Anti-interference measurement method for angle of wide and thick plate billet

Abstract

The invention relates to the technical field of steel rolling detection. The invention discloses an anti-interference measurement method for the angle of a wide and thick plate steel billet. The method comprises the following steps: S1, collecting and correcting a steel billet image in a steel rotating roller way area; s2, adaptive enhancement processing is carried out on the billet image, and a stable billet surface area pixel set is obtained; s3, constructing a minimum rectangle based on external connection on the billet surface area pixel set, and obtaining a measurement result of a billet rotation angle; and S4, smoothing the rotation angle change of the steel billet by using a queue management method to obtain continuous and stable rotation angle change of the steel billet. Interference of environmental factors on steel billet rotation angle measurement is reduced through an improved image processing algorithm and an innovative angle tracking algorithm, changes of the steel billet rotation angle in the steel rotation process are smoothly processed, support is provided for automatic steel rotation control of wide and thick plates, and accurate and stable automatic steel rolling is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of steel rolling detection, and in particular to an anti-interference measurement method for the angle of a wide and thick plate steel billet based on image processing technology. Background Art

[0002] The rolling process of wide and thick plates generally includes three rolling stages: forming, widening and extending. The slab needs to be turned before and after the widening stage. In addition to full longitudinal rolling, each slab involves 1 to 2 steel turning processes depending on the rolling strategy. The steel turning operation is an important link in the rolling production of medium and thick plates, and high-precision measurement of the angle of wide and thick plate billets is a prerequisite for automatic steel turning control. The existing technology generally uses image processing technology or manual observation and judgment to measure the angle of wide and thick plate billets. A camera is installed at a suitable position at the wide and thick plate production site, and the image of the billet is collected by the camera. Then, the rotation angle of the billet is obtained by using an image judgment and processing algorithm. However, in the actual production process, the collected billet image may be partially blocked or water-stained due to the limitations of the actual production environment. Due to the influence of environmental factors such as steam interference, traditional image judgment and processing algorithms have few considerations and are difficult to adapt to complex actual environments. They are unable to obtain continuous and stable changes in the rotation angle of the billet, resulting in the failure of the automatic steel turning control process due to the lack of accurate feedback values ​​of the billet rotation angle. This directly affects the investment rate of automatic steel turning, and therefore does not have a good use effect. Manual control of steel turning, specifically: judging the rotation angle of the billet through human observation in a complex production environment, and having the operator use a control handle to execute the steel turning operation, not only requires the operator to have high technical requirements, but also increases the labor intensity of the operator during long-term steel turning operations, making it difficult to concentrate, and then it is difficult to obtain continuous and stable changes in the billet rotation angle. Summary of the invention

[0003] The purpose of the present invention is to provide an anti-interference measurement method for the angle of wide and thick plate steel billets. Through improved image processing algorithms and innovative angle tracking algorithms, the interference of environmental factors on the measurement of the steel billet rotation angle is reduced, and the changes in the steel billet rotation angle during the steel turning process are smoothly processed, thereby providing support for automatic steel turning control of wide and thick plates and realizing precise and stable automated steel rolling.

