Ring upsetting friction coefficient determination method based on image recognition

Abstract

The application discloses a kind of ring upsetting friction coefficient determination methods based on image recognition, belongs to upsetting friction coefficient determination technical field, S1: the image of ring sample after upsetting is collected;S2: the inner contour of ring sample after upsetting in image is extracted by image edge extraction method, and the edge contour is further extracted using subpixel edge extraction method to inner contour;The inner diameter of ring after upsetting represented by edge contour is obtained;S3: based on standard friction coefficient theory calibration curve, the inner diameter of ring after upsetting, original inner diameter before upsetting, original height before upsetting and target height after upsetting are used to calculate the friction coefficient after upsetting.The application can improve detection efficiency on the basis of ensuring measurement accuracy through two rounds of edge feature extraction, thereby solving the time-consuming and laborious problem in experimental data processing when measuring friction coefficient by ring upsetting method.

Description

Technical Field

[0001] This invention belongs to the field of upsetting friction coefficient measurement technology, and more specifically, relates to a method for measuring the friction coefficient of annular upsetting based on image recognition. Background Technology

[0002] During the hot forging process, friction can lead to poor forging precision and surface quality, as well as short die life. According to relevant data, energy loss due to friction is significant. Therefore, it is crucial to strengthen the measurement of the coefficient of friction during the hot forging process.

[0003] Most methods for determining the coefficient of friction are based on the classical Coulomb's law, which involves measuring the normal force p and the frictional force f, and then calculating the coefficient of friction using the formula μ = f / p. The ring upsetting method, proposed in the 1960s, is a relatively convenient and quantitative method for determining the coefficient of friction or friction factor. However, in terms of data processing, the commonly used methods are still manual measurement and calculation, which are not only time-consuming but also produce unstable results. This falls far short of meeting the needs of practical engineering.

[0004] Thanks to advancements in digital image processing technology, rapid and accurate digital detection of the friction coefficient of upset ring samples has become possible. This not only effectively reduces waste of resources and time, but also significantly improves the repeatability of test results by enabling image recognition of the upset ring samples. Summary of the Invention

[0005] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides a method for determining the friction coefficient of an upsetting ring based on image recognition. The method involves acquiring an image of an upset annular sample; extracting the inner contour of the upset annular sample from the image using an image edge extraction method; further extracting the edge contour using a sub-pixel edge extraction method; obtaining the inner diameter of the upset annular sample characterized by the edge contour; and obtaining the friction coefficient after upsetting using the inner diameter of the upset annular sample and a standard friction coefficient theoretical calibration curve. This solves the problem of time-consuming and labor-intensive experimental data processing in the annular upsetting method for measuring the friction coefficient.

[0006] To achieve the above objectives, according to one aspect of the present invention, a method for determining the friction coefficient of annular upsetting based on image recognition is provided, comprising:

[0007] S1: Acquire an image of the annular sample after upsetting;

[0008] S2: Extract the inner contour of the upset ring sample in the image using an image edge extraction method, and further extract the edge contour using a sub-pixel edge extraction method; obtain the inner diameter of the upset ring represented by the edge contour;

[0009] S3: Based on the standard friction coefficient theoretical calibration curve, the friction coefficient after upsetting is calculated using the inner diameter of the upsetting ring after upsetting, the original inner diameter before upsetting, the original height before upsetting, and the target height after upsetting.

[0010] The standard friction coefficient theoretical calibration curve is used to characterize the mapping relationship between the friction coefficient after upsetting, the height ratio before and after upsetting, and the inner diameter compression ratio before and after upsetting.

[0011] In one embodiment, S2 includes:

[0012] S21: The pixel points of the inner diameter edge in the image are obtained by using the classical gradient operator, a continuous gray-level distribution inside the pixel is constructed by nonlinear interpolation method, the edge is extracted from the continuous gray-level distribution by gradient calculation, and the edge contour is obtained by fitting.

[0013] S22: Obtain the inner diameter of the upset annulus representing the edge contour.

[0014] In one embodiment, S22 includes:

[0015] The edge contour is divided by selecting a preset center and a preset angle step, and the distance from the preset center to the edge contour is obtained at each preset angle step. The inner diameter sequence is then used as the upset inner diameter of the annulus corresponding to each preset angle step.

[0016] In one embodiment, the preset center is the geometric center or geometric centroid of the edge contour.

