CN115908362A - Method for detecting wear resistance of skateboard wheel - Google Patents
Method for detecting wear resistance of skateboard wheel Download PDFInfo
- Publication number
- CN115908362A CN115908362A CN202211566209.5A CN202211566209A CN115908362A CN 115908362 A CN115908362 A CN 115908362A CN 202211566209 A CN202211566209 A CN 202211566209A CN 115908362 A CN115908362 A CN 115908362A
- Authority
- CN
- China
- Prior art keywords
- gray
- image
- enhanced
- closed
- gray level
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a method for detecting the wear resistance of a sliding plate wheel, belonging to the technical field of image recognition; the method comprises the following steps: obtaining a gray image of the surface of the pulley; acquiring a gray histogram of the gray image; acquiring an enhanced gray level image corresponding to each first segmentation point; acquiring the definition of a middle closed edge line of the enhanced gray level image; taking the enhanced gray level image corresponding to the maximum quality evaluation value as an optimal gray level image; and acquiring the wear resistance of the surface of the pulley according to the density of the closed regions in the optimal gray image and the uniformity of the areas of all the closed regions in the optimal gray image. According to the method, whether a large defect position exists in the surface of the sliding plate wheel or not is judged through the relationship between the edge definition of the pit hole in the enhanced image and the gray level uniformity of the pixel points in the closed edge, the uniformity of the abrasion degree of the pulley is analyzed, and therefore the abrasion resistance of the pulley is evaluated.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a method for detecting the wear resistance of a sliding plate wheel.
Background
After the pulley production, need carry out the sampling to the pulley that newly obtains and detect, detect the wearability of pulley, detect the pulley through wear-resisting tester to the wearability on pulley surface mainly, through under the time of difference, the roughness on pulley surface detects the wearability of pulley. In the prior art, after images are simply preprocessed in the process of detecting the abrasion degree of the surface, the obtained abrasion degree of the pulley is analyzed according to the number and the size of the pit holes in the surface of the pulley, but due to the fact that the colors of the pit holes and the surface of the pulley are close, the number and the area of the obtained pit holes are not accurate in the process of dividing the images, and the result is inaccurate when the abrasion resistance of the pulley is detected. In order to segment clear pit defects, a person skilled in the art firstly enhances an image, and usually adopts histogram equalization to adjust gray values by using an accumulation function so as to enhance contrast, but in the enhancing process, the gray level of the transformed image is reduced, so that weak edge details in the image disappear, and the pit edge in the pulley surface image is difficult to accurately detect.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides a method for detecting the wear resistance of a sliding plate wheel, which judges whether a larger defect position exists in the surface of the sliding plate wheel or not through the relationship between the edge definition of a pit hole in an enhanced image and the gray level uniformity of pixel points in the closed edge, analyzes the uniformity of the wear degree of a pulley and further realizes the evaluation of the wear resistance of the pulley.
The invention aims to provide a method for detecting the wear resistance of a sliding plate wheel, which comprises the following steps:
acquiring a gray image of the surface of the pulley; acquiring a gray level histogram of the gray level image;
taking each gray value in the gray image as a first segmentation point; taking the gray value with the highest occurrence frequency in the gray histogram as a second segmentation point; carrying out piecewise linear transformation on the gray value in the gray image according to each first segmentation point and each second segmentation point to obtain an enhanced gray image corresponding to each first segmentation point;
performing edge detection on the enhanced gray level image to obtain a closed area surrounded by a closed edge line and a closed edge line in the enhanced gray level image and an end point of a non-closed edge line; acquiring the consistency of the edges in the enhanced gray level image according to the average value of the distances between all adjacent end points in the enhanced gray level image and the number of the end points; acquiring the definition of the middle closed edge line of the enhanced gray level image according to the continuity of the edge in the enhanced gray level image and the contrast of the closed edge line in the enhanced gray level image;
acquiring a quality evaluation value of the enhanced gray image corresponding to each first segmentation point according to the gray uniformity degree of pixel points in all closed regions in the enhanced gray image and the definition of closed edge lines in the enhanced gray image; taking the enhanced gray level image corresponding to the maximum quality evaluation value as an optimal gray level image;
and acquiring the wear resistance of the surface of the pulley according to the density of the closed regions in the optimal gray image and the uniformity of the areas of all the closed regions in the optimal gray image.
