Profile detection device and shape measurement device

Abstract

The contour detection device is provided with: a camera that captures an image of a material to be rolled; and a processing circuit that processes the image captured by the camera. The processing circuit is configured to perform: dividing an image into a plurality of frames; specifying a plurality of target frames in which the tip of the rolled material exists from the plurality of frames; detecting a coordinate group representing the position of the rolled material in each target frame; calculating a coordinate transformation matrix for transforming the coordinate group of each object frame into a preset reference coordinate group; obtaining one composite image from the plurality of object frames transformed using the coordinate transformation matrix; and detecting the contour of the rolled material according to the brightness of the composite image.

Description

Technical Field

[0001] This disclosure relates to a contour detection device and a shape measuring device. In particular, this disclosure relates to a contour detection device for detecting the contour of a rolled material being transported in one direction on a hot rolling line, and a shape measuring device for measuring the shape of the rolled material using the contour detected by the contour detection device. Background Technology

[0002] There exists a shape measuring device for measuring the shape of rolled material after it has been rolled in a hot rolling line, such as warping, bulging, and flatness. Taking warping as an example, a warping shape measuring device is disclosed as a shape measuring device in Patent Documents 1 to 3 below.

[0003] In Patent Document 1 below, a camera is installed on the same horizontal plane as the rolled material and at a position orthogonal to the center line of the width direction of the transport table (hereinafter also referred to as the "mill center line"). The camera is used to capture images of the width direction end of the rolled material from the front side, the captured images are binarized, and the outline of the rolled material is extracted from the binarized images.

[0004] In Patent Document 2 below, a camera is installed diagonally above the stands (shell) of the finishing mill in a way that brings the space between the stands (shell) into view. The image captured by the camera is divided at a predetermined interval along the rolling direction (long side direction), and the shape of the plate width edge in each segmented image is approximated by a quadratic formula, and the warpage is quantified by curvature.

[0005] In Patent Document 3 below, a first camera is set up to photograph the rolled material from a side direction, and a second camera is set up to photograph the rolled material from a transport direction. The edge shape extracted from the image captured by the second camera is corrected based on the cropping shape extracted from the image captured by the first camera.

[0006] Existing technical documents Patent documents Patent Document 1: Japanese Patent Application Publication No. 1-285730 Patent Document 2: Japanese Patent No. 6828730 Patent Document 3: Japanese Patent No. 3724720 Summary of the Invention

[0007] The problem that the invention aims to solve However, to accurately measure warpage shape, a clear image is required. In hot rolling lines, the roll table is heated by the workpiece being heated in the furnace, and high-pressure water is sprayed onto the heated roll table from the roll cooling unit and descaling machine. This generates water vapor or water residue on the workpiece. Additionally, reflected light from the side guides located on both sides of the roll table in the width direction illuminates the workpiece. If this water vapor, residue, and reflected light are captured as interference in the image, the obtained image becomes unclear, resulting in the inability to accurately detect the contour of the workpiece.

[0008] In addition, water vapor tends to stagnate for a long time, so fans are sometimes installed to physically blow away the water vapor. However, in low-temperature environments such as winter, the amount of water vapor produced is large, and the capacity of the fans has its limits.

[0009] This disclosure was made to solve the aforementioned problems. The object of this disclosure is to provide a contour detection device that can obtain a clear image suitable for detecting the contour of a rolled material even in the presence of interference such as water vapor. Another object of this disclosure is to provide a shape measuring device capable of accurately measuring the shape of a rolled material based on the contour detected from a synthetic image by the aforementioned contour detection device.

[0010] Methods for solving problems The first view relates to a contour detection device for detecting the contour of a rolled material being transported in one direction in a hot rolling line. The contour detection device includes: a camera for capturing images of the rolled material; and a processing circuit for processing the images captured by the camera, the processing circuit being configured to: segment the images into multiple frames; determine multiple object frames from the multiple frames in which the front end of the rolled material exists; detect a coordinate group representing the position of the rolled material in each object frame; calculate a coordinate transformation matrix that transforms the coordinate group of each object frame into a pre-defined reference coordinate group; obtain a composite image from the multiple object frames obtained by the coordinate transformation matrix; and detect the contour of the rolled material based on the brightness of the composite image.

[0011] The second viewpoint, building upon the first, also possesses the following characteristics: Calculating the coordinate transformation matrix involves setting the coordinate group of a reference frame selected from multiple object frames as the reference coordinate group.

[0012] The third perspective, building upon the first, also possesses the following characteristics: Detecting the coordinate group involves detecting a coordinate group containing four coordinates for each object frame. Calculating the coordinate transformation matrix includes calculating the perspective transformation matrix as the coordinate transformation matrix.

[0013] The fourth perspective relates to a shape measuring device for measuring the shape of a rolled material being transported in one direction in a hot rolling line. The shape measuring device includes: a contour detection device as described in any one of the first to third perspectives; and a shape measuring circuit. The shape measuring circuit is configured to measure the shape of the rolled material based on the contour of the rolled material detected from a synthetic image by the contour detection device.

[0014] The fifth point, based on the fourth point, also has the following characteristics: The shape measurement circuit is configured to measure the warp direction and warp height of the front end of the rolled material.

