Large deformation of high-strength linear materials based on image recognition and A gt Measurement method

Abstract

The present invention discloses a large deformation and A gt measurement method for high-strength linear materials based on image recognition. By marking the material, two marked points are obtained; a QR code scale is placed to ensure that the QR code and the two measurement points are in the same plane; a camera device is used to record the whole process; pictures at the initial moment, any moment during the large deformation stage, and the moment of maximum load are exported, and perspective correction is performed on the captured images through image recognition and perspective transformation; according to the size and the number of pixels occupied by the QR code in the image, the pixel calibration coefficient is calculated; the positions of the marked points are respectively selected on each picture to obtain the distance dimensions between the marked points before and after deformation; relevant physical quantities are calculated. The large deformation and total elongation at maximum force A gt measurement method for high-strength linear materials based on the QR code scale and image recognition can achieve non-contact high-precision and rapid measurement with traceability throughout the whole process.

Description

Technical Field

[0001] The present invention relates to the field of material deformation measurement, and particularly to a large deformation and A measurement method of high-strength linear materials based on image recognition. gt Measurement method. Background Art

[0002] High-strength linear materials refer to linear materials with characteristics such as high strength and high elastic modulus. Generally, the strength exceeds 500 MPa, and for ultra-high-strength materials, the strength exceeds 1000 MPa or even 2000 MPa, with high breaking force and large deformation.

[0003] In the mechanical property test report of high-strength linear materials, it is necessary to clarify their stress-strain relationship, draw the full stress-strain relationship curve, and give the test result of the total elongation at maximum force (A gt ). A gt refers to the elongation of the material under the action of the maximum load during the tensile process, and is one of the important indicators to measure the ductility of the material, reflecting the deformation ability and tensile property of the material under stress. Especially for high-strength materials such as steel strands, they have a very high breaking force, and the occurrence of fracture under the maximum force is relatively sudden. A gt is a very important mechanical property index of it. The ultimate strength of steel strands is usually greater than 1700 MPa, gt usually greater than 3%.

[0004] Currently, there are three commonly used A gt measurement methods: extensometer method, testing machine method, and manual method.

[0005] I. Extensometer method: Clamp and stretch both ends of the high-strength linear material, and clamp the two measuring ends of the extensometer in the middle of the high-strength linear material and stretch along with the high-strength linear material to realize the deformation measurement of the high-strength linear material. During the process of measuring the deformation of high-strength linear materials by the extensometer method, the following deficiencies exist and need to be improved:

[0006] 1. In order to ensure the deformation measurement accuracy of the existing extensometer, the maximum deformation measurement range is usually much smaller than the large deformation of the high-strength linear material. The sudden failure of the material will seriously damage the extensometer. Therefore, the large deformation of the material cannot be obtained solely through the extensometer and needs to be inferred by combining the extensometer data with the displacement sensor data of the testing machine. Thus, the accurate large deformation of the material cannot be obtained. In addition, extensometers with a large range are expensive.

[0007] 2. Since the length range of the two measuring ends of the extensometer is limited, when the necking section of the high-strength linear material appears outside the two measuring ends of the extensometer, the deformation measurement result will be on the small side.

[0008] 3. The two measuring ends of the extensometer are in contact with the high-strength linear material for measurement. During the tensile deformation process of the high-strength linear material, slippage is likely to occur between the two measuring ends of the extensometer and the high-strength linear material, resulting in inaccurate deformation measurement data of the extensometer.

[0009] 4. The extensometer is usually arranged unilaterally, with varying degrees of eccentricity, resulting in errors in the test results.

[0010] II. Testing machine method: Clamp both ends of the high-strength linear material in the upper and lower clamps of the testing machine; among them, displacement sensors are installed inside the upper and lower clamps, and under the premise that there is no relative slippage between the measured linear material and the testing machine, the tensile deformation of the high-strength linear material can be monitored. However, due to the large ultimate load of the high-strength linear material, a large loading force is required to break the high-strength linear material, so slippage is likely to occur between the high-strength linear material and the upper and lower clamps, and thus the tensile deformation measured by the displacement sensor is inaccurate. In addition, to prevent slippage between the high-strength linear material and the upper and lower clamps, there is also a method of increasing the clamping force through physical means. Although this can reduce slippage, it will cause damage to the high-strength linear material itself due to stress concentration at the clamping position, and it is easy to generate a weak cross-section at the clamping position, resulting in the fracture of the high-strength linear material occurring first at the clamping position, and the maximum tensile force that the material can withstand cannot be measured, resulting in a smaller Agt test value.

