Highly reflective web offset measurement method for a winding system

Abstract

The application discloses a high-reflective coiled material offset measurement method suitable for a winding system, adjusts the position of an industrial camera through a cross rod connecting block and a camera clamp, so that the visual angle of the camera can cover the edge area of the measured coiled material; the angle of a lens polarizer is rotated according to the glare intensity of the surface of the coiled material; the size of a single pixel point in the real world corresponding to the image is calculated through the inherent parameters of the industrial camera and a double-telecentric lens; a statistical histogram of the image is established, the frame edge position and the inclination of the edge are located by using the histogram; and finally, the offset information is obtained. The method can be suitable for the winding system of the high-reflective coiled material, has a certain robustness to the shadow and the environmental light, and can greatly improve the environmental adaptability and the measurement accuracy of the coiled material offset detection.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of photoelectric measurement, and particularly relates to a high-reflectivity coiled material offset measurement method suitable for a coiling system. BACKGROUND

[0002] The offset detection sensor is an important link in an automatic closed-loop system for automatically correcting the lateral deviation of coiled material, and mainly serves the printing, packaging, lithium battery, photovoltaic, automobile, textile, metallurgy and other industries, and can realize the offset measurement function of coiling and uncoiling of coiled products such as film, metal foil, film, leather, non-woven fabric, and woven fabric.

[0003] The line array laser detection method based on the opposite shooting type is the mainstream method currently used in industrial sites, but is limited by the detection mechanism and has the following defects: first, the opposite shooting type deviation correction sensor is composed of a transmitting end and a receiving end, and has high installation requirements and a complex debugging process; second, in the process of detecting the edge of the measured target, the single detection range is small, which limits the movement speed of the detected object and is prone to missed detection. The image detection method based on the reflection type can realize accurate detection of edge information, has the advantages of fast system calibration and low missed detection rate, but is limited by the detection mechanism and has the following defects: first, when facing high-reflectivity coiled material, the imaging system is seriously affected by the glare of the coiled material surface, resulting in no response of the coiled material edge detection; second, it is difficult to avoid the interference of shadows and environmental light in the production site, and the image detection method based on the reflection type is very sensitive to the changes in image gray value caused by shadows and environmental light, resulting in poor detection reliability and reduced precision. SUMMARY

[0004] In order to overcome the shortcomings of the prior art, the application provides a high-reflectivity coiled material offset measurement method suitable for a coiling system, which adjusts the position of an industrial camera through a crossbar connecting block and a camera clamp, so that the viewing angle of the camera can cover the edge area of the measured coiled material; the angle of the lens polarizer is rotated according to the glare intensity of the coiled material surface; the size of a single pixel point in the image corresponding to the real world is calculated through the inherent parameters of the industrial camera and the double-telecentric lens; a statistical histogram of the image is established, and the histogram is used to locate the frame edge position and the inclination of the edge; and finally the offset information is obtained. The method can be applied to the coiling system for coiling and uncoiling high-reflectivity coiled material, has a certain robustness to shadows and environmental light, and can greatly improve the environmental adaptability and measurement accuracy of the coiled material offset detection.

[0005] The technical scheme adopted by the application to solve the technical problems is as follows:

[0006] Step 1: Construct a high-reflective web offset measurement device suitable for winding system, including an industrial camera, a double telecentric lens, a lens polarizer, a linear polarizer, a flat shadowless light source, an adjustable camera clamp, a horizontal rod, an adapter block, an adjustable light source support, a vertical rod, and a base plate;

[0007] The industrial camera and the double telecentric lens are integrally installed on the adjustable camera clamp, and the position of the camera is adjusted according to the needs of the measurement scene; the flat shadowless light source is installed on the adjustable light source support, and the linear polarizer is installed on the flat shadowless light source, keeping them parallel.