[0015] The sixth point, building upon the fifth point, also possesses the following characteristics. The shape measurement circuit is configured to perform: obtaining a black-and-white image by binarizing the synthesized image; and calculating the area of ​​white pixels in one or more regions of interest defined in the black-and-white image. Measurements of the warp direction and warp height are performed if the area of ​​the white pixels in one or more regions of interest satisfies a judgment criterion set for each region of interest.

[0016] Invention Effects According to the contour detection apparatus of this disclosure, multiple object frames containing the front end of the rolled material are synthesized using a coordinate transformation matrix. Here, even if interference such as water vapor occurs, since the interference does not exist in the same location across the multiple object frames, the interference is canceled out when the multiple object frames are synthesized. As a result, the synthesized image obtained from the multiple object frames becomes clear, enabling high-precision detection of the contour of the rolled material from the clear synthesized image.

[0017] Furthermore, according to the shape measuring apparatus of this disclosure, the shape of the rolled material can be measured with high precision based on the contour of the rolled material detected from a clear synthetic image in the aforementioned contour detection apparatus. This disclosure is particularly suitable for measuring the warp direction and warp height of the leading edge of the rolled material. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating the configuration of a warp measuring device with an embodiment of the contour detection apparatus.

[0019] Figure 2 This is a diagram illustrating an example of the hardware configuration of a processing circuit.

[0020] Figure 3 It is a flowchart representing the process executed by the processing circuit.

[0021] Figure 4 This is a diagram illustrating an example of a method for determining an object frame.

[0022] Figure 5This is a diagram illustrating an example of a front-end inspection method for rolled materials.

[0023] Figure 6 This is a diagram illustrating an example of a method for defining the plane of the rolled material in each object frame.

[0024] Figure 7 This is a diagram illustrating an example of a method for defining the plane of the rolled material in each object frame.

[0025] Figure 8 This is a diagram illustrating an example of a method for defining the plane of the rolled material in each object frame.

[0026] Figure 9 This is a diagram illustrating an example of a method for detecting the contour of rolled material from a synthetic image.

[0027] Figure 10 This is a diagram illustrating an example of a method for quantifying warp height based on a profile.

[0028] Figure 11 This is a diagram illustrating an example of a method for quantifying warp height based on a profile.

[0029] Figure 12 This is a diagram illustrating an example of a method for quantifying warp height based on a profile.

[0030] Figure 13 It is a diagram used to illustrate the perspective transformation of an outline.

[0031] Figure 14 Figures (a) and (b) are examples of methods for determining the direction of warping and methods for calculating the height of warping.

[0032] Figure 15 This is a diagram illustrating an example of a method for compensating for variations in plate width that occur at the front end of the rolled material.

[0033] Figure 16 This diagram illustrates an example of a correction method that uses a correction material to correct the vanishing point.

[0034] Figure 17 This is a diagram used to illustrate the effects of this embodiment.

[0035] Figure 18 This is a diagram illustrating other applications of this disclosure. Detailed Implementation

[0036] Hereinafter, with reference to the accompanying drawings, an embodiment of the present disclosure will be described using the case of detecting the profile of a rolled material that has been rolled by a roughing mill in a hot rolling line, and measuring the warpage shape of the rolled material based on the detected profile. That is, an embodiment of the shape measuring device will be described using a warpage measuring device as an example. Furthermore, common elements in the various figures will be labeled with the same reference numerals, and repeated descriptions will be omitted.

[0037] Figure 1 This is a schematic diagram illustrating the configuration of a warp measuring device with a contour detection apparatus according to the embodiment. The warp measuring device 1 includes a contour detection device 2 and a shape measuring circuit described later. The contour detection device 2 detects the contour of the rolled material Rm being transported in one direction on the hot rolling line RL. The contour detection device 2 also detects the contour of the rolled material Rm that has been rolled by the roughing mill 3 of the hot rolling line RL.

[0038] The contour detection device 2 includes a camera 21 as an imaging unit, a processing circuit 22, and a tracking unit 23. The processing circuit 22 also serves as the "shape measurement circuit" of the warp measurement device 1, but a shape measurement circuit with the same configuration as the processing circuit 22 can also be separately provided.

[0039] As the camera 21, a network camera with a resolution of 1280×720 pixels is preferably used, for example. The shooting mode of the camera 21 can be set to color, the frame rate can be set to 30, the exposure time can be set to fixed, and the aperture value can be set to automatic mode. The resolution can be set according to the processing capability of the processing circuit 22. The frame rate is the number of frames contained in 1 second. The exposure time and aperture value can be preset according to the predicted temperature of the rolled material Rm. In addition, the camera 21 is not limited to a network camera; for example, it can also be used by combining a visible light cutoff filter with a machine vision camera or an infrared camera. These cameras are known, so further description is omitted.

