale image is converted into a ternary image according to the clustering result, and the ternary image distinguishes diamond pixels, background pixels and busbar pixels. Further, according to the brightness characteristics of the diamond pixels in the image, the diamond pixels and non-diamond pixels in the ternary image can be accurately distinguished, so that different pixel values ​​can be set for the diamond pixels and the non-diamond pixels, so that the ternary image is converted into a binary image, completing the preprocessing process of the diamond wire image.

[0068] S1022: Determine the aggregation parameters and crushing parameters of the pre-processed image according to the diamond pixel points in the pre-processed image.

[0069] In an embodiment of the present application, in order to obtain the agglomeration parameters, the target diamond pixel density of the preprocessed image can be obtained according to the sum of the diamond pixel density of each column of pixels in the preprocessed image under M vertical sliding windows. Where M is an integer greater than 1. And the target diamond sand density of the preprocessed image can be obtained according to the sum of the sand density of each row of pixels in the preprocessed image under N horizontal windows. Where N is an integer greater than 1, and the agglomeration parameters include the target diamond pixel density and the target diamond sand density.

[0070] The vertical sliding window refers to a window determined by taking each column of pixels in the preprocessed image as the center, and the horizontal window refers to a window determined by taking each row of pixels in the preprocessed image as the center.

[0071] In the embodiment of the present application, each column of pixels has M vertical sliding windows, and the window sizes of the M vertical sliding windows can be different from each other. For example, M can be 5, and the window sizes of the 5 vertical sliding windows can be 11×H, 21×H, 31×H, 41×H, and 51×H, respectively, where H is the number of pixels in each column of pixels in the preprocessed image.

[0072] In the embodiment of the present application, each row of pixels has N vertical sliding windows, and the window sizes of the N vertical sliding windows may be different from each other. For example, N may be 5, and the window sizes of the 5 vertical sliding windows may be W×11, W×21, W×31, W×41, and W×51, respectively, where W is the number of pixels in each row of pixels in the preprocessed image.

[0073] In this way, the target diamond pixel density is determined by the sum of the diamond pixel densities under multiple windows of different sizes, and the target diamond sand density is determined by the sum of the sand density under multiple windows of different sizes, so that the target diamond pixel density and the target diamond sand density can be obtained by integrating the conditions of multiple windows, which is more accurate, thereby making the detection of whether the diamond wire image is a clustered image more accurate.

[0074] In some optional implementations of the embodiments of the present application, the process of obtaining the target diamond pixel density of the preprocessed image according to the sum of the diamond pixel density of each column of pixels in the preprocessed image under M vertical sliding windows may include:

[0075] First, for the i-th column of pixels in the preprocessed image, obtain M diamond pixel densities of the i-th column of pixels under M vertical sliding windows, and take the sum of the M diamond pixel densities corresponding to the i-th column of pixels as the sum of the diamond pixel densities of the i-th column of pixels, where i is taken from 1 to M in sequence.

[0076] Then, the target diamond pixel density is obtained according to the sum of the diamond pixel densities corresponding to all columns of pixels in the preprocessed image, wherein the target diamond pixel density is not less than the average value of the sum of the diamond pixel densities corresponding to all columns of pixels in the preprocessed image.

[0077] In the embodiment of the present application, the diamond pixel density under each vertical sliding window can be obtained by dividing the total number of diamond pixel points in the vertical sliding window by the total number of pixel points in the vertical sliding window.

[0078] In a feasible manner of the embodiment of the present application, the maximum diamond pixel density sum may be taken from the diamond pixel density sums corresponding to all column pixels as the target diamond pixel density.

[0079] In another feasible manner of the embodiment of the present application, the average value of the sum of the diamond pixel densities corresponding to all column pixels may also be calculated, and the average value is taken as the target diamond pixel density.

[0080] In the above implementation, the target diamond pixel density is determined based on the sum of the M diamond pixel densities corresponding to each column of pixels by obtaining the sum of the M diamond pixel densities corresponding to each column of pixels, and the target diamond pixel density is not less than the average value of the sum of the diamond pixel densities corresponding to all columns of pixels in the preprocessed image. The target diamond pixel density obtained in this way is more representative, thereby making the detection of whether the diamond wire image is an agglomerated image more accurate.

[0081] It can be understood that when i is less than half the window width of a vertical sliding window, there may be a situation where there are no pixels in some areas of the vertical sliding window, for example Figure 4 It should be noted that in this application Figures 4 to 7 In the diagram, each small square corresponds to a pixel, and the first digit of the number in the square represents the row number, and the last digit of the number represents the column number, for example, 12 represents the pixel in the first row and second column.