In one embodiment, the contrast of the closed edge line in the enhanced gray scale image is obtained according to the following steps:
acquiring the contrast of each closed edge line in the enhanced gray image according to the gray difference between the edge pixel point on each closed edge line in the enhanced gray image and the pixel point in the neighborhood of the edge pixel point;
and taking the average value of the contrast of all the closed edge lines as the contrast of the closed edge lines in the enhanced gray image.
In one embodiment, the density of the occlusion regions in the optimal gray image is obtained according to the following steps:
and taking the entropy value of the distance between the center of each closed region and the center of the adjacent closed region in the optimal gray image as the density degree of the closed region in the optimal gray image.
In one embodiment, the uniformity of the areas of all the closed regions in the optimal gray scale image is obtained according to the following steps:
and taking the variance of the areas of all the closed regions as the uniformity of the areas of all the closed regions in the optimal gray level image.
In an embodiment, the method for obtaining the gray uniformity of the pixels in all the closed regions in the enhanced gray image includes:
and taking the entropy values of the gray values of the pixel points in all closed areas in the enhanced gray image as the gray uniformity degree of the pixel points in all closed areas in the enhanced gray image.
In an embodiment, the enhanced grayscale image corresponding to each first segmentation point is obtained according to the following steps:
performing piecewise linear transformation on the gray value in the gray image according to each first segmentation point and each second segmentation point to obtain a transformed gray value and obtain a gray image corresponding to the transformed gray value;
and taking the gray image corresponding to the converted gray value as an enhanced gray image corresponding to each first segmentation point.
The invention has the beneficial effects that: the method comprises the steps of selecting different segmentation points, enhancing a gray level image of the surface of a pulley based on a piecewise linear variation method, improving the contrast between edge information and a background in the gray level image, obtaining a quality evaluation value of the enhanced gray level image corresponding to each first segmentation point and a second segmentation point based on the contrast between the edge information and the background in the enhanced gray level image and the gray level uniformity degree of pixels in a closed region, and constructing an evaluation on the quality of the gray level image through the characteristics of the edges of pit holes in the enhanced image, so that the first segmentation point corresponding to the highest quality evaluation value and the corresponding enhanced gray level image are obtained, the characteristics in the enhanced image are distributed more obviously, and the accuracy of detecting the wear degree of the pulley is higher in the subsequent wear performance detection process of the surface of the pulley.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic flow chart illustrating the general steps of an embodiment of a method for detecting wear resistance of a skateboard wheel according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
The main purposes of the invention are: by utilizing an image identification technology, different segmentation points perform piecewise linear transformation on the gray value in the obtained image to obtain the transformation, the contrast between the edge information and the background is improved, the edge contrast is improved, and the analysis and judgment are performed according to the edge information in the obtained image to obtain the abrasion degree of the surface of the pulley, so that the detection precision of the abrasion resistance of the pulley is improved.
The present invention is directed to the following scenarios: need detect the wear degree of pulley after pulley industrial production, the wear degree of pulley influences the life of pulley, tests the pulley through testing the wear degree of pulley on the wear detector of scooter, judges the wear degree of pulley through machine vision after the test.
The invention provides a method for detecting the wear resistance of a skateboard wheel, which is shown in figure 1 and comprises the following steps:
s1, obtaining a gray image of the surface of a pulley; acquiring a gray histogram of the gray image;
in this embodiment, a pulley to be tested is mounted on a wear tester to perform a wear test, a camera is mounted on the tester, a surface image of the tested pulley is collected by the camera, and the roughness of the pulley surface is evaluated according to the number of the pit holes and the distribution of the pit holes on the surface image after the test.
A high-resolution camera is arranged in front of a pulley on a testing machine, light sources are arranged behind and above a pulley workpiece, and surface images of the pulley workpiece are collected. And then carrying out gray processing on the surface image to obtain a gray image of the pulley surface, and obtaining a gray histogram of the gray image.
