Contour detection device and shape measuring device

Abstract

A contour detection device disclosed herein comprises: a camera for imaging a material to be rolled; and a processing circuit for processing a video captured by the camera. The processing circuit is configured to execute: dividing the video into a plurality of frames; identifying a plurality of target frames, in which the tip of the material to be rolled is present, among the plurality of frames; detecting a coordinate group representing the position of the material to be rolled in each target frame; calculating a coordinate conversion matrix for converting the coordinate group of each of the target frames into a preset reference coordinate group; obtaining one composite image from the plurality of target frames converted using the coordinate conversion matrix; and detecting the contour of the material to be rolled from the luminance 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 conveyed in one direction along 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 are shape measuring devices that measure the shape of rolled materials, such as warpage, camber, and flatness, when rolled on a hot rolling line. To explain using warpage as an example, the following Patent Documents 1 to 3 disclose warpage measuring devices as shape measuring devices.

[0003] In Patent Document 1 below, a camera is installed on the same horizontal plane as the rolled material and perpendicular to the widthwise centerline of the conveying table (hereinafter also referred to as the "mill line"). Using this camera, the widthwise end of the rolled material is imaged from the side, the captured image is binarized, and the contour of the rolled material is extracted from the binarized image.

[0004] In Patent Document 2 below, a camera is installed diagonally above the stands (housings) of a finishing rolling mill so as to capture the area between the stands. The image captured by this camera is divided at a predetermined pitch along the rolling direction (longitudinal direction), and the shape of the plate width edge in each divided image is approximated by a quadratic equation, and the amount of warping is quantified by curvature.

[0005] In Patent Document 3 below, a first camera is installed to image the rolled material from the side, and a second camera is installed to image the rolled material from the transport direction. Based on the crop shape extracted from the image captured by the second camera, the edge shape extracted from the image captured by the first camera is corrected. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 1-285730 [Patent Document 2] Japanese Patent No. 6828730 [Patent Document 3] Japanese Patent No. 3724720 [Overview of the project] [Problems that the invention aims to solve]

[0007] Incidentally, in order to accurately measure the curvature shape, it is necessary to acquire a clear image. However, in a hot rolling line, the roller table is heated by the rolled material heated in the heating furnace, and high-pressure water is sprayed onto the heated roller table from a roll cooling device or descaler. This generates water vapor and causes water to accumulate on the rolled material. In addition, reflected light reflected by the side guides provided on both sides in the width direction of the roller table is irradiated onto the rolled material. When this water vapor, accumulated water, and reflected light are captured as disturbances in the image, the acquired image becomes unclear, and as a result, the contour of the rolled material cannot be accurately detected.

[0008] Furthermore, since water vapor tends to linger for long periods, fans are sometimes installed to physically blow it away. However, in low-temperature environments such as winter, the amount of water vapor generated is large, and there are limits to the fan's capacity.

[0009] This disclosure was made to solve the problems described above. The purpose of this disclosure is to provide a contour detection device that can acquire a clear image suitable for detecting the contour of a rolled material even when disturbances such as water vapor occur. Another purpose of this disclosure is to provide a shape measuring device that can accurately measure the shape of a rolled material from the contour detected from the composite image by the above contour detection device. [Means for solving the problem]

[0010] The first aspect relates to a contour detection device for detecting the contour of a rolled material being transported in one direction along a hot rolling line. The contour detection device includes a camera for imaging the rolled material and a processing circuit for processing the images captured by the camera. The processing circuit is configured to perform the following actions: divide the image into multiple frames; identify multiple target frames from among the multiple frames in which the leading edge of the rolled material is located; detect a set of coordinates representing the position of the rolled material in each target frame; calculate a coordinate transformation matrix that transforms the coordinate set of each target frame into a pre-set set of reference coordinates; obtain a single composite image from the multiple target frames transformed using the coordinate transformation matrix; and detect the contour of the rolled material from the brightness of the composite image.

[0011] The second perspective, in addition to the first, has the following further characteristics: Calculating the coordinate transformation matrix involves setting a set of coordinates from a reference frame selected from multiple target frames as the reference coordinate set.

[0012] The third perspective, in addition to the first perspective, has the following further characteristics: Detecting coordinate sets includes detecting coordinate sets containing the coordinates of four points for each target frame. Calculating the coordinate transformation matrix includes calculating the perspective transformation matrix as the coordinate transformation matrix.

[0013] The fourth aspect relates to a shape measuring device for measuring the shape of a rolled material being conveyed in one direction along a hot rolling line. The shape measuring device comprises a contour detection device as described in any one of the first to third aspects, 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 composite image by the contour detection device.

[0014] The fifth aspect, in addition to the fourth aspect, has the following further features: The shape measurement circuit is configured to perform the measurement of the curvature direction and curvature height of the leading edge of the rolled material.