[0082] In this regard, in the embodiment of the present application, for each vertical sliding window in the i-th column of pixels, if i is less than half the window width of the vertical sliding window, the number of image columns in the vertical sliding window can be expanded by mirror flipping, and then the diamond pixel density of the i-th column of pixels under the vertical sliding window is obtained according to the image of the expanded vertical sliding window. If i is not less than half the window width of the vertical sliding window, the diamond pixel density of the i-th column of pixels under the vertical sliding window is obtained according to the image under the vertical sliding window.

[0083] For example, see Figure 5 As shown, for Figure 5 The vertical sliding window shown in FIG. 1 is mirrored, wherein the pixel points with a negative sign are the mirror images of the corresponding pixel points with corresponding numbers without a negative sign, for example, -11 is the mirror image of the pixel point 11, and -12 is the mirror image of the pixel point 12.

[0084] Similarly, in some optional implementations of the embodiments of the present application, the process of obtaining the target diamond sand density of the preprocessed image according to the sand density of each row of pixels in the preprocessed image under N horizontal windows may include:

[0085] First, for the k-th row of pixels in the preprocessed image, obtain the N sand densities of the k-th row of pixels in N horizontal sliding windows, and take the sum of the N sand densities corresponding to the k-th row of pixels as the sum of the sand densities of the k-th row of pixels, where k is taken from 1 to N in sequence.

[0086] Then, the target diamond sand density is obtained according to the sand density sum corresponding to all rows of pixels in the preprocessed image, wherein the target diamond sand density is not less than the average value of the sand density sum corresponding to all rows of pixels in the preprocessed image.

[0087] In the embodiment of the present application, the sand density under each transverse sliding window can be obtained by dividing the total number of diamonds in the transverse sliding window by the total area of ​​the transverse sliding window (which can be expressed by the total number of pixels).

[0088] In a feasible manner of the embodiment of the present application, the maximum sand density sum can be taken from the sand density sums corresponding to all row pixels as the target diamond sand density.

[0089] In another feasible manner of the embodiment of the present application, the average value of the sand density corresponding to all column pixels may also be calculated, and the average value is taken as the target diamond sand density.

[0090] The above implementation method obtains the sum of N sand densities corresponding to each column of pixels, and then determines the target diamond sand density based on the sum of N sand densities corresponding to each column of pixels, and the target diamond sand density is not less than the average value of the sum of sand densities corresponding to all columns of pixels in the preprocessed image. The target diamond sand density obtained in this way is more representative, thereby making the detection of whether the diamond wire image is a clustered image more accurate.

[0091] It can be understood that when k is less than half the window width of a certain horizontal sliding window, there may be a situation where there are no pixels in some areas of the horizontal sliding window, for example Figure 6 shown.

[0092] In this regard, in the embodiment of the present application, for each horizontal sliding window in the k-th row of pixels, if k is less than half the window width of the horizontal sliding window, the number of image rows in the horizontal sliding window can be expanded by mirror flipping, and then the diamond grit density of the k-th row of pixels under the horizontal sliding window is obtained according to the image of the expanded horizontal sliding window. If k is not less than half the window width of the horizontal sliding window, the diamond grit density of the k-th row of pixels under the horizontal sliding window is obtained according to the image under the horizontal sliding window.

[0093] For example, see Figure 7 As shown, for Figure 7 The horizontal sliding window shown in is mirrored, wherein the pixel points with a negative sign are the mirror images of the corresponding pixel points with the corresponding labels without a negative sign, for example, -11 is the mirror image of the pixel point 11, and -21 is the mirror image of the pixel point 21.

[0094] It can be understood that in the embodiment of the present application, the target diamond pixel density can be determined by a horizontal sliding window, and the target diamond sand density can be determined by a vertical sliding window; or the target diamond pixel density and the target diamond sand density can be determined simultaneously by a vertical sliding window; or the target diamond pixel density and the target diamond sand density can be determined simultaneously by a horizontal sliding window. Among them, the determination method of the target diamond pixel density and the target diamond sand density is consistent with the above embodiment, and only the relevant sliding windows are replaced.

[0095] In an embodiment of the present application, in order to obtain the crushed sand parameters, a connected domain analysis can be performed on the preprocessed image to obtain a diamond connected domain set corresponding to the diamond pixel points, and then the crushed sand ratio of the preprocessed image is obtained based on the area of ​​a single diamond and the area of ​​each connected domain in the diamond connected domain set, and the crushed sand ratio is used as the crushed sand parameter of the preprocessed image.

[0096] In the embodiment of the present application, the connected domain refers to the area formed by the aggregation of diamond pixels in the diamond wire image or the pre-processed image. Figure 8 The white areas in the binary image shown are connected domains.

[0097] In the embodiment of the present application, the area of ​​a single diamond (which can be represented by the number of single pixels corresponding to a single diamond) can be compared with the area of ​​each connected domain in the diamond connected domain set, so as to obtain the number of broken sand connected domains whose area is smaller than the area of ​​a single diamond, and then determine the broken sand ratio of the pre-processed image according to the number of broken sand connected domains and the total amount of diamonds in the pre-processed image. The broken sand connected domain refers to the connected domain determined to be broken sand.