S2, taking each gray value in the gray image as a first segmentation point; taking the gray value with the highest occurrence frequency in the gray histogram as a second segmentation point; carrying out piecewise linear transformation on the gray value in the gray image according to each first segmentation point and each second segmentation point to obtain an enhanced gray image corresponding to each first segmentation point;
when the roughness of the pulley surface is evaluated according to the number of the pits and the distribution of the pits on the gray-scale image of the pulley surface, a clear gray-scale image is required to accurately evaluate the roughness of the pulley surface. For this reason, in the present embodiment, the purpose of enhancing the pulley surface is mainly to make the edges of the pit holes on the obtained pulley surface image clearer, since the canny algorithm detects the edges, weak edges in the image are difficult to detect, and the edges in the obtained image are not clear, image enhancement is required to improve the contrast of the edges.
In the embodiment, the gray scale range in the image is segmented mainly based on the adoption of piecewise linear enhancement, the contrast of different areas, namely the background and the target area, in the image is improved, and the definition of the edge is enhanced, so that the accuracy of pit hole detection is improved. In the enhancing process, a first segmentation point and a second segmentation point of initial enhancement are selected firstly. Selecting a gray value with the highest occurrence frequency in a gray histogram as a value of a second segmentation point by the second segmentation point during enhancement, and screening the influence of the first segmentation point on the image according to different gray values; the minimum gray value in the gray image can be used as the value of the initial first segmentation point, the step length is set to be 1, the initial first segmentation point is traversed until the maximum gray value in the gray image, the gray image is enhanced through different first segmentation points and second segmentation points, and the optimal first segmentation point is selected through characteristic change of the edge in the enhanced gray image.
Specifically, the enhanced gray level image corresponding to each first segmentation point is obtained according to the following steps:
performing piecewise linear transformation on the gray value in the gray image according to each first segmentation point and each second segmentation point to obtain a transformed gray value and obtain a gray image corresponding to the transformed gray value;
and taking the gray image corresponding to the converted gray value as an enhanced gray image corresponding to each first segmentation point.
It should be noted that, it is prior art to enhance an image based on piecewise linear change, and details are not described here.
S3, performing edge detection on the enhanced gray level image to obtain a closed area surrounded by a closed edge line and a closed edge line in the enhanced gray level image and an end point of a non-closed edge line; acquiring the consistency of the edges in the enhanced gray level image according to the average value of the distances between all adjacent end points in the enhanced gray level image and the number of the end points; acquiring the definition of the closed edge line in the enhanced gray level image according to the consistency of the edge in the enhanced gray level image and the contrast of the closed edge line in the enhanced gray level image;
it should be noted that, when the pit holes in the obtained grayscale image are detected, the edge detection of the pit holes is unclear, and a plurality of pit holes are detected as one pit hole, and for a good grayscale image of the pulley surface, the edge gradient is large, the edge continuity is good, and the distribution of pixel points in the closed region is uniform. The definition of the pit hole edge in the image and the uniformity of the pixel points in the segmentation block represent the quality of the current selection threshold, the higher the definition is, the better the segmentation point is selected, the smaller the definition is, and the poorer the segmentation point is selected. The edge detection is carried out on the obtained image through a canny edge detection operator, the obtained edge curve is not clear, the weaker edge in the image is difficult to detect, and the obtained result is inaccurate when the wear resistance analysis is carried out on the distribution and the size of the pit holes, so that the quality of the image is analyzed through the change of the edge in the image after the enhancement corresponding to different first segmentation points, and the method specifically comprises the following steps:
firstly, carrying out edge detection on the enhanced gray level image according to a canny edge detection operator to obtain a closed area surrounded by closed edge lines and closed edge lines in the enhanced gray level image and end points of non-closed edge lines; and judging the consistency of the edge according to the distribution of the distances between the end points. If the number of end points in the resulting imageThe smaller the minimum distance L between the end points, the better the consistency of the edges in the corresponding image. The calculation formula for enhancing the consistency of the edges in the grayscale image is as follows:
in the formula (I), the compound is shown in the specification,representing the consistency of the edges in the enhanced gray level image; />Representing the mean of the distances between all adjacent endpoints in the enhanced gray scale image; />Indicating that a first segmentation point is->The number of end points in the corresponding enhanced gray level image; the mean value calculation formula of the distances between all adjacent end points in the enhanced gray level image is as follows: />In the formula (I), wherein,representing the ^ th or greater in the enhanced gray scale image>Coordinates of edge endpoints; />Representing the first in an enhanced grayscale imageCoordinates of edge endpoints; />Expressed as the total number of all endpoints in the enhanced gray scale image, <' > based on the intensity of the image>Representing the average of the distances between all adjacent endpoints in the enhanced gray scale image.
