Internal back surface structure detecting method, internal back surface structure detecting system, and method for manufacturing joint structure

Abstract

The objective of the present invention is to detect internal back surface structures that cannot be viewed from the front surface of an object being measured.　This internal back surface structure detecting method for detecting an internal back surface structure present in an object being measured includes: a coating step for coating the front surface of the object being measured with an identification pattern; a first acquisition step for acquiring a reference image obtained by imaging the front surface, which has been coated with the identification pattern, of the object being measured; an application step for applying a load to the object being measured to generate strain in the object being measured; a second acquisition step for acquiring a strain image obtained by imaging the front surface, which has been coated with the identification pattern, of the object being measured, in which the strain has been generated by the application of the load; a calculation step for calculating a strain distribution generated on the front surface of the object being measured on the basis of the reference image and the strain image; and a detection step for detecting the internal back surface structure present in the object being measured, with reference to a database, on the basis of the strain distribution.

Description

Internal rear surface structure detection method, internal rear surface structure detection system, and method for manufacturing bonded structure

[0001] The present invention relates to an internal back surface structure detection method, an internal back surface structure detection system, and a method for manufacturing a bonded structure. This invention claims priority from Japanese Patent Application No. 2024-025222, filed on February 22, 2024, and the contents of that application are incorporated by reference into the present application in designated states where incorporation by reference of documents is permitted.

[0002] Strain gauges are generally used to measure strains occurring in mechanical structures such as railway vehicles and construction machinery. Strain gauges can measure strain simply by being installed on the surface of the object to be measured, such as a mechanical structure, and have a strain detection limit of 5.0 × 10 －４ ~1.0 x 10 －３ This is because the measurable limit frequency is high at around several kHz.

[0003] However, since strain gauges can only measure strain at the point where they are installed, it is necessary to install many strain gauges when measuring strain distribution in an object to be measured. However, preparing and installing many strain gauges is a heavy workload and increases costs, and only discrete strain distributions can be measured.

[0004] For this reason, in recent years, digital image correlation (hereinafter referred to as DIC) has become popular as a technique for measuring strain distribution in an object to be measured without using strain gauges. In strain distribution measurement using DIC, an identification pattern that can be recognized by DIC is applied to the object to be measured in advance. Then, the amount of deformation of the identification pattern is calculated based on an image of the identification pattern taken before a load is applied to the object to be measured and an image of the identification pattern taken after the load is applied to the object to be measured, and the strain distribution occurring on the surface of the object to be measured is measured from this amount of change.

[0005] Strain distribution measurement using DIC allows continuous strain distribution measurement while reducing the workload and costs compared to methods using strain gauges.

[0006] Regarding strain distribution measurement using DIC, for example, Patent Document 1 describes a bolt axial force inspection method in which a feature pattern 5 that visually reveals strain distribution based on bolt displacement information is formed on a bolt 2, a first image of the feature pattern is captured, a known external force is applied to the bolt, a second image of the feature pattern is captured, a strain value is calculated from each of the two captured images of the feature pattern, the difference between these strain values ​​is calculated, the difference in the bolt axial force calculated from the external force and the difference in the calculated strain value are input into a calibration curve that indicates the relationship between the bolt axial force and the bolt strain value, the bolt axial force is calculated, and the calculated bolt axial force is compared with a target value for the bolt axial force, thereby determining whether or not reinspection is necessary.

[0007] Japanese Patent Application Laid-Open No. 2022-150276

[0008] The prior art such as that disclosed in Patent Document 1 is specialized in detecting the state of the surface of an object coated with a characteristic pattern, such as detecting strain distribution or cracks on the surface of the object, but is unable to detect hollow areas occurring inside the object or notches occurring on the back surface of the object. Hereinafter, in this specification, hollow areas occurring inside the object and notches occurring on the back surface of the object will be referred to as the internal back surface structure.

[0009] The present invention has been made in view of the above points, and has as its object to make it possible to detect an internal back surface structure that cannot be seen from the surface of an object to be measured.

[0010] The present application includes a number of means for solving at least part of the above problems, examples of which are as follows.