[0040] Camera 21 is configured to capture images of the rolled material Rm being transported on the roll table 4 from an obliquely upward angle. For example, camera 21 is configured to be tilted horizontally at an angle θ1 relative to a straight line Lv on the horizontal plane that is orthogonal to the mill centerline Lm, which is also the transport direction of the rolled material Rm, with point O on the mill centerline Lm as the center of view, and tilted upwards at an angle θ2 relative to the horizontal plane, with the distance from point O to camera 21 being L. cam The position of point O and the angles θ1 and θ2 are preferably set to a position where there are no obstacles between the camera 21 and the rolled material Rm, and as close as possible to the roughing mill 3. Additionally, the distance L... cam The camera is preferably configured to ensure that the field of view of the camera 21 is 5m to 10m in the direction of travel of the rolled material Rm. Distance L camIt can be adjusted according to the lens of camera 21 (not shown). The image captured by camera 21 is input to processing circuit 22.

[0041] The tracking unit 23 uses the rolling load detected by the rolling load detection unit 31 of the roughing mill 3, and the forward slip ratio predicted based on the rotational speed of the main machine 33 that rotates the rolling roll 32, the rotational speed of the roll table 4, and the reduction of the rolled material Rm, to estimate the moving distance of the front end of the rolled material Rm from the roughing mill 3.

[0042] The processing circuit 22 processes the images captured by the camera 21. The processing circuit 22 processes the images captured from the input images from the camera 21 from a predetermined start time to a predetermined end time. The start and end times of shooting are determined based on signals from the tracking unit 23, specifying the timing when the front end is at a predetermined transport position. The tracking unit 23 can be configured as part of the processing circuit 22.

[0043] Figure 2 This is a diagram illustrating an example of the hardware configuration of a processing circuit. For example... Figure 2 As shown, the processing circuit 22 may also include dedicated hardware 22a, a processor 22b, and a memory 22c. At least a portion of the processing circuit 22 may be at least one piece of dedicated hardware 22a. In this case, the processing circuit 22 may be equivalent to a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a circuit combining these. The processing circuit 22 may also include at least one processor 22b and at least one memory 22c. In this case, the functions of the contour detection device 2 are implemented by software, firmware, or a combination of software and firmware. The software and firmware are described as programs and stored in the memory 22c. The processor 22b implements the functions of each part of the contour detection device 2 by reading and executing the programs stored in the memory 22c. The processor 22b is also referred to as a CPU (Central Processing Unit), central processing unit, processing device, arithmetic unit, microprocessor, microcomputer, or DSP. The memory 22c may be equivalent to non-volatile or volatile semiconductor memories such as RAM, ROM, flash memory, EPROM, or EEPROM. In this way, the processing circuit 22 can implement the various functions of the contour detection device 2 through hardware, software, firmware, or a combination thereof.

[0044] The processing circuit 22 is configured to perform the following processing. Figure 3 This is a flowchart representing the process performed by the processing circuit 22.

[0045] The processing circuit 22 first segments the image input from the camera 21 into multiple frames (step S1). The multiple frames are, for example, color images, but are not limited to this; they can also be images after converting color images to grayscale.

[0046] Processing circuit 22 applies differential filters to multiple frames respectively. Furthermore, the timing of the differential filter application is not limited to this; it can be applied to the target frame determined in step S2 (described later), or to the synthesized image obtained in step S5 (described later). As the differential filter, a Laplacian filter performing second differentiation can be appropriately used, but it is not limited to this; a Sobel filter performing first differentiation or a Scharr filter performing third differentiation can also be used. By using such a differential filter, an edge image can be obtained where areas with abrupt brightness changes are emphasized as edges. Furthermore, the frames to which the differential filter is applied are not limited to color images; one or more channels of the RGB of a color image can be selected for application. Additionally, the differential filter can be applied to the image obtained by grayscale conversion of a color image, or to one or more channels of the RGB, and in the case of multiple channels, the images after applying the filter are synthesized for grayscale conversion.

[0047] Next, the processing circuit 22 determines the object frame from multiple frames (edge ​​images) (step S2). In step S2, it is determined whether the multiple frames match the object frame. The object frame is the frame that becomes the object of the synthesis (synthesis processing), and it is the frame in which the front end of the rolled material Rm is located within a specified range of the frame. Figure 4 This is a diagram illustrating an example of a method for determining object frames. For example... Figure 4 As shown, in frame 5 (hereinafter also referred to as "image"), a region of interest (ROI) 51 is set upstream of the mill centerline Lm, and a region of interest 52 is set downstream of the mill centerline Lm. Frame 5 is binarized into maximum and minimum brightness. If the number of pixels in region of interest 51 that become the maximum brightness is above a threshold and the number of pixels in region of interest 52 that become the maximum brightness is below a threshold, it can be determined that it is an object frame. In this case, it is determined that the leading edge of the rolled material Rm is located between the two regions of interest 51 and 52. The shapes of regions of interest 51 and 52 are, for example, rectangles (in... Figure 4 (The center is a square). Alternatively, it could be determined by referring to... Figure 5 The method described later detects the front-end position when detecting the coordinate group. If the front-end position is within a predetermined front-end detection range 53, it is determined to be an object frame. As described later, Figure 5 In the attached figure, reference numeral 54 indicates a moving window that slides along the mill centerline Lm within the front detection range 53.

[0048] Processing circuit 22 performs steps S3 to S8 on multiple object frames. Steps S3 to S5 are compositing processes for synthesizing multiple object frames to obtain a single composite image. Steps S6 to S8 are conventional and known processes performed previously.