[0098] It should be noted that in the diamond wire, there may be bonding between diamonds. For the case where multiple diamonds are bonded together, in the pre-processed image, these multiple diamonds correspond to a connected domain. Therefore, in order to more accurately obtain the total amount of diamonds in the pre-processed image, in the embodiment of the present application, the following method can be used:

[0099] First, according to the set pixel quantity, a large connected domain subset is selected from the diamond connected domain set, wherein the pixel quantity of each connected domain in the large connected domain subset is greater than the set pixel quantity.

[0100] Then, each connected domain in the large connected domain subset is split using the single pixel quantity corresponding to a single diamond, and the number of split connected domains corresponding to the large connected subset is obtained.

[0101] Finally, the total amount of diamonds in the preprocessed image is determined based on the number of remaining connected domains and the number of split connected domains, where the remaining connected domain data is the total amount of all connected domains in the diamond connected domain set except the large connected domain subset.

[0102] In this way, the proportion of broken sand in the preprocessed image can be determined based on the number of connected domains of broken sand and the total amount of diamonds.

[0103] It can be understood that since the area occupied by a single diamond (i.e., the number of pixels) will not differ too much, the aforementioned set pixel number can be set according to the number of pixels usually occupied by a single diamond. If the number of pixels of a connected domain is less than or equal to the set pixel number, it can be considered that the connected domain corresponds to a single diamond. If the number of pixels of a connected domain is greater than the set pixel number, it can be considered that the connected domain corresponds to multiple diamonds, and the connected domain needs to be split. Exemplarily, the set pixel number can be set to 40 pixels, but it is not limited to this.

[0104] In order to split the connected domains in the large connected domain subset, a feasible implementation method is: use the single pixel quantity corresponding to a single diamond to split each connected domain in the large connected domain subset to obtain the number of split connected domains corresponding to the large connected subset.

[0105] When each connected domain in the large connected domain subset is split using the single pixel quantity corresponding to a single diamond, the number of splits corresponding to the connected domain can be determined based on the number of pixels occupied by the connected domain and the single pixel quantity, and then the number of split connected domains can be determined based on the sum of the number of splits of all connected domains in the large connected domain subset.

[0106] For example, assuming that there are two connected domains in the large connected domain subset, the number of pixels is 200 and 400 respectively, and the number of single pixels is 20, then it can be determined that the number of splits of the two connected domains is 10 and 20 respectively, and then the number of split connected domains is determined to be 30.

[0107] In the above feasible implementation manner, the single pixel quantity corresponding to a single diamond can be obtained by:

[0108] Method 1: Obtain the average number of pixels corresponding to all connected domains in the diamond wire image as the first pixel quantity; determine the single pixel quantity according to the first pixel quantity.

[0109] Exemplarily, the first pixel quantity may be used as the single pixel quantity, or a preset constant may be added to or subtracted from the first pixel quantity to form the single pixel quantity.

[0110] Method 2: Obtain a specified connected domain subset whose pixel numbers are within a set pixel range in all connected domains of the diamond wire image; obtain the average pixel number corresponding to all connected domains in the specified connected domain subset as the second pixel quantity; and determine the single pixel quantity based on the second pixel quantity.

[0111] The pixel range can be set as needed. For example, the pixel range can be greater than or equal to 5 pixels; or, the pixel range can be less than or equal to 40 pixels; or, the pixel range can be greater than or equal to 5 pixels and less than or equal to 40 pixels, but this is not a limitation.

[0112] Exemplarily, the second pixel quantity may be used as a single pixel quantity.

[0113] In order to split the connected domains in the large connected domain subset, another feasible implementation method is: using the erosion and expansion algorithm to split the connected domains in the connected domain subset to obtain multiple disconnected connected domains.

[0114] In the above feasible implementation manner, different scales may be used to perform erosion operations and dilation operations on the connected domains in the connected domain subset, so as to obtain a plurality of disconnected connected domains.

[0115] After obtaining multiple disconnected connected domains, it is further possible to determine whether there are still connected domains in the multiple disconnected connected domains whose pixel quantity is greater than the set pixel quantity. If so, continue to split these connected domains.

[0116] In the embodiment of the present application, the sum of the number of remaining connected domains and the number of split connected domains can be determined as the total amount of diamonds in the preprocessed image.

[0117] In the embodiment of the present application, the ratio of the number of connected domains of broken sand to the total amount of diamonds can be determined as the broken sand ratio of the pre-processed image.

[0118] S1023: determining, according to the aggregation parameter, whether the diamond wire image is a clustering judgment result of the clustering image; and determining, according to the crushing sand parameter, whether the diamond wire image is a crushing sand judgment result of the crushing sand image.

[0119] In the embodiment of the present application, whether the diamond wire image is a clustered image can be determined based on whether the clustering parameters meet the preset clustering conditions.

