Steel belt wrapping gap measuring method and system based on curvature statistics

Abstract

The invention discloses a method and a system for measuring a steel belt wrapping gap based on curvature statistics, belongs to the technical field of image processing, and aims to solve the technical problem of how to accurately measure the steel belt wrapping gap under the condition that a production line does not stop. Comprising the following steps: collecting a steel belt wrapping image through an industrial camera under a low-angle annular light lighting condition; based on morphological operation and connected domain analysis, interference in the steel belt wrapping image is preliminarily eliminated, and a steel belt wrapping image after interference elimination is obtained; for the steel strip wrapping image after interference elimination, carrying out contour refinement on the steel strip, and carrying out steel strip edge positioning in a manner of counting the curvature condition of contour points to obtain a steel strip edge line; and based on the steel strip edge line, calculating a steel strip gap value through a proportional analysis method.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to a method and system for measuring a steel strip wrapping gap based on curvature statistics. Background Art

[0002] The steel tape armor wrapping process wraps the metal steel tape around the outer layer of the cable in a specific way to provide mechanical protection and enhance physical strength for the cable. It is an important step in cable manufacturing.

[0003] At present, the steel tape armor wrapping is mainly achieved by using special steel tape armoring machines and adjusting machine parameters according to different cable specifications. However, it is a technical difficulty to ensure the uniform distribution of the steel tape on the cable during the wrapping process, and it is impossible to completely rely on the machine to achieve accurate control of the overlap between the steel tapes throughout the entire process. In order to ensure accurate wrapping, it is necessary to manually measure the gap between the steel tape wrapping from time to time. Irregular measurements require the production line to be shut down, and the collected data is not continuous and affects production efficiency. Due to differences in operator experience and skill levels, there will also be human measurement errors; using tools such as vernier calipers to measure there is a risk of damaging the cable.

[0004] How to accurately measure the gap between steel strip wrapping without stopping the production line is a technical problem that needs to be solved. Summary of the invention

[0005] The technical task of the present invention is to address the above shortcomings and provide a method and system for measuring the gap between steel strips wrapped based on curvature statistics to solve the technical problem of how to accurately measure the gap between steel strips wrapped without stopping the production line.

[0006] In a first aspect, the present invention provides a method for measuring a steel strip wrapping gap based on curvature statistics, comprising the following steps:

[0007] Image acquisition: Under low-angle ring light conditions, the cable passes through the ring light, and the steel tape wrapping image is captured by an industrial camera;

[0008] Interference elimination: Based on morphological operations and connected domain analysis, the interference in the steel strip wrapping image is initially eliminated to obtain the steel strip wrapping image after interference elimination. The interference is understood as: there is a layer of black coating on the surface of the steel strip, and the wear and tear during the production process of the steel strip will cause thin strips or blocks of scratches on its surface. The scratched area will be illuminated by the ring light to form interference;

[0009] Steel strip edge positioning: For the steel strip wrapping image after interference removal, the steel strip contour is refined, and the steel strip edge is positioned by counting the curvature of the contour points to obtain the steel strip edge line;

[0010] Steel belt gap measurement: Based on the steel belt edge line, the steel belt gap value is calculated by proportional analysis method.

[0011] Preferably, interference elimination includes the following steps:

[0012] Gaussian blur processing is performed on the steel strip wrapping image to smooth the noise interference caused by industrial camera imaging, and image I is obtained. G , for image I G A fixed threshold is set and binarization is performed to obtain a binary image B containing the highlight area illuminated by the ring light. The grayscale value of the area illuminated by the ring light is 255, where the two-dimensional Gaussian function expression is as follows:

[0013]

[0014] σ is used as the standard deviation to control the blur degree of the steel strip wrapping image;

[0015] The binary image B is processed by morphological closing operation, wherein the morphological closing operation includes two steps: dilation stage and erosion stage. The dilation stage performs dilation operation to increase the area of ​​the white foreground area. The adjacent small scratches in the binary image B are connected together due to the dilation operation, which reduces the number of connected domains and facilitates the subsequent connected domain analysis. The independent scratches with smaller areas are completely eliminated after the erosion stage.

[0016] Extract the connected area S in the binary image 1 ,s 2 ,…s n}, for the connected domain, the width w i and height h i At the same time, the connected domains that meet the aspect ratio condition are retained, and the block scratches and the contours that do not belong to the edge of the steel strip in aspect ratio are removed to obtain the binary image B. ′ , the aspect ratio conditions are as follows:

[0017]

[0018] T L Indicates the lower limit of the aspect ratio, T H Indicates the upper limit of the aspect ratio, D cable is the cable pixel diameter;

[0019] The calculation method of cable pixel diameter is as follows:

[0020] The grayscale histogram of the original steel strip wrapping image is calculated. According to the characteristics that the dark background in the original steel strip wrapping image occupies a large pixel area and the grayscale value of the cable area pixel is widely distributed, the grayscale value on the right side of the continuous peak of the histogram is selected as the segmentation threshold T;

[0021] Based on the segmentation threshold T, the original image is segmented to obtain the binary image B cable, binary image B cable The pixel position with a gray value of 255 is the position of the cable in the entire image, and the approximate pixel diameter D of the cable is obtained. cable .