[0011] III. Manual method: Clamp and stretch both ends of the high-strength linear material, mark the gauge length on the high-strength linear material, and manually measure the distance and read the value. This measurement method has low accuracy, and errors will be introduced during the manual distance measurement and reading process, which has a greater impact on the Agt measurement result.

[0012] At the same time, since the measurement processes of the three methods cannot be strictly traced, the relationship between deformation and load change cannot be corresponded in real time. When abnormal detection results occur or there are disputes about the results, the problem of whether the measurement results are affected during the detection process cannot be traced. Summary of the Invention

[0013] The technical problem to be solved by the present invention is to provide a large deformation and A measurement method of high-strength linear materials based on image recognition in view of the above-mentioned deficiencies of the prior art. gt The large deformation and A measurement method of high-strength linear materials based on image recognition gt uses a two-dimensional code as a scale, takes pictures and records the whole process of the deformation of the material test containing the two-dimensional code, corrects the picture and calibrates the pixels with the help of the two-dimensional code, selects the marked points in the OpenCV interactive interface to obtain the length, and calculates the material deformation and A gt value.

[0014] To solve the above technical problems, the technical solution adopted by the present invention is:

[0015] A large deformation and A measurement method of high-strength linear materials based on image recognition, including the following steps. gt The measurement method includes the following steps.

[0016] Step 1, making marking points: Select a high-strength linear material with a length of L, and make two marking points on the same side of the high-strength linear material; wherein, both ends of the high-strength linear material are formed as clamping parts; then the two marking points are arranged adjacent to the two clamping parts, and the distance between the two marking points is greater than L / 2.

[0017] Step 2, clamping the high-strength linear material: Clamp the two clamping parts of the high-strength linear material in the fixture of the testing machine, and both marking points face outward.

[0018] Step 3, placing a QR code: Place a QR code with a fixed position on the testing machine on one side of the high-strength linear material, and the QR code and the two marking points are located in the same plane M; meanwhile, record the original side length dimension of the QR code.

[0019] Step 4, installing a camera: Install a camera outside the high-strength linear material, and each frame of the captured image of the camera needs to include the QR code and the two marking points.

[0020] Step 5, stretching the high-strength linear material: Start the testing machine, axially stretch the clamped high-strength linear material until the high-strength linear material breaks; during the stretching process, the testing machine records the change curve of the applied force over time in real time.

[0021] Step 6, taking deformation pictures: Before the testing machine starts, the camera takes a picture of the high-strength linear material first, as the original picture before stretching; while the testing machine starts stretching, the camera starts recording the entire stretching process until the high-strength linear material breaks, obtaining the deformation stretching video of the high-strength linear material.

[0022] Step 7, selecting deformation pictures: First, find the corresponding moment A of the set applied force from the change curve of the applied force over time in Step 5; then, find the deformation picture corresponding to moment A from the deformation stretching video in Step 6; wherein, the set applied force includes the maximum applied force.

[0023] Step 8, image correction: Using the QR code in Step 3 as a scale, perform image correction on both the original picture before stretching and each selected deformation picture in Step 7.

[0024] Step 9, calculating the pixel calibration coefficient: According to the number of pixel points occupied by the side length of the QR code in the corrected deformation picture and the original side length dimension of the QR code recorded in Step 3, calculate the pixel calibration coefficient.

[0025] Step 10, measuring large deformation and A gt, specifically including the following steps:

[0026] Step 10-1, measure the original length L0 before deformation: Import the original picture before stretching corrected in Step 8 into the OpenCV interactive interface, select the positions of two marking points, and obtain the length L0 between the two marking points at the initial moment.

