Brake disc polishing process detection method and system based on machine vision

Abstract

The invention relates to the technical field of machine vision, and discloses a brake disc polishing process detection method and system based on machine vision, and the method comprises the steps: obtaining the original image data of a polished brake disc; performing region division according to the original image data to obtain a plurality of sub-images; performing edge detection according to the plurality of sub-images to obtain an edge contour point and a minimum circumcircle; calculating the Euclidean distance from the edge contour point to the center of the minimum circumcircle to obtain circularity; calculating a brightness evaluation parameter according to the gray value of the pixel point, and combining the brightness evaluation parameter with circularity calculation to obtain a roughness evaluation parameter; according to the Euclidean distance from each edge contour point to the center of the minimum circumcircle, roundness evaluation parameters are calculated; and performing calculation according to the quality evaluation parameter, the roughness evaluation parameter and the roundness evaluation parameter to obtain a process evaluation parameter of the whole brake disc. The method can improve the detection precision of the brake disc polishing process.

Description

Technical Field

[0001] The present invention relates to the technical field of machine vision, and in particular, to a method and system for detecting the grinding process of brake discs based on machine vision. Background Art

[0002] Currently, in the manufacturing process of brake discs, the blanks need to be ground multiple times to remove defects and residual burrs during the processing, so as to obtain high-quality and high-performance brake discs. The grinding process has an important impact on the performance of brake discs. By grinding, parameters such as the surface roughness, roundness, and surface quality of the brake discs can be controlled, thereby improving the friction performance and safety performance of the brake discs. Therefore, during the production process, it is necessary to detect and evaluate the grinding process to ensure that the product quality requirements are met.

[0003] Traditional methods for evaluating the surface quality of brake discs mainly include manual visual inspection method, roughness meter method, etc. These methods have problems such as strong subjectivity and low accuracy. With the development of machine vision technology, people have begun to use image processing technology for quality evaluation, which has the advantages of high accuracy and high efficiency compared with the manual visual inspection method and the roughness meter method. When performing image processing, common algorithms include region segmentation, edge detection, texture analysis, etc. For example, surface contour features can be obtained through edge detection, and quality evaluation can be carried out by combining gray-scale distribution and geometric features; surface roughness features can be obtained through texture analysis, and quality evaluation can be carried out by combining statistical features and root mean square values; roundness features can be obtained through Hough circle detection, and quality evaluation can be carried out by combining roundness and ellipticity. These methods have improved the accuracy and speed of evaluation to a certain extent, but there are also some limitations. For example, in texture analysis, for brake discs of different materials, the corresponding quality evaluation methods need to be adjusted accordingly; in edge detection, for brake discs of different shapes, the corresponding algorithms need to be processed differently.

[0004] There is a problem of insufficient detection accuracy in the prior art. Summary of the Invention

[0005] The present invention provides a method and system for detecting the grinding process of brake discs based on machine vision, which uses machine vision technology and multi-dimensional index evaluation method to accurately evaluate and analyze the surface quality and grinding process quality of brake discs, so as to solve the problem of insufficient detection accuracy.

[0006] In a first aspect, to solve the above technical problems, the present invention provides a method for detecting the grinding process of brake discs based on machine vision, including: Obtaining the original image data of the ground brake disc; Performing region division according to the original image data to obtain a plurality of sub-images; Edge detection is performed based on the multiple sub-images to obtain edge contour points and the minimum circumscribed circle; Calculate the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, determine the Euclidean dispersion coefficient based on the Euclidean distance, and determine the circularity of the sub-image based on the Euclidean dispersion coefficient; Extract the gray values of the pixel points from the multiple sub-images, calculate the brightness evaluation parameter according to the gray values of the pixel points, and calculate the quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image; Calculate the roughness evaluation parameter according to the minimum circumscribed circle, the gray values of the pixel points, and a preset weight coefficient; Calculate the roundness evaluation parameter according to the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle; Calculate based on the quality evaluation parameter, the roughness evaluation parameter, and the roundness evaluation parameter to obtain the process evaluation parameter of the entire brake disc; When the process evaluation parameter is greater than a preset standard threshold, it is determined that the brake disc grinding process inspection is qualified.

[0007] Preferably, the obtaining of the multiple sub-images includes: Obtain the original image data of the brake disc after grinding; According to the original image data of the brake disc after grinding, perform noise reduction processing on each image using the mean filtering method; Perform segmentation on each image using the Otsu threshold segmentation method to segment the noise-reduced image into a brake disc area and a background; Divide the image into multiple continuous non-overlapping sub-images of the same size, where the area of the sub-image is 1 / 100 of the surface area of the brake disc, to obtain multiple sub-images.