[0015] According to a sixth aspect, in addition to the fifth aspect, the following feature is further provided. The shape measurement circuit is configured to perform: obtaining a black-and-white image by performing binarization processing on the composite image; and calculating an area of white pixels in one or more regions of interest set in the black-and-white image. The measurement of the warpage direction and the warpage height is performed when the area of white pixels in the one or more regions of interest satisfies a determination criterion set for each region of interest.

Effects of the Invention

[0016] According to the contour detection apparatus of the present disclosure, a plurality of target frames in which the tip end of a material to be rolled is present are combined using a coordinate transformation matrix. Here, even if disturbance such as water vapor occurs, the disturbance does not exist at the same position in the plurality of target frames, so the disturbance is canceled out when combining the plurality of target frames. As a result, a composite image obtained from the plurality of target frames becomes clear, and the contour of the material to be rolled can be accurately detected from the clear composite image.

[0017] Further, according to the shape measurement apparatus of the present disclosure, the shape of the material to be rolled can be accurately measured based on the contour of the material to be rolled detected from the clear composite image by the above contour detection apparatus. The present disclosure can be suitably applied particularly when measuring the warpage direction and warpage height of the tip end of the material to be rolled.

Brief Description of Drawings

[0018] [Figure 1] It is a schematic diagram schematically showing the configuration of a warpage measurement apparatus including the contour detection apparatus according to an embodiment. [Figure 2] It is a diagram showing an example of the hardware configuration of a processing circuit. [Figure 3] It is a flowchart showing the flow of processing executed by the processing circuit. [Figure 4] It is a diagram for explaining an example of a method of specifying a target frame. [Figure 5] It is a diagram for explaining an example of a tip detection method for a material to be rolled. [Figure 6]This diagram illustrates an example of a method for defining the plane of the rolled material in each target frame. [Figure 7] This diagram illustrates an example of a method for defining the plane of the rolled material in each target frame. [Figure 8] This diagram illustrates an example of a method for defining the plane of the rolled material in each target frame. [Figure 9] This figure illustrates an example of a method for detecting the contour of a rolled material from a composite image. [Figure 10] This diagram illustrates an example of a method for quantifying curvature height from the contour. [Figure 11] This diagram illustrates an example of a method for quantifying curvature height from the contour. [Figure 12] This diagram illustrates an example of a method for quantifying curvature height from the contour. [Figure 13] This is a diagram to explain the perspective transformation of contours. [Figure 14] (a) and (b) are diagrams showing examples of methods for determining the direction of warping and calculating the warping height. [Figure 15] This figure shows an example of a method for compensating for plate width fluctuations that occur at the leading edge of a rolled material. [Figure 16] This figure shows an example of a method for calibrating vanishing points using calibration material. [Figure 17] This is a diagram illustrating the effects of this embodiment. [Figure 18] This figure shows other application examples of this disclosure. [Modes for carrying out the invention]

[0019] The embodiments of this disclosure will be described below with reference to the drawings, using as an example a case in which the contour of a rolled material rolled in a roughing mill on a hot rolling line is detected and the warp shape of the rolled material is measured from the detected contour. That is, embodiments of a shape measuring device will be described using a warp measuring device as an example. In each figure, elements common to each other are denoted by the same reference numerals, and redundant explanations are omitted.

[0020] Figure 1 is a schematic diagram illustrating the configuration of a warpage measuring device equipped with a contour detection device according to an embodiment. The warpage measuring device 1 comprises a contour detection device 2 and a shape measurement circuit, which will be described later. The contour detection device 2 detects the contour of the rolled material Rm being conveyed in one direction on the hot rolling line RL. The contour detection device 2 detects the contour of the rolled material Rm that has been rolled by the roughing mill 3 on the hot rolling line RL.

[0021] The contour detection device 2 comprises a camera 21 as an imaging means, a processing circuit 22, and a tracking unit 23. The processing circuit 22 also serves as the "shape measurement circuit" of the warpage measurement device 1, but a separate shape measurement circuit with the same configuration as the processing circuit 22 can also be provided.

[0022] As camera 21, for example, a network camera with a resolution of 1280 x 720 pixels can be suitably used. The shooting mode of camera 21 can be set to color, the frame rate to 30, the exposure time to fixed, and the aperture value to automatic mode. The resolution can be set according to the processing capacity of the processing circuit 22. The frame rate is the number of frames contained in one second. The exposure time and aperture value can be set in advance according to the predicted temperature of the rolled material Rm. Furthermore, camera 21 is not limited to a network camera; for example, a machine vision camera or an infrared camera can be used in combination with a visible light cut filter, etc. Since these cameras are well known, further explanation will be omitted.

