Detection method for abnormal deformation degree of wear scar based on boundary distribution characteristics

Abstract

The detection method of abnormal deformation degree of wear scar based on boundary distribution characteristics provided by the present invention includes the following steps: Step 1, segment the obtained four-ball friction wear scar image to obtain a wear scar segmentation map; Step 2, obtain a boundary pixel map according to the wear scar segmentation map; Step 3, calculate the deformation degree of the wear scar according to the boundary pixel map; Step 4, detect the abnormal deformation of the wear scar according to the obtained deformation degree of the wear scar; The present invention can realize the rapid comparison of the deformation degrees of multiple wear scars and the detection of the abnormal deformation degree of the wear scar. This method has the advantages of simplicity, high efficiency, and no need for manual assistance throughout the process.

Description

Technical Field

[0001] The present invention relates to the technical field of image test data analysis and processing, and particularly to a detection method for the abnormal deformation degree of wear scars based on boundary distribution characteristics. Background Art

[0002] The research results of the consulting project of the Chinese Academy of Engineering, "Research on the Current Situation and Development Strategy of Tribology Science and Engineering Applications", show that: In 2006, the losses caused by friction and wear in China reached about 950 billion yuan, accounting for 4.5% of the gross domestic product (GDP) of that year. In addition, if wear intensifies, it will cause component failure and machine breakdown, and even lead to catastrophic consequences. A lubricant with good performance is a lubricating medium used to reduce the frictional resistance of the friction pair and slow down its wear, and can play roles such as cooling, cleaning, and preventing pollution to the friction pair. Therefore, it is particularly important to timely and accurately measure the performance of lubricants for protecting machinery and reducing energy consumption. Due to characteristics such as convenient operation, simple structure, short test cycle, small oil consumption, and low cost, the four-ball friction and wear tester is widely used in the test for measuring the friction coefficient of lubricating oil.

[0003] The standards of the petrochemical industry in China (GB-T 12583-1998 and H-T 0762-2005) stipulate the test process of the friction coefficient of lubricating oil and clarify the observation method of the wear scar morphology characteristics: The wear scar shape is generally circular or elliptical. When the wear scar deformation is severe, it cannot be used in the test process of the friction coefficient of lubricating oil, and a new running-in test needs to be carried out. Therefore, quickly and accurately determining the shape characteristics of the wear scar is very important for judging the wear scar morphology and test effectiveness. However, at present, there is no research on the automatic detection of the wear scar deformation in the four-ball friction test, and it is still qualitatively judged by the tester based on experience, inevitably resulting in subjective judgment errors, which are neither scientific nor objective and have little practical guiding significance. Summary of the Invention

[0004] The purpose of the present invention is to propose a detection method for the abnormal deformation degree of wear scars based on boundary distribution characteristics, which solves the above-mentioned deficiencies in the prior art.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] The detection method for the abnormal deformation degree of wear scars based on boundary distribution characteristics provided by the present invention includes the following steps:

[0007] Step 1: Segment the obtained four-ball friction wear scar image to obtain a wear scar segmentation map;

[0008] Step 2: Obtain a boundary pixel map according to the wear scar segmentation map;

[0009] Step 3, calculate the deformation degree of the wear scar according to the boundary pixel map;

[0010] Step 4, detect the abnormal deformation of the wear scar according to the obtained deformation degree of the wear scar.

[0011] Preferably, in step 2, the boundary pixel map is obtained according to the wear scar segmentation map. The specific method is:

[0012] Calculate the position size of the wear scar segmentation map;

[0013] According to the position size obtained in step 1, the boundary pixel map is obtained by combining the following formula:

[0014]

[0015] Where A is the wear scar boundary map; A(x, y) = 1 indicates that the pixel (x, y) is a boundary pixel; A(x, y) = 0 indicates that the pixel (x, y) is a non-boundary pixel; k and l respectively represent the index variables of rows and columns, both of which are integers, and the values are -1, 0, and 1 respectively.

[0016] Preferably, in step 3, the deformation degree of the wear scar is calculated according to the boundary pixel map. The specific method is:

[0017] Calculate the boundary smoothness of the boundary pixel map, the full range of deviation of the boundary pixel map, the dispersion characteristic of the deviation of the boundary pixel map, and the number of boundary pixels corresponding to the preset deformation amount of the boundary pixel map respectively;

[0018] Calculate the deformation degree of the wear scar corresponding to the boundary pixel map according to the obtained boundary smoothness of the boundary pixel map, the full range of deviation of the boundary pixel map, the dispersion characteristic of the deviation of the boundary pixel map, and the number of boundary pixels corresponding to the preset deformation amount of the boundary pixel map.