[0008] Preferably, the performing of edge detection based on the multiple sub-images to obtain edge contour points and the minimum circumscribed circle includes: According to the multiple sub-images, convert the pixel points of the sub-images into gray values to form a gray image, and perform gray level equalization on each gray image; Use the sobel operator to perform edge detection on the gray image and denoise the edge detection result; Perform circle detection on the denoised edge detection result using the Hough transform method, and perform drawing to obtain the minimum circumscribed circle; Obtain the edge contour of the sub-image, use the center of the minimum circumscribed circle as the coordinate origin, draw a contour histogram, and obtain the edge contour points of the sub-image.

[0009] Preferably, calculating the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, and determining the Euclidean dispersion coefficient according to the Euclidean distance includes: Calculating the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle; Obtaining the mean and standard deviation of all the Euclidean distances, calculating the ratio between the standard deviation and the mean, and taking the ratio as the Euclidean dispersion coefficient of all the edge contour points of the sub-image to the center of the minimum circumscribed circle.

[0010] Preferably, extracting the gray values of the pixel points from the multiple sub-images, calculating the brightness evaluation parameter according to the gray values of the pixel points, and calculating the quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image includes: Extracting the gray values of the pixel points from the multiple sub-images, and calculating the mean of all the gray values of the pixel points, the maximum of all the gray values of the pixel points, and the minimum of all the gray values of the pixel points; Calculating the mean of all the gray values of the pixel points according to the maximum and minimum of all the gray values of the pixel points, and constructing the brightness equalization parameter with the mean as the initial value; Calculating the brightness evaluation parameter according to the mean of all the gray values of the pixel points and the brightness equalization parameter; Calculating the quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image, in combination with a preset weight coefficient; Among them, the mean of all the gray values of the pixel points, the brightness evaluation parameter, and the quality evaluation parameter of the sub-image are calculated through the following formulas: ； ; ; In the formula, represents the mean of all the gray values of the pixel points, represents the gray value of the sub-image, represents the coordinates of the pixel point, represents the total number of pixel points in the sub-image, represents the area of the sub-image; represents the quality evaluation parameter of the sub-image, represents the circularity of the sub-image, and represent the preset weight coefficient, represents the brightness evaluation parameter, represents the brightness equalization parameter.

[0011] Preferably, the roughness evaluation parameter calculated according to the minimum circumscribed circle, the gray value of the pixel point and a preset weight coefficient includes: The roughness evaluation parameter of the sub-image is calculated by the following formula: ; Wherein, represents the roughness evaluation parameter of the sub-image, represents the minimum value of the distance from the center of the minimum circumscribed circle to the edge line, is the length of the sub-image, , are respectively weight coefficients, represents the maximum value of the distance from the center of the minimum circumscribed circle to the edge line.

[0012] Preferably, the roundness evaluation parameter calculated according to the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle includes: Calculate the range of the Euclidean distance according to the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle; Take the range of the Euclidean distance as the roundness evaluation parameter.

[0013] Preferably, the roughness evaluation parameter of the sub-image and the roundness evaluation parameter of the sub-image are used to obtain the process evaluation parameter of the entire brake disc through the process evaluation parameter calculation formula, including: The process evaluation parameter of the sub-image can be calculated by the following formula: ; Wherein, represents the process evaluation parameter of the i-th sub-image, represents the quality evaluation parameter of the i-th sub-image, represents the roughness evaluation parameter of the i-th sub-image, represents the roundness evaluation parameter of the i-th sub-image, , , respectively represent the weights of the three parameters; The process evaluation parameter of the entire brake disc can be calculated by the following formula: ; represents the process evaluation parameter of the entire brake disc, represents the ratio of the sub-image, and N represents the number of sub-images.

[0014] In a second aspect, the present invention provides a brake disc grinding process detection system based on machine vision, including: An image acquisition module for acquiring image data of a polished brake disc; A region division module that divides regions according to the image data of the polished brake disc to obtain multiple sub-images, including: dividing regions according to the original image data to obtain multiple sub-images.