[0120] Exemplarily, the agglomeration condition may include: the target diamond pixel density is greater than a preset diamond pixel density threshold, or the target diamond sand amount density is greater than a preset diamond sand amount density threshold. When the target diamond pixel density is greater than the preset diamond pixel density threshold, or the target diamond sand amount density is greater than the preset diamond sand amount density threshold, a agglomeration judgment result that the diamond wire image is agglomeration image can be generated; when the target diamond pixel density is less than or equal to the preset diamond pixel density threshold, and the target diamond sand amount density is less than or equal to the preset diamond sand amount density threshold, agglomeration judgment result that the diamond wire image is not agglomeration image can be generated. Among them, the preset diamond pixel density threshold and the preset diamond sand amount density threshold can be determined by collecting historical agglomeration images and counting the minimum diamond pixel density and the minimum diamond sand amount density in these historical agglomeration images, but it is not a limitation.

[0121] In the embodiment of the present application, whether the diamond wire image is a crushed sand image can be determined according to whether the crushed sand parameters meet the preset crushed sand conditions.

[0122] Exemplarily, the crushed sand condition may include: the crushed sand ratio is greater than a preset crushed sand ratio threshold. When the crushed sand ratio corresponding to the diamond wire image is greater than the preset crushed sand ratio threshold, a crushed sand judgment result that the diamond wire image is a crushed sand image can be generated; when the target diamond pixel density is less than or equal to the preset crushed sand ratio threshold, a clustering judgment result that the diamond wire image is not a crushed sand image can be generated. Among them, the preset crushed sand ratio threshold can be determined by collecting historical cluster images, counting the average value of the crushed sand ratio of these historical cluster images, or counting the maximum value of the crushed sand ratio of these historical cluster images, but it is not limited to it. For example, the preset crushed sand ratio threshold can be set to a value between 0.13 and 0.16.

[0123] S103: according to the agglomeration judgment result and the crushed sand judgment result of each diamond wire image, obtaining the ratio of the number of images determined as agglomeration images and crushed sand images in the diamond wire image set.

[0124] S104: Determine whether the image quantity ratio exceeds the set qualified ratio, and obtain the ratio determination result.

[0125] S105: Determine the quality status of the diamond wire coil to be tested according to the proportion judgment result.

[0126] In the embodiment of the present application, the number L3 of double defect images in the diamond wire image set that are both agglomeration images and crushed sand images can be obtained according to the agglomeration judgment result of each diamond wire image and the crushed sand judgment result of each diamond wire image;

[0127] Then, according to the number of double-defect images L3 and the total number of diamond wire image sets L, the image quantity proportion P is determined.

[0128] For example, the ratio of the number of double defect images to the total number of diamond wire image sets may be calculated, and then the ratio is used as the image quantity ratio, that is, P=L3 / L.

[0129] In this way, since the number of double-defect images represents the number of diamond wire images with both substandard agglomeration and substandard sand crushing, L3 / L represents the ratio of diamond wire images with both substandard agglomeration and substandard sand crushing to the total number of images. Based on this ratio, it is possible to more objectively evaluate whether the quality of the diamond wire is qualified.

[0130] As another example, in the embodiment of the present application, the number of clustered images L1 in the diamond wire image set that are clustered images can be obtained according to the clustering judgment result of each diamond wire image, and the number of broken sand images L2 in the diamond wire image set that are broken sand images can be obtained according to the broken sand judgment result of each diamond wire image. Thereafter, the number of defective images in the diamond wire image set can be determined according to the number of clustered images L1, the number of broken sand images L2, and the number of double defective images L3. Then, the ratio of the number of defective images to the total number of diamond wire image sets is calculated, and the ratio is used as the image number ratio P. That is, P = (L1 + L2 - L3) / L.

[0131] In this way, the number of defective images (L1+L2-L3) can represent the number of diamond wire images with at least substandard agglomeration or substandard sand crushing, and the ratio of the number of defective images to the total number of diamond wire image sets L represents the ratio of substandard images to the total number of images. Based on this ratio, the quality of the diamond wire can be evaluated more objectively.

[0132] In the embodiment of the present application, the qualified ratio can be set according to actual needs, for example, it can be set to 4%.

[0133] In an embodiment of the present application, when the ratio judgment result is that the image quantity ratio exceeds the set qualified ratio, the quality status of the diamond wire roll to be tested is determined to be unqualified; on the contrary, when the ratio judgment result is that the image quantity ratio does not exceed the set qualified ratio, the quality status of the diamond wire roll to be tested is determined to be qualified.

