A Visual Measurement System for Film Size on a Conveyor Belt and Its Measurement Method

Abstract

The present invention provides a visual measurement system for the size of a film on a conveyor belt and a measurement method thereof. The measurement system includes a main frame, a conveyor belt assembly, an infrared LED backlight, a camera, a precision sliding table, and a control device. The conveyor belt assembly is installed on the main frame, and an infrared LED backlight is arranged directly below the conveying surface of the conveyor belt assembly. A camera is arranged directly above the conveying surface of the conveyor belt assembly. The camera is connected to the control device, and the acquisition part of the camera is arranged facing the conveying surface and the infrared LED backlight. The camera is installed on the precision sliding table, and the precision sliding table is installed on the main frame. The measurement method uses this measurement system for measurement. While having a high size measurement accuracy, it is more convenient and fast, and easy to operate. The present invention can solve the technical problems of poor imaging quality of the detected film, slow measurement speed, and complex calibration method in the prior art.

Description

Technical Field

[0001] The present invention relates to the field of machine vision, especially vision measurement technology, and particularly to the technology of real-time measuring the size of an object. Specifically, it is a visual measurement system for the size of a film on a conveyor belt and its measurement method. Background Art

[0002] After the rubber sheet used in the tire production process goes through the cutting process, it is necessary to measure the geometric dimensions of the film, including geometric parameters such as length, width, and angle, to determine whether it meets the production requirements. Good dimensional accuracy of the film is of great significance to the quality of the finished tire. In the existing production line, manual sampling inspection is mostly used, with low detection efficiency and poor accuracy, which cannot meet the needs of enterprises to produce high-quality tires.

[0003] With the improvement of production efficiency, the working speed of the film conveyor belt has increased. The existing visual measurement system has problems such as the reflection light source being unable to provide sufficient incident light and poor image acquisition quality; using a 3D sensor for image acquisition results in a high system complexity and slow measurement speed. At the same time, the calibration and calibration of the camera in the existing visual measurement method are relatively complex, and there are problems such as a large number of auxiliary devices being required during the calibration process, high cost of the devices, and difficulty in operation. How to design a simple and easy-to-use film measurement system and method while ensuring imaging quality, measurement speed, and accuracy has become an urgent problem to be solved. Summary of the Invention

[0004] In view of the above, it is necessary to provide a visual measurement system for the size of a film on a conveyor belt and its measurement method to solve the technical problems of poor imaging quality, slow measurement speed, and complex calibration method in the existing technology.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] A visual measurement system for the size of a film on a conveyor belt includes a main frame, a conveyor belt assembly, an infrared LED backlight, a camera, a precision sliding table, and a control device; the conveyor belt assembly is installed on the main frame, and an infrared LED backlight is arranged directly below the conveying surface thereof. A camera is arranged directly above the conveying surface of the conveyor belt assembly. The camera is connected to the control device. The acquisition part of the camera is arranged facing the conveying surface and the infrared LED backlight. The camera is installed on the precision sliding table, and the precision sliding table is installed on the main frame.

[0007] Preferably, the camera is an industrial line array camera, and the control device is an industrial control computer.

[0008] Preferably, the precision sliding table is installed on the main frame through a camera support.

[0009] In addition, the present invention also provides a method for visually measuring the size of a film on a conveyor belt, which uses the above-mentioned visual measurement system for the size of a film on a conveyor belt to perform the measurement. The measurement method includes the following steps:

[0010] S1. Calibrate and calibrate the camera, specifically including angle calibration and length calibration. The steps are as follows:

[0011] S1.1. Perform angle calibration:

[0012] S1.1.1. The conveyor belt assembly runs at a low speed, and a checkerboard calibration board is placed on its conveying surface. The checkerboard calibration board moves to directly below the camera acquisition part as the conveyor belt assembly runs;

[0013] S1.1.2. Adjust the camera to rotate around the X and Y axes through the precision slide. When the checkerboard calibration board is located at the center of the image and the imaging is clear, the plane where the camera optical axis is located is parallel to the conveying surface of the conveyor belt assembly;

[0014] S1.1.3. Coarsely adjust the precision slide to rotate within the range of ±5 degrees along the Z axis, rotate 1 degree each time. After each rotation, the camera acquires the image information of the checkerboard calibration board once and transmits it to the control device; then, in the control device, use the corner detection algorithm to obtain the pixel coordinates of the corner points U, V, and W on the checkerboard calibration board, calculate the distances between the UV connection line, VW connection line, and UW connection line according to the obtained pixel coordinates, and then use the cosine theorem Calculate the included angle θ between the UV connection line and the VW connection line; select two images with the included angle θ closest to 90 degrees in the image information and record the corresponding rotation angles of the precision slide as x1 and x2.