[0049] In step S3, the processing circuit 22 detects coordinate groups as feature quantities of the rolled material Rm in each object frame. In this step S3, the coordinate group of the endpoints of the quadrilateral that exists on the same horizontal plane as the front end position P defined with reference to the front end position P of the rolled material Rm is detected as a feature quantity of the rolled material Rm.

[0050] Figure 5 This is a diagram illustrating an example of a front-end inspection method for the rolled material Rm. (See diagram for example.) Figure 5 As shown, a front-end detection range 53 is defined within the object frame 5 corresponding to the field of view of camera 21, and a moving window 54 is defined that slides along the mill centerline Lm within the front-end detection range 53. The dimensions of the front-end detection range 53 and the moving window 54 in the plate width direction are set to be constant. Furthermore, the sliding position of the moving window 54 is independent of its sliding position within the front-end detection range 53; its long side is parallel to the plate width direction of the rolling mill, and its short side is parallel to the long side direction of the rolling mill. While the moving window 54 is sliding, the average brightness value within the moving window 54 is calculated, and the sliding position of the moving window 54 and the average brightness value are correspondingly stored in memory 22c. The sliding position with the largest average brightness value in the average brightness values ​​stored in memory 22c is detected as the front-end position.

[0051] Figure 6 , Figure 7 as well as Figure 8 This is a diagram illustrating an example of a method for defining the plane of the rolled material Rm in each object frame. For example... Figure 6 , Figure 7 and Figure 8 As shown, the plane of the rolled material Rm in object frame 5 is defined by the vertices of the quadrilateral (points a to d, these 4 points), and the coordinate group containing the coordinates of these 4 points is detected as a feature quantity.

[0052] exist Figure 6 In the example shown, taking the front end position P on the mill centerline Lm as a reference, points moved by a distance Lw from the front end position P towards the positive side (inside in the figure) or negative side (near the front in the figure) in the width direction are defined as points a and b, respectively. Furthermore, points moved by a and b towards the tail end in the long side direction by a distance Ll are defined as points d and c, respectively. Alternatively, points moved by a and b towards the front end in the long side direction by a distance Ll can also be defined as points d and c, respectively.

[0053] exist Figure 7In the example shown, taking the front end position P on the mill centerline Lm as a reference, edge detection is performed on a straight line extending in both directions from point Q, which has been moved a distance Ll from the front end position P towards the tail end in the long side direction. The points at the wide end of the rolled material Rm obtained through edge detection are defined as points c and d, respectively. Furthermore, the points that have been moved a distance Ll from points c and d towards the front end (outside the rolled material Rm) in the long side direction are defined as points b and a, respectively.

[0054] exist Figure 8 In the example shown, point c or point d (point c in this example) overlaps with the side guide 41, which is part of the rolling mill, so point c, which is the width end of the rolled material Rm, cannot be detected. In this case, the coordinates of point c are detected using the following method. That is, point c is defined as the point that has moved a distance Lb along the width direction from point d, which is normally detected as the width end of the rolled material Rm. This moving distance Lb can be set as the plate width of the rolled material Rm calculated and measured, or the distance Lb can be gradually and significantly varied, set to be equal to the width of the rolled material Rm. Figure 7 The same method was used to detect the maximum length from point d to point c (equivalent to the plate width) and twice the length from point d to point Q.

[0055] In step S4, processing circuit 22 calculates the coordinate transformation matrix. In this step S4, the perspective transformation matrix is ​​calculated as the coordinate transformation matrix. Processing circuit 22 first sets (selects) one reference frame 5r from multiple object frames 5, and uses the coordinate group (feature quantity) of four points (points a to d) of the reference frame 5r as the reference coordinate group. Next, processing circuit 22 calculates the perspective transformation matrix M that transforms the coordinate group of four points (points a to d) of each object frame 5 other than the reference frame 5r into the reference coordinate group of four points (points a to d) of the reference frame 5r. Furthermore, the reference coordinate group does not necessarily need to be set on the reference frame 5r, and can also be preset on the object frames 5. In this case, in step S5 described later, instead of overlapping the perspective-transformed image with the reference frame 5r, the perspective-transformed images (object frames) are overlapped to obtain the composite image 6.

[0056] Set the coordinates of the 4 points on the nth object frame 5n as Set the coordinates of the 4 points of the reference frame 5r as When the perspective transformation matrix Mn is a 3×3 matrix as specified in equation (1), the perspective transformation matrix Mn is a 3×3 matrix as specified in equation (1).

[0057] Here, i r It is the X-axis position (X coordinate) of a point in the reference frame 5r, j r It is the Y-axis position (Y coordinate) of a point in the reference frame 5r, in It is the X-axis position (X coordinate) of a point in object frame 5n, j n is the Y-axis position (Y coordinate) of a point in object frame 5n. h is a component of the perspective transformation matrix Mn, and k is the position of the point group (k=a, b, c, d).

