JP7835232B2 - Method for manufacturing backing plates, method for designing backing plates, welding method for metal members, method for manufacturing column members, method for manufacturing column-beam joints, shape measurement method, shape measurement program, shape measurement device, and backing plate manufacturing system. - Google Patents

Description

This invention relates to a method for manufacturing a backing plate, a method for designing a backing plate, a welding method for metal members, a method for manufacturing a column member, a method for manufacturing a column-beam joint, a shape measurement method, a shape measurement program, a shape measurement device, and a manufacturing system for backing plates.

When welding metal components such as steel together, there is a method that involves melting and solidifying the outer surface of the end portion, including the boundary between the metal components. In this case, to prevent welding defects even when welding only one side in a full penetration weld, a backing plate is used (see, for example, Non-Patent Document 1). The backing plate is attached to the back surface of the metal component before welding, specifically to the side opposite the welding location of the metal component.

Architectural Institute of Japan, “Technical Guidelines for Steel Frame Construction/Construction Site Construction Edition”, 7th edition (January 15, 2018), p. 296-298

Incidentally, since the backing plate is positioned along the opposite side of the welded area of the metal component, it must be individually designed to match the opposite side of the welded area of the metal component being used. Therefore, there was a need for technology to efficiently design the backing plate.

This invention has been made in view of the above-mentioned problems, and its objective is to provide a method for manufacturing a backing plate, a method for designing a backing plate, a welding method for metal members, a method for manufacturing column members, a method for manufacturing column-beam joints, a shape measurement method, a shape measurement program, a shape measurement device, and a backing plate manufacturing system that enable the efficient design and manufacture of backing plates.

In order to solve the aforementioned problems and achieve the objective,
[1] The present invention relates to a method for manufacturing a backing plate used for welding metal members together, and includes a processing step of manufacturing the backing plate by processing a predetermined material based on the shape of the backing plate location detected based on image data obtained by photographing a marker plate attached to the backing plate location and having a pattern formed on it that serves as a reference for measuring length.

[2] Furthermore, the backing plate design method according to the present invention is a backing plate design method used for welding metal members together, and includes a design step of creating backing plate design data based on the shape of the backing plate, which is detected based on image data of a marker plate attached to the backing plate and having a pattern formed on it that serves as a reference for length measurement, in at least one of the metal members.

[3] Furthermore, the welding method for metal members according to the present invention is a welding method for welding metal members together using a backing plate, and includes a welding step of: placing the backing plate on the metal member based on the location where the backing plate is to be placed on the metal member, and the shape of the location where the backing plate is to be placed, which is detected based on image data of a marker plate attached to the location where the backing plate is to be placed and which has a pattern formed on it that serves as a reference for length measurement; and welding the metal member to another metal member on the side opposite to the location where the backing plate is to be placed.

[4] Furthermore, the method for manufacturing a column member according to the present invention is a method for manufacturing a column member equipped with a diaphragm, wherein the column member is the metal member, and the method includes a welding step of welding the metal member and the diaphragm using the backing plate manufactured by the method for manufacturing the backing plate described in the above invention [1].

[5] Furthermore, the method for manufacturing a column-beam joint according to the present invention is a method for manufacturing a column-beam joint comprising a column member and a beam member, and includes the step of manufacturing the column member using the method for manufacturing a column member described in the above invention [4].

[6] Furthermore, the shape measurement method according to the present invention is a shape measurement method for measuring the shape of a cross-section of a metal member, and includes a shape detection step of detecting the shape of the cross-section based on image data of the cross-section and a marker plate attached to the cross-section on which a pattern used as a reference for length measurement is formed.

[7] Furthermore, the shape measurement method according to the present invention, in the above invention [6], includes a binarization step of creating a binarized image based on the channel intensity of a predetermined color component at each pixel position in the image data, and further includes an edge detection step of detecting the shape of the edges of the cross section based on the binarized image.

[8] Furthermore, the shape measurement method according to the present invention, in the above invention [7], further includes a projection transformation step in which a plurality of patterns are formed on the marker plate, the position of the patterns in the image is detected, and the image is projectively transformed based on the detection result, and the binarization step is to create the binarized image using the image after the projection transformation.

[9] Furthermore, the shape measurement program according to the present invention is a shape measurement program for measuring the shape of a cross-section of a metal member, and causes a computer to execute a shape detection step of detecting the shape of the cross-section based on image data of the cross-section of the metal member and a marker plate attached to the cross-section and on which a pattern used as a reference for length measurement is formed.

[10] Furthermore, the shape measuring device according to the present invention is a shape measuring device for measuring the shape of a cross-section of a metal member, and comprises a shape detection unit that detects the shape of the cross-section based on image data of the cross-section of the metal member and a marker plate attached to the cross-section and on which a pattern used as a reference for length measurement is formed.

[11] Furthermore, the shape measuring device according to the present invention is equipped with an imaging unit for capturing image data in the above invention [10].

[12] Furthermore, the method for manufacturing a backing plate according to the present invention is a method for manufacturing a backing plate used for welding metal members together, and includes: a shape detection step of detecting the shape of the backing plate placement location on at least one metal member, and image data of a marker plate attached to the backing plate placement location and having a pattern formed on it that serves as a reference for length measurement; a design step of creating design data for the backing plate based on the detection result of the shape detection step; and a processing step of manufacturing the backing plate by processing a predetermined material according to the design data.

[13] The present invention also relates to a manufacturing system for a backing plate used for welding metal members together, comprising: an imaging unit that photographs a location on at least one metal member where a backing plate is to be placed, and a marker plate attached to the location where the backing plate is to be placed and on which a pattern for measuring length is formed; a shape detection unit that detects the shape of the location where the backing plate is to be placed based on the image data captured by the imaging unit; a design unit that creates design data for the backing plate based on the detection result by the shape detection unit; and a processing unit that processes a predetermined material according to the design data to manufacture the backing plate.
[14] Furthermore, the manufacturing system for the backing plate according to the present invention [13] comprises a marker plate on which a pattern to be used as a reference for measuring length is formed.

According to the present invention, it is possible to efficiently design and manufacture backing plates used, for example, during welding of metal components.