[0015] S1.1.4. Within the interval of x1 and x2, use a micrometer to finely adjust the rotation of the precision slide around the X and Y axes, rotate 0.1 degree each time. After each rotation, the camera acquires the image information of the checkerboard calibration board once and transmits it to the control device; then, use the calculation method in step S1.1.3 to calculate the included angle θ between the UV connection line and the VW connection line on the checkerboard calibration board; select two images with the included angle θ closest to 90 degrees in the image information and record the corresponding rotation angles of the precision slide as x3 and x4;

[0016] S1.1.5. Within the interval of x3 and x4, use a micrometer to finely adjust the rotation of the precision slide around the X and Y axes, rotate 0.01 degree each time. After each rotation, the camera acquires the image information of the checkerboard calibration board once and transmits it to the control device; then, use the calculation method in step S1.1.3 to calculate the included angle θ between the UV connection line and the VW connection line on the checkerboard calibration board; select the rotation angle of the precision slide with the included angle closest to 90 degrees in the image information as the angle calibration result;

[0017] S1.2. Perform length calibration:

[0018] S1.2.1. Adjust the conveyor belt assembly to the rated operating speed;

[0019] S1.2.2. Calculate the distortion of the image in the velocity direction using the cross-ratio invariance of the image: Substitute the world coordinates of any four collinear grid corner points A, B, C, and D on the checkerboard calibration plate plane into the formula Calculate the cross-ratio of the world coordinates;

[0020] S1.2.3. Extract the image coordinates A1, B1, C1, and D1 corresponding to the four grid corner points A, B, C, and D from the image information, and calculate the cross-ratio of their image coordinates CR0(A1, B1; C1, D1);

[0021] S1.2.4. Calculate the difference C between the current cross-ratio of the image coordinates CR0(A1, B1; C1, D1) and the cross-ratio of the world coordinates CR(A, B; C, D).

[0022] S1.2.5. Perform frequency division processing on the camera signal acquisition frequency. When C < 0, increase the signal acquisition frequency by 0.1 times each time to stretch the image; when C > 0, decrease the signal acquisition frequency by 0.1 times each time to compress the image until the absolute value of the cross-ratio difference C is minimized, and complete the length calibration;

[0023] S1.3. Calculate the ratio of the length and width of the checkerboard calibration plate in the pixel coordinate system to that in the world coordinate system as the scale factor K1 for size calculation;

[0024] S2. The conveyor belt assembly operates to start conveying the film;

[0025] S3. The infrared LED backlight emits infrared light to illuminate the film background;

[0026] S4. The camera acquires the image information of the film on the conveyor belt assembly and sends the acquired image information to the control device;

[0027] S5. The control device performs image processing on the image information, calculates the size information of the film, and then displays the obtained result.

[0028] Further, in step S5, the control device calculates the size information of the film through the following steps:

[0029] S5.1. Filter the image information to remove image noise;

[0030] S5.2. Perform binarization processing on the image information;

[0031] S5.3. Use the Prewitt operator to perform edge detection on the image information to obtain the single-pixel edge of the film.

[0032] S5.4. Use the Hough transform to fit the edge lines to obtain the line equations.

[0033] S5.5. Solve the line equations to obtain the intersection coordinates, and then perform corner detection to remove the redundant intersection coordinates.

[0034] S5.6. Calculate the distances between the intersections based on the remaining intersection coordinates, and then obtain the film size in the pixel coordinate system.

[0035] S5.7. The product of the film size in the pixel coordinate system and the scale factor K1 is the actual film size.

[0036] S5.8. Calculate the angles between the lines based on the slopes of the line equations, and then calculate the interior angles of the film.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. The visual measurement system for the film size on the conveyor belt provided by the present invention can meet the requirement of real-time measurement of the film size when the conveyor belt is running at high speed by using a high-speed linear array camera in cooperation with an infrared LED backlight, which is beneficial to subsequent production and the improvement of product quality.

[0039] 2. The measurement system of the present invention does not need to modify the existing production line and can be directly installed, which has good economic benefits and a good market prospect.