[0011] In order to solve the above problem, one embodiment of the present invention provides an internal back surface structure detection method for detecting an internal back surface structure present in an object to be measured, and includes: an application step for applying an identification pattern to the surface of the object to be measured; a first acquisition step for acquiring a reference image of the surface of the object to be measured on which the identification pattern has been applied; an application step for applying a load to the object to generate strain in the object to be measured; a second acquisition step for acquiring a strain image of the surface of the object to be measured on which the identification pattern has been applied, where the strain has occurred due to the application of the load; a calculation step for calculating the strain distribution generated on the surface of the object to be measured based on the reference image and the strain image; and a detection step for detecting the internal back surface structure present in the object to be measured by referring to a database based on the strain distribution.

[0012] According to the present invention, it is possible to detect an internal back surface structure that cannot be seen from the surface of an object to be measured.

[0013] Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments.

[0014] FIG. 1 is a diagram showing an example of the configuration of an internal back surface structure detection system according to an embodiment of the present invention. FIG. 2 is a diagram showing an example of imaging when the imaging unit is configured with one camera. FIG. 3 is a diagram showing an example of imaging when the imaging unit is configured with two cameras. FIG. 4 shows an example of a case where a hollow portion exists inside the object to be measured. FIG. 5 is a diagram showing an example of a case where a tensile load is applied to an object to be measured that has a hollow portion inside. FIG. 6 is a diagram showing an example of a strain image captured of the surface of an object to be measured that has no internal back surface structure and in which strain has occurred due to the application of a tensile load. FIG. 7 is a diagram showing an example of a strain image captured of the surface of an object to be measured that has a hollow portion inside and in which strain has occurred due to the application of a tensile load. FIG. 8 is a diagram showing an example of a strain distribution map of an object to be measured that has a hollow portion inside and in which strain has occurred due to the application of a tensile load. FIG. 9 is a diagram showing an example of a case where a compressive load is applied to an object to be measured that has a hollow portion inside. FIG. 10 shows an example of a strain image captured on the surface of a measured object having a hollow portion, where strain has occurred due to the application of a compressive load. FIG. 11 shows an example of a strain distribution map of a measured object having a hollow portion, where strain has occurred due to the application of a compressive load. FIG. 12 shows an example of a case where a moment load is applied to a measured object having a hollow portion, where strain has occurred due to the application of a moment load. FIG. 13 shows an example of a strain image captured on the surface of a measured object having a hollow portion, where strain has occurred due to the application of a moment load. FIG. 14 shows an example of a strain distribution map of a measured object having a hollow portion, where strain has occurred due to the application of a moment load. FIG. 15 shows an example of a case where a notch is present on the back surface of the measured object. FIG. 16 shows an example of a case where a tensile load is applied to a measured object having a notch on the back surface. FIG. 17 shows an example of a strain image captured on the surface of a measured object having a notch on the back surface, where strain has occurred due to the application of a tensile load. Fig. 18 shows an example of a strain distribution map of an object having a notch on the back surface, where strain is generated by application of a tensile load. Fig. 19 shows an example of a strain distribution map of an object having a notch on the back surface, where strain is generated by application of a compressive load.FIG. 20 is a diagram showing an example of a strain image obtained by capturing an image of the surface of a measured object having a notch on its back surface, where strain has occurred due to the application of a compressive load. FIG. 21 is a diagram showing an example of a strain distribution map of a measured object having a notch on its back surface, where strain has occurred due to the application of a compressive load. FIG. 22 is a diagram showing an example of a case where a moment load is applied to a measured object having a notch on its back surface. FIG. 23 is a diagram showing an example of a strain image obtained by capturing an image of the surface of a measured object having a notch on its back surface, where strain has occurred due to a moment load. FIG. 24 is a diagram showing an example of a strain distribution map of a measured object having a notch on its back surface, where strain has occurred due to a moment load. FIG. 25 is a flowchart explaining an example of DB (database) generation processing. FIG. 26 is a flowchart explaining an example of internal back surface structure processing. FIG. 27 is a diagram showing an example of internal back surface structure detection. FIG. 28 is a diagram showing an example of internal back surface structure detection.

