Wear-resistant material testing methods

Abstract

The present application relates to the field of data processing, and in particular to a method for detecting wear-resistant materials, the method comprising: obtaining a static image and a dynamic image of the bonding surface between the coating of the wear-resistant material and the coated substrate, the dynamic image being a microscopic image of the bonding surface after each friction, and the static image being an initial microscopic image of the bonding surface; graying the static image to obtain a first grayscale image and calculating the dispersion of the wear-resistant material; graying the dynamic image to obtain a second grayscale image and calculating the amount of wear of the wear-resistant material after each friction; calculating the dynamic friction loss, the dynamic friction loss is proportional to the cumulative sum of the number of pixels in the scratch area and the number of pixels in the debris area, and is inversely proportional to the stability of the wear-resistant material; the ratio of the dynamic friction loss to the dispersion is used as the friction loss. The present application is applied in the testing of wear-resistant materials with small friction scratches, and has the effect of quantifying the degree of wear on the surface of the wear-resistant material.

Description

Technical Field

[0001] The invention relates to the field of data processing, and in particular to a wear-resistant material detection method. Background Art

[0002] The wear resistance of the surface of wear-resistant materials can be detected with the help of machine learning methods. By acquiring images of the material surface and performing data processing, features are extracted, the wear area in the image is analyzed, and a model is established through a machine learning algorithm, thereby realizing automated detection of the wear resistance of the surface of wear-resistant materials.

[0003] In the prior art, the patent document with the authorization number CN114445407B discloses a method and device for detecting the wear resistance of materials. The method tests and analyzes the friction performance by acquiring the depth image of the material after friction. The depth image can provide height information of the surface of the wear-resistant material, including the depth change caused by friction or scratches. By processing and analyzing the depth image, features related to the wear resistance of the material can be extracted. For example, parameters such as the depth, length and shape of friction or scratches can be calculated to evaluate the wear resistance of the material.

[0004] The method of friction testing and analysis through depth images itself has the problem of difficulty in extracting depth features, and is not suitable for testing wear-resistant materials with small friction scratches after wear resistance experiments. For testing wear-resistant materials with small friction scratches, the existing technology cannot accurately reflect the interface between the coating and the base material of the wear-resistant material, cannot quantify the degree of wear on the surface of the wear-resistant material, and cannot accurately evaluate the wear resistance of the material. Summary of the invention

[0005] In order to quantify the degree of wear on the surface of wear-resistant materials and accurately evaluate the wear resistance of materials in the test of wear-resistant materials with small friction scratches, the present application provides a method for detecting wear-resistant materials.

[0006] The invention discloses a method for detecting wear-resistant materials, comprising the following steps: obtaining a static image and a dynamic image of the bonding surface between a coating of a wear-resistant material and a coated substrate, wherein the dynamic image is a microscopic image of the bonding surface after each friction, and the static image is an initial microscopic image of the bonding surface; graying the static image to obtain a first gray image and calculating the dispersion of the wear-resistant material; graying the dynamic image to obtain a second gray image and calculating the wear amount of the wear-resistant material after each friction; performing regression analysis and calculation based on the wear amount and the number of frictions to obtain a regression line for the relationship between the wear amount and the number of frictions, and obtaining the stability of the wear-resistant material based on the regression line, wherein the stability is the probability that the image of the relationship between the wear amount of the wear-resistant material and the number of frictions is a straight line; obtaining a scratch area and a debris area in the second gray image, and calculating the dynamic friction loss of the wear-resistant material, wherein the dynamic friction loss is proportional to the cumulative sum of the number of pixels in the scratch area and the number of pixels in the debris area, and is inversely proportional to the stability of the wear-resistant material; and taking the ratio of the dynamic friction loss to the dispersion as the friction loss.

[0007] The beneficial effects are: obtaining the changes in the wear-resistant material after each friction, obtaining a dynamic image, using machine learning methods to calculate the stability of the wear-resistant material, combining the static microscopic image of the bonding surface of the wear-resistant material coating and the substrate to obtain the wear resistance of the wear-resistant material. For the test of wear-resistant materials with smaller friction scratches, it can capture the subtle changes in the scratches and accurately reflect the interface conditions between the coating and the substrate material. It solves the problem that traditional methods have subjective judgments and are difficult to accurately quantify the degree of wear on the material surface, and accurately evaluates the wear resistance of the material.

[0008] Optionally, the dispersion is calculated as: ,in, is the diffusion, is the normalized grayscale dependency matrix. The number of rows in the grayscale dependency matrix is ​​8, and the number of columns in the grayscale dependency matrix is ​​9. express Middle Line The value of the column.

[0009] Optionally, the method for constructing a grayscale dependency matrix is: calculating the dependency value, the dependency value is the number of pixels in the eight-neighborhood range of the pixel point in the first grayscale image that have the same pixel value as the pixel point; setting the grayscale level, and constructing a grayscale dependency matrix, in which the rows represent the pixel values ​​of the image and the columns represent the dependency values.