[0134] The quality evaluation method of the diamond wire roll provided in the embodiment of the present application obtains each diamond wire image corresponding to the diamond wire roll to be tested, and determines the agglomeration parameter and crushed sand parameter of the diamond wire image, thereby determining whether the diamond wire image is agglomerated image and whether it is a crushed sand image based on the agglomeration parameter and crushed sand parameter, respectively, and then determines the quality status of the diamond wire roll to be tested based on whether the proportion of the number of images determined as agglomerated images and crushed sand images in the diamond wire image set exceeds the set qualified proportion. In this way, the quality detection and evaluation of the diamond wire roll is realized, and automatic detection can be realized on the machine based on the image, with high detection efficiency and low cost.

[0135] Based on the same inventive concept, the present application also provides a quality evaluation device 900 for a diamond wire roll. Fig. 9 As shown, Fig. 9 Shows the use of Figure 1 The quality evaluation device of the diamond wire roll according to the method shown. It should be understood that the specific functions of the device 900 can be referred to the description above, and the detailed description is appropriately omitted here to avoid repetition. The device 900 includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the operating system of the device 900. Specifically:

[0136] See also Fig. 9 As shown, the device 900 includes: an acquisition module 901, an image processing module 902, a ratio determination module 903, a judgment module 904 and a quality status determination module 905. Among them:

[0137] An acquisition module 901 is used to acquire a diamond wire image set corresponding to the diamond wire coil to be tested; wherein the diamond wire image set includes a plurality of diamond wire images;

[0138] The image processing module 902 is used for: for each diamond wire image in the diamond wire image set: preprocessing the diamond wire image to obtain a preprocessed image; wherein the preprocessed image includes a plurality of background pixels and a plurality of diamond pixels; determining agglomeration parameters and crushed sand parameters of the preprocessed image according to the diamond pixels in the preprocessed image; determining agglomeration judgment result of whether the diamond wire image is agglomeration image according to the agglomeration parameters; and determining a crushed sand judgment result of whether the diamond wire image is a crushed sand image according to the crushed sand parameters;

[0139] A proportion determination module 903 is used to obtain the proportion of the number of images determined as agglomerated images and broken sand images in the diamond wire image set according to the agglomeration judgment result and the broken sand judgment result of each diamond wire image;

[0140] A judgment module 904 is used to judge whether the image quantity ratio exceeds a set qualified ratio, and obtain a ratio judgment result;

[0141] The quality status determination module 905 is used to determine the quality status of the diamond wire coil to be tested according to the proportion judgment result.

[0142] In a feasible implementation of the embodiment of the present application, the image processing module 902 is specifically used for:

[0143] Obtaining a target diamond pixel density of the preprocessed image according to the sum of diamond pixel densities of each column of pixels in the preprocessed image under M vertical sliding windows; wherein M is an integer greater than 1;

[0144] According to the sum of the sand density of each row of pixels in the preprocessed image under N horizontal windows, the target diamond sand density of the preprocessed image is obtained, wherein N is an integer greater than 1, and the agglomeration parameters include the target diamond pixel density and the target diamond sand density.

[0145] In the above feasible implementation manner, the image processing module 902 is specifically used for:

[0146] For the i-th column of pixels in the preprocessed image, obtain M diamond pixel densities of the i-th column of pixels under the M vertical sliding windows, and take the sum of the M diamond pixel densities corresponding to the i-th column of pixels as the sum of the diamond pixel densities of the i-th column of pixels, where i is taken from 1 to M in sequence;

[0147] The diamond pixel density is obtained according to the sum of the diamond pixel densities corresponding to all columns of pixels in the preprocessed image, wherein the diamond pixel density is not less than an average value of the sum of the diamond pixel densities corresponding to all columns of pixels in the preprocessed image.

[0148] In the above feasible implementation manner, the image processing module 902 is specifically used for:

[0149] For each vertical sliding window in the i-th column of pixels, if i is less than half the window width of the vertical sliding window, the number of image columns in the vertical sliding window is expanded by mirror flipping;

[0150] According to the image of the expanded vertical sliding window, the diamond pixel density of the i-th column of pixels under the vertical sliding window is obtained;

[0151] If i is not less than half the window width of the vertical sliding window, the diamond pixel density of the i-th column of pixels under the vertical sliding window is obtained according to the image under the vertical sliding window.

[0152] In a feasible implementation of the embodiment of the present application, the image processing module 902 is specifically used for:

[0153] For the k-th row of pixels in the preprocessed image, obtain N sand density of the k-th row of pixels in the N horizontal sliding windows, and take the sum of the N sand density corresponding to the k-th row of pixels as the sum of the sand density of the k-th row of pixels, where k is taken from 1 to N in sequence;

[0154] The target diamond sand density is obtained according to the sum of the sand density corresponding to all rows of pixels in the preprocessed image; wherein the target diamond sand density is not less than the average value of the sum of the sand density corresponding to all rows of pixels in the preprocessed image.