[0058] In each object frame 5 other than the reference frame 5r, perspective transformation is performed using the calculated perspective transformation matrix M. At this time, the image I before perspective transformation... in Each pixel of the object is transformed into the perspective-transformed image I using the components of the perspective transformation matrix M of frame 5, as shown in equation (2). trans Each pixel. The perspective transformation matrix M represents the image I after perspective transformation. trans Each pixel of the image before perspective transformation in Which pixel corresponds to which matrix?

[0059] In equation (2) above, i is the image I before perspective transformation. in The X-axis position of the pixel in the image, where j is the image before perspective transformation. in The Y-axis position of the pixel in the image is given by m, where m is the image channel and h is a component of the perspective transformation matrix M. However, in the image I after perspective transformation... trans The pixels do not contain the image before perspective transformation I in In the case of pixels, the image I after perspective transformation trans The brightness of the pixels is set to 0.

[0060] Here, the matching of the detected front-end position P in each object frame 5 can also be verified based on conditions, and object frames 5 that do not meet the conditions can be excluded from the composite object. For example, a statistical value can be calculated based on the relationship between the front-end position P of each object frame 5 and the average brightness within the moving window 54, and object frames 5 whose statistical values ​​deviate from the distribution of all object frames 5 can be excluded from the composite object. As another method to confirm the matching based on the front-end position P of each object frame 5, the following method can be cited. That is, when the recording start timing and end timing are adjusted to keep the travel direction of the rolled material Rm in the image constant, the detected front-end position P travels in the same direction between object frames 5. Based on this relationship, if the front-end position P of the object frame 5 is not located between the front-end positions P of the object frames before and after it, the object frame 5 can be excluded from the composite object. Alternatively, the actual values ​​of the tracking unit 23 from the start to the end of the recording can be collected and compared with the front-end positions P of each object frame 5, and if the difference is large, the object frame 5 can be excluded from the composite object.

[0061] In step S5, the processing circuit 22 obtains a composite image 6 by overlaying all the images of the object frame that have undergone perspective transformation using the perspective transformation matrix M with the reference frame 5r. The brightness of each pixel in the composite image 6 is represented by the following formula (3).

[0062] In the above equation (3), I out This is a composite image 6, where i is the X-axis position of the pixel, j is the Y-axis position of the pixel, m is the image channel, and I... trans This is the image after perspective transformation (object frame 5), where n is the object frame number and N is the total number of object frames.

[0063] Steps S6 to S8 are processes for obtaining the information used in step S13 (described later), and are the same as conventional processes. In step S6, the processing circuit 22 detects the portion with large brightness changes (the boundary between maximum and minimum brightness) in the edge image of each object frame after applying a differential filter as a contour. Based on the detected contour, the processing circuit 22 determines the warping direction (upward or downward warping) (step S7) and calculates the warping height (step S8). The warping height calculated for each object frame is temporarily stored in memory 22c and is read out during the processing in step S13 (described later). The conventional processes of steps S6 to S8 and the process of step S13, which performs warping measurement using statistical processing, are known, including the disclosures of Patent Documents 1 to 3, and therefore detailed descriptions are omitted here. After processing all object frames in step S8, the process proceeds to step S9.

[0064] However, when the rolled material Rm warps downwards and to a large degree, the leading edge of the rolled material Rm sometimes collides with the roller table 4 and bounces up and down. If such bouncing occurs, an afterimage is generated at the leading edge of the rolled material Rm in the composite image 6. As a result, the brightness of this leading edge in the composite image 6 decreases, making it impossible to accurately detect the contour of the leading edge of the rolled material Rm, and it may be mistakenly measured as upward warping. The inventors of this application conducted in-depth research and found that, in the case of afterimages caused by bouncing, the bouncing portion also becomes white in the black and white image obtained by binarizing the composite image 6, thus increasing the number of white pixels.

[0065] Based on the above findings, in step S9, a black and white image (binarized image) is obtained by binarizing the synthesized image. Next, the area of ​​white pixels in the black and white image is calculated (step S10). In step S10, at least one (or more) regions of interest (e.g., regions of interest) pre-defined in the black and white image (i.e., the synthesized image) are calculated. Figure 4The area of ​​the white pixels in the interest regions 51 and 52 shown. Here, at least one interest region can be set to include the portion affected by bouncing, for example, taking into account past performance and reference coordinate groups. Furthermore, when multiple interest regions are set, the multiple interest regions can be set to be adjacent to each other or set to be separate.

[0066] Next, it is determined whether the area of ​​the white pixels calculated in step S10 meets the determination criteria (step S11). When multiple interest regions are set, a determination criterion can be set for each interest region. As a determination criterion, cases such as exceeding the criterion value and falling below the criterion value can be exemplified. The determination criterion including the criterion value is set accordingly with the interest region. For example, if two interest regions are set, and the area of ​​the white pixels in the first interest region exceeds the criterion value, while the area of ​​the white pixels in the second interest region is below the criterion value, it can be determined that the determination criterion is met. In this way, the determination criterion can be arbitrarily set according to the interest region.

