A vision-based cable sheath elongation measurement device and use method

Abstract

The present invention discloses a vision-based cable protective sleeve elongation measurement device, including a tensile machine, a pair of workpiece mounting slots are provided on the tensile machine, and a dumbbell is installed between the pair of workpiece mounting slots, a tension sensor is installed at the lower end of the workpiece mounting slot, and an industrial camera is also included, the industrial camera is connected to a host through a network port electrical signal, the dumbbell is provided with two marking lines, and the industrial camera is used to shoot the two marking lines; the tensile machine electrical signal is connected to a controller. The present invention is different from other stretching length usage methods. By collecting images of the stretching process of the cable material, when the cable material breaks during the stretching experiment, the software automatically identifies the workpiece break and stops collecting data. By pre-marking straight lines on the cable sheath, a system for automatically identifying the elongation at break is designed. The system can also realize online detection of multiple test pieces according to the parameters of the industrial camera and the shape of the workpiece slot.

Description

Technical Field

[0001] The invention relates to the technical field of cable protective sleeve measurement, and in particular to a vision-based cable protective sleeve elongation measurement device and a use method. Background Art

[0002] At present, there are two traditional methods for using the stretching length of cable materials. One is to overlap the marking line with the zero scale line of the ruler to try to ensure that the measured distance is the distance of the critical instantaneous marking line. The other is to use the two contacts of the contact tension sensor to overlap with the marking line. When the workpiece is stretched, one contact does not move and the other contact moves with the marking line. Although this method can ensure that the measured distance is relatively accurate, the contact of the sensor contact with the workpiece to be measured will generate force, which may cause the workpiece to be measured to break at the marking line, causing the experiment to fail. Therefore, the two traditional methods of using the stretching length do not have high practical value in actual applications.

[0003] With the development of machine vision technology, some scholars have proposed a measurement method based on machine vision. However, due to the variety of colors of actual cable sheaths, the color of the test specimens is uncertain, and the length of the cable sheath tensile fracture is also uncertain, but no specific solutions have been given. Summary of the invention

[0004] The purpose of the present invention is to provide a device for measuring the elongation of a cable protective sheath based on vision, so as to solve the problem that it is difficult to accurately measure the elongation of a cable protective sheath.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a vision-based cable protective sleeve elongation measurement device, comprising a tensile testing machine, wherein the tensile testing machine is provided with a pair of workpiece mounting slots, and a dumbbell is installed between the pair of workpiece mounting slots, and a tensile sensor is installed at the lower end of the workpiece mounting slot, and also comprises an industrial camera, wherein the industrial camera is connected to a host via an electrical signal of a network port, the dumbbell is provided with two marking lines, and the industrial camera is used to shoot the two marking lines; the electrical signal of the tensile testing machine is connected to a controller.

[0006] The present invention also provides the following method of use, comprising the following steps:

[0007] A. First, use the control button on the controller to move the lower workpiece mounting slot downward, put the dumbbell into the workpiece mounting slots on both sides and clamp the lower side of the dumbbell. Then, click the button on the controller to move the lower workpiece mounting slot upward to the appropriate position and clamp the upper side of the dumbbell.

[0008] B. After the dumbbell is clamped, the controller is operated to start the tension sensor to work. The tension sensor starts to drive the workpiece mounting slot on the lower side to move downward. At the same time, the industrial camera collects images of the dumbbell during its stretching process and transmits the collected images to the host for image processing in real time to determine whether the dumbbell is broken.

[0009] C. If the dumbbell has not broken, the industrial camera will continue to collect data until the dumbbell breaks. At the same time, the tension sensor detects the instantaneous tension and stops moving downward. The host controls the industrial camera to stop collecting data and takes the previous frame as the critical frame to measure the length between the two marking lines, thereby calculating the elongation at break of the dumbbell.

[0010] Compared with the prior art, the beneficial effects of the present invention are: proposing to first detect the specific color of the test piece, and to divide the acquired colors into two categories of black and white, to classify black, blue and red into the black category, and to classify yellow, green and white into the white category, and to extract and lift the precise region of interest algorithm suitable for test pieces with different stretching lengths, thereby solving practical problems in industrial applications.