[0015] In one implementation, according to the original image data of the polished brake disc, each image is denoised using the mean filter method; each image is segmented using the Otsu threshold segmentation method, and the denoised image is segmented into a brake disc region and a background; the image is divided into multiple continuous non-overlapping sub-images of the same size, where the area of each sub-image is 1 / 100 of the surface area of the brake disc, to obtain multiple sub-images; An edge detection module for performing edge detection on the multiple sub-images to obtain sub-image edge contour points and a minimum circumscribed circle, including: converting the pixel points of the sub-images into gray values to form a gray image according to the multiple sub-images, and equalizing the gray levels of each gray image; performing edge detection on the gray image using the sobel operator and denoising the edge detection result; performing circle detection on the denoised edge detection result using the Hough transform method and drawing to obtain the minimum circumscribed circle; obtaining the edge contour of the sub-image, taking the center of the minimum circumscribed circle as the coordinate origin, and drawing a contour histogram to obtain sub-image edge contour points; A circularity calculation module that calculates the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, determines the Euclidean dispersion degree coefficient according to the Euclidean distance, and determines the circularity of the sub-image based on the Euclidean dispersion degree coefficient, including: calculating the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle; obtaining the mean and standard deviation of all the Euclidean distances, calculating the ratio between the standard deviation and the mean, and taking the ratio as the Euclidean dispersion degree coefficient of all the edge contour points of the sub-image to the center of the minimum circumscribed circle; A quality evaluation module extracts the gray values of pixel points from the multiple sub-images, calculates a brightness evaluation parameter according to the gray values of the pixel points, and calculates a quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image, including: extracting the gray values of pixel points from the multiple sub-images, calculating the mean of all the gray values of the pixel points, the maximum value of all the gray values of the pixel points, and the minimum value of all the gray values of the pixel points; calculating the mean of all the gray values of the pixel points according to the maximum value and the minimum value of all the gray values of the pixel points, and constructing a brightness equalization parameter with the mean as the initial value; calculating a brightness evaluation parameter according to the mean of all the gray values of the pixel points and the brightness equalization parameter; calculating a quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image, in combination with a preset weight coefficient; Among them, the mean value of the gray values of all pixel points, the brightness evaluation parameter, and the quality evaluation parameter of the sub-image are calculated by the following formulas: ; ; ; In the formula, represents the mean value of the gray values of all pixel points, represents the gray value of the sub-image, represents the coordinates of the pixel point, represents the total number of pixel points in the sub-image, represents the area of the sub-image; represents the quality evaluation parameter of the sub-image, represents the circularity of the sub-image, and represent preset weight coefficients, represents the brightness evaluation parameter, represents the brightness balance parameter; The roughness evaluation module calculates the roughness evaluation parameter according to the minimum circumscribed circle, the gray value of the pixel point and the preset weight coefficient; The circularity evaluation module calculates the circularity evaluation parameter according to the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, including: The roughness evaluation parameter of the sub-image is calculated by the following formula: ; Wherein, represents the roughness evaluation parameter of the sub-image, represents the minimum value of the distance from the center of the minimum circumscribed circle to the edge line, is the length of the sub-image, , are respectively weight coefficients, represents the maximum value of the distance from the center of the minimum circumscribed circle to the edge line; The process evaluation module calculates based on the quality evaluation parameter, the roughness evaluation parameter and the circularity evaluation parameter to obtain the process evaluation parameter of the entire brake disc, including: calculating the range of the Euclidean distances according to the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle; taking the range of the Euclidean distances as the circularity evaluation parameter.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a detection method for the grinding process of brake discs based on machine vision, including obtaining the original image data of the ground brake discs; performing region division according to the original image data to obtain a plurality of sub-images; performing edge detection according to the plurality of sub-images to obtain edge contour points and a minimum circumscribed circle; calculating the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle, and determining the Euclidean dispersion degree coefficient according to the Euclidean distance, and determining the circularity of the sub-image based on the Euclidean dispersion degree coefficient; there is a negative correlation between the Euclidean dispersion degree and the circularity of the sub-image; extracting the gray values of pixel points for the plurality of sub-images, calculating a brightness evaluation parameter according to the gray values of pixel points, and calculating a quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image; calculating a roughness evaluation parameter according to the minimum circumscribed circle, the gray values of pixel points and a preset weight coefficient; calculating a roundness evaluation parameter according to the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle; calculating according to the quality evaluation parameter, the roughness evaluation parameter and the roundness evaluation parameter to obtain a process evaluation parameter of the entire brake disc; when the process evaluation parameter is greater than a preset standard threshold, it is determined that the detection of the brake disc grinding process is qualified. The surface quality and grinding process quality of the brake disc can be accurately evaluated and analyzed. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic flowchart of a detection method for a brake disc grinding process provided by the first embodiment of the present invention; Figure 2 is a schematic structural diagram of a detection system for a brake disc grinding process provided by the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] Referring to Figure 1 , the first embodiment of the present invention provides a detection method for a brake disc grinding process based on machine vision, including the following steps: S11, obtaining the original image data of the ground brake disc; S12, performing region division according to the original image data to obtain a plurality of sub-images; S13, performing edge detection according to the plurality of sub-images to obtain edge contour points and a minimum circumscribed circle; S14. Calculate the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, determine the Euclidean dispersion coefficient based on the Euclidean distance, and determine the circularity of the sub-image based on the Euclidean dispersion coefficient. S15. Extract the gray values of the pixel points for the multiple sub-images, calculate the brightness evaluation parameter based on the gray values of the pixel points, and calculate the quality evaluation parameter of the sub-image based on the brightness evaluation parameter and the circularity of the sub-image. S16. Calculate the roughness evaluation parameter based on the minimum circumscribed circle, the gray values of the pixel points, and a preset weight coefficient. S17. Calculate the roundness evaluation parameter based on the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle. S18. Perform calculations based on the quality evaluation parameter, the roughness evaluation parameter, and the roundness evaluation parameter to obtain the process evaluation parameter of the entire brake disc.