[0017] In one embodiment, S22 includes:

[0018] The edge profile is fitted to a complete circle, and the inner diameter data of the complete circle is used as the inner diameter of the upset annulus representing the edge profile.

[0019] In one embodiment, S1 includes: performing distortion correction and pixel accuracy calibration on the industrial camera.

[0020] In one embodiment, S3 includes:

[0021] Calculate the ratio of the inner diameter of the upset ring to the original inner diameter before upset to obtain the inner diameter reduction rate;

[0022] Based on the standard friction coefficient theoretical calibration curve, the current corresponding friction coefficient after upsetting is calculated using the inner diameter reduction rate, the original height before upsetting, and the target height after upsetting.

[0023] According to another aspect of the present invention, an image recognition-based device for measuring the friction coefficient of annular upsetting is provided, comprising:

[0024] The acquisition module is used to acquire images of the annular sample after upsetting;

[0025] The acquisition module is used to extract the inner contour of the upset ring sample in the image using an image edge extraction method, and further extract the edge contour using a sub-pixel edge extraction method to obtain the edge contour; and acquire the inner diameter of the upset ring represented by the edge contour.

[0026] The measurement module is used to calculate the friction coefficient after upsetting based on the standard friction coefficient theoretical calibration curve, using the inner diameter of the upset ring after upsetting, the original inner diameter before upsetting, the original height before upsetting, and the target height after upsetting; wherein, the standard friction coefficient theoretical calibration curve is used to characterize the mapping relationship between the friction coefficient after upsetting, the height ratio before and after upsetting, and the inner diameter compression ratio before and after upsetting.

[0027] According to another aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0028] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0029] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0030] (1) This invention provides a method for determining the friction coefficient of an upsetting ring based on image recognition. The method involves acquiring an image of the upsetting ring sample; extracting the inner contour of the upsetting ring sample from the image using an image edge extraction method; further extracting the edge contour using a sub-pixel edge extraction method; obtaining the inner diameter of the upsetting ring characterized by the edge contour; and obtaining the friction coefficient after upsetting using the inner diameter of the upsetting ring and the standard friction coefficient theoretical calibration curve. This invention improves the detection efficiency while ensuring measurement accuracy through two rounds of edge feature extraction, thereby solving the problem of time-consuming and labor-intensive experimental data processing in the ring upsetting method for measuring the friction coefficient.

[0031] (2) The pixel points of the inner diameter edge in the image are obtained by using the classical gradient operator. The continuous gray-level distribution inside the pixel is constructed by nonlinear interpolation method. The sub-pixel level edge is extracted from the continuous gray-level distribution by gradient calculation and the edge contour is obtained by fitting. This breaks through the limitation of camera resolution, improves the edge extraction accuracy, and thus improves the accuracy of image edge detection, thereby improving the accuracy of the inner diameter calculation of the ring.

[0032] (3) By performing distortion correction and pixel accuracy calibration on the industrial camera, the image of the upsetting ring sample is acquired by the industrial camera to extract the sub-pixel level edge, thus breaking through the limitation of camera resolution and improving the edge extraction accuracy. Attached Figure Description

[0033] Figure 1 This is a flowchart of the method for determining the friction coefficient of annular upsetting based on image recognition in Example 1.

[0034] Figure 2 This is an application scenario diagram of the image recognition-based method for determining the friction coefficient of annular upsetting in Example 3.

[0035] Figure 3 This is an application scenario diagram of using an industrial camera to capture images of the upset annular sample in Example 6.

[0036] Figure 4 This is a dimensional diagram of the ring sample after upsetting provided in Example 6.

[0037] Figure 5 This is a schematic diagram of the theoretical calibration curve of the standard friction coefficient in Example 7. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0039] Example 1

[0040] like Figure 1 As shown, an image recognition-based method for determining the friction coefficient of annular upsetting is provided, comprising:

[0041] S1: Image of the upsetting ring sample.

[0042] S2: Extract the inner contour of the upset ring sample in the image using the image edge extraction method, and further extract the edge contour using the sub-pixel edge extraction method; obtain the inner diameter of the upset ring characterized by the edge contour;

[0043] S3: Based on the standard friction coefficient theoretical calibration curve, the friction coefficient after upsetting is calculated using the inner diameter of the ring after upsetting, the original inner diameter before upsetting, the original height before upsetting, and the target height after upsetting.