[0011] The present invention is different from other methods of using the stretching length. By collecting images of the stretching process of the cable material, when the cable material breaks during the stretching test, the software automatically identifies the workpiece break and stops collecting data. By pre-marking a straight line on the cable sheath, a system for automatically identifying the elongation at break is designed. The system can also realize online detection of multiple test pieces according to the parameters of the industrial camera and the shape of the workpiece slot. After comparative analysis of the system experiment and the traditional experiment, the system exceeds the traditional detection system in terms of efficiency, accuracy and intelligence. At the same time, the proposed solution also solves the actual engineering problem well. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a schematic diagram of the structure of the measuring device of the present invention.

[0013] Figure 2 It is a schematic diagram of the dumbbell specimen structure of the present invention.

[0014] Figure 3 It is a schematic diagram of the structure of the small dumbbell specimen of the present invention.

[0015] Figure 4 This is the critical frame diagram of the present invention.

[0016] Figure 5 It is the effective area segmentation diagram of the present invention.

[0017] Figure 6 This is a diagram of the cable sheath fracture of the present invention.

[0018] Figure 7This is a critical frame region of interest diagram of the present invention.

[0019] Figure 8 It is a color recognition flow chart of the present invention.

[0020] Fig. 9 This is a binary image of the black dumbbell-like parts of the present invention.

[0021] Fig.10 This is the corrosion image of the black dumbbell-like piece of the present invention.

[0022] Fig.11 The black dumbbell-like binary image of the present invention is a difference image with the eroded image.

[0023] Fig.12 It is the black dumbbell-like filtered image of the present invention.

[0024] Fig.13 This is a diagram of the white dumbbell resisting illumination of the present invention.

[0025] Fig.14 It is the white dumbbell-like binary image of the present invention.

[0026] 1. Tensile testing machine; 2. Workpiece mounting slot; 3. Marking line; 4. Dumbbell; 5. Tensile sensor; 6. LED strip light; 7. Network port; 8. Industrial camera; 9. Host; 10. Controller. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] See also Figure 1 The present invention provides a technical solution: a vision-based cable sheath elongation measurement device, comprising a tensile machine 1, a pair of workpiece mounting slots 2 are provided on the tensile machine 1, a dumbbell 4 is installed between the pair of workpiece mounting slots 2, a tension sensor 5 is installed at the lower workpiece mounting slot 2, and an industrial camera 8 is also included. The industrial camera 8 is connected to a host 9 through a network port 7 electrical signal, two marking lines 3 are provided on the dumbbell 4, and the industrial camera 8 is used to shoot the two marking lines 3; the tensile machine 1 is electrically connected to a controller 10. A pair of LED strip lights 6 are provided between the industrial camera 8 and the two marking lines 3 to ensure stable lighting conditions during the experiment.

[0029] The present invention also provides a method for using the measuring device, comprising the following steps:

[0030] A. First, use the control button on the controller 10 to move the lower workpiece mounting slot 2 downward, put the dumbbell 4 between the workpiece mounting slots 2 on both sides and clamp the lower side of the dumbbell 4, then click the button of the controller 10 to move the lower workpiece mounting slot 2 upward to a suitable position, and clamp the upper side of the dumbbell 4.

[0031] B. After the dumbbell 4 is clamped, the controller 10 is operated to start the tension sensor 5 to work. The tension sensor 5 starts to drive the lower workpiece mounting slot 2 to move downward. At the same time, the industrial camera 8 collects images of the stretching process of the dumbbell 4, and transmits the collected images to the host 9 in real time for image processing to determine whether the dumbbell is broken.

[0032] C. If the dumbbell 4 has not broken, the industrial camera 8 will continue to collect data until the dumbbell 4 breaks. When the dumbbell 4 breaks, the tension sensor 5 detects that the instantaneous tension is 0 and stops moving downward. The host 9 controls the industrial camera 8 to stop collecting data and takes the previous frame as the critical frame to measure the length between the two marking lines 3, thereby calculating the elongation at break of the dumbbell 4.

[0033] The test is carried out at a temperature of 23±5℃. The moving speed of the tensile machine during the test should be 250±50mm / min; when the insulating material is PE or PP (PE is polyethylene, PP is polypropylene), the moving speed of the tensile machine should be 25±5mm / min.