[0155] In the above feasible implementation manner, the image processing module 902 is specifically used for:

[0156] For each horizontal sliding window in the k-th row of pixels in the preprocessed image, if k is less than half the window width of the horizontal sliding window, the number of image rows in the horizontal sliding window is expanded by mirror flipping;

[0157] According to the image of the expanded horizontal sliding window, the sand density of the k-th row of pixels under the horizontal sliding window is obtained;

[0158] If k is not less than half the window width of the horizontal sliding window, the sand density of the k-th row of pixels under the horizontal sliding window is obtained according to the image under the horizontal sliding window.

[0159] In a feasible implementation of the embodiment of the present application, the image processing module 902 is specifically used for:

[0160] Performing connected domain analysis on the preprocessed image to obtain a diamond connected domain set corresponding to diamond pixel points;

[0161] According to the area of ​​a single diamond and the area of ​​each connected domain in the diamond connected domain set, the crushed sand ratio of the pre-processed image is obtained, and the crushed sand ratio is used as the crushed sand parameter of the pre-processed image.

[0162] In the above feasible implementation manner, the image processing module 902 is specifically used to: compare the area of ​​the single diamond with the area of ​​each connected domain in the diamond connected domain set to obtain the number of connected domains of broken sand whose area is smaller than the area of ​​the single diamond; and determine the proportion of broken sand in the preprocessed image based on the number of connected domains of broken sand and the total amount of diamonds in the preprocessed image.

[0163] In the above feasible implementation manner, the image processing module 902 is specifically used for:

[0164] According to a set pixel quantity, a large connected domain subset is selected from the diamond connected domain set, wherein the number of pixels contained in each connected domain in the large connected domain subset is greater than the set pixel quantity;

[0165] Using the number of single pixels corresponding to a single diamond, split each connected domain in the large connected domain subset to obtain the number of split connected domains corresponding to the large connected subset;

[0166] Determine the total amount of diamonds in the preprocessed image according to the number of remaining connected domains and the number of split connected domains; wherein the remaining connected domain data is the total amount of all connected domains in the diamond connected domain set except the large connected domain subset;

[0167] The proportion of broken sand in the preprocessed image is determined according to the number of connected domains of broken sand and the total amount of diamonds.

[0168] In a feasible implementation manner of the embodiment of the present application, the proportion determination module 903 is specifically used to: obtain the number of double-defect images that are both agglomeration images and broken sand images in the diamond wire image set based on the agglomeration judgment result of each diamond wire image and the broken sand judgment result of each diamond wire image; determine the proportion of the number of images based on the number of double-defect images and the total number of the diamond wire image set.

[0169] In an optional example of the above feasible implementation mode, the proportion determination module 903 is specifically used to: calculate the ratio of the number of double-defect images to the total number of the diamond wire image set; and use the ratio as the image quantity proportion.

[0170] In another optional example of the above-mentioned feasible implementation mode, the proportion determination module 903 is specifically used to: obtain the number of clustered images in the diamond wire image set that are clustered images according to the clustering judgment result of each diamond wire image; obtain the number of broken sand images in the diamond wire image set that are broken sand images according to the broken sand judgment result of each diamond wire image; determine the number of defective images in the diamond wire image set according to the number of clustered images, the number of broken sand images and the number of double defect images; calculate the ratio of the number of defective images to the total number of the diamond wire image set, and use the ratio as the proportion of the number of images.

[0171] It should be understood that, for the sake of brevity, some of the contents described in the method embodiments will not be repeated in the device embodiments.

[0172] Based on the same inventive concept, the present application embodiment provides an electronic device, see Fig.10 As shown, it includes a processor 1001 and a memory 1002. Among them:

[0173] The processor 1001 is used to execute one or more programs stored in the memory 1002 to implement the quality evaluation method of the diamond wire roll.

[0174] It is understood that the processor 1001 may be a processor core or a processor chip, or other circuits capable of program configuration and operation. The memory 1002 may be a RAM (Random Access Memory), a ROM (Read-Only Memory), a flash memory, etc., but is not limited thereto.

[0175] It is understandable. Fig.10 The structure shown is for illustration only. The electronic device may also include Fig.10 More or fewer components as shown, or with Fig.10 For example, it may also have an internal communication bus for realizing communication between the processor 1001 and the memory 1002; for another example, it may also have an external communication interface, such as a USB (Universal Serial Bus) interface, a CAN (Controller Area Network) bus interface, etc., connected to an image acquisition device, so as to automatically acquire the image of the diamond wire to be detected; for another example, it may also have an information display component such as a display screen, but this is not a limitation.

[0176] Based on the same inventive concept, this embodiment also provides a computer-readable storage medium, such as a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash disk, an SD (Secure Digital Memory Card) card, an MMC (Multimedia Card) card, etc., in which one or more programs for implementing the above steps are stored, and the one or more programs can be executed by one or more processors to implement the quality evaluation method of the diamond wire roll. No further details will be given here.