[0027] Step 10-2, measure the length after deformation: Import each deformed picture corrected in Step 8 into the OpenCV interactive interface; and select the positions of two marking points in each deformed picture to obtain the length L1 between the two marking points at the moment A corresponding to the set loading force; among them, the length between the two marking points at the maximum loading force is L2.

[0028] Step 10-3, measure the large deformation ΔL, then ΔL = L1 - L0:

[0029] Step 10-4, measure A gt , and the specific calculation formula is: A gt = (L2 - L0) / L0 × 100%.

[0030] In Step 1, each marking point is a marking line perpendicular to the axis of the high-strength linear material and in an arc or circular shape, formed by spraying paint through a strip hole; the distance l2 from each marking point to the corresponding end of the high-strength linear material is greater than the length l1 of the clamping part; in Step 10, when calculating L0, L1, and L2, it is the average distance between the two marking lines.

[0031] In Step 2, the QR code is printed on cardboard and inserted into the acrylic board. The size of the QR code is a × a, then a = 0.2 - 0.8(L - 2l2); after the QR code is placed, use a laser collimation device for alignment to ensure that the two marking points and the QR code are on the same plane.

[0032] In Step 4, assume that the actual length and width of each frame of the camera shooting image are x × y, then:

[0033] y > 1.2a

[0034] (1 + α0 + 10%)(L - 2l2) < x < 1.2(L - 2l2)

[0035] Among them, α0 is the estimated elongation rate of the high-strength linear material, not greater than 8%.

[0036] In Step 5, in the curve of the change of the loading force with time, the accuracy of time is 0.05 s, and the accuracy of the load is 0.01 kN; in Step 6, the recording of the deformation stretching video starts simultaneously with the loading of the testing machine, and the accuracy of the selected moment is also 0.05 s.

[0037] In Step 8, image correction includes the following steps:

[0038] Step 8-1, Decode the QR code: Decode the QR code images in both the original image before stretching and each deformed image selected in Step 7 using the pyzbar package to obtain the coordinates of the four corner points in each QR code image.

[0039] Step 8-2, Image correction: Calculate the perspective transformation matrix using the cv2.getPerspectiveTransform function based on the coordinates of the four corner points of the QR code image obtained in Step 8-1; use the perspective transformation matrix to correct each deformed image in Step 8-1; after correction, the QR code images in each deformed image are converted into front-facing quadrilaterals; the pixels of each corrected deformed image are x’_pixel×y’_pixel, then:

[0040] x’_pixel = x_pixel×cosθ

[0041] y’_pixel = y_pixel×cosθ

[0042] where x_pixel and y_pixel are the pixel precisions of the camera.

[0043] θ is the angle between the camera shooting screen and the QR code plane.

[0044] In Step 4, the number of pixels in the camera shooting screen depends on the pixel precisions x_pixel and y_pixel of the shooting device; x_pixel and y_pixel need to satisfy that, on the premise that both the QR code and the two marker points are complete, a close-up zoom shot should be taken to make the high-strength linear material fill the shooting screen as much as possible, then x_pixel needs to satisfy: x_pixel≥(1.2 + α0) / 0.02α0cosθ.

[0045] In Step 9, the calculation formula for the pixel calibration coefficient pixel_coefficient is:

[0046] pixel_coefficient = a / mark’_pixel

[0047] And

[0048] pixel_coefficient<0.02α0(L - 2l2)

[0049] In the formula, mark’_pixel is the number of pixels occupied by the side length of the QR code in the corrected image.

[0050] θ is less than or equal to 45 degrees.

[0051] In step 10, the two calibration points are calibration point one and calibration point two respectively. The deformation image corresponding to the maximum load in step 7 is called the maximum load deformation image. The pixel coordinates of calibration point one and calibration point two in the original image before stretching after correction are (x 0-1 , y 0-1 ) and (x 0-2 , y 0-2 ); The pixel coordinates of calibration point one and calibration point two in the maximum load deformation image after correction are (x 1-1 , y 1-1 ) and (x 1-2 , y 1-2 ). The calculation formulas for L0 and L1 are respectively:

[0052]

[0053] In the formula, pixel_coefficient is the pixel calibration coefficient.