It should be noted that the smaller the number of end points in the enhanced gray-scale image and the smaller the distance between the end points, the better the continuity of the edge in the obtained enhanced gray-scale image, the better the detection effect of the edge of the pit hole in the enhanced gray-scale image, and the edge in the enhanced gray-scale image after the edge detection by canny is a closed area, and the end points do not exist in the closed area. In addition, the number of the end points in the obtained enhanced gray level image is less, the detection effect of the pit holes in the enhanced gray level image is better, and canny has a poor weak edge detection effect, so that the edge in the obtained enhanced gray level image is disconnected, the end points are generated, and the smaller the distance between the two separated end points is, the better the edge detection effect of the corresponding pit holes is.
Secondly, the contrast of the closed edge line in the enhanced gray level image is obtained according to the following steps: acquiring the contrast of each closed edge line in the enhanced gray image according to the gray difference between the edge pixel point on each closed edge line in the enhanced gray image and the adjacent pixel point; and taking the average value of the contrast of all the closed edge lines as the contrast of the closed edge lines in the enhanced gray image. The calculation formula for enhancing the contrast of each closed edge line in the gray scale image is as follows:
in the formula (I), the compound is shown in the specification,fifth ÷ based on mean represented as closing edges in an enhanced gray scale image>The gray value of each of the edge points,expressed as a fifth ÷ in an enhanced gray scale image>Fifth ^ in 8 neighborhoods of edge points>The gray value of each pixel point is greater or less>Represents the total number of pixel points on a closed edge in an enhanced gray scale image, and->Expressed as the mean value of the difference between all edge pixel points and neighborhood pixel points in the enhanced t-th closed edge region in the gray level image, the difference value->The larger the pixel is, the clearer the corresponding closed edge pixel point is; based on the obtained first segmentation point being->And if the average value of the edge contrast of all closed edges in the corresponding enhanced gray level image is obtained, the average value calculation formula of the contrast of all closed edge lines is as follows: />Wherein is present>Indicating that the first segmentation point is->The mean value of the contrast of all closed edge lines in the time-corresponding enhanced gray image is taken as the first segmentation point which is greater or less than>The contrast of the closed edge line in the corresponding enhanced gray level image is obtained; t denotes the number of closed edges.
It should be noted that the larger the mean value of the difference between all edge pixel points and neighborhood pixel points in the t-th closed edge region in the computed enhanced gray scale image is, the larger the contrast of the corresponding closed edge is, and thus the clearer the corresponding closed edge is.
And finally, acquiring the definition of the closed edge line in the enhanced gray level image according to the continuity of the edge in the enhanced gray level image and the contrast of the closed edge line in the enhanced gray level image, wherein the definition calculation formula of the closed edge line in the enhanced gray level image is as follows:
in the formula (I), the compound is shown in the specification,representing the consistency of the edge in the enhanced gray level image; />Indicating that a first segmentation point is->The contrast of the closed edge line in the corresponding enhanced gray level image is obtained; />Indicating that a first segmentation point is->And the definition of the closed edge line in the gray level image is correspondingly enhanced.
It should be noted that the smaller the distance between the edge pixel points in the enhanced gray image obtained by calculation is, the larger the difference between the edge pixel point and the neighboring pixel point is, and the clearer the obtained edge is. When the first segmentation point isCorresponding coherency of an edge in an enhanced gray scale image>The greater the value of (a) and the difference between the edge and the neighborhood->The larger the value of (a), the better the segmentation edge in the resulting enhanced grayscale image.