[0022] Preferably, the steel strip edge positioning comprises the following steps:

[0023] Extract the binary image B through the sobel operator ′ The vertical edge in the image is obtained by ′ The vertical edge of each connected domain in the image is B, and the edge image with a pixel width of 1 is obtained. edge , the pixel grayscale value at the edge position is 255, where the Sobel operator expression is as follows:

[0024]

[0025] Extract B edge The coordinates of all pixel points on the edge line are calculated using the three-point parametric equation method to calculate the curvature corresponding to all pixel points on each edge line. The curvature calculation formula is as follows:

[0026]

[0027] Among them, x ′ (t) and y ′ (t) is the first-order derivative at each pixel location, and x″(t) and y″(t) are the second-order derivatives;

[0028] Set the curvature threshold κ T , if the curvature of an edge line is greater than κ T The number of pixels of the edge line is greater than 1 / P of the total number of pixels of the edge line, which proves that the edge line is the edge of the steel strip. Otherwise, it is scratch interference. The image after scratch removal is B edge ′ .

[0029] Preferably, the steel strip gap measurement includes the following steps:

[0030] Based on the determined steel strip edge line, randomly select three adjacent steel strip edges, and the gap value d is calculated as follows:

[0031]

[0032] D is the actual physical width of the steel strip, a and b are the horizontal pixel intervals between the adjacent edges of the three steel strip edges, and a and b are extracted as follows: cable Get the pixel position of the center of the cable, take a horizontal line with the selected pixel coordinates as the vertical axis, and the horizontal line forms three intersections A, B, and C with the edge of the steel strip, where a is the pixel interval between AB and b is the pixel interval between BC.

[0033] In a second aspect, the present invention provides a steel strip wrapping gap measurement system based on curvature statistics, which is used to measure the steel strip wrapping gap by using a steel strip wrapping gap measurement method based on curvature statistics as described in any one of the first aspects, and includes an image acquisition module, an interference rejection module, a steel strip edge positioning module, and a steel strip gap measurement module;

[0034] The image acquisition module is used to perform the following: under the condition of low-angle ring light illumination, the cable passes through the ring light, and the steel belt wrapping image is acquired by the industrial camera;

[0035] The interference removal module is used to perform the following: preliminarily remove interference from the steel strip wrapping image based on morphological operations and connected domain analysis to obtain the steel strip wrapping image after interference removal, where interference is understood as: there is a layer of black coating on the surface of the steel strip, and the wear and tear during the production process of the steel strip will cause thin strips or blocks of scratches on its surface, and the scratched area will be illuminated by the ring light to form interference;

[0036] The steel strip edge positioning module is used to perform the following: for the steel strip wrapping image after interference removal, the steel strip contour is refined, and the steel strip edge is positioned by counting the curvature of the contour points to obtain the steel strip edge line;

[0037] The steel strip gap measurement module is used to perform the following: based on the steel strip edge line, the steel strip gap value is calculated by a proportional analysis method.

[0038] Preferably, the interference removal module is used to perform the following operations:

[0039] Gaussian blur processing is performed on the steel strip wrapping image to smooth the noise interference caused by industrial camera imaging, and image I is obtained. G , for image I G A fixed threshold is set and binarization is performed to obtain a binary image B containing the highlight area illuminated by the ring light. The grayscale value of the area illuminated by the ring light is 255, where the two-dimensional Gaussian function expression is as follows:

[0040]

[0041] σ is used as the standard deviation to control the blur degree of the steel strip wrapping image;

[0042] The binary image B is processed by morphological closing operation, wherein the morphological closing operation includes two steps: dilation stage and erosion stage. The dilation stage performs dilation operation to increase the area of ​​the white foreground area. The adjacent small scratches in the binary image B are connected together due to the dilation operation, which reduces the number of connected domains and facilitates the subsequent connected domain analysis. The independent scratches with smaller areas are completely eliminated after the erosion stage.