[0044] Among them, the standard friction coefficient theoretical calibration curve is used to characterize the mapping relationship between the friction coefficient after upsetting, the height ratio before and after upsetting, and the inner diameter compression ratio before and after upsetting.

[0045] Specifically, the ring sample is upset on an upsetting machine. The original inner diameter and height of the ring before upsetting, as well as the target height after upsetting, are already determined. An industrial camera can be used to acquire images of the upset ring sample, and the radius of the inner ring can be measured using image methods. Based on the standard friction coefficient curve, the friction coefficient corresponding to the current inner ring radius is calculated. The standard friction coefficient theoretical calibration curve is used to characterize the mapping relationship between the friction coefficient after upsetting, the height ratio before and after upsetting, and the inner diameter compression ratio before and after upsetting.

[0046] Example 2

[0047] Step S2 includes: S21: Using the classical gradient operator to obtain the pixel points of the inner diameter edge in the image, constructing a continuous gray-level distribution inside the pixel through a nonlinear interpolation method, extracting the edge from the continuous gray-level distribution using gradient calculation, and fitting the edge contour; S22: Obtaining the upset inner diameter of the ring characterized by the edge contour.

[0048] Example 3

[0049] like Figure 2 As shown, step S22 includes: selecting a preset center and a preset angle step to divide the edge contour, obtaining the distance from the preset center to the edge contour under each preset angle step, and obtaining the inner diameter sequence as the inner diameter of the upset ring corresponding to each preset angle step.

[0050] Specifically, using the geometric center of all edges as the center of a circle, the distance from the center to the edge is calculated as a radius data sequence according to a certain angle sequence. For example, it is calculated in 1° increments from 0° to 360°, in 30° increments, or in 45° increments.

[0051] Example 4

[0052] The preset center is the geometric center or geometric centroid of the edge contour.

[0053] Example 5

[0054] Step S22 includes: fitting the edge profile to a complete circle, and using the inner diameter data of the complete circle as the inner diameter of the upset annulus representing the edge profile.

[0055] Example 6

[0056] Step S1 includes: performing distortion correction and pixel accuracy calibration on the industrial camera; and using the industrial camera to acquire images of the upset annular sample. Figure 3 This is an application scenario diagram showing the use of an industrial camera to capture images of the upset annular sample in this embodiment. Figure 4 This is a dimensional diagram of the ring sample after upsetting provided in this embodiment, where D is the outer diameter, d is the inner diameter, and h is the height.

[0057] Example 7

[0058] Step S3 includes: calculating the ratio of the inner diameter of the upsetting ring to the original inner diameter before upsetting to obtain the inner diameter reduction rate; and calculating the corresponding friction coefficient after upsetting based on the standard friction coefficient theoretical calibration curve, using the inner diameter reduction rate, the original height before upsetting, and the target height after upsetting.

[0059] This embodiment mentions the standard friction coefficient theoretical calibration curve, such as... Figure 5 The figure shows the relationship curves between the friction coefficient and the height compression ratio and inner diameter reduction rate, obtained from experiments. The horizontal axis represents the height compression ratio, which is the ratio of the upset height to the original height. The vertical axis represents the inner diameter reduction rate, which is the ratio of the upset inner diameter to the original inner diameter. On a two-dimensional plane, connecting the coordinate points corresponding to the same friction coefficient for both the height compression ratio and inner diameter reduction rate forms a curve. These multiple non-intersecting curves on the two-dimensional plane constitute the theoretical calibration curves for the standard friction coefficient. Figure 5 This is a dimensional diagram of the annular sample provided in this embodiment, where D is the outer diameter, d is the inner diameter, and h is the height.

[0060] As described in S1 in this embodiment, since the height of the sample is fixed before and after upsetting—that is, the height compression ratio is determined—and the original inner diameter of the sample is fixed, the friction coefficient is a univariate function of the inner diameter of the sample after upsetting. Multiple friction coefficients corresponding to the inner diameters after upsetting are obtained through experiments, and polynomial fitting is performed on the test data to obtain a fitting formula for calculating the friction coefficient using the inner diameter after upsetting.

[0061] Further, in step S3, based on the inner diameter data or inner diameter sequence obtained in step S2, the corresponding friction coefficient or friction coefficient sequence can be obtained using the standard friction coefficient fitting formula. The friction coefficient sequence allows us to obtain the friction coefficient of the sample in various directions.