Figure 1 is a block diagram showing the configuration of a backing plate manufacturing system according to one embodiment of the present invention. Figure 2 shows an example of metal members to be joined. Figure 3 is a diagram illustrating the backing plate that is placed during welding. Figure 4 is a flowchart showing a method for manufacturing a backing plate according to one embodiment of the present invention. Figure 5 shows a marker plate used when photographing metal components. Figure 6 shows the arrangement of the marker plate on the metal member. Figure 7 is a flowchart showing the flow of the shape measurement process. Figure 8 is a diagram (part 1) illustrating the pattern coordinate position detection process. Figure 9 is a diagram illustrating the fitting of a backing plate to a metal component. Figure 10 is a diagram (part 2) illustrating the pattern coordinate position detection process. Figure 11 is a flowchart showing the flow of the shape detection process. Figure 12 is a diagram (part 1) illustrating the binarization process in shape detection. Figure 13 is a diagram (part 2) illustrating the binarization process in shape detection. Figure 14 is a diagram illustrating the edge extraction image in the edge detection process . Figure 15 shows an example of the detection result of the opposite side of a welded area of a metal component using shape measurement processing. Figure 16 is a diagram (part 1) illustrating another example of the welding method. Figure 17 is a diagram (part 2) illustrating another example of the welding method. Figure 18 is a diagram (part 3) illustrating another example of the welding method. Figure 19 is a diagram (number 4) illustrating another example of the welding method.

The following describes embodiments of the backing plate design method, backing plate manufacturing method, backing plate design program, and backing plate design apparatus according to the present invention. Note that the present invention is not limited to these embodiments.

Figure 1 is a block diagram showing the configuration of a backing plate manufacturing system according to one embodiment of the present invention. The backing plate manufacturing system 1 comprises an imaging device 2 that captures an image of the metal member to which the backing plate is to be placed, a design device 3 that designs the backing plate based on the image captured by the imaging device 2, and a processing device 4 that manufactures the backing plate by processing a predetermined material based on the information designed by the design device 3. A backing plate is a temporary material used, for example, when welding metal members together, to prevent welding defects even when welding only one side in a full penetration weld. Furthermore, the backing plate of the present invention is used on the side of the metal member opposite to the welding location.

The devices in the backing plate manufacturing system 1 are connected via a network N, etc. Network N is, for example, a public communication network such as the Internet, and consists of one or more combinations of, for example, a LAN (Local Area Network), a WAN (Wide Area Network), a telephone communication network such as a mobile phone network, a public line, a VPN (Virtual Private Network), and a dedicated line. Network N may also appropriately combine wired and wireless communication.

The imaging device 2 comprises an imaging unit 21, an image processing unit 22, a display unit 23, a communication unit 24, an input unit 25, a control unit 26, and a storage unit 27.

The imaging unit 21 captures an image of the end face of the metal member from which the backing plate is to be fabricated, and generates an imaging signal. The imaging unit 21 is configured using, for example, a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor.

Here, the metal members and backing plates will be explained with reference to Figures 2 and 3. Figure 2 shows an example of the metal members to be joined. Figure 3 is a diagram illustrating the backing plate that is placed during welding.

In this embodiment, the metal member 100 is a tubular member with a hollow rectangular prism shape and open ends in the longitudinal direction. This metal member 100 has an opening 101 that forms a rectangular prism-shaped hollow space. The metal member 100 is formed using a metallic material. Here, "metallic material" refers to a weldable material. In this embodiment, for example, steel is used for the metal member 100. Furthermore, in this embodiment, an example of joining metal members 100 by welding their outer surfaces, including the connection portion, while aligning them in the axial direction of the pipe, will be described.

The backing plate 110 is an annular member disposed in the opening 101. The shape of the backing plate 110's outer surface is rectangular, corresponding to the inner surface of the opening 101. The backing plate 110 is, for example, provided on the inner surfaces of the two metal members 100 to be joined, at a position including the connection portion (boundary). In this case, the backing plate 110 is disposed so as to abut against the inner surface of the metal member 100. The backing plate 110 is formed, for example, using the same material as the metal member 100. Note that while Figure 3 illustrates a backing plate 110 with an opening, it may also be configured as a flat plate.

The image processing unit 22 acquires the image signal generated by the imaging unit 21, performs predetermined image processing on the acquired image signal, and outputs it to the display unit 23, communication unit 24, and storage unit 27. The image processing unit 22 applies known image processing to the image signal, such as simultaneity processing, gradation correction processing, and color correction processing. Simultaneity processing, for example, synchronizes the image data of each RGB color component.

The display unit 23 is composed of, for example, a display or a speaker microphone, and displays the image formed by the optical system of the imaging unit 21, or the captured image. A liquid crystal display or plasma display may be used as the display. The display unit 23 is configured to display information input from the control unit 26. Furthermore, the display unit 23 may also include various output functions, such as a light source like an LED, or a printer that prints predetermined information onto paper.

The communication unit 24 includes, for example, a LAN interface board connected to the network N, a wireless communication circuit for wireless communication, etc., and communicates with the design device 3 and other external devices by connecting to the network N.

The input unit 25 consists of, for example, a touch panel that detects touch operations on a keyboard or display panel, and input buttons. The input unit 25 may also include an audio input device that enables communication with an external source.

The control unit 26 comprehensively controls each part of the imaging device 2. The control unit 26 is composed of a processor such as a CPU (Central Processing Unit), DSP (Digital Signal Processor), FPGA (Field-Programmable Gate Array), and RAM (Random Access Memory) and ROM (Read Only Memory).

The storage unit 27 stores the operating system (OS), various programs, various tables, various databases, etc., for executing the operation of the imaging device 2. The storage unit 27 is configured using storage media such as volatile memory (RAM), non-volatile memory (ROM), EPROM (Erasable Programmable ROM), hard disk drive (HDD), solid state drive (SSD), and removable media. Removable media include, for example, USB (Universal Serial Bus) memory, or disk recording media such as CD (Compact Disc), DVD (Digital Versatile Disc), or BD (Blu-ray® Disc). Alternatively, the storage unit 27 may be configured using a computer-readable recording medium such as an externally insertable memory card. Furthermore, various programs can be recorded on computer-readable recording media such as hard disks, flash memory, CD-ROMs, DVD-ROMs, and flexible disks for widespread distribution.

The design device 3 comprises a communication unit 31, a shape measurement unit 32, a design unit 33, an input unit 34, a control unit 35, and a storage unit 36.

The communication unit 31 includes, for example, a LAN interface board connected to the network N, a wireless communication circuit for wireless communication, etc., and communicates with the imaging device 2, the processing device 4, and other external devices by connecting to the network N.