[0034] In the process of dividing the marking line 3, the workpieces to be tested include a dumbbell specimen and a small dumbbell specimen. The length between the marking lines of the dumbbell specimen is 20 mm, and the width of the tensile part in the middle of the specimen is 4 mm, as shown below Figure 2 The size diagram is shown in Figure 1. The length between the marking lines of the small dumbbell specimen is 10 mm, and the width of the tensile part in the middle of the specimen is the same as that of the dumbbell specimen (such as Figure 3 shown).

[0035] In the selection process of industrial camera 8, since the camera resolution in width and height should be kept moderate, and in order to improve the stability of the system, 3 to 4 pixels should correspond to a certain length as much as possible, and the resolution in width should meet the requirements, so a 1280×960 camera can be selected. With the discovery of semiconductors, CMOS sensors appeared around us. By integrating their integrated circuits, CMOS industrial cameras are formed. As a key part of the machine vision system, industrial cameras are divided into CCD cameras and CMOS cameras. At present, CCD cameras have less image noise and relatively better quality. Compared with CMOS cameras, CCD cameras are more expensive. However, with the continuous development of technology, the quality of CMOS sensor imaging is not inferior, and the frame rate is also very high, with a very good cost performance. With the continuous enhancement and improvement of CMOS chip functions, CMOS chips will become the leader in the field of industrial cameras in the future. Considering various factors, a CMOS camera was selected, and the lens model M0814-MP2 was selected while ensuring a suitable working distance and field of view.

[0036] In the actual cable sheath tensile test, due to the endless emergence of cable sheaths of various materials, a slight difference in cable thickness will result in a large difference in the length when the cable breaks during the tensile test. It is very important to obtain the appropriate cutting width.

[0037] Assume that the required width is h

[0038]

[0039] Where p(i,j) is the pixel value with coordinates (i,j) in the binary image

[0040]

[0041] in p (i, j) is the pixel value of the coordinate (i, j) in the binary image. Assuming that a whole column of 0 appears when the horizontal coordinate is x, the required appropriate width h = x-30 (the purpose is to remove the excess width), as follows Figure 4 As shown in the picture, there is a slot on the right side of the picture, and all the tensile testing machines have similar structures on both sides for clamping the workpiece to be tested. Figure 5 Shown is a region of interest cropped to an appropriate width.

[0042] In the tensile test of the cable sheath material, the video is converted into frames of pictures, and then the pictures are processed in sequence. For each frame of the picture, the region of interest is first intercepted and binarized. It is assumed that the resolution of the region of interest intercepted by each picture is W*H. Set the fracture width threshold width=z.

[0043]

[0044] Where p(i,j) is the pixel value with coordinates (i,j) in the binary image. The width of the entire column of white pixels is:

[0045]

[0046] If the width w of the entire column of white pixels is greater than or equal to z, the current frame is considered to be a broken frame. Assuming the number of frames is N, the critical frame required is the N-1th frame. Figure 6 The Nth frame is shown. Figure 7 The N-1 frame is shown.

[0047] Since various dumbbells have different colors, there are red, white, black, yellow, green, blue and other colors. After segmenting the image of the region of interest, the color of the workpiece should be identified, and the workpieces of different colors should be classified and processed, with blue, black and red being classified into one category and white, yellow and green being classified into another category. Since the workpiece to be measured is fixed in the same position each time, when performing color recognition, the rectangular region of interest is first intercepted. The size of the workpiece to be measured is known, and the appropriate rectangle width can be selected. Since the position is fixed, it can be guaranteed that the intercepted rectangular frame is full of the workpiece.

[0048] Assuming that the resolution of the captured region of interest is W*H, whether its H, S, and V values ​​meet the color region range condition is:

[0049]

[0050] Where l = (h, s, v), p (i, j) [k] (i = 0, 1, 2 ..., W; j = 0, 1, 2 ..., H; k = 1, 2, 3) represent the values ​​of the H, S, V channels of a certain pixel point. Define the colors of the workpiece to be identified: red, yellow, blue, green, white, black, and initialize them to Red = 0, Yellow = 0, Blue = 0, Green = 0, White = 0, Black = 0. Table 1 shows the color threshold range.