[0019] Preferably, the boundary smoothness of the boundary pixel map is calculated according to the following formula:

[0020]

[0021] Where Z1 represents the smoothness of the wear scar boundary; num(A) is the number of pixels of the boundary pixel map; 2πr is the perimeter of the wear scar segmentation map.

[0022] Preferably, the full range of deviation of the boundary pixel map is calculated. The specific method is:

[0023] Calculate the center distance and deviation of the boundary pixel map;

[0024] Calculate the full range of deviation of the boundary pixel map according to the obtained center distance and deviation of the boundary pixel map.

[0025] Preferably, the dispersion characteristic of the boundary deviation is calculated according to the following formula:

[0026]

[0027]

[0028] Among them, Z3 is the average value of the boundary deviation; Z4 is the standard deviation of the boundary deviation.

[0029] Preferably, the number of boundary pixels corresponding to the preset deformation amount of the boundary pixel map is calculated according to the following formula:

[0030]

[0031] Among them, Z5 is the deformation ratio of the preset deformation amount; P α is the number of boundary pixels when the preset deformation amount is α; num(A) is the total number of boundary pixels; α is the preset deformation amount, and according to expert experience, the value range of α is 0 < α < 1.

[0032] Preferably, the deformation degree of the wear scar is calculated according to the following formula:

[0033]

[0034] Among them, H is the deformation degree of the wear scar; γ k is the weight coefficient of the kth feature.

[0035] Preferably, in step 4, the abnormal deformation degree of the wear scar is detected according to the following formula:

[0036]

[0037] Among them, fl is the mark of abnormal deformation. When fl = 1, it means that the shape of the wear scar has deformed and is non-circular. When fl = 0, it means that the shape of the wear scar has not deformed and is circular; β is the deformation threshold.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] The detection method of abnormal deformation degree of wear scar based on boundary distribution characteristics provided by the present invention uses an image acquisition device or a scanning electron microscope (both with magnification functions) to collect wear scar images. The use of image analysis and processing technology can achieve automatic segmentation of the wear scar area, improving the speed and accuracy of segmentation, and greatly reducing the workload of manual segmentation at the same time. Starting from the boundary pixels of the wear scar, the boundary pixels of the wear scar are defined and the center distance of the boundary pixels is calculated, laying a foundation for the subsequent representation of the distribution characteristics of the boundary pixels. Multiple image features are used to quantitatively describe the quantity and distribution characteristics of the boundary pixels, realizing the establishment and expression of feature indicators such as boundary smoothness and deviation degree. Based on multiple image features, a determination method for the deformation degree of the wear scar is established. The method can achieve rapid comparison of the deformation degrees of multiple wear scars and detection of the abnormal deformation degree of the wear scar. This method has the advantages of simplicity, high efficiency, and no need for manual assistance throughout the process, and is very suitable for developing into automatic analysis and processing software for four-ball friction test data. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is the wear scar image f;

[0041] Figure 2 is the wear scar segmentation map R;

[0042] Figure 3 For (x O , y O ) being (365.2125, 465.3137), the marked map E O ;

[0043] Figure 4 For (x O , y O ) being (465.2125, 485.3137), the marked map E O ;

[0044] Figure 5 is the illustration of the boundary pixel set A. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] The present invention will be further described in detail below with reference to the accompanying drawings.

[0046] The specific implementation steps are as follows:

[0047] Step S0: Acquisition of wear scar images. After the four-ball friction test, wear scars will be formed on the surface of the steel balls. The surface wear scars (≤1 mm) are not visible to the naked eye, and a dedicated acquisition device such as a high-power microscope or a scanning electron microscope is required to collect the wear scar image f, as Figure 1 shown.

[0048] Step S1: Segmentation of the wear scar image. The segmentation of the wear scar image includes manual segmentation and automatic segmentation, and its purpose is to distinguish the wear scar from the background. Compared with the manual segmentation method, the automatic segmentation method has the advantages of high segmentation accuracy and fast speed. The automatic segmentation method specifically performs an initial segmentation of the wear scar by establishing a double-extremum filtering difference metric operator, and uses gray-scale and distance double-constrained boundary refinement to accurately segment the boundary of the wear scar area, including steps such as gray-scale conversion, denoising, initial segmentation of the wear scar area, and double-constrained boundary refinement. The algorithm has a good segmentation effect on the boundary and a fast running speed.