[0020] In step S11, obtain the original image data of the polished brake disc.

[0021] It should be noted that after the polishing process, the surface is photographed by an image acquisition device (such as a high-resolution industrial camera) to obtain the unprocessed image information. These image data contain features such as the texture, defects, and scratches on the surface of the brake disc, which are the basis for subsequent quality inspection and analysis.

[0022] In step S12, perform region division based on the original image data to obtain multiple sub-images.

[0023] In one implementation, according to the original image data of the polished brake disc, use the mean filter method to perform noise reduction processing on each image; use the Otsu threshold segmentation method to segment each image, and segment the noise-reduced image into the brake disc region and the background; divide the image into multiple continuous non-overlapping sub-images of the same size, where the area of the sub-image is 1 / 100 of the surface area of the brake disc, to obtain multiple sub-images.

[0024] It should be noted that the multiple sub-images are obtained by separating the polished brake disc image into individual regions according to the gray value differences of the pixel points based on the original image data of the polished brake disc, and dividing the polished brake disc image into multiple sub-images.

[0025] In step S13, perform edge detection on the multiple sub-images to obtain edge contour points and the minimum circumscribed circle.

[0026] In one embodiment, according to the multiple sub-images, the pixel points of the sub-images are converted into gray values to form a gray image, and gray level equalization is performed on each gray image; the Sobel operator is used to perform edge detection on the gray image, and the edge detection result is denoised; according to the denoised edge detection result, the Hough transform method is used for circle detection, and drawing is performed to obtain the minimum circumscribed circle; the edge contour of the sub-image is obtained, and with the center of the minimum circumscribed circle as the coordinate origin, a contour histogram is drawn to obtain the edge contour points of the sub-image.

[0027] It should be noted that the edge contour points are the pixel points of the set of the contour edge lines of the sub-image obtained by performing edge detection on the sub-image; the minimum circumscribed circle is the circle with the minimum radius and center that covers all the effective detection points of the brake disc, and the minimum circumscribed circle of the edge contour points is recursively solved through the Welzl algorithm.

[0028] In step S14, the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle is calculated, and the Euclidean dispersion degree coefficient is determined according to the Euclidean distance, and the circularity of the sub-image is determined based on the Euclidean dispersion degree coefficient.

[0029] In one embodiment, the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle is calculated; the mean value and standard deviation of all the Euclidean distances are obtained, the ratio between the standard deviation and the mean value is calculated, and the ratio is used as the Euclidean dispersion degree coefficient of all the edge contour points of the sub-image to the center of the minimum circumscribed circle.

[0030] It should be noted that the Euclidean dispersion degree coefficient is a statistical index used to measure the dispersion degree of data distribution. It reflects the degree of dispersion of data by calculating the Euclidean distance between each data point in the sample and the mean value; circularity is a geometric property used to evaluate the closeness between an actual circle and an ideal circle. It is an index for measuring the similarity between a two-dimensional contour shape and a perfect circle, and is usually used in quality inspection, precision machining, and geometric measurement; there is a negative correlation between circularity and the Euclidean dispersion degree coefficient of the edge points to the center of the circle. The Euclidean dispersion degree of the edge points to the center of the circle needs to be calculated by calculating the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle; the mean value and standard deviation of all the Euclidean distances are obtained, the ratio between the standard deviation and the mean value is calculated, and the ratio is used as the Euclidean dispersion degree coefficient of all the edge contour points of the sub-image to the center of the minimum circumscribed circle.

[0031] In step S15, the pixel point gray values of the multiple sub-images are extracted, the brightness evaluation parameter is calculated according to the pixel point gray values, and the quality evaluation parameter of the sub-image is calculated according to the brightness evaluation parameter and the circularity of the sub-image.

[0032] In one embodiment, pixel grayscale values of the multiple sub-images are extracted, and the mean value, the maximum value, and the minimum value of all pixel grayscale values are calculated; the mean value of all pixel grayscale values is calculated based on the maximum value and the minimum value of all pixel grayscale values, and a brightness equalization parameter is constructed as an initial value; a brightness evaluation parameter is calculated based on the mean value of all pixel grayscale values and the brightness equalization parameter; a quality evaluation parameter of the sub-image is calculated based on the brightness evaluation parameter and the circularity of the sub-image, in combination with a preset weight coefficient. Among them, the mean value of all pixel grayscale values, the brightness evaluation parameter, and the quality evaluation parameter of the sub-image are calculated by the following formulas: ; ; ; In the formula, represents the mean value of all pixel grayscale values, represents the grayscale value of the sub-image, represents the coordinates of the pixel point, represents the total number of pixel points in the sub-image, represents the area of the sub-image; represents the quality evaluation parameter of the sub-image, represents the circularity of the sub-image, and represent the preset weight coefficients, represents the brightness evaluation parameter, represents the brightness equalization parameter.