[0051]

[0052] Table 1

[0053] If p(i,j)[k](k=1,2,3) satisfies a certain color range at the same time, each time it is satisfied, the value of the corresponding color variable is the current value plus 1. After accumulation, the maximum value is obtained through max{Red,Yellow,Blue,Green,White,Black}, and the maximum value is compared with each color variable in turn to obtain the actual color of the workpiece to be tested. The process is as follows Figure 8 shown.

[0054] Analysis of black artifacts:

[0055] In the cable sheath tensile test, for the three colors of blue, black and red workpieces to be tested, after binarization through the maximum inter-class variance, the workpieces to be tested are all black, so these three colors can be classified into one category for processing. By marking a white straight line on the blue, black and red workpieces to be tested, the image is grayed out, and after binarization, the resulting image will show a break in the middle. Then the image is corroded left and right, and the broken area of ​​the workpiece to be tested is connected to the workpieces on the left and right sides. By subtracting the corroded left and right images from the uncorroded binary image, the marked straight line is obtained. Assume that the binary image is x k , the left and right eroded images are x k+1 , then the difference image is:

[0056] Δx k =x k+1 -x k (6)

[0057] as follows Fig.11 The following is a differential image. After filtering the differential image, only a clear white outline is obtained, as shown below Fig.12 As shown, by extracting the contour of the image and then finding the centroid between the two contours, the distance between the marked lines can be obtained.

[0058] The differential image is filtered to obtain only a clear white outline. By extracting the outline of the image and then finding the centroid between the two outlines, the distance between the marking lines can be obtained.

[0059] White workpiece analysis:

[0060] In the tensile test of the white, yellow and green workpiece, the segmented region of interest is grayed to obtain a grayscale image A. Assuming its resolution is K*G, assuming its average grayscale value is:

[0061]

[0062] Divide the grayscale image A into sub-blocks of size P*Q. Assume that there are n such sub-blocks. Set each sub-block image to I k(k=1,2…,n) then the average gray value of each sub-block is:

[0063]

[0064] Subtract the average grayscale value of the sub-block from the average grayscale image A, and the difference is:

[0065]

[0066] Where k = 1, 2…, n, Δ k (k=1,2…,n) are combined into a difference matrix E, and the resolution of the matrix E is changed to image B with a resolution of K*G by interpolation. Then the pixel value of image C after illumination compensation is:

[0067] C(i,j)=A(i,j)-B(i,j) (10)

[0068] Where C(i,j), A(i,j), B(i,j) are the pixel values ​​of the coordinates (i,j) in Figure C, the pixel values ​​of the coordinates (i,j) in Figure A, and the pixel values ​​of the coordinates (i,j) in Figure B. Binarizing the image after illumination compensation, we can clearly see that the outline of the marker line is very clear, as shown below Fig.13 , 14 Shown are the resist illumination map and binarization map of the green artifact.