[0010] Optionally, the wear calculation formula is:

[0011] ,in, is the amount of wear, Indicates the grayscale image after grayscale level division. Line The grayscale value of the column position, Represents the rows of the grayscale image, Columns representing grayscale images, It is the grayscale image Line The grayscale value of the column position, the grayscale level is set to indivual.

[0012] Optionally, dynamic friction loss can be calculated as: ,in, is the dynamic friction loss, is the number of pixels in the scratch area, is the number of pixels in the debris area, is the first coefficient set, is the second coefficient set, For the stability of wear-resistant materials.

[0013] Optionally, obtaining the scratch area and the debris area in the second grayscale image includes the steps of:

[0014] Performing threshold segmentation on the second grayscale image to obtain a foreground area including scratches and debris;

[0015] The length and width of the maximum circumscribed rectangle of the connected domain in the foreground area are calculated, and the scratch area, debris area and background area are distinguished according to the value of the aspect ratio.

[0016] The wear-resistant material detection method of this application has the following technical effects:

[0017] 1. Obtain the changes of wear-resistant materials after each friction, obtain dynamic images, use machine learning methods to calculate the stability of wear-resistant materials, and combine the static microscopic images of the bonding surface of wear-resistant material coating and substrate to obtain the wear resistance of wear-resistant materials. For wear-resistant materials with smaller friction scratches, it can capture the slight changes in scratches and accurately reflect the interface between coating and substrate material, solving the problem of subjective judgment and difficulty in accurately quantifying the degree of wear on the material surface in traditional methods.

[0018] 2. By calculating the friction loss, the quantitative friction test results are generated, and the friction loss value is output as the wear resistance test result. The uniformity of the wear-resistant material coating is quantified according to the dispersion, and the wear of the wear-resistant material after each friction is quantified according to the dynamic friction loss. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easily understood. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-restrictive manner, and the same or corresponding numbers represent the same or corresponding parts.

[0020] Figure 1 It is a method flow chart of the wear-resistant material detection method of the embodiment of the present application.

[0021] Figure 2 It is a flow chart of the method for calculating dispersion in the wear-resistant material detection method in the embodiment of the present application. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0023] It should be understood that when the terms "first", "second", etc. are used in the claims, specification and drawings of the present application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0024] The embodiment of the present application discloses a wear-resistant material detection method for detecting the wear-resistant performance of the wear-resistant material, which can quantify the degree of wear on the surface of the wear-resistant material in the test of the wear-resistant material with small friction scratches, referring to Figure 1 , including steps S1-S6, which are as follows:

[0025] S1: Obtain static images and dynamic images of the bonding surface between the wear-resistant material coating and the coated substrate. The dynamic image is a microscopic image of the bonding surface after each friction, and the static image is an initial microscopic image of the bonding surface.

[0026] This application uses SEM (Scanning Electron Microscope) to collect microscopic images. The bonding surface image is a side-view image of the coating and the substrate. The coating is subjected to multiple friction tests, and the dynamic image is a microscopic image of the bonding surface after each friction. The static image is a microscopic image of the bonding surface between the wear-resistant material coating and the substrate before the friction test.

[0027] S2: grayscale the static image to obtain a first grayscale image and calculate the dispersion of the wear-resistant material.

[0028] Reference Figure 2 , the calculation of the diffusion degree includes steps S20-S22, which are as follows:

[0029] S20: Calculate the dependency value.

[0030] The dependency value is the number of pixels with the same pixel value as the pixel within the eight-neighborhood range of the pixel in the first grayscale image. The eight-neighborhood refers to the neighborhood consisting of eight pixels around a pixel. Specifically, it includes the neighboring pixels above, below, left, right and four diagonal directions of the pixel.

[0031] S21: Set the grayscale level and construct the grayscale dependency matrix.

[0032] In one embodiment, in order to reduce the amount of calculation, the gray level is set to 8, and the gray level dependency matrix is ​​obtained: , It is a matrix with 8 rows and 9 columns. The rows represent the pixel values ​​of the image, and the columns represent how many pixels in the eight neighborhoods around each pixel value have the same pixel value.

[0033] In the gray-scale dependency matrix, the rows represent the number of gray levels and the columns represent the dependency values. The gray-scale dependency matrix is ​​a matrix used to describe the dependency relationship between elements in a system. It is an existing technology and will not be described in detail here.

[0034] S22: Calculate the entropy of the gray-level dependence matrix to obtain the dispersion of the wear-resistant material.

[0035] Dispersion is used to measure the dispersion or discreteness of data. In one embodiment, the calculation formula of dispersion is: ,in, is the diffusion, is the normalized grayscale dependency matrix. The number of rows in the grayscale dependency matrix is ​​8, and the number of columns in the grayscale dependency matrix is ​​9. express Middle Line The value of the column.