[0015] An embodiment of the present invention will be described below with reference to the drawings. The embodiments are merely examples for explaining the present invention, and appropriate omissions and simplifications have been made for clarity of explanation. The present invention can be implemented in various other forms. Unless otherwise specified, each component may be singular or plural. The position, size, shape, and scope of each component shown in the drawings may not represent the actual position, size, shape, and scope in order to facilitate understanding of the invention. In all drawings used to explain the embodiments, identical components are generally designated by the same reference numerals, and repeated description thereof will be omitted. Furthermore, in the following embodiments, a component (including an element step, etc.) is not necessarily essential unless otherwise specified or considered to be clearly essential in principle. Furthermore, when a term "consists of A," "composed of A," "having A," or "including A" is used, it does not exclude other elements unless otherwise specified, such as when referring to only that element. Similarly, in the following embodiments, when referring to the shape, positional relationship, etc. of a component, etc., it includes those that are substantially similar or similar to that shape, etc., unless otherwise specified or considered to be clearly essential in principle. Furthermore, "obtaining" includes, as specific examples, at least the subject generating, calculating, or receiving from outside.

[0016] 1 shows a configuration example of an internal back surface structure detection system 10 according to an embodiment of the present invention. The internal back surface structure detection system 10 includes a load application device 20, an imaging device 30, and an image processing device 40.

[0017] The internal back surface structure detection system 10 detects internal back surface structures such as hollow areas that occur inside the object to be measured 1, which are components that make up mechanical structures such as railway vehicles and construction machinery, and cut areas that occur on the back surface (the surface opposite to the surface being imaged).

[0018] An identification pattern P that can be recognized by DIC is applied to the surface of the object 1 near the area where the internal back surface structure is expected to occur. The identification pattern P is generally a random pattern, and is formed, for example, by applying a base coat of white paint and then overlaying it with dots of black paint. By using a random pattern as the identification pattern P, it is possible to prevent the corresponding positions between the reference image and the strain image from being recognized.

[0019] The identification pattern P may be applied using any tool, such as a spray, paintbrush, roller, or brush. The identification pattern P may be applied manually or using a specified device. Furthermore, the identification pattern P may be formed by reversing the black and white colors and applying white dots on a black base. Furthermore, if the surface of the object 1 to be measured has no gloss, the identification pattern P may be formed by applying only dots without applying a base coat.

[0020] The load application device 20 applies to the object 1 either a tensile load that applies a force in a direction that stretches the object 1, a compressive load that applies a force in a direction that contracts the object 1, or a moment load that applies a force in a direction that bends the object 1, thereby generating strain in the object 1.

[0021] The imaging device 30 generates a reference image by capturing an image of the surface of the workpiece 1 on which the identification pattern P is applied before a load is applied by the load application device 20. The imaging device 30 also generates a strain image by capturing an image of the surface of the workpiece 1 on which the identification pattern P is applied after a load is applied by the load application device 20. The imaging device 30 and the image processing device 40 may be connected by wire or wirelessly.

[0022] The image processing device 40 acquires a reference image and a strain image from the imaging device 30, calculates the strain distribution on the surface of the object to be measured 1 based on the reference image and the strain image, and detects the internal back surface structure of the object to be measured 1 based on the strain distribution.

[0023] The image processing device 40 includes a processing unit 41 and a storage unit 42. The image processing device 40 is realized by a general computer such as a personal computer or a server computer. The computer includes a processor such as a central processing unit (CPU), a memory such as a dynamic random access memory (DRAM), a storage such as a hard disk drive (HDD) or a solid state drive (SSD), input devices such as a keyboard, a mouse, and a media drive, an output device such as a display, and a communication module such as an Ethernet (trademark) card or a Wi-Fi (trademark) adapter.

[0024] The processing unit 41 is realized by a processor of a computer that constitutes the image processing device 40. Each functional block of the processing unit 41, that is, the image acquisition unit 411, the load control unit 412, the strain distribution calculation unit 413, the DB generation unit 414, and the internal back surface structure detection unit 415, is realized by the processor executing a predetermined program stored in the memory.

[0025] The predetermined program executed by the processor may be stored in memory in advance, or may be downloaded from a predetermined server or the like via a removable medium (CD-ROM, flash memory, etc.) or a network such as the Internet, stored in storage, which is a non-transitory storage medium, and read from the storage when needed. For this reason, it is preferable that the computer has an interface for reading data from removable media.

[0026] Furthermore, the image processing device 40 may be realized by one physical or logical computer, or by two or more physical or logical computers, which may be distributed over a network.