[0004] In order to achieve the above object, the present invention adopts the following technical solution: An anti-interference measurement method for a wide and thick plate billet angle comprises the following steps: Step S1: using an image acquisition device to acquire an image of a steel billet in a steel roller area, and then using an affine transformation to correct the acquired steel billet image; Step S2: Adaptively enhance the billet image corrected in step S1 to obtain a stable pixel set of the billet surface area. The specific steps of image adaptive enhancement are as follows: Step S2.1: Set the pixel segmentation threshold to g1, the pixel stretching threshold to g2, and the maximum grayscale value to 255, where g1 and g2 are both smaller than the maximum grayscale value; Step S2.2: setting image P1, wherein image P1 is initialized to be a rectified steel billet image; Step S2.3: The contrast of the image P1 is enhanced at a maximum ratio to obtain an image P2, wherein the maximum grayscale value of the image P2 is 255; Step S2.4: Use g1 to perform threshold segmentation on the image P2, and use the pixels in the image P2 whose grayscale values ​​are greater than g2 as the segmentation area. If the obtained segmentation area is close to a rectangle, the adaptive enhancement of the steel billet image is completed and step S3 is executed. Otherwise, step S2.5 is executed. Step S2.5: setting the grayscale value of pixels in the image P2 whose grayscale value is greater than g2 to g2; Step S2.6: Set P2 → P1 and execute step S2.3; Step S3: using a method based on a minimum circumscribed rectangle, rectangularize the pixel set on the surface area of ​​the steel billet obtained in step S2, fit the surface shape of the steel billet, and obtain side length, angle, and coordinate information, which are used as the angle measurement result of the steel billet. The specific fitting process is as follows: set the detection interval angle, perform detection within the range of 0° to 90°, use a horizontal circumscribed rectangle to fit the boundary of the pixel set each time the detection is performed, and record the parameters of the minimum area of ​​all circumscribed rectangles. The circumscribed rectangle with the minimum area is the minimum circumscribed rectangle of the corresponding pixel set on the surface area of ​​the steel billet being found, and the rotation angle of the circumscribed rectangle with the minimum area is the rotation angle of the steel billet; Step S4: Track the billet rotation angle using a first-in-first-out queue management method, and smooth the billet rotation angle change based on the logical relationship of the billet rotation angles at different times. The specific processing steps are as follows: Step S4.1: Establish a tracking queue for storing the billet rotation angle values ​​at different moments in the steel transfer process; Step S4.2: The initial angle of the steel billet is set to be close to 0°. If the angle is greater than 0°, the rotation direction of the steel billet is counterclockwise. If the angle is less than 0°, the rotation direction of the steel billet is clockwise. Step S4.3: Using the method based on the minimum circumscribed rectangle, measure the current billet rotation angle α, extract the most recently measured billet rotation angle α1 from the tracking queue, and due to the influence of environmental factors, it is necessary to evaluate whether α is reasonable, that is, to determine the degree of proximity between α, α+90°, and α-90° and α1, select the angle value closest to α1 as the current billet rotation angle α, and determine whether α1 and α are less than the threshold △α. If the conditions are met, add it to the tracking queue and jump to step 4.5. If the conditions are not met, it is considered that the current billet rotation angle α detection fails, and execute step S4.4; Step S4.4: If it is within the protection time △t, the current billet rotation angle α is changed to the billet rotation angle α1 measured most recently in the tracking queue; if it exceeds the protection time △t, the angle value with the smallest difference from the billet rotation angle α1 measured most recently in the tracking queue is selected from the three angle values ​​of α, α+90°, and α-90° as the current billet rotation angle α, and the changed current billet rotation angle α is added to the tracking queue; Step S4.5: If the steel transfer control operation is not completed, return to and execute step S4.3; otherwise, clear the queue and end the angle smoothing process.

[0005] Furthermore, in the step S1, the specific steps of the affine transformation are as follows: Step S1.1: reading the image of the steel billet in the steel transfer roller area on a computer; Step S1.2: setting four image coordinate points according to the size range of the steel billet in the image; Step S1.3: Calculate the corresponding four actual coordinate points according to the actual size of the steel billet based on the four image coordinate points obtained in step S1.2; Step S1.4: Calculate the affine transformation matrix using the relationship between the image coordinate points and the actual coordinate points; Step S1.5: Apply the affine transformation matrix to correct the billet image.

[0006] Furthermore, the image acquisition device is an industrial camera.