[0040] 3. The visual measurement method for the film size on the conveyor belt provided by the present invention can accurately measure the geometric size of the film and detect unqualified films, which is convenient for subsequent production and beneficial to the improvement of product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic structural diagram of a film size measurement system on a conveyor belt according to the present invention;

[0042] Figure 2 is a schematic working diagram of an infrared LED backlight;

[0043] Figure 3 is a schematic structural diagram of a precision sliding table;

[0044] Figure 4 is a working flow chart of a measurement method of a film size measurement system on a conveyor belt according to the present invention;

[0045] Figure 5 is a flow chart of the angle calibration steps of the camera;

[0046] Figure 6 Flow chart of the length calibration steps for the camera

[0047] Figure 7 Schematic diagram of the checkerboard calibration board

[0048] Figure 8 Schematic diagram of the principle of dimension calibration

[0049] Figure 9 Schematic diagram of the measured dimensions of the film

[0050] Explanation of the main component symbols

[0051] In the figure: 1 - control device, 2 - image acquisition device, 3 - precision slide table, 4 - camera bracket, 5 - conveyor belt, 6 - film, 7 - infrared LED backlight

[0052] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings Specific embodiments

[0053] Please refer to Figures 1 to 3 , in a preferred embodiment of the present invention, a visual measurement system for the size of a film on a conveyor belt includes a main frame, a conveyor belt assembly 5, an infrared LED backlight 7, a camera 2, a precision slide table 3, and a control device 1; the conveyor belt assembly 5 is installed on the main frame, and an infrared LED backlight 7 is arranged directly below the conveying surface thereof. A camera 2 is arranged directly above the conveying surface of the conveyor belt assembly 5. The camera 2 is connected to the control device 1. The acquisition part of the camera 2 is arranged facing the conveying surface and the infrared LED backlight 7. The camera 2 is installed on the precision slide table 3, and the precision slide table 3 is installed on the main frame, preferably installed on the main frame through a camera bracket 4

[0054] In the present invention, the rotation of the precision slide table 3 in the X, Y, and Z directions can realize the adjustment of the pitch angle, yaw angle, and roll angle of the camera 2, so as to adjust the position and attitude of the camera 2 to ensure accurate imaging without deformation. The camera 2 is used to collect the image information of the film 6 and send the collected image information to the control device 1. The control device 1 processes the image information and calculates the size of the film 6, thereby judging whether the size of the film 6 is qualified and displaying the unqualified film 6. Preferably, in this embodiment, the camera 2 is an industrial line array camera, and the control device 1 is an industrial control computer, and the industrial line array camera is connected to the industrial control computer

[0055] In addition, it should be noted that most existing vision measurement systems use visible light as the backlight or reflection light source for illumination. However, as the acquisition speed of camera 2 increases, the exposure time of camera 2 has been shortened to the microsecond level. Due to the blockage of the conveying surface of the conveyor belt assembly 5 and the light absorption characteristics of the rubber film 6, visible light cannot provide sufficient incident light to meet the shooting needs of camera 2 whether it is used as a backlight or a reflection light source, and it is easily affected by the dirt on the conveyor belt, making it difficult to ensure the measurement accuracy. In contrast to visible light, infrared light has better penetrability, and camera 2 has a similar absorption rate for infrared light and visible light. Under the blockage of the conveying surface of the conveyor belt assembly 5, the infrared LED backlight 7 can still provide sufficient incident light for camera 2, meeting the shooting needs when the exposure time of camera 2 is at the microsecond level. That is, the infrared LED backlight 7 of the present invention is arranged directly below the conveying surface of the conveyor belt assembly 5, and the infrared light emitted by it can pass through the conveying surface of the conveyor belt assembly 5 but cannot pass through the film 6 and enter the lens of camera 2, and the influence of the dirt on the conveyor belt on camera 2 when obtaining images is relatively small, so that the edge of the film in the collected image information is clearer, which is beneficial to reducing the complexity of the later algorithm.

[0056] Please refer to Figures 4 - 9 , based on the above-mentioned vision measurement system for the film size on a conveyor belt, the present invention provides a vision measurement method for the film size on a conveyor belt, which uses the above-mentioned vision measurement system for the film size on a conveyor belt for measurement. The measurement method includes the following steps:

[0057] S1. Calibrate and calibrate the camera 2, specifically including angle calibration and length calibration. The steps are as follows:

[0058] S1.1. Perform angle calibration:

[0059] S1.1.1. The conveyor belt assembly 5 runs at a low speed, and a checkerboard calibration board is placed on its conveying surface. The checkerboard calibration board moves to directly below the acquisition part of camera 2 as the conveyor belt assembly 5 runs.