[0049] In this embodiment, the automatic segmentation method is adopted, and the obtained wear scar segmentation map R is as Figure 2 shown. The white and black areas in the figure are the wear scar and the background respectively.

[0050] Step S2: Extraction of the size parameters of the wear scar segmentation map. The segmented wear scar is equivalent to a standard circle, so the size parameters of the wear scar segmentation map can be described by the size parameters of the circle: the position of the center C of the wear scar segmentation map, the area S, and the radius r. The position of the center C of the wear scar segmentation map is represented by the two-dimensional row and column coordinates of the pixel coordinate system as (x C , y C ), where x C and y C are the row number and column number respectively; the area refers to the number of pixels in the wear scar area, denoted by S; the radius refers to the radius value of the wear scar segmentation map, denoted by r, and the expression is as shown in (1-4):

[0051]

[0052]

[0053] S = num(R) (3)

[0054]

[0055] where (x, y) is the coordinate of any pixel in the wear scar segmentation map; M and N are the maximum row number and maximum column number of the wear scar segmentation map R respectively; R(x, y) represents the value of the pixel (x, y) of the wear scar segmentation map R, and R is a binary image; num() is a statistical operator; num(R) is the total number of pixels in the wear scar area of the wear scar segmentation map R.

[0056] In this embodiment, the size parameters of the wear scar segmentation map are:

[0057] x C = 365.2125, y C = 465.3137, S = 396979 and r = 355.48.

[0058] Step S3: Extract the boundary pixels of the wear scar. This invention patent starts from the boundary pixels of the wear scar, considers their quantity and distribution characteristics, etc., and determines the deformation degree of the wear scar. Therefore, the extraction of the wear scar boundary pixels is the foundation. The wear scar boundary pixels are defined as that at least one pixel in the eight-neighborhood of the pixel in the wear scar segmentation map R is a non-wear scar, and its expression is:

[0059]

[0060] where A is the wear scar boundary map, which is a binary map; A(x, y)=1 indicates that the pixel (x, y) is a boundary pixel; A(x, y)=0 indicates that the pixel (x, y) is a non-boundary pixel; k and l respectively represent the index variables of the row and column, both are integers, and their values are -1, 0, 1 respectively.

[0061] In this embodiment, the wear scar boundary map A is as Figure 3 shown.

[0062] Step S4: Smoothness of the wear scar boundary.

[0063] On the basis of extracting the wear scar boundary pixels, establish the indexes and their characteristics related to the wear scar deformation degree. The smoothness of the wear scar boundary refers to the ratio of the perimeter of the wear scar boundary to the perimeter of the standard circle boundary, which is a measurement index parameter of the wear scar shape deformation. The larger its value is, the less smooth the wear scar shape is, that is, the higher the deformation degree of the wear scar:

[0064]

[0065] where Z1 represents the smoothness of the wear scar boundary; num(A) is the number of pixels of the wear scar boundary; 2πr is the perimeter of the standard circle boundary.

[0066] In this embodiment,

[0067] Step S5: Distance from the boundary pixels of the wear scar to the center and deviation degree.

[0068] The distance from the boundary pixels to the center is defined as the distance between the boundary pixels and the center C. Compared with the radius r, when the distance from the boundary pixels to the center is too large or too small, it means that the boundary pixels deviate from their proper positions, which can be used to quantitatively measure the deviation degree of the boundary pixels. The expressions of the distance from the boundary pixels to the center and the deviation degree are as shown in (7-8):

[0069]

[0070]

[0071] where d(x, y) is any boundary pixel( satisfying A(x, y)=1) and the center C(x C , y C) The distance between them; b(x, y) is the deviation degree of any boundary pixel. The larger its value, the farther the pixel is from the boundary. When its value is 0, the pixel is exactly on the boundary of the wear scar segmentation map.

[0072] In this embodiment, for any boundary pixel (6, 550), d(6, 550) = 369.06, and the corresponding b(6, 550) = 3.82%.

[0073] Step S6: The range of the boundary pixel deviation degree.