[0069] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A visual-based cable sheath elongation measurement device, comprising a tensile machine (1), wherein the tensile machine (1) is provided with a pair of workpiece mounting slots (2), a dumbbell (4) is installed between the pair of workpiece mounting slots (2), and a tensile force sensor (5) is installed at the lower end of the workpiece mounting slot (2), characterized in that: It also includes an industrial camera (8), the industrial camera (8) is connected to a host (9) via an Internet port (7) via an electrical signal, the dumbbell (4) is provided with two marking lines (3), and the industrial camera (8) is used to photograph the two marking lines (3); the tensile machine (1) is connected to a controller (10) via an electrical signal; and a pair of LED strip lights (6) are provided between the industrial camera (8) and the two marking lines (3); When in use, firstly, the control button on the controller (10) is used to move the workpiece mounting slot (2) on the lower side downward, the dumbbell (4) is placed between the workpiece mounting slots (2) on both sides and the lower side of the dumbbell (4) is clamped, and then the button of the controller (10) is clicked to move the workpiece mounting slot (2) on the lower side upward to a suitable position, and the upper side of the dumbbell (4) is clamped; After the dumbbell (4) is clamped, the controller (10) is operated to start the tension sensor (5) to start working, and the tension sensor (5) starts to drive the workpiece mounting slot (2) on the lower side to move downward, and at the same time, the industrial camera (8) collects images of the stretching process of the dumbbell (4), and transmits the collected images to the host (9) in real time for image processing to determine whether the dumbbell is broken; If the dumbbell (4) has not yet broken, the industrial camera (8) will continue to collect data until the dumbbell (4) breaks. When the dumbbell (4) breaks, the tension sensor (5) detects that the instantaneous tension is 0 and stops moving downward. The host (9) controls the industrial camera (8) to stop collecting data and takes the previous frame as a critical frame to measure the length between the two marking lines (3), thereby calculating the elongation at break of the dumbbell (4). And because the cable sheaths of various materials are different, it is necessary to determine the appropriate cutting width: Assume that the required width is h Where p(i,j) is the pixel value with coordinates (i,j) in the binary image in p (i, j) is the pixel value with coordinate (i, j) in the binary image; assuming that a whole column is 0 when the horizontal coordinate is x, the required width is h = x-30; The industrial camera (8) collects images of the stretching process of the dumbbell (4), and transmits the collected images to the host (9) in real time for image processing to determine whether the dumbbell is broken: By converting the video into frames of pictures, and then processing the pictures in sequence, for each frame of the picture, first intercept the region of interest, and perform binary processing on the region of interest. Assuming that the resolution of the region of interest intercepted by each picture is W*H, set the fracture width threshold width=z; Where p(i,j) is the pixel value with coordinates (i,j) in the binary image; the width of the entire column of white pixels is: If the width w of the entire column of white pixels is greater than or equal to z, the current frame is considered to be a broken frame. Assuming the number of frames is N, the critical frame required is the N-1th frame.

2. A method for using the visual-based cable sheath elongation measurement device according to claim 1, characterized in that: The following steps are included A. First, use the control button on the controller (10) to move the lower workpiece mounting slot (2) downward, put the dumbbell (4) between the two sides of the workpiece mounting slot (2) and clamp the lower side of the dumbbell (4), then click the button of the controller (10) to move the lower workpiece mounting slot (2) upward to a suitable position, and clamp the upper side of the dumbbell (4); B. After the dumbbell (4) is clamped, the controller (10) is operated to start the tension sensor (5) to start working, and the tension sensor (5) starts to drive the lower workpiece mounting slot (2) to move downward, and at the same time, the industrial camera (8) collects images of the stretching process of the dumbbell (4), and transmits the collected images to the host (9) in real time for image processing to determine whether the dumbbell is broken; C. If the dumbbell (4) has not yet broken, the industrial camera (8) will continue to collect data until the dumbbell (4) breaks. When the dumbbell (4) breaks, the tension sensor (5) detects that the instantaneous tension is 0 and stops moving downward. The host (9) controls the industrial camera (8) to stop collecting data and takes the previous frame as a critical frame to measure the length between the two marking lines (3), thereby calculating the elongation at break of the dumbbell (4); And because the cable sheaths of various materials are different, it is necessary to determine the appropriate cutting width: Assume that the required width is h Where p(i,j) is the pixel value with coordinates (i,j) in the binary image in p (i, j) is the pixel value with coordinate (i, j) in the binary image; assuming that a whole column is 0 when the horizontal coordinate is x, the required width is h = x-30; The industrial camera (8) collects images of the stretching process of the dumbbell (4), and transmits the collected images to the host (9) in real time for image processing to determine whether the dumbbell is broken: By converting the video into frames of pictures, and then processing the pictures in sequence, for each frame of the picture, first intercept the region of interest, and perform binary processing on the region of interest. Assuming that the resolution of the region of interest intercepted by each picture is W*H, set the fracture width threshold width=z; Where p(i,j) is the pixel value with coordinates (i,j) in the binary image; the width of the entire column of white pixels is: If the width w of the entire column of white pixels is greater than or equal to z, the current frame is considered to be a broken frame. Assuming the number of frames is N, the critical frame required is the N-1th frame.

3. The method for using the visual-based cable sheath elongation measurement device according to claim 2, characterized in that: The test was carried out at a temperature of 23±5℃.

4. The method for using the visual-based cable sheath elongation measurement device according to claim 2, characterized in that: During the test, the moving speed of the tensile machine should be 250±50mm / min; when the insulating material is PE or PP, the moving speed of the tensile machine should be 25±5mm / min.

Images