[0060] S1.1.2. Adjust the rotation of camera 2 around the X and Y axes through the precision slide 3. When the checkerboard calibration board is at the center of the image and the imaging is clear, the plane where the optical axis of camera 2 is located is parallel to the conveying surface of the conveyor belt assembly 5.

[0061] S1.1.3. Coarsely adjust the precision slide 3 to rotate within the range of ±5 degrees along the Z axis, rotate 1 degree each time. After each rotation, camera 2 acquires the image information of the checkerboard calibration board once and transmits it to the control device 1. Then, in the control device 1, use the corner detection algorithm to obtain the pixel coordinates of the corner points U, V, and W on the checkerboard calibration board, calculate the distances between the UV connection line, VW connection line, and UW connection line according to the obtained pixel coordinates, and then use the cosine theorem Calculate the included angle θ formed between the UV connection line and the VW connection line; select two images in the image information where the included angle θ is closest to 90 degrees and record the corresponding rotation angles x1 and x2 of the precision slide table 3.

[0062] S1.1.4. Within the range of x1 and x2, use a micrometer to finely adjust the rotation of the precision slide table 3 around the X and Y axes by 0.1 degree each time. After each rotation, the camera 2 acquires the image information of the checkerboard calibration plate once and transmits it to the control device 1; then, use the calculation method in step S1.1.3 to calculate the included angle θ formed between the UV connection line and the VW connection line on the checkerboard calibration plate; select two images in the image information where the included angle θ is closest to 90 degrees and record the corresponding rotation angles x3 and x4 of the precision slide table 3.

[0063] S1.1.5. Within the range of x3 and x4, use a micrometer to finely adjust the rotation of the precision slide table 3 around the X and Y axes by 0.01 degree each time. After each rotation, the camera 2 acquires the image information of the checkerboard calibration plate once and transmits it to the control device 1; then, use the calculation method in step S1.1.3 to calculate the included angle θ formed between the UV connection line and the VW connection line on the checkerboard calibration plate; the rotation angle of the precision slide table 3 where the included angle is closest to 90 degrees in the image information is the angle calibration result.

[0064] S1.2. Perform length calibration:

[0065] S1.2.1. Adjust the conveyor belt assembly 5 to the rated operating speed.

[0066] S1.2.2. Use the cross - ratio invariance of the image to calculate the distortion of the image in the velocity direction: Substitute the world coordinates of any four collinear grid corner points A, B, C, and D on the checkerboard calibration plate plane into the formula Calculate the cross - ratio of the world coordinates.

[0067] S1.2.3. Extract the image coordinates A1, B1, C1, and D1 corresponding to the four grid corner points A, B, C, and D in the image information, and calculate the cross - ratio of the image coordinates CR0(A1, B1; C1, D1). The cross - ratio of the world coordinates of these four grid corner points should be equal to the cross - ratio of the image coordinates when the image is undistorted, that is, CR0(A1, B1; C1, D1) = CR(A, B; C, D).

[0068] S1.2.4. Calculate the difference C between the current cross - ratio of the image coordinates CR0(A1, B1; C1, D1) and the cross - ratio of the world coordinates CR(A, B; C, D).

[0069] S1.2.5. Divide the signal acquisition frequency of the camera 2. When C < 0, increase the signal acquisition frequency by 0.1 times each time to stretch the image; when C > 0, decrease the signal acquisition frequency by 0.1 times each time to compress the image until the absolute value of the cross-ratio difference C is minimized, and complete the length calibration;

[0070] S1.3. Calculate the ratio of the length and width of the checkerboard calibration plate in the pixel coordinate system to that in the world coordinate system as the scale factor K1 for size calculation.

[0071] S2. The conveyor belt assembly 5 operates to start conveying the film 6.

[0072] S3. The infrared LED backlight 7 emits infrared light to illuminate the background of the film 6.

[0073] S4. The camera 2 collects the image information of the film 6 on the conveyor belt assembly 5 and sends the collected image information to the control device 1.

[0074] S5. The control device 1 performs image processing on the image information, calculates the size information of the film 6, and then displays the obtained result. In this step, the control device 1 calculates the size information of the film 6 through the following steps:

[0075] S5.1. Filter the image information to remove image noise;

[0076] S5.2. Perform binarization processing on the image information;

[0077] S5.3. Use the Prewitt operator to perform edge detection on the image information to obtain the single-pixel edge of the film 6;

[0078] S5.4. Use the Hough transform to fit the edge straight line to obtain the straight line equation;

[0079] S5.5. Solve the straight line equation to obtain the intersection coordinates, and then perform corner detection to remove the redundant intersection coordinates;

[0080] S5.6. Calculate the distance between the intersections through the remaining intersection coordinates, and then obtain the size of the film 6 in the pixel coordinate system;

[0081] S5.7. The product of the size of the film 6 in the pixel coordinate system and the scale factor K1 is the actual size of the film 6;

[0082] S5.8. Calculate the included angle between the straight lines through the slope of the straight line equation, and then calculate the interior angle value of the film 6.