[0043] Extract the connected domain S in the binary image 1 ,s 2 ,…s n}, for the connected domain, the width w i and height h i At the same time, the connected domains that meet the aspect ratio condition are retained, and the block scratches and the contours that do not belong to the edge of the steel strip in aspect ratio are removed to obtain the binary image B. ′ , the aspect ratio conditions are as follows:

[0044]

[0045] T L Indicates the lower limit of the aspect ratio, T H Indicates the upper limit of the aspect ratio, D cable is the cable pixel diameter;

[0046] The calculation method of cable pixel diameter is as follows:

[0047] The grayscale histogram of the original steel strip wrapping image is calculated. According to the characteristics that the dark background in the original steel strip wrapping image occupies a large pixel area and the grayscale value of the cable area pixel is widely distributed, the grayscale value on the right side of the continuous peak of the histogram is selected as the segmentation threshold T;

[0048] Based on the segmentation threshold T, the original image is segmented to obtain the binary image B cable , binary image B cable The pixel position with a gray value of 255 is the position of the cable in the entire image, and the approximate pixel diameter D of the cable is obtained. cable .

[0049] Preferably, the steel strip edge positioning module is used to perform the following operations:

[0050] Extract the binary image B through the sobel operator ′ The vertical edge in the image is obtained by ′ The vertical edge of each connected domain in the image is B, and the edge image with a pixel width of 1 is obtained. edge , the pixel grayscale value at the edge position is 255, where the Sobel operator expression is as follows:

[0051]

[0052] Extract B edge The coordinates of all pixel points on the edge line are calculated using the three-point parametric equation method to calculate the curvature corresponding to all pixel points on each edge line. The curvature calculation formula is as follows:

[0053]

[0054] Among them, x ′ (t) and y ′ (t) is the first-order derivative at each pixel location, and x″(t) and y″(t) are the second-order derivatives;

[0055] Set the curvature threshold κ T , if the curvature of an edge line is greater than κ T The number of pixels of the edge line is greater than 1 / P of the total number of pixels of the edge line, which proves that the edge line is the edge of the steel strip. Otherwise, it is scratch interference. The image after scratch removal is B edge ′ .

[0056] Preferably, the steel strip gap measurement module is used to perform the following operations:

[0057] Based on the determined steel strip edge line, randomly select three adjacent steel strip edges, and the gap value d is calculated as follows:

[0058]

[0059] D is the actual physical width of the steel strip, a and b are the horizontal pixel intervals between the adjacent edges of the three steel strip edges, and a and b are extracted as follows: cable Get the pixel position of the center of the cable, take a horizontal line with the selected pixel coordinates as the vertical axis, and the horizontal line forms three intersections A, B, and C with the edge of the steel strip, where a is the pixel interval between AB and b is the pixel interval between BC.

[0060] The steel strip wrapping gap measurement method and system based on curvature statistics of the present invention have the following advantages:

[0061] 1. No manual participation is required in the measurement, which reduces the risk of product damage during the measurement process and greatly reduces labor costs;

[0062] 2. In view of the special curvature of the edge during the steel strip wrapping process, low-angle annular light is used to measure the gap based on curvature statistics to ensure pixel-level measurement accuracy;

[0063] 3. Computer vision-based method cooperates with industrial cameras to continuously measure the wrapping gap in real time without stopping the machine for measurement, effectively improving the efficiency of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0065] The present invention is further described below in conjunction with the accompanying drawings.

[0066] Figure 1 This is a flowchart of a method for measuring a steel strip wrapping gap based on curvature statistics in Example 1;

[0067] Figure 2 In the embodiment 1, a method for measuring the gap between steel strips wrapped based on curvature statistics is used to obtain an image I by Gaussian blurring. G Schematic diagram;

[0068] Figure 3 This is a schematic diagram of a binary image B of a steel strip wrapping gap measurement method based on curvature statistics in Example 1;

[0069] Figure 4 This is a schematic diagram of an image obtained after a morphological closing operation in a method for measuring a steel strip wrapping gap based on curvature statistics in Example 1;

[0070] Figure 5 The binary image B of a method for measuring the gap between steel strips wrapped based on curvature statistics in Example 1 ′ Schematic diagram;

[0071] Figure 6 This is a schematic diagram of a grayscale histogram calculated in a method for measuring a steel strip wrapping gap based on curvature statistics in Example 1;

[0072] Figure 7 The binary image B of a method for measuring the gap between steel strips wrapped based on curvature statistics in Example 1 cable Schematic diagram of

[0073] Figure 8 The middle image B is a method for measuring the gap between steel strips wrapped based on curvature statistics in Example 1. edge Schematic diagram of

[0074] Fig. 9 The middle image B is a method for measuring the gap between steel strips wrapped based on curvature statistics in Example 1. edge ′ Schematic diagram of

[0075] Fig.10 This is a schematic diagram of calculating the pixel interval in a method for measuring the gap between steel strip wrappings based on curvature statistics in Example 1. DETAILED DESCRIPTION

[0076] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. However, the embodiments are not intended to limit the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments may be combined with each other.