The shape measurement unit 32 measures the shape of the placement location of the backing plate using the image of the end face of the metal member acquired by the imaging device 2. The shape measurement unit 32 is configured using a processor such as a CPU, DSP, or FPGA. The shape measurement unit 32 corresponds to a shape measurement device equipped with a shape detection unit.

The design unit 33 designs the shape of the backing plate based on the shape detected by the shape measurement unit 32. The design unit 33 is composed of a processor such as a CPU, DSP, or FPGA.
The design system is comprised of the shape measurement unit 32 and the design unit 33.

The input unit 34 is comprised of, for example, a touch panel that detects touch operations on a keyboard or display panel, and input buttons. The input unit 34 may also include an audio input device that enables communication with an external source.

The control unit 35 comprehensively controls each part of the design apparatus 3. The control unit 35 is composed of a processor such as a CPU, DSP, or FPGA, and memory such as RAM or ROM.

The storage unit 36 stores the OS, various programs, various tables, various databases, etc., necessary for executing the design device 3. The storage unit 36 is configured using, for example, volatile memory such as RAM, non-volatile memory such as ROM, EPROM, HDD, SSD, and removable media. Alternatively, the storage unit 36 may be configured using a computer-readable recording medium such as an externally insertable memory card. Furthermore, various programs can be recorded on computer-readable recording media such as hard disks, flash memory, CD-ROMs, DVD-ROMs, and flexible disks for wider distribution.

The processing apparatus 4 includes a communication unit 41, a molding unit 42, an input unit 43, a control unit 44, and a storage unit 45.

The communication unit 41 includes, for example, a LAN interface board connected to the network N, a wireless communication circuit for wireless communication, etc., and communicates with the design device 3 and other external devices by connecting to the network N.

The molding unit 42 manufactures the backing plate based on the information designed by the design device 3. The molding unit 42 is constructed using a 3D printer, an NC (Numerical Control) machine tool, or the like.

The input unit 43 is comprised of, for example, a touch panel that detects touch operations on a keyboard or display panel, and input buttons. The input unit 43 may also include an audio input device that enables communication with an external source.

The control unit 44 comprehensively controls each part of the design apparatus 3. The control unit 44 is composed of a processor such as a CPU, DSP, or FPGA, and memory such as RAM or ROM.

The storage unit 45 stores the OS, various programs, various tables, various databases, etc., for executing the operation of the processing device 4. The storage unit 45 is configured using, for example, volatile memory such as RAM, non-volatile memory such as ROM, EPROM, HDD, SSD, and removable media. Alternatively, the storage unit 45 may be configured using a computer-readable recording medium such as an externally insertable memory card. Furthermore, various programs can be recorded on computer-readable recording media such as hard disks, flash memory, CD-ROMs, DVD-ROMs, and flexible disks for wider distribution.

Next, a method for manufacturing a backing plate will be described. Figure 4 is a flowchart showing a method for manufacturing a backing plate according to one embodiment of the present invention. The method for manufacturing a backing plate begins by photographing the end face of the metal member 100 using the imaging device 2 (Step S1: Photography Step). At this time, the process is performed using one of the two metal members 100 to be joined. Alternatively, the process may be performed using both of the two metal members 100 to be joined. Furthermore, if there are three or more items to be joined, at least one metal member 100 can be appropriately selected for processing. The selection should be appropriate according to the condition of the items to be joined and the purpose of joining.

When the imaging device 2 photographs the end face of the metal member 100, the marker plate 200 is attached to the end face of the metal member 100 to be measured. This end face is the end face that will be joined to another metal member 100. This end face includes the cross-section, i.e., the cut surface, formed by cutting the metal member 100 before welding to adjust its length. Furthermore, the end face of the metal member 100 that will be joined to another metal member 100 may be colored to indicate the welded end face. In this embodiment, red coloring is used as an example. However, other colors such as green or blue may also be used.

Figure 5 shows a marker plate used when photographing a metal member. The marker plate 200 is for measuring the length of the end face of the metal member 100 and has a rectangular hollow flat plate body 201. That is, the marker plate 200 has an opening 201a in the center that penetrates in the thickness direction. Also, patterns 202 to 204 with known lengths and shapes, which are used as a reference for length measurement, are drawn on the surface of the body 201. Patterns 202 to 204 are provided, for example, at different corners of the body 201. Preferably, these patterns 202 to 204 are square patterns with one or more sides of known length, or AR markers (e.g., Aruco) which are markers for three-dimensional surveying. The marker plate 200 is attached to the end face or surface of the metal member 100 with a magnet or the like so that it can be easily attached and detached. Furthermore, in order to improve the accuracy of the projection transformation described later, it is preferable to have as many patterns as possible drawn on the marker plate 200. Moreover, the plane on which the end face of the metal member 100 is located and the plane through which patterns 202 to 204 pass are arranged to be as close as possible, preferably on the same plane. Furthermore, under normal operating conditions, it is preferable to make the marker plate 200 small so that the entire marker plate 200 can be photographed when photographed from the location where the imaging device 2 is located.
Note that the shape and pattern position of the marker plate 200 are not limited to those shown in Figure 5.

Figure 6 shows how the marker plate is arranged on the metal member. For example, the marker plate 200 is attached to the end face of the metal member 100 that is joined to another metal member 100 (the right side in Figure 6). In this case, the opening 101 of the metal member 100 is exposed through the opening 201a.

In the imaging step (step S1), an image including the marker plate 200 and the opening 101 is captured. That is, the imaging step consists of a marker plate placement step and a cross-sectional imaging step.

After the shooting step, the design apparatus 3 performs a shape measurement step (step S2). The shape measurement step is performed by the shape measurement unit 32.

Figure 7 is a flowchart showing the flow of the shape measurement process. In the shape measurement step, the shape measurement unit 32 first performs pattern coordinate position detection processing (step S21: pattern position coordinate detection step).

In the pattern position coordinate detection step, the shape measurement unit 32 detects the positions of patterns 202 to 204 on the marker plate 200 from the image data and obtains the image coordinates of each pattern (hereinafter referred to as "image coordinates"). Here, as many patterns as possible are detected from the image data, and the image coordinates of the positions (center or vertices) of the successfully detected patterns are listed.