S4, acquiring a quality evaluation value of the enhanced gray image corresponding to each first segmentation point according to the gray uniformity degree of pixel points in all closed regions in the enhanced gray image and the definition of closed edge lines in the enhanced gray image; taking the enhanced gray level image corresponding to the maximum quality evaluation value as an optimal gray level image;
the method for obtaining the gray uniformity degree of the pixel points in all closed areas in the enhanced gray image comprises the following steps:
and taking the entropy values of the gray values of the pixels in all closed regions in the enhanced gray image as the gray uniformity degree of the pixels in all closed regions in the enhanced gray image. Then, the calculation formula for enhancing the gray uniformity of the pixel points in all the closed regions in the gray image is as follows:
in the formula (I), the compound is shown in the specification,expressing the frequency of the occurrence of the gray values of the ith pixel points in all closed areas in the enhanced gray image; represents a base 2 logarithmic function; />Indicating that a first segmentation point is->And entropy values of the gray values of the pixels in all closed regions in the corresponding enhanced gray image are used as the gray uniformity degrees of the pixels in all closed regions in the enhanced gray image. It should be noted that, the more uniform the distribution of the gray values of the pixels in the calculated closed region is, the greater the probability that the closed region is a pit hole is; the entropy of the calculated gray value is determined if the gray value distribution in the closed region is more or less homogeneous>The larger the value of (a) is, the more the closed region is, the more the pits are present. The reason for this is the edge of the pit holeThe gradient is small, and the contrast of the edge and other areas in the gray image can be effectively enhanced by selecting different first segmentation points to judge and analyze the poor weak edge detection effect in the image in the canny edge detection process. Therefore, the entropy value of the gray value of the pixel point in all the closed areas in the gray image is enhanced>The smaller the image, the better the effect after edge detection of the image.
It should be noted that, the more uniform the pixel value distribution of the pixel points in all the closed regions, the better the segmentation effect on the closed regions in the corresponding enhanced gray scale image, and the better the contrast between the segmentation blocks in the enhanced gray scale image, therefore, the first segmentation point is defined asAnd evaluating the image, wherein the higher the evaluation value according to the quality of the obtained image is, the better the effect of the corresponding selected segmentation point is.
The weak edge is not detected in the enhanced gray scale image, so that the distribution of pixel points in a closed region in the obtained enhanced gray scale image is not uniform, the judgment of the area of a pit hole in the enhanced gray scale image is influenced, the quality of the image needs to be evaluated according to the definition of the edge in the enhanced gray scale image and whether the distribution of the pixel points in the closed region is uniform, the definition of the edge in the image is mainly influenced, and therefore the definition of the edge occupies more area.
Therefore, in this embodiment, a quality evaluation value of the enhanced gray image corresponding to each first segmentation point is obtained according to the gray uniformity degree of pixel points in all closed regions in the enhanced gray image and the definition of a closed edge line in the enhanced gray image; then, the calculation formula of the quality evaluation value of each first segmentation point corresponding to the enhanced grayscale image is as follows:
in the formula (I), the compound is shown in the specification,indicating that the first segmentation point is->A quality evaluation value corresponding to the enhanced gray level image; />Indicating that a first segmentation point is->Entropy values of gray values of pixel points in all closed regions in the corresponding enhanced gray image are gray uniformity degrees of the pixel points in all closed regions in the enhanced gray image; />Indicating that the first segmentation point is->The definition of a closed edge line in the gray level image is correspondingly enhanced; />、/>Indicating that the first segmentation point is->The weight ratio of the gray uniformity degree and the definition in the corresponding enhanced gray image is calculated; in the present embodiment, is>,/>The proportion of each part can be adjusted according to the requirement; it should be noted that the advantage of evaluating the image according to the obtained definition of the closed edge line in the enhanced gray image and the gray uniformity of the pixels in all the closed regions in the enhanced gray image is that the more uniform the gray distribution of the pixels in all the closed regions is, the better the segmentation effect on the closed regions in the corresponding enhanced gray image is, and the better the contrast between the segmented blocks in the enhanced gray image is; due to the fact that the entropy value of the gray value of the pixel point in all the closed regions in the gray image is enhanced>The smaller the image, the better the effect after edge detection of the image; the first segmentation point is->The larger the definition value of the closed edge line in the corresponding enhanced gray level image is, the better the segmentation edge in the enhanced gray level image is; the image is evaluated according to the gray level uniformity degree and the definition of the closed edge line, so that the quality of the image can be better evaluated, and the higher the evaluation value is, the clearer the edge information of the image is; in addition, what primarily affects the image quality is the sharpness of the edge in the image, so the sharpness of the edge is set ≦ ≦ for>The ratio of (A) to (B); and an entropy setting of the gray value->The ratio of the gray scale to the gray scale is calculated, and the quality evaluation of the enhanced gray scale image is more accurate under different ratios according to the definition of the closed edge line in the enhanced gray scale image and the gray scale uniformity degree of the pixel points in all closed areas in the enhanced gray scale image.