[0008] Step 2: Adjust the measurement system according to the position of the measured web, adjust the position of the industrial camera through the horizontal rod connecting block and the camera clamp, so that the view angle of the camera can cover the edge area of the measured web; adjust the position of the light source support, so that the flat shadowless light source can cover the area to be measured of the web, adjust the linear polarizer to keep it parallel to the flat shadowless light source, and completely cover the light source area;

[0009] Step 3: After the linear polarizer and the flat shadowless light source are installed and fixed, rotate the angle of the lens polarizer according to the glare intensity of the web surface, reduce the influence of the web surface glare, and ensure that the camera can clearly capture the edge area of the web surface;

[0010] Step 4: Calculate the size of a single pixel point in the real world corresponding to the image through the inherent parameters of the industrial camera and the double telecentric lens;

[0011] Step 5: Before the winding system is started, take an initial frame polarized intensity image of the web surface by the industrial camera, obtain an initial frame edge image by using an edge detection algorithm, binarize the edge pixel points in the image, establish a statistical histogram of the initial frame image, and locate the edge position of the initial frame by using the histogram;

[0012] Step 6: After the winding system is started, take the i-th frame polarized intensity image of the web surface by the industrial camera, obtain the i-th frame web surface edge image by using an edge detection algorithm, binarize the edge pixel points in the image, establish a statistical histogram of the i-th frame image, judge whether the edge is tilted by using the histogram, if not, directly locate the edge position of the i-th frame, if tilted, estimate the edge position of the i-th frame by calculating the average edge position, and calculate the tilt degree of the edge at the same time;

[0013] Step 7: Perform difference operation on the edge pixel position of the i-th frame and the edge pixel position of the initial frame to obtain the incremental information of the pixel position, and perform multiplication operation on the incremental information and the size of a single pixel in the real world obtained in step 4 to obtain the offset information.

[0014] Further, step 4 specifically involves: obtaining the pixel size u of the industrial camera, obtaining the magnification η of the dual telecentric lens, and calculating the size in the real world corresponding to a single pixel in the image using the formula x = u / η.

[0015] Furthermore, step 5 specifically includes:

[0016] Step 5-1: Before the winding system is turned on, the initial frame polarization intensity image of the roll surface is captured by an industrial camera. The intensity image is convolved using templates at angles of 45°, 90° and 135° in the Sobel operator. The edge image is obtained by weighted calculation of the three sets of discrete data.

[0017] Step 5-2: Binarize the edge image by setting the intensity T as the threshold. Pixel values ​​below the threshold T are assigned a value of 0, and those above the threshold T are assigned a value of 255. Plot a statistical histogram with the column pixels of the image as the x-axis and the number of pixels with a value of 255 in each column as the y-axis. Assuming that the edge of the initial frame position is in a 90° vertical state, the k0th column of the x-axis corresponding to the point with the largest value in the histogram is the edge position of the initial frame.

[0018] Furthermore, step 6 specifically includes:

[0019] Step 6-1: After the winding system is turned on, the edge image is calculated using the same method as before the winding system is turned on. The intensity T is set as the threshold to binarize the edge image. Pixel values ​​below the threshold T are assigned a value of 0, and pixels above the threshold T are assigned a value of 255.

[0020] Step 6-2: Plot a statistical histogram with the column pixels of the image as the x-axis and the number of pixels with a value of 255 in each column as the y-axis. Read the k-th pixel in the x-axis corresponding to the largest value in the histogram. max Let f(k) be a column. max ) represents the kth x-coordinate max The total number of pixels corresponding to the column is 255. Let S represent k. max Let S be the total number of pixels with a value of 255 in each of two neighboring regions. all This represents the total number of pixels with a value of 255 in the histogram. A threshold T′ is set to determine whether there is a skew. This means that the edge has no tilt, at which point the k-th edge... max The column represents the edge position; when This indicates that the edge is tilted. Let k′ be the smallest column pixel whose ordinate is greater than the threshold λ. min The maximum column pixel is k′ max Then the edge position is the average of the two. Instead, the tilt at this time Where H represents the total number of pixels in the vertical direction of the image.

[0021] Further, the step 7 is specifically: when no tilt is detected, the edge pixel position k of the i-th frame is subtracted from the edge pixel position k0 of the initial frame to obtain the incremental information a of the pixel position, and the incremental information is multiplied by the size x in the real world corresponding to the pixel obtained in step 4 to obtain the offset information. max Or when the tilt is detected, the edge pixel position k of the i-th frame is subtracted from the edge pixel position k0 of the initial frame to obtain the incremental information a of the pixel position, and the incremental information is multiplied by the size x in the real world corresponding to the pixel obtained in step 4 to obtain the offset information.