Figures 8 to 10 illustrate the pattern coordinate position detection process. Figure 8 shows an example of an image of the marker plate 200 and the end face (opening 101) of the metal member 100, captured in the shooting step S1. Specifically, image G1 depicts the end face image of the metal member 100, G100, which has a red-colored colored region Grm, the hollow space G101 formed by the opening 101, and an image G200 of the marker plate 200. Image G200 depicts images G202 to G204 of each pattern (patterns 202 to 204). The image processing for pattern 202 will be described below, but the same process applies to the other patterns.

The image coordinates of pattern 202 are expressed by the number of pixels along each axis, with the top left corner of the image as the origin, the right direction of the image as the positive x-axis, and the bottom direction of the image as the positive y-axis. For example, if the top left vertex P11 of pattern 202 in Figure 8 is successfully detected, its image coordinates are determined as follows: Starting from the top left origin of the image data, moving x11 pixels along the x-axis and y11 pixels along the y-axis leads to the top left vertex P11. Therefore, the image coordinates of the top left vertex P11 are (x11, y11). Similarly, for each of the patterns k = 1, 2, ..., K that were successfully detected, the image coordinates (xkn, ykn) of all successfully detected centers or vertices Pkn (n = 1, 2, ..., Nk) are listed. Note that in the case of the marker plate 200 shown in Figure 5, the maximum value of K is 4.

Furthermore, pattern 202 also has the function of controlling the orientation of the manufactured backing plate 110. When the cross-section of the metal member 100 is a highly symmetrical shape such as a square or an H-shape, this symmetry is not perfect, and there are often differences in side lengths, slight irregularities, and distortions. Therefore, when fitting the metal member 100 to the backing plate 110, if care is not taken to avoid rotation around the axis of symmetry or reversal of the front and back sides, fitting may not be possible, or even if fitting is possible, the gap may be of a different width than designed. Since the welding performance of the metal member 100, especially the welding strength, depends on the width of the gap between the metal member 100 and the backing plate 110, it is preferable to correctly control the orientation of the metal member 100 and the backing plate 110 during fitting. In other words, when fitting the backing plate 110 to the metal member 100, the orientation of the backing plate 110 should be correctly aligned with the metal member 100.

As a concrete example, Figure 9 shows the fitting patterns of a fabricated backing plate 110 to a metal member 100 having a square cross-section. In the example shown in Figure 9, the shape of the backing plate 110 to be fabricated is point-symmetric with the intersection of the two diagonals of the square as the axis of symmetry, and there are four possible fitting directions for the backing plate 110 to the metal member 100. Furthermore, considering the two cases where the front and back of the backing plate 110 are reversed, there are 4 × 2 = 8 fitting methods. However, as mentioned above, each side of the metal member 100 and the backing plate 110 has slight irregularities. Therefore, as shown in Figure 9, although there is a certain gap in the correct fitting direction, the size of the gap varies depending on the position when the fitting direction is different, and there is a high possibility that fitting will not be possible depending on the orientation.

Therefore, in this invention, it is preferable to make it possible to distinguish between the patterns 202 arranged on the marker plate 200. In this case, it becomes possible to correlate the direction of the shape when the opening 101 of the metal member 100 is photographed with the direction of the shape of the designed backing plate 110. Based on the aforementioned directional correlation information, a mark can be placed on the surface of the manufactured backing plate 110 at a position that does not affect the welding performance, indicating the orientation and front/back of the member, and the fitting direction can be immediately grasped. From the above, it can be said that errors in the fitting direction can also be suppressed according to the method of this invention.

After the processing in step S21, the projection transformation process is executed (step S22: projection transformation step). In the projection transformation step, first, before estimating the projection transformation parameters, the resolution r in the image data after projection transformation is set as a parameter. The resolution r in the image data after projection transformation is preferably coarser than the resolution of the original image data and smaller than the allowable measurement error; for example, it is preferably set to 0.2 [mm]. During the projection transformation, the actual coordinates of the pattern vertices Pkn (hereinafter referred to as "object coordinates") are set in advance. The following explanation will refer to Figure 10.

The object coordinates of pattern 202 are expressed as lengths (units are, for example, [mm]) along each axis, with the upper left corner of the marker plate 200 as the origin, the right direction of the marker plate 200 as the positive X-axis, and the downward direction of the marker plate 200 as the positive Y-axis. For example, the object coordinates of the upper left vertex P11 in pattern 202 in Figure 10 are determined as follows: Starting from the upper left origin of the marker plate 200, moving X11 [mm] along the X-axis and Y11 [mm] along the Y-axis leads to the upper left vertex P11. Therefore, the object coordinates of the upper left vertex P11 are (X11, Y11). Similarly, for each of the successfully detected patterns k=1, 2, ..., K, the object coordinates (Xkn, Ykn) of all successfully detected centers or vertices Pkn (n=1, 2, ..., Nk) are listed.

Based on the above, the correspondence between object coordinates (Xkn, Ykn) and image coordinates (xkn, ykn) has been clarified for all detected vertices Pkn. From these correspondences, the projection transformation parameters will be calculated. Note that in the following explanation, object coordinates (Xkn, Ykn) and image coordinates (xkn, ykn) may be simply referred to as object coordinates (X, Y) and image coordinates (x, y).

Projection transformation is a type of image transformation process. Projection transformation transforms an image by moving the image coordinates (x, y) to another image coordinate system (x', y') using equations (1) and (2) below.

Here, in equations (1) and (2) above, A, B, C, D, E, F, G, and H are parameters for the projection transformation.

By applying a projection transformation with appropriate parameters to the captured image data, the correspondence between one pixel of the transformed image data and its actual length—that is, the resolution r of the transformed image data—can be arbitrarily determined. Using this, by applying a projection transformation to image data of the metal component 100, it becomes possible to measure the length of the metal component 100 directly in the image.

To calculate the projection transformation parameters, at least four "points" placed on the same plane are photographed, and the correspondence between their object coordinates (X, Y) and image coordinates (x, y) is obtained. In this embodiment, the correspondence between object coordinates and image coordinates is obtained by arranging the centers and vertices of the patterns in a predetermined configuration. Therefore, in the photography step, it is preferable to photograph not only the metal member 100, but also four or more centers or vertices Pkn of patterns 202 to 204 so that they are included in the image data.