[0067] If the determination criteria are met in step S11, it is determined that no bounce has occurred, and the process proceeds to step S12. Conversely, if the determination criteria are not met, it is determined that a bounce has occurred, and the process proceeds to step S13. In step S13, using the same method as in the conventional example, the warp direction and warp height of the multiple unsynthesized object frames are statistically processed to measure the warp. In the statistical processing, for example, the median value of the warp height calculated for the multiple object frames is obtained, but it is not limited to this. The maximum value can also be obtained, or the median value of the result (warp height) after removing the maximum and minimum values ​​can also be obtained.

[0068] In step S12, the processing circuit 22 detects the contour Ed of the rolled material Rm from the composite image 6. Figure 9This is a diagram illustrating an example of a method for detecting the contour Ed of the rolled material Rm from the composite image 6. First, the leading edge position (leading edge point) P of the rolled material Rm in the composite image 6 is detected. Next, a region of interest (ROI) 55 is defined, extending from the leading edge point P on the mill centerline Lm, with a length Lsw (mm) in the plate width direction and a length Lsl (mm) in the long side direction. Then, a moving window 56 is defined inside the ROI 55, moving along the long side direction, having the same length Lsw (mm) in the plate width direction as the ROI 55, and an arbitrary length Lml (mm) smaller than Lsl in the long side direction. Furthermore, a moving window 57 is defined inside the moving window 56, moving along the plate width direction, having the same length Lml (mm) in the long side direction as the moving window 56, and an arbitrary length Lmw (mm) smaller than Lsw in the plate width direction. At each moving position along the long side of the moving window 56, the average brightness value within the moving window 57 when it moves along the width of the plate is stored in the memory 22c corresponding to the moving position. At each moving position along the long side of the moving window 56, the center point C of the moving window 57 when the average brightness value within the moving window 57 is at its maximum is detected as the profile Ed of the rolled material Rm.

[0069] Here, a composite image 6, synthesized by performing perspective transformation on the edge images of each object frame 5, is used to detect the contour Ed. However, this is not the only method used; the same result can be obtained through other processing methods. For example, a binarized image can be obtained by binarizing the composite image 6, and the center position of the pixel with the maximum brightness in the plate width direction at each long side position within the region of interest 55 can be detected as the contour Ed of the rolled material Rm. Additionally, a noise removal filter can be applied to the composite image 6 as needed.

[0070] Next, refer to Figures 10-14 The processing circuit 22, which is part of the shape measurement circuit, quantifies the warpage height based on the detected contour Ed. Firstly, as... Figure 10 As shown, the inner contour Ed1 in the width direction is extracted from the contour Ed of the rolled material Rm. A perspective transformation is performed on the extracted contour Ed1, and the warping direction (upward or downward warping) is determined based on the characteristics of the transformed contour Ed2. Then, the warping height is quantified according to the conditions for each warping direction. The perspective transformation of contour Ed1, the determination of the warping direction, and the calculation of the warping height performed by the processing circuit 22 are explained below.

[0071] (Perspective transformation of the outline) To convert the unit [pix] of warpage height in the image to [mm], an image of the rolled material Rm viewed from the front side is obtained using the method described below. Specifically, the processing circuit 22 defines the plane containing the inner contour Ed1 in the width direction, i.e., the plane for calculating the warpage height, using the following method. For example... Figure 11 As shown, the extension of the non-warped portion of contour Ed1 is defined as the baseline Lr. The intersection of baseline Lr and the predetermined straight lines t3-t4 (described later) is detected as point r1. The intersection of baseline Lr and the predetermined straight lines t1-t2 (described later) is detected as point r2. Then, as... Figure 12 As shown, the intersection of the line connecting point r1 and vanishing point VP3 (described later) with the predetermined line t8-t7 is detected as point r4. Then, the intersection of the line connecting point r2 and vanishing point VP3 with the predetermined line t5-t6 is detected as point r3. Plane Pn1 is defined by the points r1, r2, r3, and r4 detected through the above processing. The short side of plane Pn1 extends along the plate thickness direction, and the long side extends along the long side direction. Then, the perspective transformation of plane Pn1 is calculated to have… Figure 13 The matrix M of the plane Pn2 (defined by points r1', r2', r3', r4') of the rectangle with predetermined side lengths (in the length direction) shown is... meas .like Figure 13 As shown, the correction material Cm has a known thickness dimension Tc [mm] and a long side dimension Lc [mm]. When the short side of the rectangle is set to Tc / SDes [pix] and the long side to Lc / SDes [pix] based on these dimensions Tc, Lc, and the desired pixel size transformation coefficient SDes [mm / pix], in the perspective-transformed plane Pn2, since the line in the thickness direction is orthogonal to the line in the long side direction, it becomes the desired pixel size transformation coefficient SDes [mm / pix] at any position in the image. That is, one pixel size transformation coefficient SDes can be used to transform the unit [pix] of warpage height to [mm].

[0072] (Determination of warping direction) Next, the processing circuit 22 utilizes M meas A perspective transformation is performed on the group of points that constitute the inner contour Ed1 in the width direction, to obtain... Figure 14 The outline Ed2 after perspective transformation as shown. Figure 14This diagram illustrates an example of a method for determining the direction of warping. From the points on the perspective-transformed contour Ed2, the highest point p along the Y-axis (vertical direction in the diagram) is defined as the peak point, and the point q that will become the foremost point of contour Ed2 is defined as the foreground point. At this point, if the distance d along the long side from the peak point p to the foreground point q is less than a pre-set threshold, it is determined to be upward warping (see reference). Figure 14 (a) is considered a downward warp if it exceeds the threshold (see reference). Figure 14 (b)). The threshold can be set appropriately based on past performance.