[0033] It should be noted that the quality evaluation parameter is an evaluation parameter for the brake disc process. The quality evaluation parameter consists of two parts: the brightness evaluation parameter and the circularity, where the brightness evaluation parameter is calculated based on the mean value of all pixel grayscale values and the brightness equalization parameter.

[0034] In step S16, a roughness evaluation parameter is calculated based on the minimum circumscribed circle, the pixel grayscale value, and the preset weight coefficient.

[0035] In one embodiment, the roughness evaluation parameter of the sub-image is calculated by the following formula: ; Among them, represents the roughness evaluation parameter of the sub-image, represents the minimum value of the distance from the center of the minimum circumscribed circle to the edge line, is the length of the sub-image, , are respectively Weight coefficient Represents the maximum value of the distance from the center of the minimum circumscribed circle to the edge line.

[0036] It should be noted that the roughness evaluation parameter is an evaluation parameter for the brake disc grinding process, which is calculated from the minimum circumscribed circle, the gray value of the pixel points, and a preset weight coefficient.

[0037] In step S17, according to the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, a roundness evaluation parameter is calculated.

[0038] In one implementation, according to the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, the range of the Euclidean distance is calculated; the range of the Euclidean distance is used as the roundness evaluation parameter.

[0039] It should be noted that the roundness evaluation parameter is an evaluation parameter for the brake disc grinding process, which is calculated from the range of the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle.

[0040] In step S18, according to the quality evaluation parameter, the roughness evaluation parameter, and the roundness evaluation parameter, a process evaluation parameter for the entire brake disc is calculated.

[0041] In one implementation, the process evaluation parameter of the sub-image can be calculated by the following formula: ; Where Represents the process evaluation parameter of the i-th sub-image, Represents the quality evaluation parameter of the i-th sub-image, Represents the roughness evaluation parameter of the i-th sub-image, Represents the roundness evaluation parameter of the i-th sub-image, , , Respectively represent the weights of the three parameters; represents the process evaluation parameter of the entire brake disc, represents the proportion of the sub-image, and N represents the number of sub-images.

[0042] It should be noted that the process evaluation parameter is a comprehensive evaluation parameter for the brake disc grinding process, which is composed of the quality evaluation parameter, the roughness evaluation parameter, and the roundness evaluation parameter. This process evaluation parameter can improve the accuracy and efficiency of the brake disc grinding process detection.

[0043] The working process of the present invention is described below by taking a relatively common scenario as an example. In an automotive parts manufacturing factory, the quality of brake discs is directly related to the safety performance of vehicles. Therefore, strict inspection of the production process of brake discs is an essential step. The factory has introduced a brake disc grinding process inspection system based on machine vision to improve inspection accuracy and efficiency. On the production line, the ground brake discs enter the inspection area through a conveyor belt. The image acquisition module is activated, and a high-definition industrial camera is used to capture the front of each brake disc, generating high-resolution raw image data. The system processes the acquired raw image data using the region division module: eliminating noise through the mean filter method to enhance image clarity; using the Otsu threshold segmentation method to divide the image into the brake disc region and the background region; dividing the brake disc image into 100 sub-images of the same size, with each sub-image corresponding to 1 / 100 of the area of the brake disc surface. The edge detection module of the system performs edge detection on each sub-image: converting the sub-image into a grayscale image and performing grayscale equalization processing on the grayscale image; using the Sobel operator to extract the edge information of the sub-image and removing the noise in the detection; applying the Hough transform method to detect the circular contour formed by the edges, drawing the minimum circumscribed circle of each sub-image, and calculating the relevant data of the edge contour points and the minimum circumscribed circle. The circularity calculation module of the system calculates the Euclidean dispersion coefficient based on the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle, and then evaluates the circularity of the sub-image. The quality evaluation module of the system extracts the pixel grayscale values of each sub-image, calculates the mean, maximum, minimum, and brightness evaluation parameters of the grayscale values, and combines with the circularity to calculate the quality evaluation parameters. The roughness evaluation module of the system comprehensively calculates the roughness evaluation parameters by combining the minimum circumscribed circle, grayscale values, and weight coefficients. The roundness evaluation module calculates the roundness evaluation parameters through the range of the Euclidean distance from the edge contour points to the center of the minimum circumscribed circle. The process evaluation module performs weighted calculation on the quality evaluation parameters, roughness evaluation parameters, and roundness evaluation parameters of each sub-image to obtain the process evaluation parameters of the sub-image. Finally, the process evaluation parameters of all sub-images are summarized to calculate the process evaluation parameters of the entire brake disc. The system compares the process evaluation parameters of the brake disc with a preset standard threshold: if the evaluation parameter is greater than the threshold, it is determined that the grinding process of the brake disc is qualified and allowed to enter the next process; if the evaluation parameter is lower than the threshold, the brake disc is marked as a defective product and sent to the rework area for reprocessing. Through this system, the accuracy of brake disc inspection is significantly improved, and it can effectively identify quality problems caused by uneven grinding, excessive roughness, or poor roundness. At the same time, the machine vision inspection process is automated, reducing the subjective errors and labor costs of manual inspection, and significantly improving the production efficiency and product quality of the factory.