The parameters for the projection transformation are calculated as follows: First, a linear transformation is performed on the object coordinates (Xkn, Ykn) of all vertices Pkn using appropriate coefficients a and constants b and c. This determines the coordinates (xkn', ykn') (in units of [pixels]) where vertex Pkn should be in the image data after the projection transformation. The linear transformation is performed using equations (3) and (4) below.

Here, the coefficient 'a' is the reciprocal of the resolution 'r' in the image data after projection transformation. Therefore, it is preferable to pre-determine the resolution 'r' in the image data after projection transformation and then set the value of coefficient 'a' to the reciprocal of 'r'. Furthermore, the constants 'b' and 'c' are the x and y coordinates of the origin of the object coordinate system in the image data after projection transformation, respectively. Therefore, it is preferable to pre-determine the coordinates (x0, y0) of the origin of the object coordinate system in the image data after projection transformation and then set the values of constants 'b' and 'c' to 'b = x0' and 'c = y0', respectively.

For example, consider the case where the resolution r in the projected image data is 0.5 [mm/px], and the origin of the object coordinates in the projected image data is (0 [px], 100 [px]). In this case, the coefficient a should be a = 1/r = 1/0.5 [mm/px] = 2 [px/mm], and the constants b and c should be 0 [px] and 100 [px], respectively.

Next, calculate the values of parameters A, B, C, D, E, F, G, and H in equations (1) and (2) above. The values of each parameter are determined so that the image coordinates (xkn, ykn) of vertex Pkn in the captured image data are transformed by equations (1) and (2) above to the coordinates (xkn', ykn') that vertex Pkn should be in the image data after projection transformation.

There are several methods for calculating each parameter. For example, if the number of vertices Pkn is exactly four, substituting the coordinate values mentioned above into equations (1) and (2) above will result in the number of equations matching the number of unknowns, so it is sufficient to solve the system of equations for parameters A, B, C, D, E, F, G, and H. If the number of vertices Pkn is more than four, the values of parameters A, B, C, D, E, F, G, and H are determined using known mathematical methods such as the least squares method. Below, the combination of estimated projection transformation parameters (parameters A, B, C, D, E, F, G, and H) will be described as parameter M.
Note that the specific methods of projective transformation are not limited to those shown here.

Next, the shape measurement unit 32 performs a projection transformation on the image captured in the shooting step S1 using the estimated projection transformation parameters M. The image after projection transformation is generated by retaining the luminance or RGB values from the original image, and replacing the coordinates (x, y) in the original image with the coordinates (x', y') calculated using equations (1) and (2) above.

After the projection transformation step, the shape measurement unit 32 performs shape detection processing (step S23: shape detection step). In the shape detection step, the area occupied by the end face of the metal member 100 is detected from the image after projection transformation. For example, if the object is a long steel pipe, it is preferable to convert the image to grayscale and then apply a threshold based on the brightness value. This takes advantage of the fact that the cross section (especially the cut surface) is a bright area with a metallic luster, while the inside of the steel pipe is covered with an oxide film or is in shadow, making it relatively darker. Note that the method of detecting the cross section is not limited to that shown here. Below, an example of processing when the end face of the metal member 100 is colored red will be described.

Figure 11 is a flowchart illustrating the flow of the shape detection process. Figures 12 and 13 are diagrams illustrating the binarization process in the shape detection process.

In the shape detection process, the shape measurement unit 32 first reads the image after projection transformation (step S201). The shape measurement unit 32 reads the object coordinates (X, Y) and image coordinates (x, y) as well as the width (maximum value of the X coordinate) and height (maximum value of the Y coordinate) as information of the image after projection transformation. In this case, if it is not necessary to read the data by referring to the storage unit 36, for example, when processing the data after the projection transformation step, this step can be omitted. The maximum value of the X coordinate is used in the calculation of x _MAX described later, and the maximum value of the Y coordinate is used in the calculation of y _MAX .
Hereafter, the image loaded at this time will be referred to as the shape detection image.

The shape detection image, for example, as shown in the shape detection image G10 in Figure 12, is an end face image of the metal member 100, depicting the end face image G100 having a colored region Grm colored in red, and the hollow space G102 formed by the opening 101. Note that some images may also include an image of the marker plate 200.

Subsequently, the shape measurement unit 32 performs initialization of the shape detection image (step S202). The shape measurement unit 32 initializes the image by setting all pixels of the shape detection image to black.

After initialization, the shape measurement unit 32 first sets the coordinates (x, y) of the object to be processed to x = 1, y = 1, i.e., (x, y) = (1, 1) (step S203).

After setting the coordinates, the shape measurement unit 32 sets the color for the coordinates (x, y). First, the shape measurement unit 32 acquires the color intensity of the coordinates (x, y) (step S204). At this time, the shape measurement unit 32 acquires the channel intensity of red (R) (0 to 255), the channel intensity of green (G) (0 to 255), and the channel intensity of blue (B) (0 to 255).
In this embodiment, an example in which pixel values are represented by 256 gradations is described, but the invention is not limited to this.

The shape measurement unit 32 acquires the color intensity of the coordinates (x, y) and then performs a determination process on the channel intensity of R. First, the shape measurement unit 32 compares the channel intensity of R (hereinafter sometimes simply referred to as "R") with the value obtained by multiplying the channel intensity of B by a coefficient α (B × α: hereinafter sometimes simply referred to as "B × α") (step S205). In this case, the coefficient α is set to, for example, 1.25, and is preferably 1 or greater. By setting this coefficient α to a larger value of 1 or greater, it is possible to detect a red color closer to the primary color.

If the shape measurement unit 32 determines that R ≤ B × α (step S205: No), it proceeds to step S209. Conversely, if the shape measurement unit 32 determines that R > B × α (step S205: Yes), it proceeds to step S206.

In step S206, the shape measurement unit 32 then compares R with the value obtained by multiplying the channel strength of G by a coefficient α (G × α; hereinafter sometimes simply referred to as "G × α"). If the shape measurement unit 32 determines that R ≤ G × α (step S206: No), it proceeds to step S209. Conversely, if the shape measurement unit 32 determines that R > G × α (step S206: Yes), it proceeds to step S207.

In step S207, the shape measurement unit 32 then compares R with the first threshold value TL. Here, the first threshold value TL is a value that sets the minimum intensity of the red channel to be extracted, and is set, for example, to 50 or less. If the shape measurement unit 32 determines that R ≤ TL (step S207: No), it proceeds to step S209. Conversely, if the shape measurement unit 32 determines that R > TL (step S207: Yes), it proceeds to step S208.