[0054] The present invention has the following beneficial effects:

[0055] 1. The present invention can cope with the measurement of large deformations, overcoming the limitations of the range and measurement scope of existing measuring instruments.

[0056] 2. The present invention can achieve non-contact measurement based on images, avoiding errors caused by relative slippage between the instrument and the material and damage to the material.

[0057] 3. The video image evidence of the present invention and the basic information of the corresponding high-strength linear material can be stored in the two-dimensional code, and the two-dimensional code is adhered to the corresponding test report for retention; at the same time, the load information and the deformation information are corresponding in real time, thus ensuring the accuracy and traceability of the test results.

[0058] 4. The present invention uses a two-dimensional code calibrator to automatically achieve perspective correction and pixel calibration, solves the problem of imaging distortion, improves the accuracy of image measurement, does not require manual participation, the process is automated, and reduces human errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 Shows a schematic diagram of the principle of the large deformation and A gt measurement method of high-strength linear materials based on image recognition of the present invention; among them, Figure (a) shows a schematic diagram of the dimension marking of the original image before stretching after correction; Figure (b) shows a schematic diagram of the dimension marking of the maximum load deformation image after correction.

[0060] Figure 2Shows a schematic diagram of the QR code used in the present invention; among them, Figure (a) shows a schematic diagram of the overall image of the QR code; Figure (b) shows a schematic diagram of the functional modules in the QR code image; Figure (c) shows the data module that functions as data storage in the QR code image; Figure (d) shows a schematic diagram of the four corner markers in Figure (b); Figure (e) shows a schematic diagram of the guiding module connecting the four corner points in Figure (b).

[0061] Figure 3 Shows a schematic diagram of the camera zoom shooting screen in the present invention; among them, Figure (a) shows a schematic diagram of the original shooting cross-section; Figure (b) shows a schematic diagram of 1x zoom; Figure (c) shows a schematic diagram of 5x zoom.

[0062] Figure 4 Is the automatic perspective correction flow chart of the present invention; among them, Figure (a) shows a schematic diagram of the QR code decoding coordinates; Figure (b) shows a schematic diagram of the QR code perspective correction; Figure (c) shows a schematic diagram of the deformed picture before correction; Figure (d) shows a schematic diagram of the deformed picture after correction.

[0063] Figure 5 Is a schematic diagram of the interactive distance measurement method of the present invention; among them, Figure (a) shows a schematic diagram of the distance between two marker points before stretching; Figure (b) shows a schematic diagram of the distance between two marker points after stretching.

[0064] Among them are:

[0065] 10. Marker point; 20. Steel strand; 30. Testing machine. Specific embodiments

[0066] The present invention will be further described in detail below in conjunction with the drawings and specific preferred embodiments.

[0067] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by terms such as "left side", "right side", "upper part", "lower part", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. "First", "second", etc. do not represent the importance of components, so they cannot be understood as limitations to the present invention. The specific dimensions adopted in this embodiment are only for illustrating the technical solution by way of example and do not limit the protection scope of the present invention.

[0068] As Figure 1 and Figure 5 shown, a large deformation and A gt measurement method of high-strength linear materials based on image recognition includes the following steps.

[0069] Step 1, making marking points: select a high-strength linear material (preferably a steel strand 20) with a length of L, and make two marking points 10 on the same side of the high-strength linear material; wherein both ends of the high-strength linear material are formed into a clamping portion with a length of l1; then the two marking points are arranged adjacent to the two clamping portions, and the spacing between the two marking points is greater than L / 2. The advantage of this arrangement is that the necking section (i.e., the position where the maximum deformation occurs) of the high-strength linear material can be between the two marking points, so that the maximum deformation can be detected.

[0070] Furthermore, in step 1, each marking point is a marking line that is perpendicular to the axis of the high-strength linear material and is in an arc or ring shape, and is formed by spraying paint through strip holes; the distance l2 from each marking point to the corresponding end of the high-strength linear material is greater than the length l1 of the clamping portion; in step 10, L0, L1 and L2 are calculated as the average distance between the two marking lines.