Sequentially calculating and obtaining quality evaluation values of the enhanced gray level images corresponding to different first segmentation points, and taking the enhanced gray level image corresponding to the maximum quality evaluation value as an optimal gray level image; and taking the first segmentation point corresponding to the maximum quality evaluation value as the optimal first segmentation point.
And performing piecewise linear transformation on the gray value in the gray image through the calculated optimal first segmentation point and the second segmentation point to obtain an enhanced gray image serving as the optimal gray image. Therefore, the wear resistance of the pulley is detected according to the optimal gray image, and the accuracy of the subsequent analysis of the wear degree condition of the pulley surface is improved.
And S5, acquiring the wear resistance of the surface of the pulley according to the density of the closed regions in the optimal gray image and the uniformity of the areas of all the closed regions in the optimal gray image.
The abrasion resistance of the surface of the pulley is evaluated according to the area and the density of the closed area in the optimal gray level image, when the abrasion resistance of the pulley is good, after the surface of the pulley is tested by an abrasion resistance tester, the abrasion resistance of the pulley is better, the area size of the pit holes in the obtained optimal gray level image is uniform, the obtained pit holes are densely distributed, and the abrasion resistance of the pulley is better; the obtained optimal gray image has large single pit hole area and uneven distribution among the segmentation blocks, which indicates that the abrasion resistance of the surface of the pulley is poor, and the abrasion pits on the surface of the pulley have uneven size and uneven distribution position in the test process.
Specifically, the density of the closed region in the optimal gray image is obtained according to the following steps:
and taking the entropy value of the distance between the center of each closed region and the center of the adjacent closed region in the optimal gray image as the density degree of the closed region in the optimal gray image.
In this embodiment, the distance between the center points of the closed edge regions in the optimal gray image is obtained by performing canny algorithm edge detection on the optimal gray image, so as to obtain the distance between the center points of the adjacent closed regionsThe minimum of the center distances of all adjacent closing regions is found->And a maximum value->And counts the frequency of occurrence of different distances>(ii) a Obtaining the entropy of the coordinates according to the frequency of the central point coordinate position of the closed region in the optimal gray level image to represent the density degree of the closed region in the optimal gray level image: the density calculation formula of the closed region in the optimal gray image is as follows:
in the formula (I), the compound is shown in the specification,means that the distance between the center of each closed area and the center of its neighboring closed area is->The frequency of occurrence of time; />、/>Respectively representing the minimum and maximum of the distances between the centers of all adjacent closed regions; counting the distance between the central points of adjacent segmentation blocks in the optimal gray level image to obtain the occurrence frequency; />An entropy value representing the distance between the center of each closed region and the center of the adjacent closed region in the optimal gray level image is the density degree of the closed region in the optimal gray level image, and the value is based on the intensity value of the closed region in the optimal gray level image>Greater values of (a) indicate a more diffuse distribution between pit holes in the image, greater>The smaller the value of (A), the denser the distribution of pit holes.
Further, the uniformity of the areas of all closed regions in the optimal gray-scale image is obtained according to the following steps:
and taking the variance of the areas of all closed regions as the uniformity of the areas of all closed regions in the optimal gray-scale image. Then the equation for calculating the uniformity of the areas of all closed regions in the optimal gray scale image is as follows:
in the formula (I), the compound is shown in the specification,expressed as the ^ th or greater in the optimal gray scale image>Area of a closure zone>Expressed as the mean of the area of the closed region in the optimal gray image, <' >>Expressed as the total number of closed regions in the optimal gray scale image, <' > based on the gray scale image>Expressing the uniformity of the areas of all closed regions in the optimal gray level image; the calculation of the homogeneity of the areas of all closing regions in the optimal gray-scale image is based primarily on the degree of homogeneity of the distribution of the closing regions, where->The smaller the value of (c) is, the more uniform.