[0083] In step S5, after the control device 1 calculates the size information of the film 6, it can determine whether the film 6 is qualified according to whether the size information exceeds the tolerance requirements. When the tolerance requirements are exceeded, the film 6 is determined to be unqualified. The control device 1 can display the unqualified film 6 for easy selection.

[0084] Based on the above measurement method, the present invention has carried out an example verification. First, the precision slide table 3 is used to cooperate with steps S1.1.1 - S1.1.5 to calibrate the angle of the camera 2. When calibrating the angle, the calculated included angle θ between the UV connection line and the VW connection line on the checkerboard calibration plate is 90.069°, while the actual theoretical value is 90°, and the relative angle error is 0.07%; the precision is ±0.035°. This relative error and precision meet the measurement requirements, and the angle calibration of the camera 2 is completed. Secondly, the internal frequency divider of the camera 2 is used to cooperate with steps S1.2.1 - S1.2.5 to calibrate the length of the camera 2. When calibrating the length, the calculated cross-ratio CR0(A1, B1; C1, D1) of the image coordinates of the four grid corner points A1, B1, C1, D1 on the checkerboard calibration plate is 1.33396, while the theoretical value of the world coordinate cross-ratio CR(A, B; C, D) is 1.33333, and the relative length error is 0.05%, meeting the measurement requirements, and the length calibration of the camera 2 is completed. Then, the camera 2 cooperates with the infrared LED backlight 7 to clearly collect the image information of the film 6 on the conveying surface of the conveyor belt assembly 5 when the film 6 moves at high speed with the conveying assembly. The camera 2 sends the collected image information to the control device 1, and the control device 1 processes the image information and calculates the width measurement value h = 284.52 mm, the hypotenuse measurement value a = 328.53 mm, and the angle measurement value θ = 61° of the film 6 as shown in Figure 9. The actual width value of the film 6 is h = 285 mm, the theoretical hypotenuse value is a = 330 mm, and the actual angle value is θ = 60°; then the absolute value of the width error of the film 6 is 0.48 mm, the error rate is 0.17%, the absolute value of the hypotenuse error is 1.47 mm, the error rate is 0.45, the angle error is 1°, and the error rate is 1.67%. The errors are all small and within the allowable error range. Therefore, the measurement method given by the present invention can measure the size of the film 6 more accurately, is more practical and convenient for subsequent production.

[0085] The above description is a detailed description of the preferred feasible embodiment of the present invention, but the embodiment is not used to limit the patent application scope of the present invention. Any equivalent changes or modifications completed under the technical spirit prompted by the present invention should fall within the patent scope covered by the present invention.

Claims

1. A method for visually measuring the size of a film on a conveyor belt, which is measured by a visual measurement system for the size of a film on a conveyor belt, characterized in that, A visual measurement system for film size on a conveyor belt includes a main frame, a conveyor belt assembly, an infrared LED backlight, a camera, a precision slide table, and a control device; the conveyor belt assembly is installed on the main frame, and an infrared LED backlight is arranged directly below the conveying surface thereof. A camera is arranged directly above the conveying surface of the conveyor belt assembly. The camera is connected to the control device. The acquisition part of the camera is arranged facing the conveying surface and the infrared LED backlight. The camera is installed on the precision slide table, and the precision slide table is installed on the main frame; The measurement method includes the following steps: S1. Calibrate and calibrate the camera, specifically including angle calibration and length calibration. The steps are as follows: S1.

1. Perform angle calibration: S1.1.

1. The conveyor belt assembly runs at a low speed, and a checkerboard calibration plate is placed on its conveying surface. The checkerboard calibration plate moves to directly below the acquisition part of the camera as the conveyor belt assembly runs; S1.1.

2. Adjust the rotation of the camera around the X and Y axes through the precision slide table. When the checkerboard calibration plate is at the center of the image and the imaging is clear, the plane where the optical axis of the camera is located is perpendicular to the conveying surface of the conveyor belt assembly; S1.1.