[0071] The above-mentioned method of spraying bright paint on round holes is preferably: use a flexible spray paint mold with strip holes, place it close to the steel strand, spray the mold with bright white paint or black paint, and let it stand for one hour to ensure that the mark is clear and recognizable; to ensure that the material within the mark is effectively deformed, the distance between the mark point and the end of the material is l2≥l1.

[0072] Step 2, clamping the high-strength linear material: clamp the two clamping parts of the high-strength linear material in the fixture of the testing machine 30, and the two marking points are facing outward.

[0073] Step 3, placing the QR code: Place a fixed QR code on the testing machine on one side of the high-strength linear material, and the QR code and the two marking points are located in the same plane M; at the same time, record the original side length of the QR code.

[0074] The position of the two-dimensional code is fixed, that is, it is in a stationary state relative to the fixed end of the testing machine, so it will not be deformed due to the stretching of high-strength linear materials such as steel strands.

[0075] like Figure 2 As shown, the QR code is preferably printed on cardboard and inserted into an acrylic board. The size of the QR code is a×a, then:

[0076] a=0.2~0.8(L-2l2)

[0077] After the QR code is placed, use a laser alignment device to align it to ensure that the two marking points are on the same plane as the QR code.

[0078] The QR code includes a data module and a functional module, and has the characteristics of large information storage capacity, fast scanning, error correction and adaptability to various data formats.

[0079] The above functional modules are used for positioning, subsequent screen correction, and pixel calibration, enabling the scanning device to accurately locate and analyze QR codes, and can be further subdivided into marking and guiding modules.

[0080] The above data module is used to store the basic information of the measured material and the captured video data; each time it is used, the functional module remains unchanged, and a new data module is customized to store the information of different material components.

[0081] Step 4: Install the camera: Install the camera on the outside of the high-strength linear material. Each frame of the captured image by the camera should include a QR code and two marking points.

[0082] Assume the actual length and width of each frame of the captured image by the camera are x×y, then:

[0083] y>1.2a

[0084] (1+α0+10%)(L-2l2)<x<1.2(L-2l2)

[0085] Where α0 is the estimated elongation rate of the high-strength linear material, not greater than 8%.

[0086] The number of pixels in the captured image by the above camera depends on the pixel accuracy x_pixel and y_pixel of the capturing device; x_pixel and y_pixel need to meet the condition that when the QR code and the two marking points are both complete, close-range zoom shooting should be carried out to make the high-strength linear material fill the captured image as much as possible. Then x_pixel needs to meet:

[0087] x_pixel≥(1.2+α0) / 0.02α0cosθ

[0088] In the formula, θ is the angle between the captured image plane of the camera and the QR code plane, and θ is preferably less than or equal to 45 degrees.

[0089] As Figure 3 shown, the resolution of the captured image determines the pixel calibration coefficient, which directly affects the measurement accuracy. Increasing the number of pixels occupied by the target can improve the subsequent pixel calibration coefficient. Therefore, using an advanced photographing device or adopting image super-resolution reconstruction technology can enhance the measurement accuracy of this method. For example, the HUAWEI nova 6 mobile phone released in 2019 has an image resolution of 2736×3648 in the normal mode and has a zoom function. When zooming in on a ruler and shooting, at 1x zoom, the measurement accuracy can reach 70 / 2736 = 0.0256mm, and at 5x zoom, the measurement accuracy can reach 20 / 2736 = 0.00731mm.

[0090] Step 5: Stretch the high-strength linear material: Start the testing machine and axially stretch the clamped high-strength linear material until it fractures. During the stretching process, the testing machine records the curve of the applied force varying with time in real time. Among them, in the curve of the applied force varying with time, the time accuracy is preferably 0.05 s, and the load accuracy is 0.01 kN.

[0091] Step 6: Take deformation pictures: Before starting the testing machine, the camera takes a picture of the high-strength linear material first, which serves as the original picture before stretching. While starting to stretch the testing machine, the camera starts to record the entire stretching process until the high-strength linear material fractures, obtaining the deformation stretching video of the high-strength linear material. The moment accuracy that can be selected in the deformation stretching video is also 0.05 s.