Further, the wear resistance of the surface of the pulley is obtained according to the density of the closed regions in the optimal gray image and the uniformity of the areas of all the closed regions in the optimal gray image. The wear resistance calculation formula of the surface of the pulley is as follows:
in the formula (I), the compound is shown in the specification,expressing the entropy value of the distance between the center of each closed region and the center of the adjacent closed region in the optimal gray level image, namely the density of the closed regions in the optimal gray level image; />Expressing the uniformity of the areas of all closed regions in the optimal gray level image; />Expressed as the degree of wear resistance of the sheave surface; the more poor the wear resistance corresponds to the larger the pits generated on the surface of the tested pulley, the denser the distribution of the pits indicates that the wear resistance of the surface of the pulley is more uniform, the larger the area of the enhanced divided block is, the more discrete the distribution is, the less the wear resistance of the surface of the corresponding pulley is, the smaller the area of the divided block is, the more uniform the distribution is, and the better the wear resistance of the pulley is.
The embodiment can evaluate the abrasion performance of the surface of the pulley according to the area and the density of the closed area in the optimal gray level image, when the abrasion resistance of the pulley is good, after the surface of the pulley is tested by an abrasion resistance tester, the abrasion resistance of the pulley is better, the area size of the pit holes in the obtained optimal gray level image is uniform, the distribution of the obtained pit holes is dense, and the abrasion resistance of the pulley is better; the obtained optimal gray level image has large single pit hole area and uneven distribution among the segmentation blocks, which indicates that the wear resistance of the surface of the pulley is poor, and the wear pits on the surface of the pulley have uneven size and uneven distribution positions in the test process. Therefore, the operator can judge the wear resistance of the pulley according to the wear resistance of the surface of the pulley.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions, improvements, etc. within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A method for detecting the wear resistance of a skateboard wheel is characterized by comprising the following steps:
acquiring a gray image of the surface of the pulley; acquiring a gray level histogram of the gray level image;
taking each gray value in the gray image as a first segmentation point; taking the gray value with the highest occurrence frequency in the gray histogram as a second segmentation point; carrying out piecewise linear transformation on the gray value in the gray image according to each first segmentation point and each second segmentation point to obtain an enhanced gray image corresponding to each first segmentation point;
performing edge detection on the enhanced gray level image to obtain a closed area surrounded by a closed edge line and a closed edge line in the enhanced gray level image and an end point of a non-closed edge line; obtaining the consistency of the edges in the enhanced gray level image according to the average value of the distances between all adjacent end points in the enhanced gray level image and the number of the end points; acquiring the definition of the closed edge line in the enhanced gray level image according to the consistency of the edge in the enhanced gray level image and the contrast of the closed edge line in the enhanced gray level image;
acquiring a quality evaluation value of the enhanced gray image corresponding to each first segmentation point according to the gray uniformity degree of pixel points in all closed regions in the enhanced gray image and the definition of closed edge lines in the enhanced gray image; taking the enhanced gray level image corresponding to the maximum quality evaluation value as an optimal gray level image;
and acquiring the wear resistance of the surface of the pulley according to the density of the closed regions in the optimal gray image and the uniformity of the areas of all the closed regions in the optimal gray image.
2. The method for detecting the wear resistance of the skateboard wheel according to claim 1, wherein the contrast of the closed edge line in the enhanced gray image is obtained by the following steps:
acquiring the contrast of each closed edge line in the enhanced gray image according to the gray difference between the edge pixel point on each closed edge line in the enhanced gray image and the adjacent pixel point;
and taking the average value of the contrast of all the closed edge lines as the contrast of the closed edge lines in the enhanced gray image.
3. The method for detecting the wear resistance of the skateboard wheel according to claim 1, wherein the density of the closed area in the optimal gray image is obtained by:
and taking the entropy value of the distance between the center of each closed region and the center of the adjacent closed region in the optimal gray level image as the density degree of the closed regions in the optimal gray level image.
4. The method for detecting the wear resistance of the skateboard wheel according to claim 1, wherein the uniformity of the areas of all the closed areas in the optimal gray image is obtained by the following steps:
and taking the variance of the areas of all closed regions as the uniformity of the areas of all closed regions in the optimal gray-scale image.
5. The method for detecting the wear resistance of the skateboard wheel according to claim 1, wherein the method for obtaining the gray uniformity of the pixel points in all the closed areas in the enhanced gray image comprises:
and taking the entropy values of the gray values of the pixel points in all closed areas in the enhanced gray image as the gray uniformity degree of the pixel points in all closed areas in the enhanced gray image.