[0027] The image acquisition unit 411 controls the imaging device 30 and acquires a reference image and a strain image from the imaging device 30. The load control unit 412 controls the load application device 20. The strain distribution calculation unit 413 calculates the strain distribution on the surface of the object 1 using DIC based on the reference image and the strain image. The DB generation unit 414 generates a DB 421 that associates the type of load and the strain distribution for each combination of the position of the internal back surface structure of the object 1, which is known to have an internal back surface structure, and at least one of the material of the object 1, the size of the object 1, the type of internal back surface structure (e.g., internal hollow portion, notched portion on the back surface), and the size of the internal back surface structure. The internal back surface structure detection unit 415 references the DB 421 in the storage unit 42 and detects the internal back surface structure T of the object 1 based on the strain distribution on the surface of the object 1.

[0028] The storage unit 42 is realized by a computer storage. The storage unit 42 stores the DB 421. Note that the storage unit 42 may store information and data other than the DB 421.

[0029] 2 shows an example of how imaging is performed when the imaging device 30 is configured with one camera 31 (hereinafter referred to as a single camera system). In the single camera system, it is not possible to detect the imaging distance L between the imaging device 30 and the object 1 to be measured from the image captured by the camera 31. For this reason, when imaging, it is necessary to measure and manage the imaging distance L in advance.

[0030] 3 shows an example of how images are captured when the imaging device 30 is configured with two cameras 31 and 32 (hereinafter referred to as a stereo camera system). Even in the case of the stereo camera system, it is desirable to measure and manage the imaging distance L between the imaging device 30 and the object 1 to be measured in advance. However, in the case of the stereo camera system, since the imaging conditions including the imaging distance L can be obtained from the images captured by the two cameras 31 and 32, it is also possible to correct the images based on the imaging conditions. Note that even in the case of a single camera system, image correction is possible by using the imaging distance L.

[0031] <Strain distribution when a hollow portion T1 exists inside the object to be measured 1> Figure 4 shows a schematic example of an example of a state in which an identification pattern P is applied to an object to be measured 1 that has a hollow portion T1 inside, with the upper part of the figure showing the top surface (imaging surface) of the object to be measured 1 and the lower part of the figure showing the AA' cross section of the object to be measured 1.

[0032] 5 shows a state in which a tensile load 21 is applied by a load application device 20 to a measured object 1 having a hollow portion T1 therein. When the tensile load 21 is applied, the measured object 1 deforms in the in-plane direction (the left-right direction in the drawing), so the imaging distance L can be considered to be approximately constant. Therefore, whether the imaging device 30 is a single-camera type or a stereo-camera type, correction of the captured image is not necessary.

[0033] 6 shows an example of a strain image captured of the surface of the object 1 under test, which does not have an internal back surface structure, in which strain has occurred due to the application of a tensile load 21. As shown in the figure, when a tensile load 21 is applied to the object 1 under test, which does not have an internal back surface structure, strain occurs uniformly (evenly).

[0034] Fig. 7 shows an example of a strain image captured of the surface of the object 1 under test, which has a hollow region T1 inside and in which strain has occurred due to the application of a tensile load 21. Fig. 8 shows the strain distribution on AA' of the object 1 under test, calculated based on the strain image and a reference image. In Fig. 8, the horizontal axis represents the position on AA' of the object 1 under test, and the vertical axis represents the magnitude of strain, with positive values ​​representing expansion and negative values ​​representing contraction.

[0035] 8, the strain distribution of the object 1 having a hollow portion T1 therein, in which strain has been generated by the application of a tensile load 21, is positive throughout. Note that the strain is uniform in the portion where the hollow portion T1 is not present, but the absolute values ​​of the strain on both sides of the portion where the hollow portion T1 is present are maximum values, and the absolute value of the strain on the surface immediately adjacent to the hollow portion T1 is minimum values ​​that are slightly smaller than those on both sides.

[0036] 9 shows a state in which a compressive load 22 is applied by a load application device 20 to a measured object 1 having a hollow portion T1 therein. When a compressive load 22 is applied, the measured object 1 deforms in the in-plane direction (the left-right direction in the drawing), so the imaging distance L can be considered to be approximately constant. Therefore, whether the imaging device 30 is a single-camera type or a stereo-camera type, correction of the captured image is not necessary.