[0177] In the embodiments provided in this application, various embodiments and implementation methods can be freely combined to obtain new embodiments without conflict.

[0178] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0179] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0180] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0181] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0182] As used herein, a plurality refers to two or more than two.

[0183] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for evaluating the quality of a diamond wire coil, characterized in that: include: Acquire a diamond wire image set corresponding to the diamond wire coil to be tested; wherein the diamond wire image set includes a plurality of diamond wire images; For each diamond wire image in the diamond wire image set: preprocessing the diamond wire image to obtain a preprocessed image; wherein the preprocessed image includes a plurality of background pixels and a plurality of diamond pixels; determining agglomeration parameters and crushed sand parameters of the preprocessed image according to the diamond pixels in the preprocessed image; determining agglomeration judgment result of whether the diamond wire image is agglomeration image according to the agglomeration parameters; and determining a crushed sand judgment result of whether the diamond wire image is a crushed sand image according to the crushed sand parameters; According to the agglomeration judgment result and the crushed sand judgment result of each diamond wire image, the ratio of the number of images determined as agglomeration images and crushed sand images in the diamond wire image set is obtained; Determine whether the image quantity ratio exceeds a set qualified ratio, and obtain a ratio determination result; According to the proportion judgment result, the quality status of the diamond wire reel to be tested is determined.

2. The quality evaluation method of the diamond wire coil according to claim 1, characterized in that: Determining agglomeration parameters of the preprocessed image according to diamond pixels in the preprocessed image includes: Obtaining a target diamond pixel density of the preprocessed image according to the sum of diamond pixel densities of each column of pixels in the preprocessed image under M vertical sliding windows; wherein M is an integer greater than 1; According to the sum of the sand density of each row of pixels in the preprocessed image under N horizontal windows, the target diamond sand density of the preprocessed image is obtained, wherein N is an integer greater than 1, and the agglomeration parameters include the target diamond pixel density and the target diamond sand density.

3. The quality evaluation method of the diamond wire coil according to claim 2, characterized in that: Obtaining a target diamond pixel density of the preprocessed image according to the sum of diamond pixel densities of each column of pixels in the preprocessed image under M vertical sliding windows, comprising: For the i-th column of pixels in the preprocessed image, obtain M diamond pixel densities of the i-th column of pixels under the M vertical sliding windows, and take the sum of the M diamond pixel densities corresponding to the i-th column of pixels as the sum of the diamond pixel densities of the i-th column of pixels, where i is taken from 1 to M in sequence; The diamond pixel density is obtained according to the sum of the diamond pixel densities corresponding to all columns of pixels in the preprocessed image, wherein the diamond pixel density is not less than an average value of the sum of the diamond pixel densities corresponding to all columns of pixels in the preprocessed image.

4. The quality evaluation method of the diamond wire coil according to claim 3, characterized in that: For the i-th column of pixels in the preprocessed image, obtaining M diamond pixel densities of the i-th column of pixels in the M vertical sliding windows, including: For each vertical sliding window in the i-th column of pixels, if i is less than half the window width of the vertical sliding window, the number of image columns in the vertical sliding window is expanded by mirror flipping; According to the image of the expanded vertical sliding window, the diamond pixel density of the i-th column of pixels under the vertical sliding window is obtained; If i is not less than half the window width of the vertical sliding window, the diamond pixel density of the i-th column of pixels under the vertical sliding window is obtained according to the image under the vertical sliding window.

5. The quality evaluation method of the diamond wire coil according to claim 3, characterized in that: According to the sum of the sand density of each row of pixels in the preprocessed image under N horizontal windows, the target diamond sand density of the preprocessed image is obtained, including: For the k-th row of pixels in the preprocessed image, obtain N sand density of the k-th row of pixels in the N horizontal sliding windows, and take the sum of the N sand density corresponding to the k-th row of pixels as the sum of the sand density of the k-th row of pixels, where k is taken from 1 to N in sequence; The target diamond sand density is obtained according to the sum of the sand density corresponding to all rows of pixels in the preprocessed image; wherein the target diamond sand density is not less than the average value of the sum of the sand density corresponding to all rows of pixels in the preprocessed image.

6. The quality evaluation method of the diamond wire coil according to claim 5, characterized in that: For the k-th row of pixels in the preprocessed image, obtaining N sand density of the k-th row of pixels in the N horizontal sliding windows includes: For each horizontal sliding window in the k-th row of pixels in the preprocessed image, if k is less than half the window width of the horizontal sliding window, the number of image rows in the horizontal sliding window is expanded by mirror flipping; According to the image of the expanded horizontal sliding window, the sand density of the k-th row of pixels under the horizontal sliding window is obtained; If k is not less than half the window width of the horizontal sliding window, the sand density of the k-th row of pixels under the horizontal sliding window is obtained according to the image under the horizontal sliding window.