[0092] Step 7: Select deformation pictures: First, find the corresponding moment A of the set applied force from the curve of the applied force varying with time in Step 5. Then, find the deformation picture corresponding to moment A from the deformation stretching video in Step 6. Among them, the set applied force includes the maximum applied force, and the deformation picture corresponding to the moment of the maximum applied force is called the maximum loading deformation picture.

[0093] Step 8: Image correction: Using the two-dimensional code in Step 3 as a scale, perform image correction on both the original picture before stretching and each selected deformation picture in Step 7.

[0094] As Figure 4 shown, the above image correction preferably includes the following steps.

[0095] Step 8-1: Decode the two-dimensional code: Decode the two-dimensional code images in both the original picture before stretching and each selected deformation picture in Step 7 using the pyzbar package to obtain the coordinates of the four corner points in each two-dimensional code image.

[0096] Step 8-2: Image correction: Compare the coordinates of the four corner points of the two-dimensional code image obtained in Step 8-1 with the coordinates of the four corner points of the functional modules in the actual two-dimensional code in Step 2, and use the cv2.getPerspectiveTransform function to calculate the perspective transformation matrix.

[0097] Use the perspective transformation matrix to correct each deformation picture in Step 8-1. After correction, the two-dimensional code images in each deformation picture are converted into front-view quadrilaterals. The pixels of each corrected deformation picture are x’_pixel×y’_pixel, then:

[0098] x’_pixel = x_pixel×cosθ

[0099] y’_pixel = y_pixel×cosθ

[0100] Among them, x_pixel and y_pixel are the pixel precisions of the camera.

[0101] Step 9: Calculate the pixel calibration coefficient: According to the number of pixel points occupied by the side length of the QR code in the corrected deformed image and the original side length size of the QR code recorded in Step 3, the pixel calibration coefficient is calculated.

[0102] Pixel calibration is completed by comparing the actual size of the QR code with the number of pixels occupied by the corrected QR code in the image.

[0103] The calculation formula of the pixel calibration coefficient pixel_coefficient is preferably:

[0104] pixel_coefficient = a / mark’_pixel

[0105] And

[0106] pixel_coefficient < 0.02α0(L - 2l2)

[0107] In the formula, mark’_pixel is the number of pixel points occupied by the side length of the QR code in the corrected image.

[0108] Step 10: Measure the large deformation and A gt , which specifically includes the following steps.

[0109] Step 10 - 1: Measure the original length L0 before deformation: Import the original image before stretching corrected in Step 8 into the OpenCV interactive interface, select the positions of two marker points, and obtain the length L0 between the two marker points at the initial moment.

[0110] Step 10 - 2: Measure the length after deformation: Import each deformed image corrected in Step 8 into the OpenCV interactive interface; and select the positions of two marker points in each deformed image to obtain the length L1 between the two marker points at the moment corresponding to the set loading force A; among them, the length between the two marker points at the maximum loading force is L2.

[0111] When calculating the above L0, L1, and L2, they are all the average distances between the two marker lines to prevent calculation deviations or errors caused by the rotation of the steel strand, etc.

[0112] Step 10 - 3: Measure the large deformation ΔL, then ΔL = L1 - L0.

[0113] As Figure 5 shown, the two calibration points are calibration point one and calibration point two respectively. The deformed image corresponding to the maximum loading force in Step 7 is called the maximum loading deformation image; then the pixel coordinates of calibration point one and calibration point two in the original image before stretching after correction are (x 0-1, y 0-1 ), and (x 0-2 , y 0-2 ); The pixel coordinates of the first calibration point and the second calibration point in the maximum load deformation image after correction are (x 1-1 , y 1-1 ) and (x 1-2 , y 1-2 ), then the calculation formulas for L0 and L1 are respectively:

[0114]

[0115] In the formula, pixel_coefficient is the pixel calibration coefficient.

[0116] Step 10-4, measure A gt , and the specific calculation formula is: A gt = (L2 - L0) / L0 × 100%.