[0037] A strain image (not shown) of the surface of the object 1 to be measured, which has no internal back surface structure and in which strain has occurred due to the application of a compressive load 22, is similar to the strain image (Figure 6) of the surface of the object 1 to be measured, which has no internal back surface structure and in which strain has occurred due to the application of a tensile load 21, and therefore is not shown in the figure.

[0038] Fig. 10 shows an example of a strain image captured of the surface of the object 1 under test, which has a hollow region T1 inside and in which strain has occurred due to the application of a compressive load 22. Fig. 11 shows the strain distribution on AA' of the object 1 under test, calculated based on the strain image and a reference image. In Fig. 11, the horizontal axis represents the position on AA' of the object 1 under test, with positive values ​​representing expansion and negative values ​​representing contraction.

[0039] 11 , the strain distribution when compressive load 22 is applied to object 1, in which strain is generated by the application of compressive load 22, is negative throughout. Note that the strain is uniform in the area where hollow portion T1 does not exist, but the absolute values ​​of the strain on both sides of the area where hollow portion T1 exists are maximum values, and the absolute value of the strain on the surface immediately adjacent to hollow portion T1 is a minimum value that is slightly smaller.

[0040] FIG. 12 shows a state in which a moment load 23 is applied by a load application device 20 to a measured object 1 having a hollow portion T1 therein. When the moment load 23 is applied, the measured object 1 deforms in the out-of-plane direction (the up-down direction in the drawing), and the imaging distance L changes. Therefore, if the imaging device 30 is a single-camera type, an error occurs in the strain calculation. On the other hand, with a stereo camera type, it is possible to detect changes in the imaging distance L due to out-of-plane deformation, and therefore it is possible to acquire the strain distribution without correcting the image. Note that even with a single-camera type, the strain distribution can be calculated by correcting the image based on the amount of change in the imaging distance L.

[0041] A strain image (not shown) of the surface of the object 1 to be measured, which has no internal back surface structure and in which strain has occurred due to the application of moment load 23, is similar to the strain image (Figure 6) of the surface of the object 1 to be measured, which has no internal back surface structure and in which strain has occurred due to the application of tensile load 21, and therefore is not shown in the figure.

[0042] Fig. 13 shows an example of a strain image captured of the surface of the object 1 under test, which has a hollow portion T1 inside and in which strain has occurred due to the application of moment load 23. Fig. 14 shows the strain distribution on AA' of the object 1 under test, calculated based on the strain image and a reference image. In Fig. 14, the horizontal axis represents the position on AA' of the object 1 under test, with positive values ​​representing expansion and negative values ​​representing contraction.

[0043] 14, the strain distribution of the object 1 having a hollow portion T1 therein, in which strain has been generated by the application of moment load 23, is positive throughout. Note that the strain is uniform in the portion where hollow portion T1 is not present, but the absolute values ​​of the strain on both sides of the portion where hollow portion T1 is present are maximum values, and the absolute value of the strain on the surface immediately adjacent to hollow portion T1 is a minimum value that is slightly smaller.

[0044] <Strain distribution when a cut-out portion T2 is present on the back surface of the object to be measured 1> Figure 15 shows a schematic example of an example of a state in which an identification pattern P is applied to an object to be measured 1 having a cut-out portion T2 on the back surface, where the upper part of the figure shows the top surface (imaging surface) of the object to be measured 1 and the lower part of the figure shows the AA' cross section of the object to be measured 1.

[0045] 16 shows a state in which a tensile load 21 is applied by a load application device 20 to a measured object 1 having a notch T2 on its back surface. When the tensile load 21 is applied, the measured object 1 deforms in the in-plane direction (the left-right direction in the drawing), so the imaging distance L can be considered to be approximately constant. Therefore, whether the imaging device 30 is a single-camera type or a stereo-camera type, correction of the captured image is not necessary.

[0046] Fig. 17 shows an example of a strain image captured of the surface of the object 1 under test, which has a notch T2 on its back surface and in which strain has occurred due to the application of a tensile load 21. Fig. 18 shows the strain distribution along AA' of the object 1 under test, calculated based on the strain image and a reference image. In Fig. 18, the horizontal axis represents the position along AA' of the object 1 under test, with positive values ​​representing expansion and negative values ​​representing contraction.