[0073] Thus, near the front end of the rolled material Rm, the warping direction is determined based on the distance d along the long side from point p to point q of the inner contour Ed2 in the width direction, but it is not limited to this. For example, near the front end of the rolled material Rm, the warping direction can be determined by performing the same treatment on the near-front contour in the width direction. In this case, the warping height calculation described later can be performed only if the warping directions near the front and the inner side in the width direction are consistent. If the warping directions determined for the inner and near-front sides in the width direction are inconsistent, the warping height is calculated using the method described later.

[0074] (Calculation of warpage height) In step S12 above, the processing circuit 22 calculates the warpage height based on the inner contour Ed2 in the width direction. In the case of upward warpage, such as Figure 14 As shown in (a), the lowest point in the Y-axis direction of the non-warped profile Ed2 is defined as the lowest point r, and the difference h between the height of the peak point p and the lowest point r in the plate thickness direction is calculated as the warpage height. In the case of downward warpage, as... Figure 14 As shown in (b), the difference h between the heights of the peak point p and the front point q in the thickness direction is calculated as the warpage height. The height of the upper warpage is a positive value, and the height of the lower warpage is a negative value.

[0075] Of course, the image synthesis and warpage measurement described above can be performed after one rolling pass of the roughing mill 3, and the control corresponding to the measured warpage (e.g., change in the rotational speed of the upper and lower rolling rolls 32) can be applied to the next rolling pass. However, by performing image synthesis and warpage measurement during the rolling pass, it can also be applied to the current rolling pass, thus achieving real-time performance. When performed during the current rolling pass, it can also be configured such that if a predetermined number of object frames 5 are not obtained, data from past rolling passes is read from memory 22c to obtain the synthesized image 6.

[0076] (Compensation method for plate width variation) Reference Figure 15A method for quantifying the warpage height when a width variation occurs at the leading edge of the rolled material Rm is described. When a width variation occurs at the leading edge of the rolled material Rm, and the aforementioned warpage direction determination is applied to both the inner contour Ed2 in the width direction and the contour near the leading edge in the width direction, the warpage direction determination results may sometimes be inconsistent. For example, as... Figure 15 As shown, when the width near the front end of the rolled material Rm is narrow, based on the inner contour Ed2 and distance da in the width direction, it is possible to falsely detect downward warping. On the other hand, based on the contour Ed3 near the front in the width direction and distance db, it is possible to falsely detect excessive upward warping. Therefore, in order to compensate for the width variation at the front end of the rolled material Rm and accurately measure the warping height, the following method can be used. That is, at each position in the long side direction (hereinafter referred to as "each long side position"), a line segment extending along the width direction on the rolled material Rm is detected, and the point Pc, which is the center point in the width direction of each line segment, is detected as the center point, and the center line La connecting the center points Pc of each long side position is obtained. Using this center line La to replace the contours Ed2 and Ed3 on both sides in the width direction, the above-mentioned perspective transformation, warping direction determination, and warping height calculation are performed. As a result, the width variation at the front end of the rolled material Rm can be compensated, and the warping height can be measured with high accuracy. In addition, this compensation method is based on the premise that the width variation of the front end of the rolled material Rm is generated approximately equally on the inner side and near the front side in the width direction.

[0077] (Correction method) Figure 16This diagram illustrates an example of a correction method using a correction material Cm to correct vanishing points VP1, VP2, and VP3. The correction material Cm has a known plate thickness Tc, plate width Wc, and length Lc. The correction material Cm is positioned such that its width-direction center aligns with the mill centerline Lm, and an image of the correction material Cm is taken. Points t1 to t6 are detected on the captured image. The point where the line connecting points t1 and t2 intersects with the line connecting points t3 and t4 is defined as vanishing point VP1. Similarly, the point where the line connecting points t1 and t4 intersects with the line connecting points t2 and t3 is defined as vanishing point VP2. Likewise, the point where the line connecting points t1 and t5 intersects with the line connecting points t2 and t6 is defined as vanishing point VP3. If the X or Y coordinates, or both, of these vanishing points VP1, VP2, and VP3 exceed a predetermined value sufficiently large compared to the number of pixels in the frame, it is assumed that the vanishing point does not exist, and lines parallel to the aforementioned straight lines are designated as straight lines in their respective axial directions. The reference straight lines are the line connecting points t1 and t5 in the plate thickness direction, the line connecting points t1 and t2 in the plate width direction, and the line connecting points t1 and t4 in the long side direction. Next, the intersection of the lines (t7-VP1), (t5-VP2), and (t4-VP3) is detected as point t8. The coordinates of the vanishing points VP1 to VP3 and points t1 to t8 defined through the above operations are stored in memory 22c and used in the detection of the aforementioned coordinate group (feature quantity), calculation of warpage height, etc.