[0022] The beneficial effects of the present application are as follows:

[0023] The method of the present application can be applied to the winding system of collecting and releasing high-reflective coiled materials, has certain robustness to shadows and environmental light, can greatly improve the accuracy of the offset detection of the coiled material, and overcomes the problems of high installation requirements of the emitting type correction sensor, complex debugging process, small single detection range, low detection accuracy and poor reliability of the reflecting type correction sensor in the face of high-reflective materials, shadows and environmental light interference. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The flow chart of the method of the present application is shown in the figure;

[0025] Figure 2 The schematic diagram of the measuring device of the present application is shown in the figure;

[0026] Figure 3 The surface polarization intensity image of the copper-coated coiled material in the embodiment of the present application is shown in the figure;

[0027] Figure 4 The initial frame edge detection image in the embodiment of the present application is shown in the figure;

[0028] Figure 5 The second frame edge detection image in the embodiment of the present application is shown in the figure;

[0029] Figure 6 The initial frame statistical histogram in the embodiment of the present application is shown in the figure;

[0030] Figure 7 The second frame statistical histogram in the embodiment of the present application is shown in the figure;

[0031] Figure 8 The third frame edge detection image in the embodiment of the present application is shown in the figure;

[0032] Figure 9 The third frame statistical histogram in the embodiment of the present application is shown in the figure.

[0033] In the figure: 1 industrial camera, 2 double telecentric lens, 3 lens polarizer, 4 linear polarizer, 5 plane shadowless light source, 6 adjustable camera clamp, 7 horizontal rod, 8 adapter block, 9 adjustable light source support, 10 vertical rod, 11 bottom plate. DETAILED DESCRIPTION ​

[0034] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0035] The purpose of this invention is to provide a method for measuring the offset of highly reflective roll materials suitable for winding systems, in order to overcome the problems of high installation requirements, complex debugging process, and small single detection range of through-beam correction sensors, and low detection accuracy and poor reliability of reflective correction sensors under conditions of highly reflective materials, shadows, and ambient light interference.

[0036] The measuring device includes an industrial camera, a dual telecentric lens, a lens polarizer, a linear polarizer, a plane shadowless light source, an adjustable camera fixture, a crossbar, an adapter block, an adjustable light source bracket, a vertical pole, and a base plate.

[0037] like Figure 1 As shown, a method for measuring the offset of highly reflective roll material suitable for winding systems includes the following steps:

[0038] S1. Adjust the measurement system according to the position of the roll material being measured. Adjust the crossbar connecting block and camera fixture to position the industrial camera so that the camera's field of view can cover the edge area of ​​the roll material being measured. Adjust the light source bracket to position so that the planar shadowless light source can cover the area of ​​the roll material to be measured. Adjust the linear polarizer to keep it parallel to the planar shadowless light source and completely cover the light source area.

[0039] S2. After the linear polarizer and the planar shadowless light source are installed and fixed, rotate the angle of the lens polarizer according to the glare intensity of the roll surface to reduce the influence of glare on the roll surface and ensure that the camera can clearly capture the edge area.

[0040] S3. Calculate the real-world size of a single pixel in the image using the inherent parameters of the industrial camera and the dual telecentric lens.

[0041] Obtain the pixel size u of the industrial camera and the magnification η of the dual telecentric lens. Calculate the real-world size corresponding to a single pixel in the image using the formula x = u / η.

[0042] S4. Before the winding system is turned on, the initial frame polarization intensity image of the roll surface is captured by an industrial camera. The edge image of the initial frame is obtained by using an edge detection algorithm. The edge pixels in the image are binarized to establish a statistical histogram of the initial frame image. The edge position of the initial frame is located by using the histogram.