3. Coarsely adjust the precision stage to rotate within the range of ±5 degrees along the Z-axis, with each rotation being 1 degree. After each rotation, the camera acquires the image information of the checkerboard calibration board once and transmits it to the control device. Then, in the control device, use the corner detection algorithm to obtain the pixel coordinates of the corner points U, V, and W on the checkerboard calibration board, calculate the distances between the UV connection line, VW connection line, and UW connection line based on the obtained pixel coordinates, and then use the cosine theorem Calculate the included angle θ formed between the UV connection line and the VW connection line; select two images with the included angle θ closest to 90 degrees in the image information and record the corresponding rotation angles x1 and x2 of the precision stage S1.1.

4. In the x1, x2 interval, finely adjust the rotation of the precision slide table around the X and Y axes using a micrometer, rotating 0.1 degree each time. And after each rotation, the camera acquires the image information of the checkerboard calibration plate once and transmits it to the control device; then, use the calculation method in step S1.1.3 to calculate the included angle θ formed between the UV connection line and the VW connection line on the checkerboard calibration plate; select two images with the included angle θ closest to 90 degrees in the image information and record the corresponding rotation angles of the precision slide table as x3 and x4; S1.1.

5. In the x3, x4 interval, finely adjust the rotation of the precision slide table around the X and Y axes using a micrometer, rotating 0.01 degree each time. And after each rotation, the camera acquires the image information of the checkerboard calibration plate once and transmits it to the control device; then, use the calculation method in step S1.1.3 to calculate the included angle θ formed between the UV connection line and the VW connection line on the checkerboard calibration plate; take the rotation angle of the precision slide table with the included angle closest to 90 degrees in the image information as the angle calibration result; S1.

2. Perform length calibration: S1.2.

1. Adjust the conveyor belt assembly to the rated working speed; S1.2.

2. Calculate the distortion of the image in the velocity direction using the cross-ratio invariance of the image: Substitute the world coordinates of any four collinear grid corner points A, B, C, and D on the checkerboard calibration plate plane into the formula Calculate the cross-ratio of the world coordinates; S1.2.

3. Extract the image coordinates A1, B1, C1, D1 of the corresponding four grid corner points A, B, C, D on the image information, and calculate the cross-ratio CR0(A1, B1; C1, D1) of the image coordinates; S1.2.

4. Calculate the difference C between the current cross-ratio CR0(A1, B1; C1, D1) of the image coordinates and the cross-ratio CR(A, B; C, D) of the world coordinates; S1.2.

5. Perform frequency division processing on the camera signal acquisition frequency. When C < 0, increase the signal acquisition frequency by 0.1 times each time to stretch the image; when C > 0, decrease the signal acquisition frequency by 0.1 times each time to compress the image until the absolute value of the cross-ratio difference C is the smallest, and complete the length calibration; S1.

3. Calculate the ratio of the length and width of the checkerboard calibration plate in the pixel coordinate system to that in the world coordinate system as the scale factor K1 for size calculation. S2. Operate the conveyor belt assembly to start conveying the film. S3. The infrared LED backlight emits infrared light to illuminate the film background. S4. The camera acquires the image information of the film on the conveyor belt assembly and sends the acquired image information to the control device. S5. The control device performs image processing on the image information, calculates the size information of the film, and then displays the obtained result.

2. The visual measurement method for the film size on a conveyor belt according to claim 1, wherein: The camera is an industrial line array camera, and the control device is an industrial control computer.

3. The visual measurement method for the size of the film on the conveyor belt according to claim 1, characterized in that: The precision sliding table is installed on the main frame through a camera bracket.

4. A method for visually measuring the size of a film on a conveyor belt according to claim 1, characterized in that: In step S5, the control device calculates the size information of the film through the following steps: S5.

1. Filter the image information to remove image noise. S5.

2. Perform binarization processing on the image information. S5.

3. Use the Prewitt operator to perform edge detection on the image information to obtain the single-pixel edge of the film. S5.

4. Use the Hough transform to fit the edge straight line to obtain the straight line equation. S5.

5. Solve the straight line equation to obtain the intersection coordinates, and then perform corner detection to remove the redundant intersection coordinates. S5.

6. Calculate the distance between the intersections through the remaining intersection coordinates, and then obtain the film size in the pixel coordinate system. S5.

7. The product of the film size in the pixel coordinate system and the scale factor K1 is the actual size of the film. S5.

8. Calculate the included angle between the straight lines through the slope of the straight line equation, and then calculate the inner angle value of the film.

Images