[0078] As explained above, according to this embodiment, a coordinate transformation matrix M is used to synthesize multiple object frames 5 at the front end of the rolled material Rm. Figure 17 As shown, even if interference such as water vapor or stagnant water occurs, since the interference does not exist in the same location in multiple (four in the figure) object frames 51-54, the interference is canceled out when multiple object frames 51-54 are synthesized. As a result, the synthesized image 6 obtained from multiple object frames 51-54 becomes clear, and the contour Ed of the rolled material Rm can be detected with high precision from the clear synthesized image 6. In addition, based on the contour Ed of the rolled material detected from the clear synthesized image 6, the warpage of the rolled material Rm can be measured with high precision.

[0079] The embodiments of this disclosure have been described above, but this disclosure is not limited to the above-described embodiments, and various modifications can be made to implement it without departing from the spirit of this disclosure. When numerical values ​​such as the number, quantity, amount, and range of each element are mentioned in the above embodiments, the present invention is not limited to the mentioned numerical values, except where specifically stated or explicitly determined in principle. Furthermore, the structures described in the above embodiments are not essential to the present invention, except where specifically stated or explicitly determined in principle.

[0080] In the above embodiments, the example described is based on measuring the warp direction and warp height of the rolled material Rm, but the shape of the rolled material is not limited to warping. This disclosure can also be applied to, for example, Figure 18 The arch shown in (a) Figure 18 The shape shown in (b) is a wavy surface shape called flatness, which is measured based on the profile Ed of the rolled material Rm.

[0081] In the above embodiment, the coordinate group of the four points at the front end of the rolled material Rm is detected as a feature quantity. However, when the rolled material Rm is a short dimension, the coordinate group of the four points at the rear end of the rolled material Rm can also be detected as a feature quantity.

[0082] In the above embodiment, the example described is the calculation of the perspective transformation matrix M used to perform perspective transformation on the object frame 5 so that the shape and size of the rolled material Rm in the reference frame 5r are consistent with the shape and size of the rolled material Rm in the object frame 5, but it is not limited to this. When the camera 21 is set at a position far from the mill centerline Lm, and the rolled material Rm in multiple object frames 5 is the same size and can be regarded as similar, that is, when the size change of the rolled material Rm in multiple object frames 5 can be ignored, a composite image can be obtained by parallel movement. In this case, the coordinates of 3 points are detected for each object frame 5, and the affine transformation matrix is ​​calculated as the coordinate transformation matrix. Since the affine transformation matrix does not need to include rotation, the affine transformation matrix can be represented as a 3×2 matrix.

[0083] In the above embodiments, the example described is the detection of the profile Ed of the rolled material Rm that has been rolled by the roughing mill 3, but it is not limited thereto. This disclosure can be applied, for example, to the detection of the profile of the rolled material Rm that is being transported in one direction in the hot rolling line RL, as in the case of detecting the profile of the rolled material Rm that has been rolled by the finishing mill.

[0084] Explanation of reference numerals in the attached figures RL… hot rolling line, Rm… rolled material, 1… shape measuring device, 2… contour detection device, 21… camera, 22… processing circuit (shape measuring circuit), 23… tracking unit, 3… roughing mill, 31… rolling load detection unit, 32… rolling roll, 33… main machine, 4… roll table, 41… side guide, 5… object frame, 5r… reference frame, 6… composite image.

Claims

1. A profile detection device that detects a profile of a rolled material being transported in a direction in a hot rolling line, comprising: a camera that captures the rolled material; and a processing circuit that processes an image captured by the camera, the processing circuit configured to: divide the image into a plurality of frames; determine a plurality of object frames in which a leading end of the rolled material is present from among the plurality of frames; detect a group of coordinates that indicates a position of the rolled material in each of the object frames; calculate a coordinate transformation matrix that transforms the group of coordinates of each of the object frames into a reference group of coordinates that is set in advance; obtain one composite image from the plurality of object frames that are transformed using the coordinate transformation matrix; and detect the profile of the rolled material based on luminance of the composite image.

2. The profile detection device according to claim 1, wherein calculating the coordinate transformation matrix includes setting the group of coordinates of a reference frame selected from among the plurality of object frames as the reference group of coordinates.

3. The profile detection device according to claim 1, wherein detecting the group of coordinates includes detecting, for each of the object frames, the group of coordinates that includes 4-point coordinates, and calculating the coordinate transformation matrix includes calculating a perspective transformation matrix as the coordinate transformation matrix.

4. A shape measurement device that measures a shape of a rolled material being transported in a direction in a hot rolling line, comprising: the profile detection device according to any one of claims 1 to 3; and a shape measurement circuit configured to measure the shape of the rolled material based on the profile of the rolled material detected from the composite image by the profile detection device.

5. The shape measurement device according to claim 4, wherein the shape measurement circuit is configured to perform measurement of a warping direction and a warping height of a leading end of the rolled material.

6. The shape measurement device according to claim 5, wherein the shape measurement circuit is configured to: obtain a black-and-white image by performing a binarization process on the composite image; and calculate an area of white pixels in one or a plurality of regions of interest set in the black-and-white image, the measurement of the warping direction and the warping height being performed when the area of white pixels in the one or the plurality of regions of interest satisfies a determination criterion set for each of the regions of interest. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​

Images