[0077] The embodiments of the present invention provide a method and system for measuring the gap between steel strips wrapped around a package based on curvature statistics, which are used to solve the technical problem of how to accurately measure the gap between steel strips wrapped around a package without stopping the production line.

[0078] Embodiment 1:

[0079] The invention discloses a method for measuring the gap between steel strips during wrapping based on curvature statistics, which comprises four steps: image acquisition, interference elimination, steel strip edge positioning and steel strip gap measurement.

[0080] Step S100: Image acquisition: under the condition of low-angle annular light illumination, the cable passes through the annular light, and the steel tape wrapping image is acquired by an industrial camera.

[0081] As a specific implementation of image acquisition, low-angle ring light is used for lighting, and the cable passes through the ring light. Since the steel belt is reflective and has a certain thickness, the edge of the steel belt will be illuminated during the wrapping process, and image acquisition is performed through an industrial camera.

[0082] Step S200 interference removal: based on morphological operations and connected domain analysis, the interference in the steel strip wrapping image is preliminarily removed to obtain the steel strip wrapping image after interference removal, wherein the interference is understood as: there is a layer of black coating on the surface of the steel strip, and the wear of the steel strip during the production process will cause thin strips or blocks of scratches on its surface, and the scratched area will be illuminated by the ring light to form interference.

[0083] There is a layer of black coating on the surface of the steel strip. The wear and tear during the production process will cause thin strips or blocks of scratches on the surface. The scratched area will be illuminated by the ring light to form interference. The interference can be initially eliminated through morphological operations and connected domain analysis. As a specific implementation of interference elimination, it includes the following steps:

[0084] (1) Gaussian blur processing is performed on the steel strip wrapping image to smooth the noise interference caused by industrial camera imaging, and image I is obtained. G (like Figure 2 As shown), for image I G A fixed threshold is set and binarization is performed to obtain a binary image B containing the highlight area illuminated by the ring light. The grayscale value of the area illuminated by the ring light is 255 (e.g. Figure 3 As shown), the two-dimensional Gaussian function expression is as follows:

[0085]

[0086] σ is used as the standard deviation to control the blur degree of the steel strip wrapping image;

[0087] (2) The binary image B is processed by a morphological closing operation, wherein the morphological closing operation includes two steps: a dilation stage and an erosion stage. The dilation stage performs a dilation operation as follows:

[0088] a. Dilation stage: The dilation operation will increase the area of ​​the white foreground region. The adjacent small scratches in the binary image B will be connected together due to the dilation operation, reducing the number of connected domains, which is convenient for subsequent connected domain analysis;

[0089] b. Erosion stage: Small independent scratches will be completely eliminated after the corrosion stage. The morphological operation results are as follows: Figure 4 As shown;

[0090] (3) Based on the morphological operation, the connected domain S = {s 1 ,s 2 ,…s n}, for the connected domain, the cable is horizontally distributed in the image, the edge of the wrapped steel belt has a small acute angle with the vertical direction, the edge height of the steel belt is large, and the aspect ratio of the minimum outer rectangular frame of its connected domain is distributed in a certain range. i and height h i At the same time, the connected domains that meet the aspect ratio condition are retained, and the block scratches and the contours that do not belong to the edge of the steel strip in aspect ratio are removed to obtain the binary image B. ′ , the aspect ratio conditions are as follows:

[0091]

[0092] T L Indicates the lower limit of the aspect ratio, T H Indicates the upper limit of the aspect ratio, D cable is the pixel diameter of the cable. Block scratches and contours whose width-to-height ratio does not belong to the edge of the steel strip will be removed through the above steps to obtain a binary image B ′ (like Figure 5 shown).

[0093] The calculation method of cable pixel diameter is as follows:

[0094] (1) Calculate the grayscale histogram of the original steel strip wrapping image (such as Figure 6 As shown in the figure, according to the characteristics that the dark background in the original steel strip wrapping image occupies a large pixel area and the grayscale value of the cable area pixel is widely distributed, the grayscale value on the right side of the continuous peak of the histogram is selected as the segmentation threshold T;

[0095] (2) Segment the original image based on the segmentation threshold T to obtain the binary image B cable (like Figure 7 As shown), the binary image B cableThe pixel position with a gray value of 255 is the position of the cable in the whole image. Figure 7 Get the approximate pixel diameter D of the cable cable .

[0096] Step S300: Positioning the edge of the steel strip: for the steel strip wrapping image after interference removal, the steel strip contour is refined, and the edge of the steel strip is positioned by counting the curvature of the contour points to obtain the edge line of the steel strip.