In step S208, the shape measurement unit 32 sets the color of the coordinate (x, y) to white. In this step 208, steps S205 to S207 determined that the coordinate (x, y) was red, and a process to set its color to white is executed.

Meanwhile, in step S209, the shape measurement unit 32 compares R with the second threshold TH. Here, the second threshold TH is a value that sets the brightness (maximum intensity) of the red color to be extracted relative to the red channel intensity, and is set, for example, to 200 or higher. In this case, if the end face exhibits specular reflection due to lighting, etc., a so-called "blown-out" effect occurs in the image. On the other hand, areas other than the end face have high diffuse reflectivity and are less prone to blown-out. That is, in step S209, the unit determines whether an area is colored or not by determining whether R is caused by blown-out, even if the red channel intensity is not high compared to other colors.

If the shape measurement unit 32 determines that R ≤ TH (step S209: No), it proceeds to step S210. Conversely, if the shape measurement unit 32 determines that R > TL (step S209: Yes), it proceeds to step S208.

In step S210, the shape measurement unit 32 sets the color of the coordinate (x, y) to black. In this step S210, it is determined that the coordinate (x, y) is not red, and the process of setting its color to black is executed.

After setting the color in step S208 or S210, the shape measurement unit 32 proceeds to step S211. In step S211, the shape measurement unit 32 increases the value of x by 1.

The shape measurement unit 32 then determines whether the value of x = x + 1 is less than or equal to x _MAX (step S212). x _MAX is the maximum value in the x-axis direction of the image and is calculated based on the maximum value of the X coordinate. In this case, x _MAX is a value used to determine whether the right edge has been reached, for example, if the left edge of the image is set to 1. If the shape measurement unit 32 determines that x ≤ x _MAX (step S212: Yes), it proceeds to step S204 and performs the color setting process described above for the updated coordinates (x, y). On the other hand, if the shape measurement unit 32 determines that x > x _MAX (step S212: No), it proceeds to step S213.

In step S213, the shape measuring unit 32 resets the value of x to 1 and increases the value of y by 1.

The shape measurement unit 32 then determines whether the value of y = y + 1 is less than or equal to y _MAX (step S214). y _MAX is the maximum value in the y-axis direction of the image and is calculated based on the maximum value of the Y coordinate. In this case, y _MAX is a value used to determine whether the lower end has been reached, for example, if the upper end of the image is set to 1. If the shape measurement unit 32 determines that y ≤ y _MAX (step S214: Yes), it proceeds to step S204 and executes the color setting process described above for the updated coordinates (x, y). On the other hand, if the shape measurement unit 32 determines that y > y _MAX (step S214: No), it terminates the shape detection process and proceeds to the edge detection process (step S24).

The above-described process generates a grayscale image through a binarization process that turns the red-colored area white and the uncolored area black. For example, as shown in the grayscale image G20 in Figure 13, the red-colored area (corresponding to the colored area Grm in Figure 12, for example) is turned into a white area GW1, the area outside the white area GW1, which is the area GB1 other than the colored part of the metal member 100, and the hollow area GB2 corresponding to the opening 101 are turned into black.

Returning to Figure 7, the shape measurement unit 32 performs edge detection based on the end face detected in step S23 (step S24: edge detection step). Specifically, the shape measurement unit 32 extracts the boundary line (edge) between the detected end face and the hollow space and obtains the image coordinates of the point cloud constituting the boundary line. For example, as shown in the edge extraction image G30 in Figure 14, an image is obtained in which the edge GE constituting the boundary between the white area GW1 and the hollow area GB2 is depicted.

In this process, the image transformed by the projection transformation step has a known actual length corresponding to one pixel, i.e., its resolution. Therefore, based on this information, the actual shape and length of the aperture can be measured.

Here, a seamless rectangular steel pipe was used to perform shape detection and measurement necessary for creating shape data for the backing plate. The target steel pipe is a macro sample (a cross-sectional sample with a thickness of 30 mm) of a rectangular steel pipe with a cross-sectional shape of 200 mm on each side and a wall thickness of 22 mm. In addition, the area including the edge on the opening side of the end face is colored. A suitable marker plate was created for this sample and placed on the end face of the rectangular steel pipe and photographed. Next, an image that allows for length measurement on the image was generated by projection transformation. Then, contour line extraction was performed using the difference in brightness between the end face and the interior, taking advantage of the fact that the inside of the rectangular steel pipe is darkened by shadow. An example of the extracted contour line is shown in Figure 15. Figure 15 is a diagram showing an example of the detection result of the opposite side of the welded area of the metal member by shape measurement processing. When the coordinates of the point cloud data of this contour line were compared with the measurement data of the actual rectangular steel pipe, the maximum error was 1.7 mm. This error is within the range where welding is possible without problems and does not significantly worsen workability.

Then, after detecting the edge, the shape measurement unit 32 completes the shape measurement step S2 and proceeds to the subsequent design step (step S3).

Returning to Figure 4, in step S3, the design unit 33 designs the backing plate design data based on the extracted edge shape.
In the design of the backing plate, the design unit 33 translates, enlarges, or reduces the edge shape in a predetermined direction according to a predetermined thickness, for example, and creates a cross-sectional view assuming that the backing plate is cut at the end face of the metal member 100.

Subsequently, the design unit 33 designs a shape by extending the created cross-sectional drawing in a direction perpendicular to the cross-section, for example, according to a predetermined extension length, and uses this as the design drawing (design data) for the backing plate. Note that, taking into account measurement errors and distortion of the steel material, the entire or a part of the aforementioned design shape may be enlarged, reduced, or subjected to other geometric deformations. Furthermore, the design method for the shape of the backing plate is not limited to the above.

After the design step, the processing step (step S4) is performed. In the processing step, the processing flow of the processing device is determined based on the design data of the backing plate designed in the design step S3, and the molding unit 42 is controlled to produce the backing plate (for example, the backing plate 110 shown in Figure 3). Here, any predetermined material can be used to process the material into the backing plate. The predetermined material is appropriately selected according to the type of metal member 100 to be joined and the purpose of joining.