[0043] Before the winding system is turned on, an initial frame polarization intensity image of the roll surface is captured by an industrial camera. To make the edge detection algorithm more sensitive to edge information in the vertical and tilt directions, the intensity image is convolved using templates at angles of 45°, 90°, and 135° in the Sobel operator. The edge image is obtained by weighted calculation of the three sets of discrete data. The intensity T is set as the threshold to binarize the edge image. Pixel values ​​below the threshold T are assigned a value of 0, and pixels above the threshold T are assigned a value of 255. A statistical histogram is plotted with the column pixels of the image as the horizontal axis and the number of pixels with a value of 255 in each column as the vertical axis. At this time, it is assumed that the edge of the initial frame position is in a 90° vertical state. Then, the k0th column of the horizontal axis corresponding to the point with the largest value in the histogram is the edge position of the initial frame.

[0044] S5. After the winding system is turned on, the i-th frame polarization intensity image of the roll surface is captured by an industrial camera. The edge detection algorithm is used to obtain the i-th frame edge image of the roll. The edge pixels in the image are binarized to establish a statistical histogram of the i-th frame image. The histogram is used to determine whether the edge is tilted. If there is no tilt, the edge position of the i-th frame is directly located. If there is a tilt, the edge position of the i-th frame is estimated by calculating the average edge position, and the tilt degree of the edge is calculated at the same time.

[0045] After the winding system is turned on, the edge image is calculated using the same method as before the winding system is turned on. The edge image is binarized with an intensity T as the threshold. Pixel values ​​below the threshold T are assigned a value of 0, and those above the threshold T are assigned a value of 255. A statistical histogram is plotted with the column pixels of the image as the x-axis and the number of pixels with a value of 255 in each column as the y-axis. The k-th pixel in the x-axis corresponding to the largest value in the histogram is then read. max Let f(k) be a column. max ) represents the kth x-coordinate max The total number of pixels corresponding to the column is 255. Let S represent k. max Let S be the total number of pixels with a value of 255 in each of two neighboring regions. all This represents the total number of pixels with a value of 255 in the histogram. A threshold T′ is set to determine whether there is a skew. This means that the edge has no tilt, at which point the k-th edge... max The column represents the edge position; when This indicates that the edge is tilted. Let k′ be the smallest column pixel whose ordinate is greater than the threshold λ. min The maximum column pixel is k′ max Then the edge position is the average of the two. Instead, the tilt at this time Where H represents the total number of pixels in the vertical direction of the image.

[0046] S6. The difference between the edge pixel position of the i-th frame and the edge pixel position of the initial frame is calculated to obtain the incremental information of the pixel position. The incremental information is multiplied by the real-world size of the pixel obtained in S3 to obtain the offset information.

[0047] The edge pixel position k of the i-th frame max (When no tilt is detected) or (When tilt is detected) The incremental information α of the pixel position is obtained by subtracting the pixel position k0 at the edge of the initial frame. The incremental information is multiplied by the real-world size x of the pixel obtained in S3 to obtain the offset information.

[0048] Example:

[0049] like Figure 2 As shown, the measurement system is adjusted according to the position of the copper-plated coil being measured. The industrial camera is positioned appropriately by adjusting the crossbar connecting block and camera fixture so that the camera's field of view can cover the edge area of ​​the coil being measured. The light source bracket is adjusted to a suitable position so that the planar shadowless light source can cover the area of ​​the coil to be measured. The linear polarizer is adjusted to be parallel to the planar shadowless light source and completely cover the light source area.

[0050] like Figure 3 As shown, after the linear polarizer and the planar shadowless light source are installed and fixed, the angle of the lens polarizer is rotated according to the intensity of the glare on the roll surface to reduce the influence of the glare on the roll surface and ensure that the imaging device can clearly capture images of the edge area.

[0051] The pixel size of the industrial camera is obtained as u = 3.45 μm, and the magnification of the dual telecentric lens is obtained as η = 0.243. The size of a single pixel in the image in the real world is calculated as x = 14.2 μm using the formula x = u / η.

[0052] like Figure 4 As shown, before the winding system is turned on, an initial frame polarization intensity image of the roll surface is captured by an industrial camera. In order to make the edge detection algorithm more sensitive to edge information in the vertical and tilt directions, the intensity image is convolved using templates at angles of 45°, 90° and 135° in the Sobel operator. The edge image is obtained by weighted calculation of the three sets of discrete data.

[0053] like Figure 5 As shown, after the winding system is turned on, the edge image of the second frame is calculated using the same method as before the winding system is turned on.