[0036] The smaller the dispersion, the more uniform the grayscale distribution of the image, that is, the less obvious the texture of the image, which means the more uniform the coating coverage; the larger the dispersion, the more uneven the grayscale distribution of the image, that is, the more obvious the texture of the image, which means the more uneven the coating coverage.

[0037] S3: grayscale the dynamic image to obtain a second grayscale image and calculate the wear amount of the wear-resistant material after each friction.

[0038] Specifically, an image is taken each time friction occurs to obtain a dynamic image sequence of the wear-resistant material, the dynamic image is converted into a second grayscale image, a grayscale image sequence is constructed, and the grayscale value of each pixel point in the subsequent image in the grayscale image sequence is subtracted from the grayscale value of each pixel point at the corresponding position in the previous image to obtain a differential sequence of grayscale images.

[0039] The energy of each second grayscale image in the differential sequence is used as the wear amount of the friction-resistant material each time. In one embodiment, the wear amount calculation formula is:

[0040] ,in, is the amount of wear, Indicates the grayscale image after grayscale level division. Line The grayscale value of the column position, Represents the rows of the grayscale image, Columns representing grayscale images, It is the grayscale image Line The grayscale value of the column position, the grayscale level is set to For example, in order to reduce the amount of calculation, the present application divides the grayscale values ​​on the grayscale image into 16 grayscale levels, namely Take 16.

[0041] S4: According to the wear amount and the number of frictions, regression analysis and calculation are performed to obtain a regression line about the relationship between the wear amount and the number of frictions. According to the regression line, the stability of the wear-resistant material is obtained. The stability is the probability that the graph of the relationship between the wear amount and the number of frictions of the wear-resistant material is a straight line.

[0042] Specifically, the wear amount of the wear-resistant material obtained in step S3 is displayed in two-dimensional coordinates, with the number of frictions as the horizontal axis and the wear amount as the vertical axis. For stable wear-resistant materials, the wear amount should be a smooth straight line or curve with steady growth, with less sudden growth. For unstable wear-resistant materials, it is a broken line or curve with more sudden growth.

[0043] In one embodiment, the method for performing regression analysis on the wear amount is: constructing a fitting straight line: ,in, is the amount of wear, is the number of frictions, and For the coefficients we need, we use the least squares method to find and The value of is fitted to get the regression line, and the regression line is Test, get the wear amount and friction times of wear-resistant materials The probability of a straight line is a straight line. The greater the probability of a straight line, the more stable the wear-resistant material is. For the stability of wear-resistant materials. The inspection is a prior art and will not be described in detail here.

[0044] S5: Obtain the scratch area and the debris area in the second grayscale image, and calculate the dynamic friction loss of the wear-resistant material. The dynamic friction loss is proportional to the cumulative sum of the number of pixels in the scratch area and the number of pixels in the debris area, and is inversely proportional to the stability of the wear-resistant material.

[0045] The dynamic image taken at the end of the friction experiment is converted into a second grayscale image, and the second grayscale image is threshold segmented to obtain the background and foreground. The foreground area includes scratches, debris and part of the background. When judging the wear resistance of wear-resistant materials, the effects of scratches and debris are different. The presence of debris can better indicate that the wear resistance of the wear-resistant material is poor than that of scratches. Therefore, it is necessary to determine which connected domains are debris and which are scratches. Threshold segmentation is a prior art and will not be described in detail here.

[0046] In the image after threshold segmentation, the connected domain of scratches is basically a straight line, and the connected domain of debris is a closed curve. Find the maximum circumscribed rectangle of the connected domain and calculate the aspect ratio of the circumscribed rectangle. The length refers to the longer side of the rectangle, and the width refers to the shorter side of the rectangle. If the aspect ratio is greater than 2, the connected domain is considered to be a scratch, and if the aspect ratio is less than or equal to 1.5, the connected domain is considered to be debris, and the aspect ratio between 1.5 and 2 is considered to be the background. In this way, the scratch area and the debris area are divided.

[0047] Calculate the number of pixels A in the scratch area and the number of pixels B in the debris area, and calculate the dynamic friction of the wear-resistant material. The calculation formula for dynamic friction loss is: ,in, is the dynamic friction loss, is the number of pixels in the scratch area, is the number of pixels in the debris area, is the first coefficient set, is the second coefficient set, For the stability of wear-resistant materials.

[0048] and It can be set according to user needs. In this application, Set to 0.4, Set to 0.6, the larger the scratch area and debris area in the image, the greater the dynamic friction loss; the smaller the stability of the wear-resistant material, the greater the dynamic friction loss.

[0049] S6: The ratio of dynamic friction loss to dispersion is taken as friction loss.