[0023] Camera 21 is positioned to capture images of the rolled material Rm being transported on the roller table 4 from an oblique angle above. For example, camera 21 is tilted horizontally by an angle θ1 with respect to a straight line Lv perpendicular to the mill center line Lm, which is also the transport direction of the rolled material Rm, and is tilted upward by an angle θ2 with respect to the horizontal plane, with point O on the mill center line Lm as the center of the field of view, and the distance from point O to camera 21 is L camIt is installed in such a manner. The position of point O and angles θ1, θ2 should preferably be set so that there are no obstacles between the camera 21 and the rolled material Rm, and it is as close as possible to the roughing mill 3. Also, distance L cam It is desirable to set the camera 21 so that its field of view is approximately 5 to 10 meters in the direction of travel of the rolled material Rm. Distance L cam This can be changed according to the lens of camera 21 (not shown). The image captured by camera 21 is input to processing circuit 22.

[0024] The tracking unit 23 estimates the distance traveled by the leading edge of the rolled material Rm from the roughing mill 3, using the rolling load detected by the rolling load detection unit 31 of the roughing mill 3, the rotational speed of the main engine 33 that rotates the rolling rolls 32, the rotational speed of the roller table 4, and the amount of reduction of the rolled material Rm.

[0025] The processing circuit 22 processes the video captured by the camera 21. The processing circuit 22 processes the video captured from the camera 21 between a predetermined start timing and an end timing. The start and end timings of the capture are specified by signals from the tracking unit 23, which indicate when the tip position is at a predetermined transport position. The tracking unit 23 can be configured as part of the processing circuit 22.

[0026] Figure 2 shows an example of the hardware configuration of a processing circuit. The processing circuit 22 may include dedicated hardware 22a, a processor 22b, and memory 22c, as shown in Figure 2. At least a part of the processing circuit 22 may be at least one dedicated hardware 22a. In this case, the processing circuit 22 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof. The processing circuit 22 may also include at least one processor 22b and at least one memory 22c. In this case, each function of the contour detection device 2 is realized by software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in memory 22c. The processor 22b realizes the functions of each part of the contour detection device 2 by reading and executing the programs stored in memory 22c. The processor 22b is also called a CPU (Central Processing Unit), central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, or DSP. Memory 22c includes, for example, non-volatile or volatile semiconductor memory such as RAM, ROM, flash memory, EPROM, and EEPROM. In this way, the processing circuit 22 can realize each function of the contour detection device 2 through hardware, software, firmware, or a combination thereof.

[0027] The processing circuit 22 is configured to perform the following tasks. Figure 3 is a flowchart showing the processes performed by the processing circuit 22.

[0028] The processing circuit 22 first divides the video 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 may also be images obtained by converting color images to grayscale.

[0029] The processing circuit 22 applies a differential filter to each of the multiple frames. The timing of application of the differential filter is not limited to this; it may also be applied to the target frame identified in step S2 described later, or to the composite image obtained in step S5 described later. A Laplacian filter that performs second derivative is preferably used as the differential filter, but it is not limited to this; a Sobel filter that performs first derivative or a Scherr filter that performs third derivative can also be used. By using such a differential filter, an edge image is obtained in which areas where the brightness changes abruptly 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 may be selected and applied. In addition, the differential filter may be applied to an image obtained by converting a color image to grayscale, or the differential filter may be applied to one or more channels of RGB, and in the case of multiple channels, the filtered images may be combined and converted to grayscale.

[0030] Next, the processing circuit 22 identifies the target frame from among multiple frames (edge ​​images) (step S2). In step S2, it is determined whether or not multiple frames correspond to the target frame. The target frame is the frame to be synthesized (synthesis processing), in which the leading edge of the rolled material Rm is located within a predetermined range of the frame. Figure 4 is a diagram illustrating an example of a method for identifying the target frame. As shown in Figure 4, in frame (hereinafter also referred to as "image") 5, a region of interest (ROI: Region of Interest) 51 is set upstream of the mill center line Lm, and a region of interest 52 is set downstream of the mill center line Lm. Frame 5 is binarized to maximum brightness and minimum brightness, and if the number of pixels that result in maximum brightness in region of interest 51 is greater than or equal to a threshold, and the number of pixels that result in maximum brightness in region of interest 52 is less than or equal to a threshold, it can be determined that it is the target 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 shape of regions of interest 51 and 52 is, for example, a rectangle (a square in Figure 4). Another method of identification is to use the tip position detected when detecting the coordinate group using the method described later with reference to Figure 5, and determine that it is the target frame if the tip position is located within a predetermined tip detection range 53. As will be described later, the reference numeral 54 in Figure 5 indicates a movable window that slides within the tip detection range 53 along the mill center line Lm.

[0031] The processing circuit 22 performs the processing in steps S3 to S8 for multiple target frames. Steps S3 to S5 are synthesis processes to combine multiple target frames to obtain a single composite image. Steps S6 to S8 are conventional processes that have been performed in the past.

[0032] In step S3, the processing circuit 22 detects a set of coordinates as a feature quantity of the rolled material Rm in each target frame. In step S3, the set of coordinates of the endpoints of a rectangle that lies on the same horizontal plane as the tip position P, which is defined with respect to the tip position P of the rolled material Rm, is detected as a feature quantity of the rolled material Rm.