[0054] like Figure 6As shown in the figure, before the winding system is opened, the intensity T=200 is set as a threshold value to binarize the edge image, the pixel value lower than the threshold value T is assigned as 0, and the pixel value higher than the threshold value T is assigned as 255. The column pixels of the image are taken as the horizontal coordinates, and the number of pixel points with the value of 255 in each column is taken as the vertical coordinates to draw a statistical histogram. At this time, it is assumed that the edge of the initial frame position is in a 90° vertical state, and the k0=310th column in the horizontal coordinates corresponding to the point with the maximum value in the histogram is the edge position of the initial frame.

[0055] As shown in the figure, Figure 7 after the winding system is opened, the polarized intensity image of the second frame of the material surface is shot by an industrial camera, the intensity image is convolved by using the 45°, 90° and 135° angle templates in the sobel operator, the edge image is obtained by performing weighted calculation on three groups of discrete data, the intensity T=200 is set as a threshold value to binarize the edge image, the pixel value lower than the threshold value T is assigned as 0, and the pixel value higher than the threshold value T is assigned as 255. The column pixels of the image are taken as the horizontal coordinates, and the number of pixel points with the value of 255 in each column is taken as the vertical coordinates to draw a statistical histogram. The k max =399th column in the horizontal coordinates corresponding to the point with the maximum value in the histogram is read out, f(k max ) represents the total number of pixel points with the value of 255 in the k max th column of the horizontal coordinates, S represents the total number of pixel points with the value of 255 in the two adjacent areas of the k max th column, and S all represents the total number of pixel points with the value of 255 in the histogram. The threshold value T'=0.8 is set to judge whether there is an inclination. Since represents that the edge does not exist inclination, and the k max =399th column is the edge position.

[0056] The difference between the edge pixel position k max =399 of the second frame and the edge pixel position k0=310 of the initial frame is calculated to obtain the increment information a=89 of the pixel position. The increment information is multiplied by the size 14.2 μm of the pixel corresponding to the real world in S3 to obtain the offset 1263.8 μm.

[0057] As shown in the figure, Figure 8 , Figure 9 after the winding system is opened, the edge image and the statistical histogram of the third frame are calculated by using the same method as before the winding system is opened.

[0058] Since S=96, represents that the edge exists inclination. The minimum column pixel with the vertical coordinate greater than the threshold value λ=3 is k' min =279, and the maximum column pixel is k' max=288, then the edge position is replaced by the average of the two, 284. The slope is then:

[0059]

[0060] Where H = 600 represents the total number of pixels in the height of the image.

[0061] Position the edge pixels of the 3rd frame The incremental information α′ = 26 of the pixel position is obtained by subtracting the initial frame edge pixel position k0 = 310. The incremental information is multiplied by the real-world size of 14.2μm corresponding to the pixel obtained in S3 to obtain the offset 369.2μm.