[0117] In this embodiment, taking the deformation of the steel strand and the measurement of Agt as an example, the actual length of the QR code used in the figure is 120 mm, the number of pixels it occupies is 227, and the calibration coefficient pixel_coefficient = 120 / 227 = 0.529 mm. The dotted line is the marked point segment selected in the OpenCV interactive interface. The pixel coordinate difference between the two marked points at the initial moment is:

[0118]

[0119] Then the length L0 between the two marked points at this time = 1417.002 × 0.529 = 749.594 mm.

[0120] Calculate the pixel coordinate difference between the two marked points at the maximum load moment in the same way:

[0121]

[0122] Then the length L2 between the two marked points at this time = 1480.066 × 0.529 = 782.955 mm.

[0123] Then the total deformation of the steel strand L2 - L0 = 782.955 – 749.594 = 33.361 mm.

[0124] Then the total elongation at maximum force of the steel strand material is preferably:

[0125] A gt = (L2 - L0) / L0 × 100% = 33.361 / 749.594 × 100% = 4.45%.

[0126] After the test, make large deformations and A of high-strength linear materialsgt Test report, which includes the curve of the loading force changing with time, and pastes the QR code storing the basic information of the high-strength linear material and the deformation stretching video in the corresponding test report for traceability.

[0127] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A large deformation and A measurement method of high-strength linear materials based on image recognition, characterized in that: gt It includes the following steps: Step 1, making marking points: Select a high-strength linear material with a length of L, and make two marking points on the same side of the high-strength linear material; wherein, both ends of the high-strength linear material are formed as clamping parts; then the two marking points are arranged adjacent to the two clamping parts, and the distance between the two marking points is greater than L / 2; Step 2, clamping the high-strength linear material: Clamp the two clamping parts of the high-strength linear material in the fixture of the testing machine, and both marking points face outward; Step 3, placing the QR code: Place a QR code with a fixed position on the testing machine on one side of the high-strength linear material, and the QR code and the two marking points are located in the same plane M; at the same time, record the original side length dimension of the QR code; Step 4, installing the camera: Install a camera outside the high-strength linear material, and each frame of the captured image of the camera needs to include the QR code and the two marking points; Step 5, stretching the high-strength linear material: The testing machine starts, and axially stretches the clamped high-strength linear material until the high-strength linear material breaks; during the stretching process, the testing machine records the change curve of the applied force over time in real time; Step 6, taking deformed pictures: Before the testing machine starts, the camera first takes a picture of the high-strength linear material as the original picture before stretching; at the same time when the testing machine starts stretching, the camera starts to record the entire stretching process until the high-strength linear material breaks, obtaining a deformed stretching video of the high-strength linear material; Step 7, selecting deformed pictures: First, find the corresponding moment A of the set applied force from the change curve of the applied force over time in Step 5; then, find the deformed picture corresponding to the moment A from the deformed stretching video in Step 6; wherein, the set applied force includes the maximum applied force; Step 8, image correction: Using the QR code in Step 3 as a scale, perform image correction on both the original picture before stretching and each deformed picture selected in Step 7; Step 9, calculating the pixel calibration coefficient: According to the number of pixel points occupied by the side length of the QR code in the corrected deformed picture and the original side length dimension of the QR code recorded in Step 3, calculate the pixel calibration coefficient; Step 10, measure large deformation and A gt , which specifically includes the following steps: Step 10-1, measuring the original length L0 before deformation: Import the original picture before stretching corrected in Step 8 into the OpenCV interactive interface, select the positions of the two marking points, and obtain the length L0 between the two marking points at the initial moment; Step 10-2, measuring the length after deformation: Import each deformed picture corrected in Step 8 into the OpenCV interactive interface; and select the positions of the two marking points in each deformed picture to obtain the length L1 between the two marking points at the corresponding moment A of the set applied force; wherein, the length between the two marking points at the maximum applied force is L2; Step 10-3, measuring the large deformation ΔL, then ΔL = L1 - L0: Step 10-4, measure A gt , and the specific calculation formula is: A gt = (L2 - L0) / L0 × 100%.