[0033] Figure 5 illustrates an example of a method for detecting the tip of a rolled material Rm. As shown in Figure 5, a tip detection range 53 is defined within a target frame 5 corresponding to the field of view of the camera 21, and a movable window 54 is defined that slides along the mill center line Lm within the tip detection range 53. The dimensions of the tip detection range 53 and the movable window 54 in the plate width direction are set to be constant. In addition, regardless of the sliding position within the tip detection range 53, the long side of the movable window 54 is parallel to the plate width direction of the rolling equipment, and the short side is parallel to the longitudinal direction of the rolling equipment. While sliding the movable window 54, the average brightness value within the movable window 54 is calculated, and the sliding position of the movable window 54 and the average brightness value are associated and stored in memory 22c. Among the average brightness values ​​stored in memory 22c, the sliding position with the maximum average brightness value is detected as the tip position.

[0034] Figures 6, 7, and 8 illustrate, respectively, an example of how to define the plane of the rolled material Rm in each target frame. As shown in Figures 6, 7, and 8, the plane of the rolled material Rm in target frame 5 is defined by the vertices of a rectangle (four points from a to d), and the coordinate set including the coordinates of these four points is detected as a feature.

[0035] In the example shown in Figure 6, points a and b are defined as points moved by a distance Lw from the tip position P on the mill center line Lm, either on the positive side (far side in the figure) or the negative side (forward side in the figure) in the width direction. Points d and c are defined as points moved by a distance Ll from points a and b toward the tail end in the longitudinal direction. Alternatively, points d and c may be defined as points moved by a distance Ll toward the tip in the longitudinal direction from points a and b.

[0036] In the example shown in Figure 7, edge detection is performed on straight lines extending in the width direction from point Q, which is a distance Ll from the tip position P on the mill center line Lm toward the tail end in the longitudinal direction, using the tip position P as a reference. The points at the width ends of the rolled material Rm obtained by edge detection are defined as points c and d, respectively. Then, the points obtained by moving a distance Ll from points c and d toward the tip side in the longitudinal direction (outside the rolled material Rm) are defined as points b and a, respectively.

[0037] In the example shown in Figure 8, point c or point d (point c in this example) overlaps with the side guide 41, which is part of the rolling equipment, making it impossible to detect point c, which is the width end of the rolled material Rm. In this case, the coordinates of point c are detected using the following method. Specifically, point c is defined as the point moved by a distance Lb in the width direction from point d, which is the width end of the rolled material Rm that was detected correctly. This distance Lb may be the plate width of the rolled material Rm calculated or measured, or it may be the maximum value of the length from point d to point c (corresponding to the plate width) detected by the same method as in the example in Figure 7, or twice the length from point d to point Q, by gradually increasing the distance Lb.

[0038] In step S4, the processing circuit 22 calculates a coordinate transformation matrix. In this step S4, a perspective transformation matrix is ​​calculated as the coordinate transformation matrix. First, the processing circuit 22 sets (selects) one reference frame 5r from among multiple target 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, the processing circuit 22 calculates a perspective transformation matrix M that transforms the coordinate group of four points (points a to d) of each target frame 5 other than the reference frame 5r to the reference coordinate group of four points (points a to d) of the reference frame 5r. Note that the reference coordinate group does not necessarily have to be set on the reference frame 5r, but may be set in advance on the target frames 5. In this case, in step S5 described later, the image after perspective transformation is not superimposed on the reference frame 5r, but rather the images after perspective transformation (target frames) are superimposed on each other to obtain a composite image 6.

[0039] TIFF0007917073000001.tif27170

number

[0040] Here, i r This is the X-axis position (X-coordinate) of the point in the reference frame 5r, j rrepresents the Y-axis position (Y coordinate) of a point in the reference frame 5r, i n represents the X-axis position (X coordinate) of a point in the target frame 5n, j n represents the Y-axis position (Y coordinate) of a point in the target frame 5n. h is a component of the perspective transformation matrix Mn, and k is the position of the point cloud (k=a,b,c,d).

[0041] In each target 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 is converted to a corresponding pixel in the image I after perspective transformation according to the following formula (2) using each component of the perspective transformation matrix M of the target frame 5 trans . The perspective transformation matrix M is a matrix representing which pixel in the image I before perspective transformation trans each pixel in the image I after perspective transformation in corresponds to, respectively. [Formula]

[0042] In the above formula (2), i is the X-axis position of a pixel in the image I before perspective transformation in , j is the Y-axis position of a pixel in the image I before perspective transformation in , m is the image channel, and h is a component of the perspective transformation matrix M. Provided that, for a pixel in the image I after perspective transformation trans , if there is no corresponding pixel in the image I before perspective transformation in , the luminance of this pixel in the image I after perspective transformation trans is set to 0.