[0047] As shown in Figure 18, the strain distribution of the object 1 to be measured, which has a cut portion T2 on its back surface and in which strain has occurred due to the application of a tensile load 21, is characterized in that the strain is uniform in areas where the cut portion T2 is not present, but the absolute value of the strain is greatest near the cut portion T2, and the sign of the strain changes from positive to negative to positive.

[0048] 19 shows a state in which a compressive load 22 is applied by a load application device 20 to a measured object 1 having a notch T2 on its back surface. When a compressive load 22 is applied, the measured object 1 deforms in the in-plane direction (the left-right direction in the drawing), so the imaging distance L can be considered to be approximately constant. Therefore, whether the imaging device 30 is a single-camera type or a stereo-camera type, correction of the captured image is not necessary.

[0049] Fig. 20 shows an example of a strain image captured of the surface of the object 1 under test, which has a cut portion T2 on its back surface and in which strain has occurred due to the application of a compressive load 22. Fig. 21 shows the strain distribution on AA' of the object 1 under test, calculated based on the strain image and a reference image. In Fig. 21, the horizontal axis represents the position on AA' of the object 1 under test, with positive values ​​representing expansion and negative values ​​representing contraction.

[0050] As shown in Figure 21, the strain distribution of the object 1 to be measured, which has a cut portion T2 on its back surface and in which strain has occurred due to the application of a compressive load 22, is characterized in that the strain is uniform in areas where the cut portion T2 is not present, but the absolute value of the strain is greatest near the cut portion T2, and the sign of the strain changes from negative to positive to negative.

[0051] 22 shows a state in which a moment load 23 is applied by a load application device 20 to a measured object 1 having a notch T2 on its back surface. When the moment load 23 is applied, the measured object 1 deforms in the out-of-plane direction (the up-down direction in the drawing), and the imaging distance L changes. Therefore, if the imaging device 30 is a single-camera type, an error occurs in the strain calculation. On the other hand, with a stereo camera type, it is possible to detect changes in the imaging distance L due to out-of-plane deformation, and therefore it is possible to acquire the strain distribution without correcting the image. Note that even with a single-camera type, the strain distribution can be calculated by correcting the image based on the amount of change in the imaging distance L.

[0052] Fig. 23 shows an example of a strain image captured of the surface of the object 1 under test, which has a notch T2 on its back surface and in which strain has occurred due to the application of moment load 23. Fig. 24 shows the strain distribution on AA' of the object 1 under test, calculated based on the strain image and a reference image. In Fig. 24, the horizontal axis represents the position on AA' of the object 1 under test, with positive values ​​representing expansion and negative values ​​representing contraction.

[0053] 24, the strain distribution of the object 1 having the cut portion T2 on the back surface, where strain has been generated by the application of moment load 23, is positive throughout. Note that the strain is uniform in the area where the cut portion T2 is not present, but the absolute values ​​of the strain on both sides of the area where the cut portion T2 is present are maximum values, and the absolute value of the strain on the surface immediately adjacent to the cut portion T2 is a minimum value that is slightly smaller.

[0054] <DB Creation Processing by Image Processing Device 40> FIG. 25 is a flowchart illustrating an example of DB creation processing by the image processing device 40. As shown in FIG.

[0055] The DB generation process is a process for generating DB421 in preparation for performing the internal back surface structure detection process (described below), and is performed using a large number of objects 1 to be measured that have internal back surface structures and whose type, position, and size are known.

[0056] The DB generation process is initiated, for example, in response to a predetermined start operation from the user. First, for example, an operator applies an identification pattern P to the object 1 under test, the type, position, and size of the internal back surface structure of which are known (step S1).

[0057] Next, the image acquisition unit 411 controls the imaging device 30 to capture an image of the surface of the object 1 to which the identification pattern P is applied, and acquires the resulting reference image (step S2).

[0058] Next, the load control unit 412 controls the load application device 20 to apply any one of a tensile load, a compressive load, or a moment load to the object 1 to be measured (step S3).

[0059] Next, the image acquisition unit 411 controls the imaging device 30 to capture an image of the surface of the object to be measured 1 on which the identification pattern P is applied while a load is being applied, and acquires the resulting strain image (step S4).

[0060] Next, the strain distribution calculation unit 413 calculates the strain distribution on the surface of the object 1 by DIC based on the reference image and the strain image (step S5).