[0074] The range refers to the difference between the maximum value and the minimum value in the data set, and is a commonly used indicator to measure the dispersion degree of the data set. Similarly, the range of the boundary pixel deviation degree can also be used to describe the width of the deviation degree of the outer boundary of the wear scar. The smaller its value, the smaller the deviation degree of the outer boundary of the wear scar, and vice versa. Usually, the minimum value of the deviation degree of the wear scar boundary pixel is often 0, so it is more reasonable to directly use the maximum value of the deviation degree to represent the range, which is represented by the symbol Z2:

[0075]

[0076] Among them, Z2 represents the range of the boundary pixel deviation degree.

[0077] In this embodiment, the range: Z2 = 0.151554321226492.

[0078] Step S7: The dispersion characteristics of the boundary deviation degree.

[0079] The range (step S6) can generally reflect the dispersion degree of the data, but it is difficult to describe the distribution form of the data. Especially when chips and other noises are misdetected as wear scars, it will cause a large discreteness error in the outer contour described by the range. Therefore, the mean and standard deviation are further established to describe the discreteness (distribution) of the outer contour. The mean and standard deviation can reflect the dispersion degree of the data from an overall perspective. Therefore, the dispersion characteristics of the boundary deviation degree are also measured by two indicators, the mean value and the standard deviation. The smaller the mean value, the smaller the average deviation degree of the outer boundary of the wear scar, and vice versa. The smaller the standard deviation, the more concentrated the deviation degree of the outer boundary of the wear scar is around the mean value, and the larger the standard deviation, the farther the deviation degree of the outer boundary of the wear scar is from the mean value:

[0080]

[0081]

[0082] Among them, Z3 is the mean value of the boundary deviation degree; Z4 is the standard deviation of the boundary deviation degree.

[0083] In this embodiment, the mean value of the boundary deviation degree Z3 = 0.02073; the standard deviation of the boundary deviation degree Z4 = 0.02188.

[0084] Step S8: Calculate the number of boundary pixels of the preset deformation amount.

[0085] Although the full range (Step S6), mean, and standard deviation (Step S7) generally reflect the distribution characteristics of the data, when there are abnormal mutation data, the error of the distribution characteristics reflected by these three index parameters will become larger. To reduce the influence of abnormal mutation data, the preset deformation amount is further used to describe the discreteness (distribution) of the outer contour. The specific process is as follows:

[0086] First, form a one-dimensional data set from the deviation data of the boundary and sort it in ascending order to obtain the sorted deviation data set.

[0087] Second, calculate the number of boundary pixels when the preset deformation amount is α, which is represented by the symbol P α For the same wear scar, on the one hand, the larger the value of α, the more boundary pixels P α included, and vice versa. On the other hand, when the value of α is the same, the larger the P α of the wear scar, the smaller the deformation of the wear scar, which can be used to quantitatively describe the deformation degree of different wear scars.

[0088] Finally, calculate the deformation ratio of the preset deformation amount:

[0089]

[0090] where Z5 is the deformation ratio of the preset deformation amount; P α is the number of boundary pixels when the preset deformation amount is α; num(A) is the total number of boundary pixels; α is the preset deformation amount, and according to expert experience, the value range of α is 0 < α < 1.

[0091] In this embodiment, when the value of α is 0.9, we get

[0092] Step S9: The deformation degree of the wear scar is determined by the shape characteristics of the wear scar boundary.

[0093] According to the H value of the wear scar, the deformation degrees of different wear scars can be quantitatively compared. The larger the H value, the greater the deformation degree of the wear scar:

[0094]

[0095] where H is the deformation degree of the wear scar; γ k is the weight coefficient of the kth feature. In the present invention, 5 feature indexes are established, so k = 1, 2, 3, 4, 5, and it satisfies: Z k is the value of the kth feature parameter.

[0096] In this embodiment, γk (where k = 1, 2, 3, 4, 5), the values are as follows: and obtain:

[0097] Step S10: Detection of the abnormal deformation degree of the wear scar.

[0098] Due to the unified characteristics, it is also possible to judge the wear scar with abnormal deformation as the basis for whether to restart the running-in test. That is, when the deformation degree is not less than the deformation threshold β, the shape of the wear scar formed by the test has deformed, and the test process needs to be re-regulated to carry out the test.

[0099]

[0100] Among them, fl is the mark of abnormal deformation. When fl = 1, it means that the shape of the wear scar has deformed and is non-circular. When fl = 0, it means that the shape of the wear scar has not deformed and is circular.