[0043] Here, the consistency of the tip position P detected in each target frame 5 may be verified according to the conditions, and target frames 5 that do not satisfy the conditions may be excluded from the synthesis target. For example, statistical values ​​may be calculated from the relationship between the tip position P of each target frame 5 and the average brightness value in the moving window 54, and target frames 5 whose statistical values ​​deviate from the distribution of all target frames 5 may be excluded from the synthesis target. Another method for checking consistency from the tip position P of each target frame 5 is as follows: That is, if the recording start timing and end timing have been adjusted so that the direction of travel of the rolled material Rm in the video is constant, the detected tip position P will travel in the same direction between target frames 5. From this relationship, if the tip position P of a target frame 5 is not located between the tip positions P of the target frames 5 before and after it, that target frame 5 is excluded from the synthesis target. Alternatively, actual values ​​of the tracking unit 23 from the start to the end of recording are collected and compared with the tip position P of each target frame 5, and if the difference is large, that target frame 5 is excluded from the synthesis target.

[0044] In step S5, the processing circuit 22 obtains a composite image 6 by superimposing all the images of the target frame, which have been perspective-transformed using the perspective transformation matrix M, onto the reference frame 5r. The brightness of each pixel in the composite image 6 is expressed by the following equation (3).

number

[0045] In equation (3) above, out Here, is the composite image 6, i is the X-axis position of the pixel, j is the Y-axis position of the pixel, m is the channel of the image, I trans is the image after perspective transformation (target frame 5), n is the sequential number of the target frame, and N is the total number of target frames.

[0046] Steps S6 to S8 are processes for acquiring information used in step S13, which will be described later, and are the same as conventional processes. In step S6, the processing circuit 22 detects areas with large changes in brightness (boundaries between maximum and minimum brightness) as contours from edge images obtained by applying a differential filter to each target frame. Based on the detected contours, the processing circuit 22 determines the direction of warping (upward or downward) (step S7) and calculates the warping height (step S8). The calculated warping height for each target frame is temporarily stored in memory 22c and read out during the processing of step S13, which will be described later. The conventional processing of steps S6 to S8, as well as the processing of step S13, which performs warping measurement using statistical processing, are publicly known, including the disclosures in the above-mentioned Patent Documents 1 to 3, so a detailed explanation is omitted here. Once the processing of step S8 is completed for all target frames, the process moves to step S9.

[0047] Incidentally, if the rolled material Rm is curved downwards and the curvature is large, the leading edge of the rolled material Rm may collide with the roller table 4 and bounce up and down. When such bouncing occurs, an afterimage is created at the leading edge of the rolled material Rm in the composite image 6. As a result, the brightness of the leading edge in the composite image 6 decreases, making it difficult to accurately detect the contour of the leading edge of the rolled material Rm, and there is a risk of mismeasurement as upward curvature. As a result of diligent research by the inventors, we have found that when an afterimage occurs due to bouncing, the bouncing part also becomes white in the grayscale image obtained by applying binarization processing to the composite image 6, resulting in an increase in white pixels.

[0048] Based on the above findings, in step S9, a grayscale image (binarized image) is obtained by applying a binarization process to the composite image. Next, the area of ​​white pixels in the grayscale image is calculated (step S10). In step S10, the area of ​​white pixels in at least one (one or more) pre-set regions of interest (for example, regions of interest 51, 52 shown in Figure 4) in the grayscale image (i.e., composite image) is calculated. Here, at least one region of interest can be set to include a part affected by bounding, for example, by considering past performance or reference coordinate sets. If multiple regions of interest are set, they may be set to be adjacent to each other or spaced apart.

[0049] Next, it is determined whether the area of ​​the white pixels calculated in step S10 above satisfies the criteria (step S11). If multiple regions of interest are set, criteria can be set for each region of interest. Examples of criteria include exceeding a reference value or being below a reference value. Criteria, including the reference value, are set according to the region of interest. For example, if two regions of interest are set, and the area of ​​the white pixels in the first region of interest exceeds the reference value, and the area of ​​the white pixels in the second region of interest is below the reference value, it can be determined that the criteria are met. In this way, the criteria can be arbitrarily set according to the region of interest.

[0050] If the criteria are met in step S11, it is determined that no bouncing has occurred, and the process proceeds to step S12. On the other hand, if the criteria are not met, it is determined that bouncing has occurred, and the process proceeds to step S13. In step S13, the curvature is measured by statistically processing the curvature direction and curvature height of multiple uncombined target frames in the same manner as in the conventional example. In the statistical processing, for example, the median value of the curvature height obtained for multiple target frames is obtained, but the process is not limited to this; the maximum value may be obtained, or the median value of the results (curvature height) after excluding the maximum and minimum values ​​may be obtained.