[0061] Next, the DB generation unit 414 associates the type of load with the strain distribution for each combination of the material, size, type of internal back surface structure (internal hollow portion, notched portion on the back surface, etc.), position of the internal back surface structure, and size of the internal back surface structure of the object 1, and registers the result in the DB 421 of the storage unit 42 (step S6).

[0062] In addition, the DB generation process is performed three times by preparing at least three objects 1 to be measured that have the same combination of material, size, type of internal back surface structure (internal hollow areas, cut areas on the back surface, etc.), position of the internal back surface structure, and size of the internal back surface structure, and changing the type of load applied each time.

[0063] <Regarding Internal Backside Structure Detection Processing by Image Processing Device 40> FIG. 26 is a flowchart illustrating an example of internal backside structure detection processing by the image processing device 40. In FIG.

[0064] As a prerequisite for the internal back surface structure detection process, it is assumed that the DB generation process has already been executed and the DB 421 has been saved in the storage unit 42 of the image processing device 40 .

[0065] The internal back surface structure detection process is started, for example, in response to a predetermined start operation from the user. First, for example, an operator applies an identification pattern P to the surface of the object 1 under test, including the area immediately adjacent to the position where the internal back surface structure is expected to occur (step S11).

[0066] Next, the image acquisition unit 411 controls the imaging device 30 to capture an image of the surface of the object 1 to be measured on which the identification pattern P is applied, and acquires the resulting reference image (step S12).

[0067] Next, the load control unit 412 controls the load application device 20 to apply either a tensile load, a compressive load, or a moment load to the object 1 (step S13). The type of load to be applied here is selected to be closest to the load that will be applied to the object 1 when the object 1 is actually used as a component of a railway vehicle, construction machine, etc.

[0068] Next, the image acquisition unit 411 controls the imaging device 30 to capture an image of the surface of the object to be measured 1 on which the identification pattern P is applied while a load is applied, and acquires the resulting strain image (step S14).

[0069] Next, the strain distribution calculation unit 413 calculates the strain distribution on the surface of the object 1 by DIC based on the reference image and the strain image (step S15).

[0070] Next, the internal back surface structure detection unit 415 refers to the DB 421 of the storage unit 42, and detects the internal back surface structure generated inside or on the back surface of the object 1 based on the strain distribution on the surface of the object 1 (step S16). This concludes the description of the internal back surface structure detection process.

[0071] 27 shows an example of an internal back surface structure that can be detected by the internal back surface structure detection process, in which a hollow portion T1 is detected inside a measured object 1 in which two metals are joined by a weld joint. In this case, the hollow portion T1 corresponds to an unwelded portion or a blowhole. Furthermore, because a weld bead 11 is formed on the front and back surfaces of the measured object 1, the thickness from the tip of the hollow portion T1 to the surface increases. However, even in such a case, the strain distribution on the surface of the measured object 1 remains the same as when a weld bead 11 is not formed, so it is possible to detect unwelded portions and blowholes caused by the weld joint.

[0072] 28 shows an example of an internal back surface structure that can be detected by the internal back surface structure detection process, in which a cut T2 is detected on the back surface of a workpiece 1 in which two metals are joined by a weld joint. In this case, the cut T2 corresponds to an unwelded area. Furthermore, because a weld bead 11 is formed on the surface of the workpiece 1, the thickness from the tip of the cut T2 to the surface increases. However, even in this case, the strain distribution on the surface of the workpiece 1 remains the same as when the weld bead 11 is not formed, so the unwelded area caused by the weld joint can be detected.

[0073] As described above, according to this embodiment, it is possible to detect the internal back surface structure that cannot be seen from the surface of the object 1 to be measured.

[0074] In the above-described embodiment, the types of loads are three, namely, tensile load, compressive load, and moment load, but the types of loads may be increased.

[0075] The present invention is not limited to the above-described embodiments, and various modifications are possible. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with or add to the configuration of another embodiment.

[0076] For example, the present invention can be applied to a manufacturing method for manufacturing a bonded structure.