[0044] Please also refer to Figure 2 , which is Figure 1Schematic diagram of the working scenario of the method.

[0045] In summary, the present invention discloses a method for detecting the grinding process of brake discs based on machine vision, including obtaining the original image data of the ground brake discs; performing region division according to the original image data to obtain multiple sub-images; performing edge detection according to the multiple sub-images to obtain edge contour points and the minimum circumscribed circle; calculating the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle, and determining the Euclidean dispersion degree coefficient according to the Euclidean distance, and determining the circularity of the sub-image based on the Euclidean dispersion degree coefficient; extracting the gray values of pixel points from the multiple sub-images, calculating the brightness evaluation parameter according to the gray values of pixel points, and calculating the quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image; calculating the roughness evaluation parameter according to the minimum circumscribed circle, the gray values of pixel points and a preset weight coefficient; calculating the roundness evaluation parameter according to the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle; calculating according to the quality evaluation parameter, the roughness evaluation parameter and the roundness evaluation parameter to obtain the process evaluation parameter of the entire brake disc; when the process evaluation parameter is greater than a preset standard threshold, it is determined that the detection of the brake disc grinding process is qualified. By using machine vision technology and various evaluation parameters instead of the old methods such as manual visual inspection, the present invention realizes the improvement of the detection accuracy of the brake disc grinding process.

[0046] Referring to Figure 2 , the second embodiment of the present invention provides a system for detecting the grinding process of brake discs based on machine vision, including: An image acquisition module for acquiring the image data of the ground brake discs; A region division module that performs region division according to the image data of the ground brake discs to obtain multiple sub-images, including: performing region division according to the original image data to obtain multiple sub-images.

[0047] In one implementation, according to the original image data of the ground brake discs, each image is denoised using the mean filter method; each image is segmented using the Otsu threshold segmentation method, and the denoised image is segmented into a brake disc region and a background; the image is divided into multiple continuous non-overlapping sub-images of the same size, where the area of the sub-image is 1 / 100 of the surface area of the brake disc, to obtain multiple sub-images; An edge detection module, which is used to perform edge detection based on the multiple sub-images to obtain the edge contour points and the minimum circumscribed circle of the sub-images, including: converting the pixel points of the sub-images into grayscale values to form a grayscale image according to the multiple sub-images, and performing grayscale equalization on each grayscale image; using the sobel operator to perform edge detection on the grayscale image and denoising the edge detection result; performing circle detection on the denoised edge detection result using the Hough transform method and drawing to obtain the minimum circumscribed circle; obtaining the edge contour of the sub-image, taking the center of the minimum circumscribed circle as the coordinate origin, drawing a contour histogram, and obtaining the edge contour points of the sub-image; A circularity calculation module, which calculates the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, determines the Euclidean dispersion degree coefficient according to the Euclidean distance, and determines the circularity of the sub-image based on the Euclidean dispersion degree coefficient, including: calculating the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle; obtaining the mean and standard deviation of all the Euclidean distances, calculating the ratio between the standard deviation and the mean, and taking the ratio as the Euclidean dispersion degree coefficient of all the edge contour points of the sub-image to the center of the minimum circumscribed circle; A quality evaluation module, which extracts the pixel point grayscale values of the multiple sub-images, calculates the brightness evaluation parameter according to the pixel point grayscale values, and calculates the quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image, including: extracting the pixel point grayscale values of the multiple sub-images, calculating the mean of all the pixel point grayscale values, the maximum value of all the pixel point grayscale values, and the minimum value of all the pixel point grayscale values; calculating the mean of all the pixel point grayscale values according to the maximum value and the minimum value of all the pixel point grayscale values, and constructing the brightness equalization parameter with the mean as the initial value; calculating the brightness evaluation parameter according to the mean of all the pixel point grayscale values and the brightness equalization parameter; calculating the quality evaluation parameter of the sub-image according to the brightness evaluation parameter and the circularity of the sub-image, in combination with a preset weight coefficient; Among them, the mean of all the pixel point grayscale values, the brightness evaluation parameter, and the quality evaluation parameter of the sub-image are calculated through the following formulas: ; ; ; In the formula, represents the mean of all the pixel point grayscale values, represents the grayscale value of the sub-image, represents the coordinates of the pixel point, represents the total number of pixel points in the sub-image, represents the area of the sub-image; represents the quality evaluation parameter of the sub-image, represents the circularity of the sub-image, and represents a preset weight coefficient, represents a brightness evaluation parameter, represents a brightness equalization parameter; The roughness evaluation module calculates a roughness evaluation parameter according to the minimum circumscribed circle, the gray value of the pixel point, and the preset weight coefficient; The roundness evaluation module calculates a roundness evaluation parameter according to the Euclidean distance from each of the edge contour points to the center of the minimum circumscribed circle, including: calculating the roughness evaluation parameter of the sub-image through the following formula: ; wherein, represents the roughness evaluation parameter of the sub-image, represents the minimum value of the distance from the center of the minimum circumscribed circle to the edge line, is the length of the sub-image, , are respectively weight coefficients, represents the maximum value of the distance from the center of the minimum circumscribed circle to the edge line; The process evaluation module calculates to obtain the process evaluation parameter of the entire brake disc according to the quality evaluation parameter, the roughness evaluation parameter, and the roundness evaluation parameter, including: calculating the range of the Euclidean distances according to the Euclidean distances from each of the edge contour points to the center of the minimum circumscribed circle; taking the range of the Euclidean distances as the roundness evaluation parameter.