[0007] Beneficial effects of the present invention: The method is suitable for billet rotation angle measurement in complex production environments. The method uses an improved image processing algorithm to realize adaptive detection of billets in images that have been partially blocked, affected by water vapor pollution and other environmental factors, and reduces the interference of environmental factors on billet rotation angle measurement. Combined with an innovative angle tracking algorithm, the billet rotation angle is smoothed according to the angle change trend, and abnormal billet angles are eliminated, so that the billet rotation angle can be measured and output continuously and stably, and accurate detection of the billet angle is achieved, providing support for automatic steel turning control of wide and thick plates, reducing manual operation procedures in the rolling process, and improving the automation and intelligence level of the rolling process. The automatic steel turning control process provides higher reproducibility and accuracy for the billet rotation angle requirements, thereby realizing accurate and stable automated steel rolling and improving production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a schematic diagram of the steel billet image in the steel turning roller area before affine transformation in an embodiment of the present invention.

[0009] Figure 2 It is a schematic diagram of the steel billet image in the steel turning roller area after affine transformation in an embodiment of the present invention.

[0010] Figure 3 It is a schematic diagram of the surface area of ​​the steel billet after fitting based on the minimum circumscribed rectangle method according to an embodiment of the present invention.

[0011] Figure 4 It is a curve comparison diagram of the billet rotation angle measured abnormally at different times before and after smoothing based on the queue management method according to an embodiment of the present invention.

[0012] Figure 5 It is a statistical table of billet processing data under automatic steel transfer control.

[0013] Figure 6 It is a statistical table of the billet processing data of manually controlled steel transfer.

[0014] The reference numerals are as follows: 101: Steel roller table; 102: Steel billet; 103: billet surface area; 104: circumscribed rectangle; A1, A2, A3, A4: image coordinate points; a1, a2, a3, a4: actual coordinate points; L1: length of the rectangle; L2: width of the rectangle; β: The rotation angle of the rectangle. DETAILED DESCRIPTION

[0015] The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings.

[0016] An anti-interference measurement method for a wide and thick plate billet angle comprises the following steps: Step S1: Use an image acquisition device to acquire an image of the steel billet in the steel roller area. The steel billet image is as follows: Figure 1 As shown, four image coordinate points are selected in the billet image according to the size range of the billet, and the four image coordinate points are marked as A1, A2, A3, and A4 respectively; then the corresponding four actual coordinate points are calculated according to the actual size of the billet, and the affine transformation matrix is ​​calculated using the relationship between the image coordinate points and the actual coordinate points; the billet image is corrected by applying the affine transformation matrix, and the corrected billet image is as shown in Figure 2 As shown, the four actual coordinate points are marked as a1, a2, a3, and a4 respectively.

[0017] It should be noted that the design basis of step S1 is as follows: due to the limitations of production site conditions, the image acquisition equipment is installed near the steel transfer roller, and the actual acquired image is tilted. In order to accurately measure the rotation angle of the billet, the horizontal axis is specified as the rolling line direction and the vertical axis is specified as the roller direction. Before measuring the rotation angle of the billet, the billet image is corrected using affine transformation, and the corrected billet image is transformed into a top view. The billet rotation angle measured on the top view is the actual angle.

[0018] Step S2: Adaptively enhance the billet image corrected in step S1 to obtain a stable set of pixels in the billet surface area. The specific steps of image adaptive enhancement are as follows: set the pixel segmentation threshold to g1=180, and the pixel stretching threshold to g2=200. After 3 iterations, a set of pixels in the billet surface area with a rectangular ratio of 0.86 is obtained.

[0019] It should be noted that, based on step S2, in view of the influence of water vapor environmental factors on the billet image during the actual rolling process, the improved billet image adaptive enhancement algorithm provided by the present invention can adaptively iterate and eliminate the dynamic water vapor interference on the billet surface, greatly improving the success rate of billet threshold segmentation, and obtaining a suitable set of pixels in the billet surface area, thereby providing data support for the subsequent measurement of the billet rotation angle.