[0050] The calculation formula for friction loss is: ,in, is the friction loss, is the dynamic friction loss, represents the dispersion, In order to avoid the friction loss denominator is 0. Dynamic friction loss The larger the friction loss of the wear-resistant material The larger the wear resistance, the greater the friction loss of the wear-resistant material. The smaller the wear-resistant material, the dispersion The larger it is, the more uniform the wear-resistant material is covered, so the coating is more wear-resistant and the friction loss of the wear-resistant material is The smaller the friction loss of the wear-resistant material, the The larger the friction loss value is, the greater the wear resistance will be.

[0051] In one embodiment, friction tests are performed on a variety of wear-resistant materials, and the friction losses of the wear-resistant materials are calculated after the tests. By comparing the friction losses of different wear-resistant materials, a wear-resistant material with better wear resistance can be obtained.

[0052] The implementation principle of the wear-resistant material detection method of the embodiment of the present application is:

[0053] The changes in the wear-resistant material after each friction are obtained, and a dynamic image is obtained. The stability of the wear-resistant material is calculated using the machine learning method, and the wear resistance of the wear-resistant material is obtained by combining the static microscopic image of the bonding surface of the wear-resistant material coating and the substrate. For the test of wear-resistant materials with smaller friction scratches, the slight changes in the scratches can be captured, and the interface conditions between the coating and the base material can be accurately reflected, which solves the problem that traditional methods have subjective judgments and are difficult to accurately quantify the degree of wear on the material surface.

[0054] By calculating the friction loss, quantitative friction test results are generated. The greater the friction loss of the wear-resistant material, the greater the dispersion of the wear-resistant material, which means that the fusion of the wear-resistant layer and the base layer is better, the more wear-resistant it is, and the smaller the friction loss of the wear-resistant material. The friction loss calculation results and the wear resistance results are generated into visual test results output to accurately evaluate the wear resistance of the material.

Claims

1. A wear-resistant material detection method, characterized in that: Includes steps: Obtain static images and dynamic images of the bonding surface between the coating of the wear-resistant material and the coated substrate, the dynamic image is a microscopic image of the bonding surface after each friction, and the static image is an initial microscopic image of the bonding surface; The static image is grayed to obtain a first gray image and the dispersion of the wear-resistant material is calculated; the dispersion calculation formula is: ,in, is the diffusion, is the normalized grayscale dependency matrix. The number of rows in the grayscale dependency matrix is ​​8, and the number of columns in the grayscale dependency matrix is ​​9. express Middle Line The grayscale dependency matrix is ​​constructed by: calculating the dependency value, which is the number of pixels in the eight neighborhoods of the pixel point in the first grayscale image that have the same pixel value as the pixel point; setting the grayscale level, and constructing the grayscale dependency matrix, in which the rows represent the pixel values ​​of the image and the columns represent the dependency values; The dynamic image is grayed to obtain the second gray image and the wear amount of the wear-resistant material after each friction is calculated; the wear amount calculation formula is: ,in, is the amount of wear, Indicates the grayscale image after grayscale level division. Line The grayscale value of the column position, Represents the rows of the grayscale image, Columns representing grayscale images, It is the first Line The grayscale value of the column position, the grayscale level is set to indivual; According to the wear amount and the number of frictions, regression analysis and calculation are performed to obtain the regression line of the relationship between the wear amount and the number of frictions. According to the regression line, the stability of the wear-resistant material is obtained. The stability is the probability that the relationship between the wear amount and the number of frictions of the wear-resistant material is a straight line. Construct the fitted line: ,in, is the amount of wear, is the number of frictions, and For the coefficients we need, we use the least squares method to find and The value of is fitted to get the regression line, and the regression line is Test, get the wear amount and friction times of wear-resistant materials The probability of it being a straight line is greater, and the greater the probability of it being a straight line, the more stable the wear-resistant material is. The scratch area and the debris area in the second grayscale image are obtained, and the dynamic friction loss of the wear-resistant material is calculated. The dynamic friction loss is proportional to the cumulative sum of the number of pixels in the scratch area and the number of pixels in the debris area, and is inversely proportional to the stability of the wear-resistant material. The calculation formula for the dynamic friction loss is: ,in, is the dynamic friction loss, is the number of pixels in the scratch area, is the number of pixels in the debris area, is the first coefficient set, is the second coefficient set, For the stability of wear-resistant materials; The ratio of dynamic friction loss to dispersion is taken as friction loss.

2. The wear-resistant material detection method according to claim 1, characterized in that: Obtaining the scratch area and the debris area in the second grayscale image includes the following steps: Performing threshold segmentation on the second grayscale image to obtain a foreground area including scratches and debris; The length and width of the maximum circumscribed rectangle of the connected domain in the foreground area are calculated, and the scratch area, debris area and background area are distinguished according to the value of the aspect ratio.

Images