[0048] It should be noted that a brake disc grinding process detection system based on machine vision provided in an embodiment of the present invention is used to execute all the process steps of a brake disc grinding process detection method based on machine vision in the above embodiment. The working principles and beneficial effects of the two correspond one by one, so they will not be elaborated here.

[0049] An embodiment of the present invention also provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as an image acquisition program. When the processor executes the computer program, the steps in each of the above embodiments of the brake disc grinding process detection method based on machine vision are implemented, such as Figure 1 the step S11 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in each of the above system embodiments are implemented, such as the image acquisition module.

[0050] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the electronic device.

[0051] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than those described above, or combine certain components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0052] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and circuits.

[0053] The memory can be used to store the computer programs and / or modules. By running or executing the computer programs and / or modules stored in the memory and invoking the data stored in the memory, the processor can implement various functions of the electronic device. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store the operating system, application programs required for at least one function (such as the sound playback function, the image playback function, etc.); the data storage area can store the data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0054] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present invention, it can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or system, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0055] It should be noted that the system embodiments described above are merely illustrative. 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 to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the system embodiments provided by the present invention, the connection relationships between the modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0056] In the specific embodiments described above, the purpose, technical solution and beneficial effects of the present invention have been further described in detail. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A brake disc grinding process detection method based on machine vision, characterized in that: Executed by a computer, including: Obtaining the original image data of the brake disc after grinding; Performing region division according to the original image data to obtain a plurality of sub-images; Perform edge detection according to the multiple sub-images to obtain edge contour points and a minimum circumscribed circle; Calculating the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle, determining the Euclidean dispersion coefficient according to the Euclidean distance, and determining the circularity of the sub-image based on the Euclidean dispersion coefficient; Extracting pixel grayscale values ​​from the multiple sub-images, calculating brightness evaluation parameters according to the pixel grayscale values, and calculating quality evaluation parameters of the sub-images according to the brightness evaluation parameters and the circularity of the sub-images; The roughness evaluation parameter is calculated according to the minimum circumscribed circle, the gray value of the pixel point and a preset weight coefficient; Calculating a roundness evaluation parameter according to the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle; Calculate according to the quality assessment parameter, the roughness assessment parameter and the roundness assessment parameter to obtain the process assessment parameter of the entire brake disc; When the process evaluation parameter is greater than a preset standard threshold, it is determined that the brake disc grinding process test is qualified.

2. The brake disc grinding process detection method based on machine vision according to claim 1 is characterized in that: The original image data of the brake disc after grinding is divided into regions to obtain multiple sub-images, including: According to the original image data of the brake disc after grinding, each image is subjected to noise reduction processing by using a mean filter method; Each image is segmented using the Otsu threshold segmentation method, and the denoised image is segmented into the brake disc area and the background; The image is divided into a plurality of continuous non-overlapping sub-images of the same size, wherein the area of ​​the sub-image is 1 / 100 of the surface area of ​​the brake disc, to obtain a plurality of sub-images.

3. The brake disc grinding process detection method based on machine vision according to claim 1 is characterized in that: The plurality of sub-images are subjected to edge detection to obtain edge contour points and minimum circumscribed circles of the sub-images, including: According to the multiple sub-images, pixels of the sub-images are converted into grayscale values ​​to form grayscale images, and grayscale equalization is performed on each grayscale image; Use the Sobel operator to perform edge detection on the grayscale image and denoise the edge detection results; According to the edge detection results after denoising, circle detection is performed using the Hough transform method, and the minimum circumscribed circle is obtained by drawing. Get the edge contour of the sub-image, take the center of the minimum circumscribed circle as the coordinate origin, draw the contour histogram, and get the edge contour points of the sub-image.