2. The large deformation and A of high-strength linear materials based on image recognition according to claim 1 gt , characterized in that: In Step 1, each marking point is a marking line perpendicular to the axis of the high-strength linear material and both are arc-shaped or annular, formed by spraying paint through a strip hole; the distance l2 from each marking point to the corresponding end of the high-strength linear material is greater than the length l1 of the clamping part; in Step 10, when calculating L0, L1 and L2, they are all the average distance between the two marking lines.

3. The large deformation and A of high-strength linear materials based on image recognition according to claim 2 gt measurement method, characterized in that: In step 2, the QR code is printed on cardboard and inserted into the acrylic board. The size of the QR code is a×a, where a = 0.2 - 0.8(L - 2l2). After the QR code is placed, a laser collimation device is used for alignment to ensure that the two marking points and the QR code are on the same plane.

4. The large deformation and A of high-strength linear materials based on image recognition according to claim 3 gt measurement method, characterized in that: In step 4, if the actual length and width of each frame of the camera's captured image are x×y, then: y > 1.2a (1 + α0 + 10%)(L - 2l2) < x < 1.2(L - 2l2) where α0 is the estimated elongation rate of the high-strength linear material, not greater than 8%.

5. The large deformation and A measurement method of high-strength linear materials based on image recognition according to claim 1, characterized in that: gt In step 5, in the curve of the loading force changing with time, the time accuracy is 0.05 s and the load accuracy is 0.01 kN. In step 6, the recording of the deformation stretching video starts simultaneously with the loading of the testing machine, and the selectable time accuracy is also 0.05 s.

6. The large deformation and A of high-strength linear materials based on image recognition according to claim 4 gt , characterized in that: In step 8, image correction includes the following steps: Step 8-1, decoding the QR code: Decode the QR code images in the original picture before stretching and each deformed picture selected in step 7 using the pyzbar package to obtain the coordinates of the four corner points in each QR code image. Step 8-2, image correction: According to the coordinates of the four corner points of the QR code image obtained in step 8-1, use the cv2.getPerspectiveTransform function to calculate the perspective transformation matrix; use the perspective transformation matrix to correct each deformed picture in step 8-1. After correction, the QR code images in each deformed picture are converted into front-view quadrilaterals. The pixels of each corrected deformed picture are x’_pixel×y’_pixel, then: x’_pixel = x_pixel×cosθ y’_pixel = y_pixel×cosθ where x_pixel and y_pixel are the pixel accuracies of the camera. θ is the angle between the camera's captured image plane and the QR code plane.

7. The large deformation and A of high-strength linear materials based on image recognition according to claim 6 gt , characterized in that: In step 4, the number of pixels in the camera's captured image depends on the pixel accuracies x_pixel and y_pixel of the capturing device; x_pixel and y_pixel need to satisfy that, on the premise that the QR code and the two marking points are both complete, close-range zoom shooting should be carried out to make the high-strength linear material fill the captured image as much as possible, then x_pixel needs to satisfy: x_pixel ≥ (1.2 + α0) / 0.02α0cosθ.

8. The large deformation and A of high-strength linear materials based on image recognition according to claim 7 gt The measurement method is characterized in that: In step 9, the calculation formula for the pixel calibration coefficient pixel_coefficient is: pixel_coefficient = a / mark’_pixel and pixel_coefficient < 0.02α0(L - 2l2) In the formula, mark’_pixel is the number of pixels occupied by the side length of the QR code in the corrected image.

9. The large deformation and A of high-strength linear materials based on image recognition according to claim 6 or 7 or 8 gt , characterized in that: θ is less than or equal to 45 degrees.

10. The large deformation and A of high-strength linear materials based on image recognition according to claim 1 gt measurement method, characterized in that: In step 10, the two calibration points are calibration point one and calibration point two respectively. The deformation picture corresponding to the maximum load force in step 7 is called the maximum load deformation picture. The pixel coordinates of calibration point one and calibration point two in the original picture before stretching after correction are (x 0-1 , y 0-1 ) and (x 0-2 , y 0-2 ); the pixel coordinates of calibration point one and calibration point two in the maximum load deformation picture after correction are (x 1-1 , y 1-1 ) and (x 1-2 , y 1-2 ). The calculation formulas for L0 and L1 are respectively: In the formula, pixel_coefficient is the pixel calibration coefficient.

Images