The fabricated backing plate is placed at the connection point (placement position) of the metal members 100 to be joined, as shown in Figure 3. After the backing plate is placed, the outer circumference opposite to the backing plate placement point is welded at the connection point of the metal members 100 (welding step). A welding jig (not shown) is used for welding. Known techniques can be used for welding jigs and welding methods, and they should be appropriately selected depending on the intended use of the metal members 100. As a welding jig, for example, a jig, device, or robot equipped with welding functions can be used. By welding the predetermined location in this welding step, a joined body is created in which the metal members 100 are joined together. At this time, the predetermined location to be welded is on the opposite side of the fabricated backing plate.

In the embodiment described above, a marker plate 200 is attached to the end face of the metal member 100, and an image captured is used to detect the edge of the opening of the metal member 100. Based on the detected shape, a backing plate is designed. According to this embodiment, since the shape detection process is performed mechanically by attaching a marker plate to the metal member 100 and capturing an image, the backing plate can be efficiently designed and manufactured.

Furthermore, according to this embodiment, the orientation of the metal member 100 is calibrated by projective transformation using multiple patterns formed on the marker plate 200, and the edges of the opening 101 are detected after calibration. This improves the degree of freedom of the angle during shooting while enabling accurate shape detection.

Furthermore, while the above-described embodiment explained an example of manufacturing a single backing plate placed at the connection point between metal members 100, it is not limited to this example and can be applied to the manufacturing of various backing plates used during welding. Figures 16 to 19 illustrate other examples of welding configurations in the welding step.

For example, as shown in Figure 16, a backing plate 120 may be placed on each metal member 100. Specifically, one backing plate 120 is placed so as to overlap the boundary portion of the metal member 100. In this state, the boundary portions of adjacent metal members 100 are welded together on the opposite side from where the backing plate 120 is placed. This backing plate 120 is manufactured by attaching a marker plate 200 to the metal member 100 to be fitted, photographing it, and performing shape detection processing on each. Note that, for example, if the metal members 100 are positioned with a large offset from each other, a backing plate 120 may be placed on each metal member 100 (two in the case of Figure 16).

Furthermore, as shown in Figure 17, in a structure where a diaphragm 130 is provided between metal members 100 and the metal members 100 and diaphragm 130 are welded together, a backing plate 140 may be provided on each metal member 100. Specifically, a backing plate 140 is provided on each metal member 100, and the diaphragm 130 is sandwiched between the backing plates 140. In this state, the metal member 100 and the diaphragm 130 are welded together on the side opposite to where the backing plates 140 are provided. These backing plates 140 are manufactured by attaching a marker plate 200 to the metal member 100 to be fitted, taking a photograph, and performing shape detection processing on each. According to the structure shown in Figure 17, a welded portion Rw, which is molten and solidified, is formed at the end of the metal member 100 on the diaphragm 130 side.

Furthermore, the backing plate manufactured according to the present invention can also be used in the manufacture of a column member 160 that will become part of the column of the structure being constructed and will also be equipped with a diaphragm 150a for connecting beams at the construction site. Figure 18 shows an example of a typical column member 160 with a diaphragm. The column member 160 in Figure 18 comprises two diaphragms 150a and three metal members 150 arranged to sandwich the diaphragms 150a. The metal members 150 here are hollow rectangular prism-shaped tubular members. The axial lengths of the metal members 150 may be the same or different depending on their intended use. When welding the metal members 150 of the column member 160 to the diaphragms 150a, a backing plate may be placed on each of the metal members 150. Specifically, a backing plate is placed on the inner surface of each metal member 150, and the diaphragm 150a is sandwiched between the backing plates. Then, similar to the case in Figure 17, the metal member 150 and the diaphragm 150a are welded together on the side opposite to where the backing plate is to be installed (welding step). This backing plate is manufactured by attaching a marker plate 200 to each metal member 150 to be installed, photographing them, and performing shape detection processing on each.

Furthermore, the backing plate manufactured according to this invention can also be used in the manufacture of a column-beam joint 180, which comprises a column member 160 that forms part of the column of the structure to be constructed, and two or more beam members 170 for connecting beams at the construction site, thereby joining the columns and beams of the structure. Figure 19 shows an example of a typical column-beam joint 180.

The column member 160 of the column-beam joint 180 in Figure 19 has the same structure as the diaphragm-equipped column member 160 in Figure 18. That is, when welding the metal member 150 and the diaphragm 150a of the column member 160, a backing plate may be placed on each metal member 150. Specifically, a backing plate is placed on the inner surface of each metal member 150, and the diaphragm 150a is sandwiched between these backing plates. Then, as in the case of Figure 17, the metal member 150 and the diaphragm 150a are welded together on the side opposite to where the backing plate is placed (welding step). These backing plates are created by attaching a marker plate 200 to each metal member 150 to be fitted, photographing them, and performing shape detection processing on each.

On the other hand, the beam member 170 is made of an H-shaped steel with two flanges 170a and one web 170b. Furthermore, in the case of Figure 19, the two beam members 170 are arranged in the same straight line via the column member 160. The number and arrangement of the beam members 170 relative to the column member 160 can be changed depending on the intended use of the column-beam joint. For example, when the column-beam joint is located at the corner of a structure, the two beam members 170 are arranged perpendicular to each other with respect to the column member 160.

Furthermore, in the case of the column-beam joint 180 in Figure 19, when welding the beam member 170 to the column member 160, a backing plate may be placed at the connection point between the diaphragm 150a and the flange 170a. Specifically, a backing plate is placed on the underside of each flange 170a. In this state, as in the case of Figure 16, the diaphragm 150a and the flange 170a are welded together on the side opposite to where the backing plate is placed (welding step). The backing plate is photographed by attaching a marker plate 200 to the diaphragm 150a or flange 160a to be placed, and shape detection processing is performed on the end face of the placement location.

In the embodiment described above, an example was explained in which a portion of the end face of the metal member 100 is colored red. However, if it is colored with another color, in steps S205 to S207 and S209 of the shape detection process shown in Figure 11, processing is performed based on the color component characteristic of the colored area. Furthermore, even if the end face of the metal member 100 is not colored, the processing according to this embodiment can be applied. In this case as well, in steps S205 to S207 and S209 of the shape detection process shown in Figure 11, processing is performed based on the color component characteristic of the end face image of the metal member 100.

Furthermore, while the above-described embodiment explained an example of performing processing as a backing plate manufacturing system 1, the system may be installed within the same building, or at least a portion of it may be installed in different buildings.