4. The brake disc grinding process detection method based on machine vision according to claim 1 is characterized in that: The calculating of the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle and determining the Euclidean dispersion coefficient according to the Euclidean distance comprises: Calculating the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle; The mean and standard deviation of all the Euclidean distances are obtained, the ratio between the standard deviation and the mean is calculated, and the ratio is used as the Euclidean dispersion coefficient from all edge contour points of the sub-image to the center of the minimum circumscribed circle.

5. The brake disc grinding process detection method based on machine vision according to claim 1 is characterized in that: Extracting pixel grayscale values ​​from the multiple sub-images, calculating brightness evaluation parameters according to the pixel grayscale values, and calculating quality evaluation parameters of the sub-images according to the brightness evaluation parameters and the circularity of the sub-images, including: Extracting pixel grayscale values ​​from the multiple sub-images, and calculating the average of all pixel grayscale values, the maximum of all pixel grayscale values, and the minimum of all pixel grayscale values; The average grayscale value of all pixels is calculated based on the maximum and minimum grayscale values ​​of all pixels, and is used as the initial value to construct the brightness balance parameter; The brightness evaluation parameter is calculated based on the mean of the grayscale values ​​of all pixels and the brightness balance parameter; The quality assessment parameter of the sub-image is calculated based on the brightness assessment parameter and the circularity of the sub-image in combination with a preset weight coefficient; The mean value of the grayscale values ​​of all pixels, the brightness evaluation parameter, and the quality evaluation parameter of the sub-image are calculated using the following formula: ； ； ； In the formula, Represents the mean gray value of all pixels. represents the gray value of the sub-image, Represents the coordinates of the pixel point. Represents the total number of pixels in the sub-image, represents the region of the sub-image; represents the quality assessment parameter of the sub-image, represents the circularity of the sub-image, and Represents the preset weight coefficient, represents the brightness evaluation parameter, Indicates brightness balance parameters.

6. The brake disc grinding process detection method based on machine vision according to claim 1 is characterized in that: The roughness evaluation parameter is calculated according to the minimum circumscribed circle, the gray value of the pixel point and a preset weight coefficient, including: The roughness evaluation parameter of the sub-image is calculated by the following formula: ； in, represents the roughness evaluation parameter of the sub-image, Indicates the minimum value of the distance from the center of the minimum circumscribed circle to the edge line. is the length of the sub-image, , They are Weight coefficient, Indicates the maximum value of the distance from the center of the minimum circumscribed circle to the edge line.

7. The brake disc grinding process detection method based on machine vision according to claim 1 is characterized in that: The roundness evaluation parameters are calculated based on the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle, including: Calculate the range of the Euclidean distance according to the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle; The range of the Euclidean distance is used as a circularity evaluation parameter.

8. The brake disc grinding process detection method based on machine vision according to claim 1 is characterized in that: The roughness evaluation parameter of the sub-image and the roundness evaluation parameter of the sub-image are used to obtain the process evaluation parameters of the entire brake disc through the process evaluation parameter calculation formula, including: The process evaluation parameters of the sub-image can be calculated by the following formula: ； in, represents the process evaluation parameter of the i-th sub-image, represents the quality assessment parameter of the i-th sub-image, represents the roughness evaluation parameter of the i-th sub-image, represents the roundness evaluation parameter of the i-th sub-image, , , They represent the weights of the three parameters respectively; they represent the process evaluation parameters of the entire brake disc, they represent the proportion of the sub-image, and N represents the number of sub-images.

9. A brake disc grinding process detection system based on machine vision, characterized in that: include: An image acquisition module, used to acquire image data of the brake disc after grinding; A region division module, performing region division according to the image data of the polished brake disc to obtain a plurality of sub-images; An edge detection module, used to perform edge detection based on the multiple sub-images to obtain sub-image edge contour points and minimum circumscribed circles; a circularity calculation module, which calculates the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle, determines the Euclidean dispersion coefficient according to the Euclidean distance, and determines the circularity of the sub-image based on the Euclidean dispersion coefficient; A quality assessment module extracts pixel grayscale values ​​from the multiple sub-images, calculates brightness assessment parameters according to the pixel grayscale values, and calculates quality assessment parameters of the sub-images according to the brightness assessment parameters and the circularity of the sub-images; A roughness evaluation module, which calculates a roughness evaluation parameter according to the minimum circumscribed circle, the gray value of the pixel point and a preset weight coefficient; A roundness evaluation module, which calculates a roundness evaluation parameter according to the Euclidean distance from each edge contour point to the center of the minimum circumscribed circle; The process evaluation module calculates according to the quality evaluation parameter, the roughness evaluation parameter and the roundness evaluation parameter to obtain the process evaluation parameter of the entire brake disc.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the brake disc grinding process detection method based on machine vision as described in any one of claims 1 to 8.

Images