[0077] 1...Object to be measured, 10...Internal back surface structure detection system, 11...Weld bead, 20...Load application device, 21...Tensile load, 22...Compressive load, 23...Moment load, 30...Imaging device, 31...Camera, 32...Camera, 40...Image processing device, 41...Processing unit, 42...Memory unit, 411...Image acquisition unit, 412...Load control unit, 413...Distribution calculation unit, 414...DB generation unit, 415...Internal back surface structure detection unit, L...Imaging distance, T...Internal back surface structure, T1...Hollow portion, T2...Notched portion, P...Identification pattern

Claims

1. An internal back surface structure detection method for detecting an internal back surface structure present in an object to be measured, comprising: an application step of applying an identification pattern to the surface of the object to be measured; a first acquisition step of acquiring a reference image of the surface of the object to be measured on which the identification pattern has been applied; an application step of applying a load to the object to generate strain in the object to be measured; a second acquisition step of acquiring a strain image of the surface of the object to be measured on which the identification pattern has been applied, in which the strain has been generated by the application of the load; a calculation step of calculating the strain distribution generated on the surface of the object to be measured based on the reference image and the strain image; and a detection step of detecting the internal back surface structure present in the object to be measured by referring to a database based on the strain distribution.

2. An internal back surface structure detection method as set forth in claim 1, wherein the detection step detects, as the internal back surface structure, at least one of a hollow portion present inside the object to be measured and a notched portion present on the back surface.

3. An internal back surface structure detection method according to claim 2, wherein the detection step detects, as the internal back surface structure, at least one of an unwelded portion of the object to be measured in which two metals are joined by a weld joint, and a blowhole.

4. An internal back surface structure detection method according to claim 1, wherein the applying step applies a random pattern as the identification pattern to the surface of the object to be measured.

5. An internal back surface structure detection method according to claim 1, wherein the application step applies the identification pattern to the surface immediately adjacent to a position where the internal back surface structure is assumed to exist.

6. The internal back surface structure detection method according to claim 1, wherein the applying step applies a tensile load, a compressive load, or a moment load to the object to be measured.

7. An internal back surface structure detection method as set forth in claim 1, comprising a registration step of registering in the database the type of load and the strain distribution calculated from the object to be measured, in association with each position of the internal back surface structure of the object to be measured, the presence of which is known in advance.

8. An internal back surface structure detection method as set forth in claim 7, wherein the registration step associates the type of load and the strain distribution calculated from the object to be measured with each combination of the position of the internal back surface structure and at least one of the material of the object to be measured, the size of the object to be measured, the type of internal back surface structure, and the size of the internal back surface structure, and registers them in the database.

9. A method for detecting an internal back surface structure as described in claim 7, wherein the strain distribution of a measured object in which the presence of a hollow portion inside as the internal back surface structure is known has strain values ​​that are either positive or negative throughout, and the absolute values ​​of strain on the surfaces immediately adjacent to both sides of the portion in which the hollow portion exists are maximum values, and the absolute value of strain on the surfaces immediately adjacent to the hollow portion is minimum values.

10. A method for detecting an internal back surface structure as described in claim 7, wherein the strain distribution of an object to be measured, which is known to have a notch on the back surface as the internal back surface structure, is such that the strain value changes between positive and negative positive, or negative and positive and negative, with the notch as the center.

11. An internal back surface structure detection system comprising a load application device, an imaging device, and an image processing device, wherein the load application device applies a load to an object to be measured to generate strain in the object to be measured, the imaging device images a surface of the object to be measured on which an identification pattern is applied to generate a reference image, and also images the surface of the object to be measured on which the identification pattern is applied and on which the strain has occurred due to the application of the load to generate a strain image, and the image processing device comprises: an image acquisition unit that acquires the reference image and the strain image, a strain distribution calculation unit that calculates the strain distribution generated on the surface of the object to be measured based on the reference image and the strain image, and an internal back surface structure detection unit that detects the internal back surface structure present in the object to be measured by referring to a database based on the strain distribution.

12. A method for manufacturing a bonded structure, comprising: a coating step of coating an identification pattern on a surface of a measured object that is the bonded structure; a first acquisition step of acquiring a reference image of the surface of the measured object on which the identification pattern has been coated; an application step of applying a load to the measured object to generate strain in the measured object; a second acquisition step of acquiring a strain image of the surface of the measured object on which the identification pattern has been coated, where the strain has been generated by the application of the load; a calculation step of calculating a strain distribution generated on the surface of the measured object based on the reference image and the strain image; and a detection step of detecting an internal back surface structure present in the measured object by referring to a database based on the strain distribution.

Images