[0051] In step S12, the processing circuit 22 detects the contour Ed of the rolled material Rm from the composite image 6. Figure 9 is a diagram illustrating an example of how to detect 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. Then, a region of interest (ROI) 55 is defined on the mill center line Lm, having a length Lsw (mm) in the width direction and a size of length Lsl (mm) in the longitudinal direction from the leading edge point P. Next, a movable window 56 is defined by moving longitudinally within the region of interest 55, having the same length Lsw (mm) in the width direction as the region of interest 55, and an arbitrary length Lml (mm) smaller than Lsl in the longitudinal direction. Furthermore, a movable window 57 is defined by moving longitudinally within the movable window 56, having the same length Lml (mm) in the longitudinal direction as the movable window 56, and an arbitrary length Lmw (mm) smaller than Lsw in the width direction. At each longitudinal movement position of the movable window 56, the average brightness value within the movable window 57 when the movable window 57 is moved in the plate width direction is stored in memory 22c in association with the movement position. At each longitudinal movement position of the movable window 56, the center point C of the movable window 57 where the average brightness value within the movable window 57 is maximum is detected as the contour Ed of the rolled material Rm.

[0052] Here, the contour Ed was detected using a composite image 6, which was created by perspective transforming and combining the edge images of each target frame 5. However, this is not the only method, and similar results can be obtained with other processing methods. For example, a binarized image can be obtained by applying a binarization process to the composite image 6, and the center position in the plate width direction of the pixel with the maximum brightness at each longitudinal position within the region of interest 55 can be detected as the contour Ed of the rolled material Rm. A noise reduction filter may also be applied to the composite image 6 as needed.

[0053] Next, referring to Figures 10 to 14, the processing circuit 22, acting as a shape measurement circuit, quantifies the warp height from the detected contour Ed. First, as shown in Figure 10, the contour Ed1 on the far side in the width direction is extracted from the contour Ed of the rolled material Rm. The extracted contour Ed1 is then transformed using perspective, and the warp direction (upward or downward) is determined from the characteristics of the transformed contour Ed2. Then, the warp height is quantified based on the conditions for each warp direction. The following describes the perspective transformation of contour Ed1, the determination of the warp direction, and the calculation of the warp height performed by the processing circuit 22.

[0054] (Perspective transformation of contours) To convert the unit of curvature height [pix] on the image to [mm], an image of the rolled material Rm viewed from the side is obtained using the method described below. That is, the processing circuit 22 defines the plane on which the contour Ed1 on the far side in the width direction exists, i.e., the plane on which the curvature height is calculated, in the following way. As shown in Figure 11, the extension of the non-curved portion of the contour Ed1 defines the reference line Lr. The intersection of the reference line Lr and the predetermined straight lines t3-t4 (described later) is detected as point r1. The intersection of the reference line Lr and the predetermined straight lines t1-t2 (described later) is detected as point r2. Then, as shown in Figure 12, the intersection of the straight line connecting point r1 and vanishing point VP3 (described later) and the predetermined straight lines t8-t7 is detected as point r4. Finally, the intersection of the straight line connecting point r2 and vanishing point VP3 and the predetermined straight lines t5-t6 is detected as point r3. The points r1, r2, r3, and r4 detected by the above process define the plane Pn1. P The shorter side of n1 extends in the thickness direction, and the longer side extends in the longitudinal direction. Then, the plane Pn1 is a rectangle (rectangular) with predetermined side lengths as shown in Figure 13. shape Matrix M that performs perspective transformation on the plane Pn2 (the plane Pn2 defined by points r1', r2', r3', r4') measHowever, this is calculated. As shown in Figure 13, the calibration material Cm has known thickness dimension Tc [mm] and longitudinal dimension Lc [mm]. From these dimensions Tc, Lc, and the desired pixel dimension conversion coefficient Sdes [mm / pix], when the shorter side of the rectangle is set to Tc / Sdes [pix] and the longer side to Lc / Sdes [pix], in the plane Pn2 after perspective transformation, the straight line in the thickness direction and the straight line in the longitudinal direction are orthogonal, so the desired pixel dimension conversion coefficient Sdes [mm / pix] is obtained at any position in the image. In other words, the unit of the warp height can be converted from [pix] to [mm] using a single pixel dimension conversion coefficient Sdes.

[0055] (Determination of the direction of curvature) Next, the processing circuit 22 processes the point cloud that constitutes the detected contour Ed1 on the far side in the width direction into M meas By performing a perspective transformation, the perspective-transformed contour Ed2 shown in Figure 14 is obtained. Figure 14 shows an example of a method for determining the direction of curvature. From among the points on the perspective-transformed contour Ed2, the point p that is the highest in the Y-axis direction (vertical direction in the figure) is defined as the peak point, and the point q that is the outermost point of contour Ed2 is defined as the tip point. At this time, if the longitudinal distance d from the peak point p to the tip point q is smaller than a predetermined threshold, it is determined to be an upward curvature (see Figure 14(a)), and if it is larger than the threshold, it is determined to be a downward curvature (see Figure 14(b)). The threshold can be set appropriately based on past results.