Claims

1. A method for measuring the offset of highly reflective roll material suitable for winding systems, characterized in that, It comprises the following steps: Step 1: Construct a high-reflective web offset measurement device suitable for a winding system, comprising an industrial camera, a double-telecentric lens, a lens polarizer, a linear polarizer, a flat shadowless light source, an adjustable camera clamp, a horizontal rod, an adapter block, an adjustable light source support, a vertical rod, and a base plate; The industrial camera and the double-telecentric lens are integrally installed on the adjustable camera clamp, and the position of the camera is adjusted according to the needs of the measurement scene; the flat shadowless light source is installed on the adjustable light source support, and the linear polarizer is installed on the flat shadowless light source, keeping them parallel; Step 2: Adjust the measurement system according to the position of the measured web, adjust the position of the industrial camera through the horizontal rod connecting block and the camera clamp, so that the viewing angle of the camera can cover the edge area of the measured web; adjust the position of the light source support, so that the flat shadowless light source can cover the area to be measured of the web, adjust the linear polarizer to be parallel to the flat shadowless light source, and completely cover the light source area; Step 3: After the linear polarizer and the flat shadowless light source are installed and fixed, rotate the angle of the lens polarizer according to the glare intensity of the web surface, reduce the influence of the web surface glare, and ensure that the camera can clearly shoot the edge area of the web surface; Step 4: Calculate the size of a single pixel point in the image corresponding to the real world through the inherent parameters of the industrial camera and the double-telecentric lens; Step 5: Before the winding system is started, shoot the initial frame polarized intensity image of the web surface through the industrial camera, obtain the initial frame edge image by using the edge detection algorithm, binarize the edge pixel points in the image, establish the statistical histogram of the initial frame image, and locate the initial frame edge position by using the histogram; Step 6: After the winding system is turned on, take a picture of the first side of the roll material using an industrial camera. i The frame polarization intensity image is obtained using an edge detection algorithm. i The edge image of the roll surface of the frame is binarized, and the edge pixels in the image are used to establish the first... i The statistical histogram of the frame image is used to determine whether the edges are tilted. If there is no tilt, the frame is directly located. i The edge position of the frame is estimated by calculating the average edge position if the frame is tilted. i The edge position of the frame is calculated, and the tilt of the edge is also calculated. The step 6 is specifically: Step 6-1: After the winding system is opened, the edge image is calculated by the same method as before the winding system is opened, and the intensity is set T The edge image is binarized with the threshold value, and the pixel value below the threshold value T is assigned a value of 0, and the value above the threshold value T is assigned a value of 255; Step 6-2: Draw a statistical histogram with the column pixel of the image as the horizontal coordinate and the number of pixel points with value 255 in each column as the vertical coordinate, read the horizontal coordinate of the point corresponding to the maximum value in the histogram Column, let represent the total number of pixel points with value 255 corresponding to the horizontal coordinate of the Column, let S represent the total number of pixel points with value 255 in the two neighborhoods of the Column, let represent the total number of pixel points with value 255 in the histogram, set the threshold value to judge whether there is a tilt, when represent that there is no tilt of the edge, at this time the Column is the edge position; when represent that there is a tilt of the edge, record the minimum column pixel with vertical coordinate greater than the threshold value as , the maximum column pixel as , then the edge position is replaced by the average of the two , at this time the tilt degree , wherein H represents the total number of pixels in the vertical direction of the image; Step 7: the edge pixel position of the first frame is subtracted from the edge pixel position of the second frame to obtain the increment information of the pixel position, and the increment information is multiplied by the size corresponding to a single pixel in the real world obtained in step 4 to obtain the offset information. i Step 7: the edge pixel position of the first frame is subtracted from the edge pixel position of the second frame to obtain the increment information of the pixel position, and the increment information is multiplied by the size corresponding to a single pixel in the real world obtained in step 4 to obtain the offset information. Step 7 specifically involves: when no tilt is detected at the edge, the... i Frame edge pixel position Or when tilt is detected, the first i Frame edge pixel position relative to the edge pixel position of the initial frame The difference operation yields the incremental information of the pixel position. The incremental information corresponds to the real-world size of the pixels obtained in step 4. x The offset information is obtained by performing a multiplication operation.

2. A method for measuring the offset of a highly reflective web material in a winding system according to claim 1, characterized in that The step 4 is specifically: obtaining the pixel size of the industrial camera u , obtaining the magnification of the double-telecentric lens , calculating the size in the real world corresponding to a single pixel point in the image through the formula .

3. A method of measuring the offset of a highly reflective web in a winding system according to claim 2, wherein The step 5 is specifically: Step 5-1: Before the winding system is started, shoot the initial frame polarized intensity image of the web surface through the industrial camera, convolve the intensity image by using the 45°, 90°, and 135° angle templates in the sobel operator, and obtain the edge image by weighted calculation of three groups of discrete data; Step 5-2: Set intensity T The edge image is binarized for threshold, the pixel value below threshold T is assigned to 0, higher than threshold T is assigned to 255, the statistical histogram is drawn with the column pixels of the image as the horizontal coordinate and the number of pixel points with value 255 in each column as the vertical coordinate; Assuming the edge of the initial frame position is 90° vertical state, the column of the horizontal coordinate corresponding to the point with the maximum value in the histogram is the initial frame edge position. column of the horizontal coordinate corresponding to the point with the maximum value in the histogram is the initial frame edge position.

Images