For example, the imaging device 2 may be equipped with an image processing unit 22 and a shape measurement unit 32. In this case, the operator can adjust the shape measurement step S2 while observing the measurement results, which is preferable as it enables more accurate shape measurement. The design device 3 may be configured to have only a design unit 33 without the shape measurement unit 32. The processing device 4 may be equipped with the design unit 33. Alternatively, the entire system may not have a design device 3. In the last example, it is preferable that the shape measurement unit 32 is located in either the remaining imaging device 2 or processing device 4, and the design unit 33 is located in the processing device 4. Furthermore, instead of the design device 3, a shape measurement device 3' specialized for shape measurement may be provided, which is the same as the design device 3 but without the design unit 33. In this case, the shape measurement device 3' may be equipped with a communication unit 31, a shape measurement unit 32, a control unit 35, an input unit 34, and a storage unit 36.

For example, the image processing unit 22, shape measurement unit 32, design unit 33, and molding unit 42 may be provided on servers located in the cloud, one or more of them. Of course, other units may also be located in the cloud as needed.

For example, a company performing the photography step, a company performing the measurement and design steps, and a company performing the processing step could each have system components installed in the respective company's building. Similarly, a company performing the photography and measurement steps, a company performing the design step, and a company performing the processing step could each have system components installed in the respective company's building. Similarly, a company performing the photography and measurement steps, a company performing the design and processing steps could each have system components installed in the respective company's building.

1. Backing plate manufacturing system 2. Imaging device 3. Design device 4. Processing device 21. Imaging unit 22. Image processing unit 23. Display unit 24, 31, 41. Communication unit 25, 34, 43. Input unit 26, 35, 44. Control unit 27, 36, 45. Memory unit 32. Shape measurement unit 33. Design unit 42. Molding unit 100, 150. Metal member 101. Opening 110, 120. Backing plate 130, 150a. Diaphragm 160. Column member (with diaphragm)
170 Beam member 170a Flange 170b Web 180 Column-beam joint 200 Marker plate 201 Main body 202-204 Pattern

Claims (11)

A method for manufacturing a backing plate used for welding metal components together,
A processing step of manufacturing the backing plate by processing a predetermined material based on the shape of the arrangement location detected based on image data of the arrangement location and a marker plate attached to the arrangement location on which a pattern used as a reference for length measurement is formed.
A method for manufacturing a backing plate that includes this. A method for designing a backing plate used for welding metal components together,
A design step of creating design data for a backing plate based on image data of a location on at least one of the metal members where a backing plate is to be placed, and the shape of the location where the backing plate is to be placed, which is detected based on image data of a marker plate attached to the location where the backing plate is to be placed and on which a pattern used as a reference for measuring the length is formed.
A design method for backing plates, including the backing plate. A welding method in which metal members are welded together using a backing plate,
A welding step comprising: placing the backing plate on the metal member at the location where the backing plate is to be placed, and the backing plate, which is manufactured based on the shape of the location where the backing plate is to be placed, as detected based on image data of a marker plate attached to the location where the backing plate is to be placed and on which a pattern used as a reference for length measurement is formed, on the metal member, and welding the metal member to another metal member on the side opposite to the location where the backing plate is to be placed,
A welding method for metal components including metal parts. A method for manufacturing a column member equipped with a diaphragm,
The column member is the metal member,
A welding step of welding the metal member and the diaphragm using the backing plate manufactured by the method for manufacturing the backing plate described in claim 1,
A method for manufacturing a column member including a column member. A method for manufacturing a column-beam joint comprising a column member and a beam member,
The steps include manufacturing the column member using the method for manufacturing a column member described in claim 4,
A method for manufacturing a column-beam joint, including the joint itself. A shape measurement method for measuring the cross-sectional shape of a metal component,
A projection transformation step involves detecting the position of the pattern in image data obtained by photographing the cross-section of the metal member and a marker plate attached to the cross-section on which multiple patterns used as a reference for length measurement are formed , and projecting the image based on the detection result.
The process includes a binarization step to create a binarized image based on the channel intensity of a predetermined color component at each pixel position in the image data after projection transformation , and a shape detection step to detect the shape of the cross-section,
An edge detection step for detecting the edges of the cross-section based on the binarized image,
A method for measuring shape, including the measurement of shape. A shape measurement program for measuring the cross-sectional shape of a metal component,
A projection transformation step involves detecting the position of the pattern in image data obtained by photographing the cross-section of the metal member and a marker plate attached to the cross-section on which multiple patterns used as a reference for length measurement are formed , and projecting the image based on the detection result.
The process includes a binarization step to create a binarized image based on the channel intensity of a predetermined color component at each pixel position in the image data after projection transformation , and a shape detection step to detect the shape of the cross-section,
An edge detection step for detecting the edges of the cross-section based on the binarized image,
A shape measurement program that is executed by a computer. A shape measuring device for measuring the cross-sectional shape of a metal component,
A shape detection unit detects the shape of the cross-section based on image data of the cross-section of the metal member and a marker plate attached to the cross-section, on which multiple patterns used as a reference for length measurement are formed.
Equipped with ,
The shape detection unit,
A projection transformation process that detects the position of the pattern in the image data and performs a projection transformation on the image based on the detection result,
The process includes a binarization process to create a binarized image based on the channel intensity of a predetermined color component at each pixel position in the image data after the projection transformation, and a shape detection process to detect the shape of the cross-section,
Based on the binarized image, an edge detection process is performed to detect the edges of the cross-section,
A shape measuring device that performs this operation . The shape measuring device according to claim 8 , further comprising an imaging unit for capturing the aforementioned image data. A method for manufacturing a backing plate used for welding metal components together,
A shape detection step in which the shape of the arrangement location is detected based on image data of the arrangement location of the backing plate on at least one of the metal members and a marker plate attached to the arrangement location on which a pattern used as a reference for length measurement is formed,
A design step which creates design data for a backing plate based on the detection results from the shape detection step,
A processing step of manufacturing the backing plate by processing a predetermined material according to the design data,
A method for manufacturing a backing plate that includes this. A manufacturing system for backing plates used in welding metal components together,
An imaging unit that photographs the location where a backing plate is provided on at least one metal component, and a marker plate attached to the location where a pattern for measuring length is formed,
Based on the image data captured by the imaging unit, a shape detection unit detects the shape of the installation location,
A design unit creates design data for a backing plate based on the detection results from the shape detection unit,
A processing unit that processes a predetermined material according to the design data to produce the backing plate,
A manufacturing system for backing plates equipped with the following features.