[0020] Step S3: After obtaining the pixel set of the billet surface area, a method based on the minimum circumscribed rectangle is used to rectangularize the pixel set of the billet surface area to fit the billet surface shape. In order to speed up the search speed of the minimum circumscribed rectangle, a preliminary search is first performed, and the rotation interval angle is set to 5°, so that the pixel set of the billet surface area is rotated within the range of 0° to 90°. After each rotation, a horizontal circumscribed rectangle is used to fit the boundary of the pixel set. The pixel set of the billet surface area after the circumscribed rectangle is fitted is as follows: Figure 3 As shown, L1 is the length of the rectangle, L2 is the width of the rectangle, β is the rotation angle of the rectangle, the circumscribed rectangle of the minimum area found in the preliminary search is located at a rotation angle of 35°, and then a detailed search is performed on this basis, the search range is set to 35°~40°, the rotation interval angle is set to 0.5°, and the circumscribed rectangle of the minimum area finally found is located at a rotation angle of 36.5°, which is the actual rotation angle of the billet; Step S4: Establish a tracking queue for the billet rotation angle, which is used to store the billet rotation angle values ​​at different times during the steel transfer process. When the current billet reaches the steel transfer position, the PLC gives a trigger event. The initial angle of the billet is specified to be close to 0°. After the steel transfer control is started, the billet rotation angle is tracked in each detection cycle according to the first-in-first-out queue management method. Before the currently measured billet rotation angle is added to the tracking queue, the currently measured billet rotation angle is first comprehensively judged using the preamble information stored in the tracking queue. If the continuous conditions are met, it is allowed to join the tracking queue. The specific billet rotation angle data tracking process is as follows: Figure 4 shown.

[0021] Please note that, please refer to Figure 4 At B and C, B is the detection difference of 90° caused by the billet being blocked, and at C is the detection failure of the billet rotation angle. Both are examples of abnormal processing of the billet rotation angle. The smoothing processing of the angle tracking algorithm of the present invention can obtain the effective billet rotation angle, avoid the angle jump caused by boundary detection errors due to occlusion of the billet and the detection failure within the protection time, and realize the adaptive processing process of billet rotation angle smoothing and detection failure.

[0022] An example of comparing the automatic steel turning control implemented according to the present invention with the manual steel turning control of the prior art is shown in FIG. Figure 5 and Figure 6Under the specifications of five aspect ratios of the steel billet, the time taken for the automatic steel turning control realized by the present invention is less than that taken for the manual control of steel turning; therefore, the present method is suitable for the measurement of steel billet rotation angle in complex production environments. Through the improved image enhancement algorithm and the innovative angle tracking algorithm, the interference of environmental factors on the measurement of steel billet rotation angle is reduced, and the change of the steel billet rotation angle during the steel turning process is smoothly processed, so that the steel billet rotation angle can be measured and output continuously and stably, and the accurate detection of the steel billet angle is realized, so as to provide support for the automatic steel turning control of wide and thick plates, reduce the manual operation procedures in the rolling process, improve the automation and intelligent level of the rolling process, and provide higher reproducibility and accuracy for the steel billet rotation angle demand of the automatic steel turning control process, thereby realizing accurate and stable automated steel rolling and improving production efficiency.