7. The quality evaluation method of the diamond wire coil according to claim 1, characterized in that: Determining the crushing parameters of the preprocessed image according to the diamond pixel points in the preprocessed image includes: Performing connected domain analysis on the preprocessed image to obtain a diamond connected domain set corresponding to diamond pixel points; According to the area of ​​a single diamond and the area of ​​each connected domain in the diamond connected domain set, the crushed sand ratio of the pre-processed image is obtained, and the crushed sand ratio is used as the crushed sand parameter of the pre-processed image.

8. The quality evaluation method of the diamond wire coil according to claim 7, characterized in that: According to the area of ​​a single diamond and the area of ​​each connected domain in the diamond connected domain set, the proportion of broken sand in the pre-processed image is obtained, including: Comparing the area of ​​the single diamond with the area of ​​each connected domain in the diamond connected domain set, obtaining the number of connected domains of crushed sand whose area is smaller than the area of ​​the single diamond; The proportion of broken sand in the preprocessed image is determined according to the number of connected domains of broken sand and the total amount of diamonds in the preprocessed image.

9. The quality evaluation method of the diamond wire coil according to claim 8, characterized in that: Determining the proportion of broken sand in the pre-processed image according to the number of connected domains of broken sand and the total amount of diamonds in the pre-processed image includes: According to a set pixel quantity, a large connected domain subset is selected from the diamond connected domain set, wherein the number of pixels contained in each connected domain in the large connected domain subset is greater than the set pixel quantity; Using the number of single pixels corresponding to a single diamond, split each connected domain in the large connected domain subset to obtain the number of split connected domains corresponding to the large connected subset; Determine the total amount of diamonds in the preprocessed image according to the number of remaining connected domains and the number of split connected domains; wherein the remaining connected domain data is the total amount of all connected domains in the diamond connected domain set except the large connected domain subset; The proportion of broken sand in the preprocessed image is determined according to the number of connected domains of broken sand and the total amount of diamonds.

10. The quality evaluation method of a diamond wire coil according to any one of claims 1 to 9, characterized in that: According to the agglomeration judgment result and the crushed sand judgment result of each diamond wire image, the ratio of the number of images determined as agglomeration images and crushed sand images in the diamond wire image set is obtained, including: According to the agglomeration judgment result of each diamond wire image and the crushed sand judgment result of each diamond wire image, the number of double-defect images in the diamond wire image set that are both agglomeration images and crushed sand images is obtained; The image quantity ratio is determined according to the number of the double-defect images and the total number of the diamond wire image set.

11. The quality evaluation method of the diamond wire coil according to claim 10, characterized in that: Determining the image quantity ratio according to the number of the double defect images and the total amount of the diamond wire image set includes: Calculating the ratio of the number of the double defect images to the total number of the diamond wire image set; The ratio is used as the image quantity ratio.

12. The quality evaluation method of the diamond wire coil according to claim 10, characterized in that: The method further comprises: According to the agglomeration judgment result of each diamond wire image, the number of agglomerated images in the diamond wire image set that are agglomerated images is obtained; According to the crushed sand judgment result of each diamond wire image, the number of crushed sand images in the diamond wire image set that are crushed sand images is obtained; Correspondingly, according to the number of the double defect images and the total number of the diamond wire image set, determining the image quantity ratio includes: Determining the number of defect images in the diamond wire image set according to the number of agglomerated images, the number of broken sand images, and the number of double defect images; The ratio of the number of defective images to the total number of the diamond wire image set is calculated, and the ratio is used as the image quantity ratio.

13. A detection device for a diamond wire image, characterized in that: include: An acquisition module, used for acquiring a diamond wire image set corresponding to the diamond wire coil to be tested; wherein the diamond wire image set includes a plurality of diamond wire images; An image processing module is used for: for each diamond wire image in the diamond wire image set: preprocessing the diamond wire image to obtain a preprocessed image; wherein the preprocessed image includes a plurality of background pixels and a plurality of diamond pixels; determining agglomeration parameters and crushed sand parameters of the preprocessed image according to the diamond pixels in the preprocessed image; determining agglomeration judgment result of whether the diamond wire image is agglomeration image according to the agglomeration parameters; and determining a crushed sand judgment result of whether the diamond wire image is a crushed sand image according to the crushed sand parameters; A proportion determination module, used for obtaining the proportion of the number of images determined as agglomerated images and broken sand images in the diamond wire image set according to the agglomeration judgment result and the broken sand judgment result of each diamond wire image; A judgment module, used to judge whether the image quantity ratio exceeds a set qualified ratio, and obtain a ratio judgment result; The quality status determination module is used to determine the quality status of the diamond wire coil to be tested according to the proportion judgment result.

14. An electronic device, characterized in that: include: Processor and memory; The processor is used to execute one or more programs stored in the memory to implement the diamond wire image detection method as described in any one of claims 1-12.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the diamond wire image detection method according to any one of claims 1 to 12.

Images