[0097] After step S200, thin strip scratches with smaller width and block scratches whose width-to-height ratio does not meet the range are removed, and it is necessary to further remove the interference of thin strip scratches with larger width. The edge of the steel strip is accurately located by refining the contour and counting the curvature of the contour points. The edge positioning of the steel strip includes the following steps:

[0098] (1) Extract the binary image B using the Sobel operator ′ The vertical edge in the image is obtained by ′ The vertical edge of each connected domain in the image is B, and the edge image with a pixel width of 1 is obtained. edge (like Figure 8 As shown), the pixel grayscale value at the edge position is 255, and the Sobel operator expression is as follows:

[0099]

[0100] (2) Extraction of B edge The coordinates of all pixel points on the edge line are calculated using the three-point parametric equation method to calculate the curvature corresponding to all pixel points on each edge line. The curvature calculation formula is as follows:

[0101]

[0102] Among them, x ′ (t) and y ′ (t) is the first-order derivative at each pixel location, and x″(t) and y″(t) are the second-order derivatives;

[0103] (3) Setting the curvature threshold κ T , if the curvature of an edge line is greater than κ T The number of pixels of the edge line is greater than 1 / P of the total number of pixels of the edge line, which proves that the edge line is the edge of the steel strip. Otherwise, it is scratch interference. The image after scratch removal is B edge ′ (like Fig. 9 shown).

[0104] Step S400: measuring the gap between the steel strips: calculating the gap between the steel strips by a proportional analysis method based on the edge line of the steel strips.

[0105] In the double-layer steel strip armoring process, the gap between the steel strips is much smaller than the actual width of the steel strips. The gap value of the steel strips can be obtained through proportional analysis. After step S300, the edge of the steel strip is located. Based on the determined edge line of the steel strip, three adjacent edges of the steel strip are randomly selected, and the gap value d is calculated as follows:

[0106]

[0107] D is the actual physical width of the steel strip, a and b are the horizontal pixel intervals between the adjacent edges of the three steel strip edges, and a and b are extracted as follows: cable Get the pixel position of the center of the cable, take a horizontal line with the selected pixel coordinate as the vertical axis, and the horizontal line and the edge of the steel strip form three intersections A, B, and C, where a is the pixel interval between AB and b is the pixel interval between BC (e.g. Fig.10 shown).

[0108] The method of this embodiment is aimed at the double-layer steel strip armoring process in the steel strip armor wrapping. It utilizes the special curvature degree of the steel strip edge and proposes a steel strip wrapping gap measurement method based on curvature statistics. It can ensure non-contact, 24-hour continuous, pixel-level precision measurement, effectively improve production line efficiency, and significantly reduce manpower and maintenance costs.

[0109] Embodiment 2:

[0110] The present invention discloses a steel strip wrapping gap measurement system based on curvature statistics, comprising an image acquisition module, an interference rejection module, a steel strip edge positioning module and a steel strip gap measurement module.

[0111] The image acquisition module is used to perform the following: under the condition of low-angle ring light illumination, the cable passes through the ring light, and the steel belt wrapping image is acquired through the industrial camera.

[0112] As a specific implementation of the image acquisition module, the module uses low-angle ring light for lighting. The cable passes through the ring light. Since the steel belt is reflective and has a certain thickness, the edge of the steel belt will be highlighted during the wrapping process, and image acquisition is performed through an industrial camera.

[0113] The interference removal module is used to perform the following: preliminarily remove interference from the steel strip wrapping image based on morphological operations and connected domain analysis to obtain the steel strip wrapping image after interference removal, where interference is understood as: there is a layer of black coating on the surface of the steel strip, and the wear and tear during the production process of the steel strip will cause thin strips or blocks of scratches on its surface, and the scratched area will be illuminated by the ring light to form interference.

[0114] There is a layer of black coating on the surface of the steel strip. The wear and tear during the production process will cause thin strips or blocks of scratches on its surface. The scratched area will be illuminated by the ring light to form interference. The interference can be initially eliminated through morphological operations and connected domain analysis. As a specific implementation of the interference elimination module, this module is used to perform the following operations:

[0115] (1) Gaussian blur processing is performed on the steel strip wrapping image to smooth the noise interference caused by industrial camera imaging, and image I is obtained. G (like Figure 2 As shown), for image I G A fixed threshold is set and binarization is performed to obtain a binary image B containing the highlight area illuminated by the ring light. The grayscale value of the area illuminated by the ring light is 255 (e.g. Figure 3 As shown), the two-dimensional Gaussian function expression is as follows:

[0116]

[0117] σ is used as the standard deviation to control the blur degree of the steel strip wrapping image;

[0118] (2) The binary image B is processed by a morphological closing operation, wherein the morphological closing operation includes two steps: a dilation stage and an erosion stage. The dilation stage performs a dilation operation as follows:

[0119] a. Dilation stage: The dilation operation will increase the area of ​​the white foreground region. The adjacent small scratches in the binary image B will be connected together due to the dilation operation, reducing the number of connected domains, which is convenient for subsequent connected domain analysis;

[0120] b. Erosion stage: Small independent scratches will be completely eliminated after the corrosion stage. The morphological operation results are as follows: Figure 4 As shown;

[0121] (3) Based on the morphological operation, the connected domain S = {s 1 ,s 2 ,…s n}, for the connected domain, the cable is horizontally distributed in the image, the edge of the wrapped steel belt has a small acute angle with the vertical direction, the edge height of the steel belt is large, and the aspect ratio of the minimum outer rectangular frame of its connected domain is distributed in a certain range. i and height h i At the same time, the connected domains that meet the aspect ratio condition are retained, and the block scratches and the contours that do not belong to the edge of the steel strip in aspect ratio are removed to obtain the binary image B. ′ , the aspect ratio conditions are as follows:

[0122]

[0123] T L Indicates the lower limit of the aspect ratio, T H Indicates the upper limit of the aspect ratio, D cable is the pixel diameter of the cable. Block scratches and contours whose width-to-height ratio does not belong to the edge of the steel strip will be removed through the above steps to obtain a binary image B ′ (like Figure 5 shown).

[0124] The calculation method of cable pixel diameter is as follows:

[0125] (1) Calculate the grayscale histogram of the original steel strip wrapping image (such as Figure 6 As shown in the figure, according to the characteristics that the dark background in the original steel strip wrapping image occupies a large pixel area and the grayscale value of the cable area pixel is widely distributed, the grayscale value on the right side of the continuous peak of the histogram is selected as the segmentation threshold T;

[0126] (2) Segment the original image based on the segmentation threshold T to obtain the binary image B cable (like Figure 7 As shown), the binary image B cable The pixel position with a gray value of 255 is the position of the cable in the whole image. Figure 7 Get the approximate pixel diameter D of the cable cable .

[0127] The steel strip edge positioning module is used to perform the following: for the steel strip wrapping image after interference removal, the steel strip contour is refined, and the steel strip edge is positioned by counting the curvature of the contour points to obtain the steel strip edge line.

[0128] Thin strip scratches with smaller width and block scratches whose width-to-height ratio does not meet the range are eliminated, and it is necessary to further eliminate the interference of thin strip scratches with larger width. The edge of the steel strip is accurately located by refining the contour and counting the curvature of the contour points. The steel strip edge positioning module is used to perform the following operations:

[0129] (1) Extract the binary image B using the Sobel operator ′ The vertical edge in the image is obtained by ′ The vertical edge of each connected domain in the image is B, and the edge image with a pixel width of 1 is obtained. edge (like Figure 8 As shown), the pixel grayscale value at the edge position is 255, and the Sobel operator expression is as follows:

[0130]

[0131] (2) Extraction of B edgeThe coordinates of all pixel points on the edge line are calculated using the three-point parametric equation method to calculate the curvature corresponding to all pixel points on each edge line. The curvature calculation formula is as follows:

[0132]

[0133] Among them, x ′ (t) and y ′ (t) is the first-order derivative at each pixel position, and x″(t) and y″(t) are the second-order derivatives;

[0134] (3) Setting the curvature threshold κ T , if the curvature of an edge line is greater than κ T The number of pixels of the edge line is greater than 1 / P of the total number of pixels of the edge line, which proves that the edge line is the edge of the steel strip. Otherwise, it is scratch interference. The image after scratch removal is B edge ′ (like Fig. 9 shown).

[0135] The steel strip gap measurement module is used to perform the following: based on the steel strip edge line, the steel strip gap value is calculated by a proportional analysis method.

[0136] In the double-layer steel belt armoring process, the gap between the steel belts is much smaller than the actual width of the steel belts. The gap value of the steel belts can be obtained through proportional analysis. Based on the determined steel belt edge line, the steel belt gap measurement module is used to randomly select three adjacent steel belt edges, and the gap value d is calculated as follows:

[0137]

[0138] D is the actual physical width of the steel strip, a and b are the horizontal pixel intervals between the adjacent edges of the three steel strip edges, and a and b are extracted as follows: cable Get the pixel position of the center of the cable, take a horizontal line with the selected pixel coordinate as the vertical axis, and the horizontal line and the edge of the steel strip form three intersections A, B, and C, where a is the pixel interval between AB and b is the pixel interval between BC (e.g. Fig.10 shown).

[0139] The system of this embodiment can execute the method disclosed in Example 1 to measure the steel strip wrapping gap.