[0056] Thus, the warp direction was determined based on the longitudinal distance d from point p to point q of the contour Ed2 on the far side in the width direction near the leading edge of the rolled material Rm, but this is not the only method. For example, the warp direction can be determined by performing the same process on the contour on the near side in the width direction near the leading edge of the rolled material Rm. In this case, the warp height calculation described later can be performed only if the warp directions on the near and far sides in the width direction coincide. If the warp directions determined on the far and near sides in the width direction do not coincide, the warp height calculation is performed using the method described later.

[0057] (Calculation of curvature height) In step S12 above, the processing circuit 22 calculates the curvature height based on the contour Ed2 on the far side in the width direction. In the case of upward curvature, as shown in Figure 14(a), the lowest point in the Y-axis direction of the contour Ed2 in the non-curved portion is defined as the lowest point r, and the difference h in the plate thickness direction between the peak point p and the lowest point r is calculated as the curvature height. In the case of downward curvature, as shown in Figure 14(b), the difference h in the plate thickness direction between the peak point p and the tip point q is calculated as the curvature height. The height for upward curvature is a positive value, and the height for downward curvature is a negative value.

[0058] While it is certainly possible to perform the image synthesis and warpage measurement described above after one rolling pass by the rough rolling mill 3 is completed and apply the control (for example, changing the rotation speed of the upper and lower rolling rollers 32) according to the measured warpage to the next rolling pass, it is also possible to perform the image synthesis and warpage measurement during the rolling pass, thereby enabling real-time performance. When performing this during the current rolling pass, if a predetermined number of target frames 5 cannot be obtained, the system may be configured to read data from past rolling passes from memory 22c to obtain the synthesized image 6.

[0059] (Method of compensating for variations in board width) Referring to Figure 15, a method for quantifying the warp height when plate width variation occurs at the leading edge of the rolled material Rm will be explained. When plate width variation occurs at the leading edge of the rolled material Rm, if the above-described warp direction determination is performed on the contour Ed2 on the far side in the plate width direction and the contour on the near side in the width direction, the results of the two warp direction determinations may not match. For example, as shown in Figure 15, if the width near the leading edge of the rolled material Rm is narrow, a downward warp may be mistakenly detected based on the contour Ed2 on the far side in the width direction and the distance da, while an excessive upward warp may be mistakenly detected based on the contour Ed3 on the near side in the width direction and the distance db. Therefore, in order to compensate for the width variation at the leading edge of the rolled material Rm and to accurately measure the warp height, the following method can be adopted. Specifically, line segments extending in the width direction on the rolled material Rm are detected at each position in the longitudinal direction (hereinafter referred to as "each longitudinal position"), and the point Pc that is the widthwise center of each line segment is detected as the center point, and a center line La is obtained by connecting the center points Pc of each longitudinal position. This center line La is used in place of the contours Ed2 and Ed3 on both sides in the width direction, and the above-mentioned perspective transformation, warp direction determination, and warp height calculation are performed. This makes it possible to compensate for the width variation at the leading edge of the rolled material Rm, and to measure the warp height with high accuracy. Note that this compensation method assumes that the width variation at the leading edge of the rolled material Rm occurs almost equally on the far side and near side in the width direction.

[0060] (Calibration method) Figure 16 shows an example of a calibration method for vanishing points VP1, VP2, and VP3 using a calibration material Cm. The calibration material Cm has a known plate thickness Tc, plate width Wc, and length Lc. The calibration material Cm is positioned so that its widthwise center coincides with the mill center line Lm, and the calibration material Cm is imaged. Points t1 to t6 are detected on the image. The position 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 position where the line connecting points t1 and t4 intersects with the line connecting points t2 and t3 is defined as vanishing point VP2. Similarly, the position 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-coordinate, Y-coordinate, or both of these vanishing points VP1, VP2, and VP3 exceed a predetermined value that is sufficiently larger than the number of pixels in the frame, the vanishing point is considered nonexistent, and lines parallel to each of the above-mentioned lines are used as the lines in each axis direction. The reference 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 longitudinal direction. Next, the intersections of the lines (t7-VP1), (t5-VP2), and (t4-VP3) are detected as point t8. The image coordinates of the vanishing points VP1~VP3 and points t1~t8 defined by the above operations are stored in memory 22c and used for detecting the above-mentioned coordinate group (feature quantity) and calculating the curvature height.

[0061] As described above, according to this embodiment, multiple target frames 5 in which the leading edge of the rolled material Rm is located are combined using a coordinate transformation matrix M. As shown in Figure 17, even if disturbances such as water vapor or stagnant water occur, the disturbances do not exist in the same location in the multiple (four in the figure) target frames 51 to 54, so the disturbances are canceled out when the multiple target frames 51 to 54 are combined. As a result, the combined image 6 obtained from the multiple target frames 51 to 54 becomes clear, and the contour Ed of the rolled material Rm can be detected with high accuracy from the clear combined image 6. Furthermore, the warpage of the rolled material Rm can be measured with high accuracy based on the contour Ed of the rolled material detected from the clear combined image 6.