Claims

1. An anti-interference measurement method for the angle of a wide and thick plate billet, characterized in that: The following steps are involved: Step S1: using an image acquisition device to acquire an image of a steel billet in a steel roller area, and then using an affine transformation to correct the acquired steel billet image; Step S2: Adaptively enhance the billet image corrected in step S1 to obtain a stable pixel set of the billet surface area. The specific steps of image adaptive enhancement are as follows: Step S2.1: Set the pixel segmentation threshold to g1, the pixel stretching threshold to g2, and the maximum grayscale value to 255, where g1 and g2 are both smaller than the maximum grayscale value; Step S2.2: setting image P1, wherein image P1 is initialized to be a rectified steel billet image; Step S2.3: The contrast of the image P1 is enhanced at a maximum ratio to obtain an image P2, wherein the maximum grayscale value of the image P2 is 255; Step S2.4: Use g1 to perform threshold segmentation on the image P2, and use the pixels in the image P2 whose grayscale values ​​are greater than g2 as the segmentation area. If the obtained segmentation area is close to a rectangle, the adaptive enhancement of the steel billet image is completed and step S3 is executed. Otherwise, step S2.5 is executed. Step S2.5: setting the grayscale value of pixels in the image P2 whose grayscale value is greater than g2 to g2; Step S2.6: Set P2 → P1 and execute step S2.3; Step S3: using a method based on a minimum circumscribed rectangle, rectangularize the pixel set on the surface area of ​​the steel billet obtained in step S2, fit the surface shape of the steel billet, and obtain the side length, angle, and coordinate information, which are used as the angle measurement result of the steel billet. The specific fitting process is as follows: set the detection interval angle, perform detection in the range of 0° to 90°, use a horizontal circumscribed rectangle to fit the boundary of the pixel set each time the detection is performed, and record the parameters of the minimum area of ​​all circumscribed rectangles. The circumscribed rectangle with the minimum area is the minimum circumscribed rectangle of the corresponding pixel set on the surface area of ​​the steel billet being found, and the rotation angle of the circumscribed rectangle with the minimum area is the rotation angle of the steel billet; Step S4: Track the billet rotation angle using a first-in-first-out queue management method, and smooth the billet rotation angle change based on the logical relationship of the billet rotation angles at different times. The specific processing steps are as follows: Step S4.1: Establish a tracking queue for storing the billet rotation angle values ​​at different moments in the steel transfer process; Step S4.2: The initial angle of the steel billet is set to be close to 0°. When the angle is greater than 0°, the rotation direction of the steel billet is counterclockwise. When the angle is less than 0°, the rotation direction of the steel billet is clockwise. Step S4.3: Using the method based on the minimum circumscribed rectangle, measure the current billet rotation angle α, extract the most recently measured billet rotation angle α1 from the tracking queue, and due to the influence of environmental factors, it is necessary to evaluate whether α is reasonable, that is, to determine the degree of proximity between α, α+90°, and α-90° and α1, select the angle value closest to α1 as the current billet rotation angle α, and determine whether α1 and α are less than the threshold △α. If the conditions are met, add it to the tracking queue and jump to step 4.

5. If the conditions are not met, it is considered that the current billet rotation angle α detection fails, and execute step S4.4; Step S4.4: If it is within the protection time △t, the current billet rotation angle α is changed to the billet rotation angle α1 measured most recently in the tracking queue; if it exceeds the protection time △t, the angle value with the smallest difference from the billet rotation angle α1 measured most recently in the tracking queue is selected from the three angle values ​​of α, α+90°, and α-90° as the current billet rotation angle α, and the changed current billet rotation angle α is added to the tracking queue; Step S4.5: If the steel transfer control operation is not completed, return to and execute step S4.3; otherwise, clear the queue and end the angle smoothing process.

2. The anti-interference measurement method for the angle of a wide and thick plate steel billet according to claim 1 is characterized in that: In step S1, the specific steps of the affine transformation are as follows: Step S1.1: reading the image of the steel billet in the steel transfer roller area on a computer; Step S1.2: setting four image coordinate points according to the size range of the steel billet in the image; Step S1.3: Calculate the corresponding four actual coordinate points according to the actual size of the steel billet based on the four image coordinate points obtained in step S1.2; Step S1.4: Calculate the affine transformation matrix using the relationship between the image coordinate points and the actual coordinate points; Step S1.5: Apply the affine transformation matrix to correct the billet image.

3. The anti-interference measurement method for the angle of a wide and thick plate steel billet according to claim 1 or 2, characterized in that: The image acquisition device is an industrial camera.

Images