[0140] The above is a detailed introduction to the steel strip wrapping gap measurement method and system based on curvature statistics provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for measuring the gap between steel strips wrapped based on curvature statistics, characterized in that: The steps include: Image acquisition: Under low-angle ring light conditions, the cable passes through the ring light, and the steel tape wrapping image is captured by an industrial camera; Interference elimination: Based on morphological operations and connected domain analysis, the interference in the steel strip wrapping image is initially eliminated to obtain the steel strip wrapping image after interference elimination. The interference is understood as: there is a layer of black coating on the surface of the steel strip, and the wear and tear during the production process of the steel strip will cause thin strips or blocks of scratches on its surface. The scratched area will be illuminated by the ring light to form interference; Steel strip edge positioning: For the steel strip wrapping image after interference removal, the steel strip contour is refined, and the steel strip edge is positioned by counting the curvature of the contour points to obtain the steel strip edge line; Steel belt gap measurement: Based on the steel belt edge line, the steel belt gap value is calculated by proportional analysis method.

2. The method for measuring the gap between steel strips wrapped based on curvature statistics according to claim 1 is characterized in that: Interference removal includes the following steps: Gaussian blur processing is performed on the steel strip wrapping image to smooth the noise interference caused by industrial camera imaging, and image I is obtained. G , for image I G A fixed threshold is set and binarization is performed to obtain a binary image B containing the highlight area illuminated by the ring light. The grayscale value of the area illuminated by the ring light is 255, where the two-dimensional Gaussian function expression is as follows: σ is used as the standard deviation to control the blur degree of the steel strip wrapping image; The binary image B is processed by morphological closing operation, wherein the morphological closing operation includes two steps: dilation stage and erosion stage. The dilation stage performs dilation operation to increase the area of ​​the white foreground area. The adjacent small scratches in the binary image B are connected together due to the dilation operation, which reduces the number of connected domains and facilitates the subsequent connected domain analysis. The independent scratches with smaller areas are completely eliminated after the erosion stage. Extract the connected domain S={s1,s2,…s n }, for the connected domain, the width w i and height h i At the same time, the connected domains that meet the aspect ratio condition are retained, and the block scratches and the contours that do not belong to the edge of the steel strip in aspect ratio are removed to obtain the binary image B. ′ , the aspect ratio conditions are as follows: T L Indicates the lower limit of the aspect ratio, T H Indicates the upper limit of the aspect ratio, D cable is the cable pixel diameter; The calculation method of cable pixel diameter is as follows: The grayscale histogram of the original steel strip wrapping image is calculated. According to the characteristics that the dark background in the original steel strip wrapping image occupies a large pixel area and the grayscale value of the cable area pixel is widely distributed, the grayscale value on the right side of the continuous peak of the histogram is selected as the segmentation threshold T; Based on the segmentation threshold T, the original image is segmented to obtain the binary image B cable , binary image B cable The pixel position with a gray value of 255 is the position of the cable in the entire image, and the approximate pixel diameter D of the cable is obtained. cable .

3. The method for measuring the gap between steel strips wrapped based on curvature statistics according to claim 1 is characterized in that: The strip edge positioning includes the following steps: Extract the binary image B through the sobel operator ′ The vertical edge in the image is obtained by ′ The vertical edge of each connected domain in the image is B, and the edge image with a pixel width of 1 is obtained. edge , the pixel grayscale value at the edge position is 255, where the Sobel operator expression is as follows: Extract B edge The coordinates of all pixel points on the edge line are calculated using the three-point parametric equation method to calculate the curvature corresponding to all pixel points on each edge line. The curvature calculation formula is as follows: Among them, x ′ (t) and y ′ (t) is the first-order derivative at each pixel position, and x″(t) and y″(t) are the second-order derivatives; Set the curvature threshold κ T , if the curvature of an edge line is greater than κ T The number of pixels of the edge line is greater than 1 / P of the total number of pixels of the edge line, which proves that the edge line is the edge of the steel strip. Otherwise, it is scratch interference. The image after scratch removal is B edge ′ .

4. The method for measuring the gap between steel strips wrapped based on curvature statistics according to claim 1, characterized in that: The steel belt gap measurement includes the following steps: Based on the determined steel strip edge line, randomly select three adjacent steel strip edges, and the gap value d is calculated as follows: D is the actual physical width of the steel strip, a and b are the horizontal pixel intervals between the adjacent edges of the three steel strip edges, and a and b are extracted as follows: cable Get the pixel position of the center of the cable, take a horizontal line with the selected pixel coordinates as the vertical axis, and the horizontal line forms three intersections A, B, and C with the edge of the steel strip, where a is the pixel interval between AB and b is the pixel interval between BC.