[0062] While embodiments of this disclosure have been described above, this disclosure is not limited to the embodiments described above and can be implemented in various modified forms without departing from the spirit of this disclosure. When the number of elements, quantities, amounts, ranges, etc., are mentioned in the embodiments described above, this invention is not limited to the number mentioned unless it is specifically stated or clearly defined in principle. Furthermore, the structures, etc., described in the embodiments described above are not necessarily essential to this invention unless they are specifically stated or clearly defined in principle.

[0063] In the above embodiment, the case of measuring the curvature direction and curvature height of the rolled material Rm was described as an example, but the shape of the rolled material is not limited to curvature. This disclosure can also be applied to the measurement of shapes measured based on the contour Ed of the rolled material Rm, such as the camber shown in Figure 18(a) or the wavy surface shape called flatness shown in Figure 18(b).

[0064] In the above embodiment, the coordinates of the four points at the leading edge of the rolled material Rm were detected as feature quantities. However, if the rolled material Rm is short, the coordinates of the four points at the trailing end of the rolled material Rm can also be detected as feature quantities.

[0065] In the above embodiment, we have described an example in which a perspective transformation matrix M is calculated to perform a perspective transformation on the target frame 5 so that the shape and size of the rolled material Rm in the reference frame 5r match the shape and size of the rolled material Rm in the target frame 5. However, the embodiment is not limited to this. When the camera 21 is installed far from the mill center line Lm, and the sizes of the rolled material Rm in multiple target frames 5 are equivalent, and the rolled material Rm can be considered similar, that is, when the size change of the rolled material Rm in multiple target frames 5 can be ignored, it is possible to obtain a composite image by translation. In this case, the coordinates of three points are detected for each target frame 5, and the affine transformation matrix is ​​calculated as a coordinate transformation matrix. Since it is not necessary to include rotation in the affine transformation matrix, the affine transformation matrix can be represented as a 3x2 matrix.

[0066] In the above embodiment, the case of detecting the contour Ed of the rolled material Rm rolled in the roughing mill 3 was described as an example, but the invention is not limited to this. This disclosure can be applied to detecting the contour of a rolled material Rm being conveyed in one direction along a hot rolling line RL, for example, when detecting the contour of a rolled material Rm rolled in a finishing mill. [Explanation of symbols]

[0067] RL...Hot rolling line, Rm...Material to be rolled, 1...Shape measuring device, 2...Contour detection device, 21...Camera, 22...Processing circuit (shape measuring circuit), 23...Tracking unit, 3...Rough rolling mill, 31...Rolling load detection unit, 32...Rolling roll, 33...Main unit, 4...Roller table, 41...Side guide, 5...Target frame, 5r...Reference frame, 6...Composite image

Claims

1. In a contour detection device for detecting the contour of a rolled material being transported in one direction along a hot rolling line, A camera for imaging the rolled material, The system includes a processing circuit for processing images captured by the aforementioned camera, The aforementioned processing circuit is Dividing the aforementioned video into multiple frames, Identifying multiple target frames from among the multiple frames in which the leading edge of the rolled material is located, To detect a set of coordinates representing the position of the rolled material in each target frame, This involves calculating a coordinate transformation matrix that converts the coordinate set of each target frame into a pre-defined reference coordinate set, and Obtaining a single composite image from the multiple target frames transformed using the coordinate transformation matrix, A contour detection device configured to detect the contour of the rolled material from the brightness of the composite image.

2. In the contour detection device according to claim 1, A contour detection device that calculates the coordinate transformation matrix, which includes setting the coordinate group of a reference frame selected from the plurality of target frames as the reference coordinate group.

3. In the contour detection device according to claim 1, Detecting the aforementioned coordinate group includes detecting the aforementioned coordinate group, which includes the coordinates of four points, for each target frame. A contour detection device that calculates the aforementioned coordinate transformation matrix, which includes calculating the perspective transformation matrix as the aforementioned coordinate transformation matrix.

4. In a shape measuring device for measuring the shape of a rolled material being transported in one direction along a hot rolling line, A contour detection device according to any one of claims 1 to 3, A shape measurement circuit configured to measure the shape of the rolled material based on the contour of the rolled material detected from the composite image by the contour detection device, A shape measuring device equipped with the following features.

5. In the shape measuring device according to claim 4, The shape measuring circuit is a shape measuring device configured to measure the curvature direction and curvature height of the leading edge of the rolled material.

6. In the shape measuring device according to claim 5, The shape measurement circuit described above is By applying a binarization process to the aforementioned composite image, a grayscale image is obtained. The system is configured to perform the following: calculate the area of ​​white pixels in one or more regions of interest set in the aforementioned grayscale image; A shape measuring device that measures the curvature direction and curvature height when the area of ​​the white pixels in one or more regions of interest satisfies the criteria set for each region of interest.

Images