[0062] Example 8

[0063] According to another aspect of the present invention, an image recognition-based device for measuring the friction coefficient of annular upsetting is provided, comprising:

[0064] The acquisition module is used to acquire images of the upsetting ring sample;

[0065] The acquisition module is used to extract the inner contour of the upset ring sample in the image using an image edge extraction method, and further extract the edge contour using a sub-pixel edge extraction method; and acquire the inner diameter of the upset ring characterized by the edge contour.

[0066] The measurement module is used to calculate the friction coefficient after upsetting based on the standard friction coefficient theoretical calibration curve, using the inner diameter of the ring after upsetting, the original inner diameter before upsetting, the original height before upsetting, and the target height after upsetting. Among them, the standard friction coefficient theoretical calibration curve is used to characterize the mapping relationship between the friction coefficient after upsetting, the height ratio before and after upsetting, and the inner diameter compression ratio before and after upsetting.

[0067] Example 9

[0068] According to another aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0069] Example 10

[0070] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0071] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for determining the friction coefficient of annular upsetting based on image recognition, characterized in that, include: S1: Acquire an image of the upsetting ring sample; S2: Extract the inner contour of the upset ring sample in the image using an image edge extraction method, and further extract the edge contour using a sub-pixel edge extraction method; obtain the inner diameter of the upset ring represented by the edge contour; S3: Based on the standard friction coefficient theoretical calibration curve, the friction coefficient after upsetting is calculated using the inner diameter of the upsetting ring after upsetting, the original inner diameter before upsetting, the original height before upsetting, and the target height after upsetting. The standard friction coefficient theoretical calibration curve is used to characterize the mapping relationship between the friction coefficient after upsetting, the height ratio before and after upsetting, and the inner diameter compression ratio before and after upsetting. S2 includes: S21: using a classical gradient operator to obtain the pixel points of the inner diameter edge in the image, constructing a continuous gray-level distribution inside the pixel through a nonlinear interpolation method, extracting the edge from the continuous gray-level distribution using gradient calculation, and fitting the edge contour; S22: obtaining the upset inner diameter of the annulus represented by the edge contour; S22 includes: selecting a preset center and a preset angle step size to divide the edge contour, obtaining the distance from the preset center to the edge contour at each preset angle step size, and obtaining an inner diameter sequence as the upset inner diameter of the annulus corresponding to each preset angle step size.

2. The method for determining the friction coefficient of annular upsetting based on image recognition as described in claim 1, characterized in that, The preset center is the geometric center or geometric centroid of the edge contour.

3. The method for determining the friction coefficient of annular upsetting based on image recognition as described in claim 1, characterized in that, S22 includes: The edge profile is fitted to a complete circle, and the inner diameter data of the complete circle is used as the inner diameter of the upset annulus representing the edge profile.

4. The method for determining the friction coefficient of annular upsetting based on image recognition as described in any one of claims 1-3, characterized in that, S1 includes: Distortion correction and pixel accuracy calibration are performed on industrial cameras; The industrial camera was used to capture images of the annular sample after upsetting.

5. The method for determining the friction coefficient of annular upsetting based on image recognition as described in any one of claims 1-3, characterized in that, S3 includes: Calculate the ratio of the inner diameter of the upset ring to the original inner diameter before upset to obtain the inner diameter reduction rate; Based on the standard friction coefficient theoretical calibration curve, the current corresponding friction coefficient after upsetting is calculated using the inner diameter reduction rate, the original height before upsetting, and the target height after upsetting.

6. A device for measuring the friction coefficient of annular upsetting based on image recognition, characterized in that, The method for determining the friction coefficient of annular upsetting based on image recognition as described in any one of claims 1-5 includes: The acquisition module is used to acquire images of the upsetting ring sample; The acquisition module is used to extract the inner contour of the upset ring sample in the image using an image edge extraction method, and further extract the edge contour using a sub-pixel edge extraction method to obtain the edge contour; and acquire the inner diameter of the upset ring represented by the edge contour. The measurement module is used to calculate the friction coefficient after upsetting based on the standard friction coefficient theoretical calibration curve, using the inner diameter of the upset ring after upsetting, the original inner diameter before upsetting, the original height before upsetting, and the target height after upsetting; wherein, the standard friction coefficient theoretical calibration curve is used to characterize the mapping relationship between the friction coefficient after upsetting, the height ratio before and after upsetting, and the inner diameter compression ratio before and after upsetting.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images