5. A steel strip wrapping gap measurement system based on curvature statistics, characterized in that: Used to measure the steel strip wrapping gap by using a steel strip wrapping gap measurement method based on curvature statistics as described in any one of claims 1 to 4, comprising an image acquisition module, an interference rejection module, a steel strip edge positioning module and a steel strip gap measurement module; The image acquisition module is used to perform the following: under the condition of low-angle ring light illumination, the cable passes through the ring light, and the steel belt wrapping image is acquired by the industrial camera; The interference removal module is used to perform the following: preliminarily remove interference from the steel strip wrapping image based on morphological operations and connected domain analysis to obtain the steel strip wrapping image after interference removal, where interference is understood as: there is a layer of black coating on the surface of the steel strip, and the wear and tear during the production process of the steel strip will cause thin strips or blocks of scratches on its surface, and the scratched area will be illuminated by the ring light to form interference; The steel strip edge positioning module is used to perform the following: for the steel strip wrapping image after interference removal, the steel strip contour is refined, and the steel strip edge is positioned by counting the curvature of the contour points to obtain the steel strip edge line; The steel strip gap measurement module is used to perform the following: based on the steel strip edge line, the steel strip gap value is calculated by a proportional analysis method.

6. The steel strip wrapping gap measurement system based on curvature statistics according to claim 5 is characterized in that: The interference removal module is used to perform the following operations: Gaussian blur processing is performed on the steel strip wrapping image to smooth the noise interference caused by industrial camera imaging, and image I is obtained. G , for image I G A fixed threshold is set and binarization is performed to obtain a binary image B containing the highlight area illuminated by the ring light. The grayscale value of the area illuminated by the ring light is 255, where the two-dimensional Gaussian function expression is as follows: σ is used as the standard deviation to control the blur degree of the steel strip wrapping image; The binary image B is processed by morphological closing operation, wherein the morphological closing operation includes two steps: dilation stage and erosion stage. The dilation stage performs dilation operation to increase the area of ​​the white foreground area. The adjacent small scratches in the binary image B are connected together due to the dilation operation, which reduces the number of connected domains and facilitates the subsequent connected domain analysis. The independent scratches with smaller areas are completely eliminated after the erosion stage. Extract the connected domain S={s1,s2,…s n }, for the connected domain, the width w i and height h i At the same time, the connected domains that meet the aspect ratio condition are retained, and the block scratches and the contours that do not belong to the edge of the steel strip in aspect ratio are removed to obtain the binary image B. ′ , the aspect ratio conditions are as follows: T L Indicates the lower limit of the aspect ratio, T H Indicates the upper limit of the aspect ratio, D cable is the cable pixel diameter; The calculation method of cable pixel diameter is as follows: The grayscale histogram of the original steel strip wrapping image is calculated. According to the characteristics that the dark background in the original steel strip wrapping image occupies a large pixel area and the grayscale value of the cable area pixel is widely distributed, the grayscale value on the right side of the continuous peak of the histogram is selected as the segmentation threshold T; Based on the segmentation threshold T, the original image is segmented to obtain the binary image B cable , binary image B cable The pixel position with a gray value of 255 is the position of the cable in the entire image, and the approximate pixel diameter D of the cable is obtained. cable .

7. The steel strip wrapping gap measurement system based on curvature statistics according to claim 5 is characterized in that: The strip edge positioning module is used to perform the following operations: Extract the binary image B through the sobel operator ′ The vertical edge in the image is obtained by ′ The vertical edge of each connected domain in the image is B, and the edge image with a pixel width of 1 is obtained. edge , the pixel grayscale value at the edge position is 255, where the Sobel operator expression is as follows: Extract B edge The coordinates of all pixel points on the edge line are calculated using the three-point parametric equation method to calculate the curvature corresponding to all pixel points on each edge line. The curvature calculation formula is as follows: Among them, x ′ (t) and y ′ (t) is the first-order derivative at each pixel position, and x″(t) and y″(t) are the second-order derivatives; Set the curvature threshold κ T , if the curvature of an edge line is greater than κ T The number of pixels of the edge line is greater than 1 / P of the total number of pixels of the edge line, which proves that the edge line is the edge of the steel strip. Otherwise, it is scratch interference. The image after scratch removal is B edge ′ .

8. The steel strip wrapping gap measurement system based on curvature statistics according to claim 5, characterized in that: The belt gap measurement module is used to perform the following operations: Based on the determined steel strip edge line, randomly select three adjacent steel strip edges, and the gap value d is calculated as follows: D is the actual physical width of the steel strip, a and b are the horizontal pixel intervals between the adjacent edges of the three steel strip edges, and a and b are extracted as follows: cable Get the pixel position of the center of the cable, take a horizontal line with the selected pixel coordinates as the vertical axis, and the horizontal line forms three intersections A, B, and C with the edge of the steel strip, where a is the pixel interval between AB and b is the pixel interval between BC.

Images