[0101] In this embodiment, the deformation threshold β is taken as 0.2, and fl = 0 is obtained, indicating that the shape of the wear scar has not deformed and is circular.

Claims

1. A detection method for the abnormal deformation degree of wear scars based on boundary distribution characteristics, characterized in that, It includes the following steps: Step 1: Segment the obtained four-ball friction wear scar image to obtain a wear scar segmentation diagram; Step 2: Obtain a boundary pixel diagram based on the wear scar segmentation diagram; Step 3: Calculate the deformation degree of the wear scar according to the boundary pixel diagram; Step 4: Detect the abnormal deformation of the wear scar according to the obtained deformation degree of the wear scar; In Step 3, to calculate the deformation degree of the wear scar according to the boundary pixel diagram, the specific method is as follows: Calculate the boundary smoothness of the boundary pixel diagram, the full range of deviation of the boundary pixel diagram, the dispersion characteristic of the deviation of the boundary pixel diagram, and the number of boundary pixels corresponding to the preset deformation amount of the boundary pixel diagram respectively; Calculate the deformation degree of the wear scar corresponding to the boundary pixel diagram according to the obtained boundary smoothness of the boundary pixel diagram, the full range of deviation of the boundary pixel diagram, the dispersion characteristic of the deviation of the boundary pixel diagram, and the number of boundary pixels corresponding to the preset deformation amount of the boundary pixel diagram; Calculate the deformation degree of the wear scar according to the following formula: where H is the deformation degree of the wear scar; γ k is the weight coefficient of the k-th feature; Z1 represents the smoothness of the wear scar boundary; Z2 represents the full range of the boundary pixel deviation degree; Z3 is the average value of the boundary deviation degree; Z4 is the standard deviation of the boundary deviation degree; Z5 is the deformation ratio of the preset deformation amount.

2. The detection method of the abnormal deformation degree of the wear scar based on the boundary distribution characteristics according to claim 1, characterized in that, In Step 2, to obtain a boundary pixel diagram according to the wear scar segmentation diagram, the specific method is as follows: Calculate the position size of the wear scar segmentation diagram; According to the position size obtained in Step 1, combine the following formula to obtain the boundary pixel diagram: where A is the wear scar boundary diagram; A(x,y)=1 indicates that the pixel (x,y) is a boundary pixel; A(x,y)=0 indicates that the pixel (x,y) is a non-boundary pixel; k and l respectively represent the index variables of rows and columns, both are integers, and the values are -1, 0, 1 respectively.

3. The detection method of the abnormal deformation degree of the wear scar based on the boundary distribution characteristics according to claim 1, characterized in that, Calculate the boundary smoothness of the boundary pixel diagram according to the following formula: where Z1 represents the smoothness of the wear scar boundary; num(A) is the number of pixels of the boundary pixel diagram; 2πr is the perimeter of the wear scar segmentation diagram.

4. The detection method for the abnormal deformation degree of the wear scar based on the boundary distribution characteristics according to claim 1, wherein Calculate the full range of deviation of the boundary pixel diagram, the specific method is as follows: Calculate the center distance and deviation of the boundary pixel diagram; Calculate the full range of deviation of the boundary pixel diagram according to the obtained center distance and deviation of the boundary pixel diagram.

5. The detection method for the abnormal deformation degree of the wear scar based on the boundary distribution characteristics according to claim 4, wherein Calculate the dispersion characteristic of the boundary deviation according to the following formula: where Z3 is the average value of the boundary deviation; Z4 is the standard deviation of the boundary deviation.

6. The detection method for the abnormal deformation degree of wear scars based on the boundary distribution characteristics according to claim 1, wherein Calculate the number of boundary pixels corresponding to the preset deformation amount of the boundary pixel diagram according to the following formula: Among them, Z5 is the deformation ratio of the preset deformation amount; P α is the number of boundary pixels when the preset deformation amount is α; num(A) is the total number of boundary pixels; α is the preset deformation amount, and according to expert experience, the value range of α is 0 < α < 1.

7. The detection method for the abnormal deformation degree of the wear scar based on the boundary distribution characteristics according to claim 1, characterized in that In Step 4, detect the abnormal deformation degree of the wear scar according to the following formula: where fl is the mark of abnormal deformation. When fl = 1, it indicates that the shape of the wear scar has deformed and is non-circular. When fl = 0, it indicates that the shape of the wear scar has not deformed and is circular; β is the deformation threshold.

Images