6. The method for detecting wear resistance of a skateboard wheel according to claim 1, wherein the enhanced grayscale image corresponding to each first segmentation point is obtained according to the following steps:
performing piecewise linear transformation on the gray value in the gray image according to each first segmentation point and each second segmentation point to obtain a transformed gray value and obtain a gray image corresponding to the transformed gray value;
and taking the gray image corresponding to the converted gray value as an enhanced gray image corresponding to each first segmentation point.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211566209.5A CN115908362A (en) | 2022-12-07 | 2022-12-07 | Method for detecting wear resistance of skateboard wheel |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211566209.5A CN115908362A (en) | 2022-12-07 | 2022-12-07 | Method for detecting wear resistance of skateboard wheel |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115908362A true CN115908362A (en) | 2023-04-04 |
Family
ID=86495326
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211566209.5A Pending CN115908362A (en) | 2022-12-07 | 2022-12-07 | Method for detecting wear resistance of skateboard wheel |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115908362A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116912254A (en) * | 2023-09-14 | 2023-10-20 | 山东博诚电气有限公司 | Cable defect identification method based on data enhancement preprocessing |
CN117495862A (en) * | 2024-01-03 | 2024-02-02 | 深圳家红齿科技术有限公司 | Denture wearability detection device |
CN117953434A (en) * | 2024-03-27 | 2024-04-30 | 广州煜能电气有限公司 | Intelligent gateway-based method and system for monitoring external damage of power transmission line |
-
2022
- 2022-12-07 CN CN202211566209.5A patent/CN115908362A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116912254A (en) * | 2023-09-14 | 2023-10-20 | 山东博诚电气有限公司 | Cable defect identification method based on data enhancement preprocessing |
CN116912254B (en) * | 2023-09-14 | 2023-12-08 | 山东博诚电气有限公司 | Cable defect identification method based on data enhancement preprocessing |
CN117495862A (en) * | 2024-01-03 | 2024-02-02 | 深圳家红齿科技术有限公司 | Denture wearability detection device |
CN117495862B (en) * | 2024-01-03 | 2024-03-12 | 深圳家红齿科技术有限公司 | Denture wearability detection device |
CN117953434A (en) * | 2024-03-27 | 2024-04-30 | 广州煜能电气有限公司 | Intelligent gateway-based method and system for monitoring external damage of power transmission line |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115908362A (en) | Method for detecting wear resistance of skateboard wheel | |
CN114937055B (en) | Image self-adaptive segmentation method and system based on artificial intelligence | |
CN115601364B (en) | Golden finger circuit board detection method based on image analysis | |
CN114627111B (en) | Textile defect detection and identification device | |
CN115294113B (en) | Quality detection method for wood veneer | |
CN116843678B (en) | Hard carbon electrode production quality detection method | |
CN114494259B (en) | Cloth defect detection method based on artificial intelligence | |
CN115345885A (en) | Method for detecting appearance quality of metal fitness equipment | |
CN114549522A (en) | Textile quality detection method based on target detection | |
CN115829883A (en) | Surface image denoising method for dissimilar metal structural member | |
CN117197141B (en) | Method for detecting surface defects of automobile parts | |
CN116645367B (en) | Steel plate cutting quality detection method for high-end manufacturing | |
CN113554629A (en) | Strip steel red rust defect detection method based on artificial intelligence | |
CN115100206B (en) | Printing defect identification method for textile with periodic pattern | |
CN115984272B (en) | Semitrailer axle defect identification method based on computer vision | |
CN114926463A (en) | Production quality detection method suitable for chip circuit board | |
CN116758083A (en) | Quick detection method for metal wash basin defects based on computer vision | |
CN116071363B (en) | Automatic change shaped steel intelligent production monitoring system | |
CN117095004B (en) | Excavator walking frame main body welding deformation detection method based on computer vision | |
CN115294159B (en) | Method for dividing corroded area of metal fastener | |
CN115719332A (en) | Welding quality detection method | |
CN114782329A (en) | Bearing defect damage degree evaluation method and system based on image processing | |
CN116523923B (en) | Battery case defect identification method | |
CN115861307B (en) | Fascia gun power supply driving plate welding fault detection method based on artificial intelligence | |
CN116645364B (en) | Alloy steel casting air hole defect detection method based on image data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |