JP7525266B2 - Inspection system and inspection method - Google Patents

Description

The present invention relates to an inspection system and an inspection method .

The technology described in Patent Document 1 is known as an inspection technology for inspection objects in structures. Patent Document 1 describes a condition measurement device that includes an input means for inputting time-series image data obtained by photographing a predetermined pattern attached to a columnar structure, a displacement calculation means for determining the displacement occurring in the columnar structure from the time-series image data, a natural frequency calculation means for determining the natural frequency of the columnar structure from the displacement, and a determination means for determining the condition of the columnar structure based on the natural frequency.

JP 2018-141663 A (especially claim 1)

In the technology described in Patent Document 1, a marker for measuring the displacement of a utility pole is attached to the utility pole as a structure. The marker formed on the utility pole is then imaged by an imaging device. However, images including markers usually have a similar composition, so it is not possible to identify the inspection target based on the image, such as what or where the inspection target is.

The problem to be solved by the present invention is to provide an inspection system and an inspection method capable of identifying an inspection object.

The inspection system of the present invention includes an imaging device that captures an image of an identification pattern formed on an object to be inspected, the identification pattern including individual identification information, and a processing device, the processing device including an identification unit that identifies the imaged object to be inspected based on the identification pattern in the image captured by the imaging device, and an image analysis unit that performs image analysis of the identification pattern in the image based on a digital image correlation method, the identification pattern further including characters. Other solutions will be described later in the description of the embodiment of the invention.

According to the present invention, it is possible to provide an inspection system and an inspection method capable of identifying an inspection object.

FIG. 1 is a schematic diagram showing an inspection system according to a first embodiment. FIG. 2 is an enlarged view of a portion A in FIG. 1, and is a schematic diagram showing an inspection target on the surface of a structure. FIG. 13 is a diagram showing an example of an identification pattern. FIG. 13 is a diagram showing an example of an identification pattern. FIG. 13 is a diagram showing an example of an identification pattern. FIG. 1 is a block diagram of an inspection system including a processor. 1 is a schematic diagram of an identification pattern formed on an inspection target before deformation. FIG. 13 is a schematic diagram of an identification pattern formed on the inspection target after deformation. FIG. FIG. 1 is a schematic diagram showing an image including an identification pattern. FIG. 13 is a schematic diagram showing a displacement distribution obtained by image analysis based on a digital image correlation method. 4 is a flowchart showing an inspection method according to the first embodiment. FIG. 13 is a schematic diagram showing an inspection system according to a second embodiment. FIG. 10 is an enlarged view of the window in FIG. 9 . FIG. 13 is a schematic diagram showing an inspection system according to a third embodiment.

The following describes a form for implementing the present invention (the present embodiment). However, the present invention is not limited to the following content and the content shown in the drawings, and can be modified as desired without significantly impairing the effects of the present invention. The present invention can be implemented by combining different embodiments. In the following description, the same components in different embodiments are given the same reference numerals, and duplicate explanations will be omitted.

Figure 1 is a schematic diagram showing an inspection system 100 of a first embodiment. The inspection system 100 uses an imaging device 3 (e.g., a camera described below) to inspect an inspection object 30 in a structure 1. Specifically, the inspection system 100 captures an image of an identification pattern 31 (Figures 2 and 3) formed on the inspection object 30. The inspection system 100 applies digital image correlation (hereinafter referred to as DIC) to measure the amount and direction of displacement (either one may be sufficient; the same applies below) of the identification pattern 31 associated with deformation of the structure 1. However, the inspection content is not limited to this example.

The structure 1 inspected by the inspection system 100 is a structure 1 inspected by DIC, and has an inspection target 30 with an identification pattern 31 (FIG. 2) including individual identification information. The inspection target 30 is, for example, the surface 10 of the structure 1, and includes three unit parts 30a in the example of FIG. 1. In the illustrated example, the structure 1 is a power generation facility, specifically a wind power generation facility. However, the structure 1 is not limited to a power generation facility, and may be a railroad vehicle or construction machine. That is, the inspection target 30 is, for example, the surface 10 of at least one of the structure 1, which is a power generation facility, a railroad vehicle, or a construction machine. As a result, the surface 10 of at least one of the structure 1, which is a power generation facility, a railroad vehicle, or a construction machine, can be inspected by DIC. However, the structure 1 is not limited to these.

For example, the structure 1, which is composed of a wind power generation facility, includes blades 20, a tower 21, a rotating shaft 22, and a nacelle 23. The blades 20 are attached to the rotating shaft 22, which is connected to a power generation device (not shown) in the nacelle 23. The power generation device (not shown) is driven by the rotation of the blades 20, as indicated by the solid arrow, to generate electricity. The nacelle 23 is supported by the tower 21, and the imaging device 3 is installed on the top surface of the nacelle 23. Note that the installation location of the imaging device 3 is not limited to the top surface of the nacelle 23, and it can be installed at any height and location, such as on the ground or the tower 21. In addition, for example, the imaging device 3 may be mounted on an unmanned aerial vehicle (not shown), and images may be captured by the imaging device 3 while the unmanned aerial vehicle is flying in synchronization with the rotation of the blades 20.

The inspection system 100 includes an imaging device 3 that images an inspection target 30 on the surface 10 of a structure 1. The imaging device 3 is a single monocular camera that captures two-dimensional images, and in the illustrated example, is installed on the top surface of a nacelle 23. The imaging device 3 captures an image of a portion where stress concentration is likely to occur, specifically, the end of the surface 10 of a blade 20 at part A in FIG. 1. The image obtained by imaging is acquired by an image acquisition unit 51 (described below) of a computing device 50.

Other examples of parts where stress concentration is likely to occur include joints that may exist in the structure 1, curved surfaces where the shape or dimensions change suddenly, cutouts, etc., and are not limited to the image capturing locations shown in Fig. 1. For example, the image capturing device 3 may capture an image of a bolt or weld that secures the tower 21 to a foundation (not shown) on the ground.
The inspection object 30 imaged by the imaging device 3 will be described with reference to FIG.

Figure 2 is an enlarged view of part A in Figure 1, and is a schematic diagram showing an inspection target 30 on the surface 10 of the structure 1. In Figure 2, to show that the structure 1 extends beyond the area shown in Figure 2, the surface 10 on the front side and the back side of the page is cut by a curved line to show a cross-sectional perspective view.

The imaging device 3 (Figure 1) images an identification pattern 31 formed on an inspection object 30. The inspection object 30 is, for example, a portion of the structure 1 where stress is likely to concentrate. The identification pattern 31 includes individual identification information that can identify the inspection object 30. The individual identification information is information that can identify at least one of the structure 1 or the inspection object 30, such as the specific position of the inspection object 30 in the structure 1, the manufacturing time of the structure 1, the manufacturer of the structure 1, the manufacturing location of the structure 1, the time when the identification pattern 31 was formed on the inspection object 30, etc.

The identification pattern 31 is formed in a color corresponding to the color of the inspection target 30 of the structure 1. For example, when the color of the inspection target 30 is matte white, the identification pattern 31 is preferably black. Also, when the color of the inspection target 30 is black, the identification pattern 31 is preferably matte black. In the case of the structure 1 exemplified as a wind power generation facility, the color of the inspection target 30 (i.e., the color of the surface 10 of the structure 1) is usually white.
A specific example of the identification pattern 31 will be described with reference to FIGS. 3A to 3C.

Figure 3A is a diagram showing an example of an identification pattern 31. The identification pattern 31 includes a point portion 32 consisting of a collection of multiple unit points 32a spreading two-dimensionally, and an information portion 33 associated with individual identification information. By including the point portion 32 and the information portion 33, DIC can be applied to the point portion 32, and the amount and direction of displacement of the unit points 32a can be measured. Furthermore, the inspection object 30 can be identified based on the information portion 33.

In the example of FIG. 3A, many small unit points 32a are arranged in a scattered manner on a two-dimensional plane. DIC measures the amount and direction of displacement of the unit points 32a caused by the deformation (distortion; same below) of the structure 1, i.e., the deformation of the inspection object 30. It is preferable that the unit points 32a are formed in a random spread without any regularity. In this way, the analysis accuracy by DIC can be improved.

In the example of FIG. 3A, the information section 33 is composed of three dots 33a arranged on the diagonal of the square identification pattern 31. The dots 33a are larger than the unit dots 32a, and the dots 33a are distinguished from the unit dots 32a. The information section 33 is associated with individual identification information. For example, in the example of FIG. 3A, by changing the location and number of the dots 33a (or only one of them may be used), the individual identification information of the test object 30 on which the identification pattern 31 including the information section 33 is formed can be determined.

Figure 3B is a diagram showing an example of the identification pattern 31. The example shown in Figure 3B is, for example, a two-dimensional digital matrix code (two-dimensional barcode) such as a QR code (registered trademark). The identification pattern 31 shown in Figure 3B also includes a dot portion 32 and an information portion 33. Specifically, the dot portion 32 is formed by arranging square unit points 32a in the XY direction on a two-dimensional plane. Therefore, in the DIC, the amount and direction of displacement of the square unit points 32a caused by the deformation of the structure 1, i.e., the deformation of the inspection target 30, are measured. In addition, by forming the identification pattern 31 according to the standard of the two-dimensional digital matrix code, the entire identification pattern 31 functions as the information portion 33. If the two-dimensional barcode is, for example, a QR code, the standard of the two-dimensional digital matrix code that constitutes the information portion 33 is ISO/IEC18004.

Figure 3C is a diagram showing an example of the identification pattern 31. In the example shown in Figure 3C, the identification pattern 31 includes characters 34. The characters 34 are arranged, for example, in the same direction at equal intervals and of equal size. By including the characters 34, the inspection object 30 can be directly identified by the characters 34. Therefore, the characters 34 function as identifiable information. Furthermore, by considering the character 34 as a single point and evaluating the distortion of the character 34 based on the DIC, the amount and direction of displacement of the point can be measured. Therefore, the character 34 also functions as an analysis object by the DIC.

Returning to FIG. 1, in the first embodiment, the inspection object 30 includes a plurality of unit parts 30a, and the imaging device 3 images each of the unit parts 30a. Thus, in the example of FIG. 1, one imaging device 3 images three unit parts 30a included in the structure 1. As described above, the unit parts 30a, which are the inspection object 30, are provided with individual identification information in the form of an identification pattern 31 (FIG. 2). Therefore, even if there are a plurality of inspection objects 30, it is possible to identify which inspection object 30 the captured image corresponds to. Furthermore, by being able to prevent inability to identify, it is possible to prevent the need to re-image (retake) the inspection object 30 after identifying it again.

In the example of FIG. 1, the inspection object 30 includes multiple unit parts 30a, but the structure 1 may include only one inspection object 30. That is, one imaging device 3 may image one inspection object 30 in one structure 1. As described above, the images obtained by the imaging device 3 often have similar structures. Therefore, even when image data is stored, the imaged inspection object 30 can be identified based on the identification pattern 31 in the image.

The imaging device 3 continuously images the multiple unit parts 30a. That is, a single fixed imaging device 3 continuously images the three unit parts 30a of the rotating blade 20. Therefore, when the blade 20 rotates once in the direction of the solid arrow, three images corresponding to the unit parts 30a, which are different inspection objects 30, are obtained. By continuously imaging the unit parts 30a, even if the number of images is large, the inspection object 30 can be identified based on the identification pattern 31 in the image.

The inspection system 100 includes a calculation processing device 50 to which an image captured by the imaging device 3 is transmitted, a display device 60, and an alarm device 61. First, the calculation processing device 50 will be described with reference to FIG. 4.

Figure 4 is a block diagram of an inspection system 100 including a processor 50. The processor 50 includes an image acquisition unit 51, a recognition unit 52, an image analysis unit 53, a display unit 54, and a notification unit 55.

The image acquisition unit 51 acquires the image obtained by the imaging device 3. The acquired image includes the identification pattern 31. The image acquired by the image acquisition unit 51 is transmitted to the identification unit 52. The image acquisition (i.e., the timing of imaging by the imaging device 3) is performed in synchronization with the rotation timing of the blade 20 so that the same unit portion 30a of the blade 20 can be imaged each time it rotates.

The identification unit 52 identifies the imaged inspection target 30 based on the identification pattern 31 in the image captured by the imaging device 3. In the first embodiment, it is identified which of the three blades 20 has been imaged based on the identification pattern 31 in the image. Since the images obtained by imaging have a similar composition, the identification unit 52 can identify the blade 20 that is the inspection target 30.

The image analysis unit 53 performs image analysis based on DIC on the identification pattern 31 in the image obtained by the imaging device 3. Using the digital images of the inspection object 30 before and after deformation obtained by DIC, the amount and direction of displacement of the surface of the inspection object 30 can be simultaneously determined based on the brightness distribution of the identification pattern 31. In other words, the image analysis unit 53 can determine the amount and direction of displacement of the surface of the structure 1 caused by the deformation of the inspection object 30. The determination of the amount and direction of displacement based on DIC will be described with reference to Figures 5A and 5B.

Figure 5A is a schematic diagram of the identification pattern 31 formed on the inspection object 30 before deformation. Also, Figure 5B is a schematic diagram of the identification pattern 31 formed on the inspection object 30 after deformation. Figures 5A and 5B illustrate the identification pattern 31 shown in Figure 3B above as an example. Also, in Figures 5A and 5B, points R and S are virtual, and in reality, the amount and direction of displacement of each unit point 32a (Figure 3B) are determined.

In the examples of Figures 5A and 5B, the luminance distribution is found within a tiny image region (subset) R centered on an arbitrary point Q in the identification pattern 31 (Figure 5A) in the image before deformation. Then, in the identification pattern 31 (Figure 5B) in the image after deformation, a region S having a luminance distribution that has the best correlation with the luminance distribution of the region R before deformation is searched for. By setting the position of the center point U of the region S to the position of point Q displaced by the deformation of the structure 1, the amount and direction of displacement of point Q can be determined simultaneously.

In this way, the image analysis unit 53 analyzes the amount and direction of displacement of an arbitrary point Q on the inspection object 30 caused by the deformation of the inspection object 30. This makes it possible to grasp the degree of deformation of the inspection object 30 and to grasp the strain distribution diagram of the inspection object 30.

Figure 6 is a schematic diagram showing an image including an identification pattern 31. Stress (strain) concentration is likely to occur when there is a geometric discontinuity, a notch, etc. For this reason, the imaging device 3 images the end of the blade 20 where stress is likely to concentrate.

Figure 7 is a schematic diagram showing the displacement distribution obtained by image analysis based on DIC. The image analysis unit 53 analyzes an image including the discrimination pattern 31 shown in Figure 6 to obtain a visualized displacement distribution map (Figure 7) including a displacement contour map 22 and a displacement vector map (not shown). The obtained displacement distribution map is transmitted to the display unit 54 and the notification unit 55 (both of which will be described later). The displacement distribution map may be recorded in a recording unit (not shown). The displacement distribution map may also be transmitted to another device, for example, via an external transmission unit (not shown) and a communication cable (which may be wireless).

Returning to FIG. 4, the display unit 54 displays the results of the image analysis performed by the image analysis unit 53 on the display device 60. Specifically, in the example of FIG. 4, the display unit 54 displays the displacement distribution map shown in FIG. 7 on the display device 60. The display device 60 is, for example, a monitor. Providing the display unit 54 allows the user to check the displacement distribution map and perform appropriate measures such as maintenance and inspection.

The notification unit 55 issues an alarm to the user when the result of the image analysis by the image analysis unit 53 deviates from a predetermined standard. The standard here is, for example, a physical quantity related to the deformation (e.g., the amount of displacement, etc.), and specifically, for example, an allowable amount of displacement indicating that the deformation of the inspection object 30 does not significantly affect the service life of the structure 1 including the inspection object 30. The allowable amount of displacement can be determined, for example, by an experiment, a trial run, a simulation, etc. The allowable amount of displacement may be a value input by an operator via an input device (e.g., a keyboard) not shown.

For example, if the displacement amount determined based on the displacement distribution map is within a displacement amount range that has little effect on the service life, the notification unit 55 will not issue an alarm. However, if the displacement amount determined based on the displacement distribution map falls outside of the displacement amount range that has little effect on the service life, that is, if the deformation of the inspection object 30 affects the service life of the structure 1 including the inspection object 30, the notification unit 55 will issue an alarm. This allows the user to understand unacceptable displacement and can urge the user to, for example, perform an emergency inspection.

The alarm is issued by driving the alarm device 61 (e.g., by flashing a red lamp, etc.). The alarm device 61 may also be used as the display device 60, so that the alarm is displayed on the display device 60. The alarm may also be set to notify, for example, of any part that is outside the displacement range (i.e., the part that needs repair).

The arithmetic processing device 50 is configured with, for example, a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), HDD (Hard Disk Drive), I/F (Interface), etc., all of which are not shown in the figure. The arithmetic processing device 50 is realized by the CPU executing a predetermined control program stored in the ROM. The arithmetic processing device 50 and the imaging device 3 are electrically connected by a wired cable or wirelessly. The arithmetic processing device 50 and the imaging device 3 are connected to power cables as appropriate.

Figure 8 is a flowchart showing the inspection method of the first embodiment. The inspection method shown in Figure 8 can be mainly executed by the processor 50 shown in Figure 4. Therefore, in the following explanation, Figure 8 will be explained with reference to Figure 4 as well.

The inspection method of the first embodiment includes a formation process S1, an imaging process S2, an image acquisition process S3, an identification process S4, an image analysis process S5, a display process S6, and a notification process S7.

The formation process S1 is a process of forming an identification pattern 31 including individual identification information on an inspection target 30, which is a portion of the surface of the structure 1 that is to be inspected. The inspection target 30 and the identification pattern 31 are the same as the inspection target 30 and the identification pattern 31 described in the description of the structure 1.

The formation of the identification pattern 31 on the inspection object 30 can be determined arbitrarily, for example, depending on the shape of the identification pattern 31. In addition, the materials and methods for constructing the identification pattern 31 can be selected and combined in a variety of ways depending on the purpose.

For example, when the identification pattern 31 is a coating, the identification pattern 31 can be formed by applying a solution containing a solvent and the constituent material of the identification pattern 31 to the inspection object 30 and drying it. The application and drying can be performed by any application device and drying device. For example, the identification pattern 31 can be formed by spray application of the constituent material of the identification pattern 31. When forming by spray application, it is preferable to use a constituent material with excellent durability such as environmental load resistance. In addition, the identification pattern 31 can be formed by forming it into any shape, such as a coating material, a thin sheet, or a sticker, and fixing it to the inspection object 30 so that it does not peel off. The application can be performed by any application device.

Another method for forming the identification pattern 31 on the inspection object 30 is to use a laser marking device to place the identification pattern 31 on the surface layer of the inspection object 30. The identification pattern 31 can be formed by heating the very surface of the inspection object 30 with a laser. As a result, the identification pattern 31 is formed on a part of the inspection object 30, and the durability (environmental load resistance, etc.) of the identification pattern 31 can be improved. In this case, a paint film surface layer is formed by painting the surface of the inspection object 30, and a more suitable contrast pattern can be formed by laser marking the paint film surface layer.

In particular, the durability of the identification pattern 31 can be improved by laser marking. This allows the structure 1 to be inspected intermittently over a long period of time without stopping the operation of the structure 1. As a result, economic losses associated with the suspension of the operation of the structure 1 can be suppressed.

The imaging process S2 is a process of imaging the identification pattern 31 formed on the inspection object 30, which includes individual identification information, by the imaging device 3. As described with reference to FIG. 1, the imaging process S2 is performed continuously for multiple unit portions 30a (FIG. 1; it may be a single unit portion) while the blade 20 is rotating.

The image acquisition process S3 is a process of acquiring an image captured by the imaging device 3. The image acquisition process S3 is performed by the image acquisition section 51. The identification process S4 is a process of identifying the imaged inspection object 30 based on the identification pattern 31 in the image captured by the imaging device 3. The identification process S4 is performed by the identification section 52. The identification process S4 makes it possible to identify which inspection object 30 the captured image represents.

The specific method of identification can be determined according to the type of identification pattern 31. For example, if the identification pattern 31 is the identification pattern 31 shown in FIG. 3A, it can be identified by the location and number of dots 33a that make up the information section 33. Also, if the identification pattern 31 is the identification pattern 31 configured by the QR code shown in FIG. 3B, it can be identified by analyzing the information section 33 configured in accordance with ISO/IEC 18004. Furthermore, if the identification pattern 31 is the identification pattern 31 shown in FIG. 3C, it can be identified by deciphering the characters 34.

The image analysis process S5 is a process of performing image analysis based on DIC on the identification pattern 31 in the image captured by the imaging device 3. The image analysis process S5 is executed by the image analysis unit 53. The image analysis process S5 makes it possible to simultaneously determine the amount and direction of displacement of the inspection object 30 caused by the deformation of the structure 1. As specific methods of DIC, for example, the methods shown in Figures 5A to 7 can be applied.

The display step S6 is a step of displaying the result of the image analysis by the image analysis unit 53 on the display device 60. The display step S6 is executed by the display unit 54. The notification step S7 is a step of notifying the user of an alarm when the result of the image analysis by the image analysis unit 53 deviates from a predetermined standard. The standard here is the same as the content explained in the explanation of the notification unit 55 above. The notification step S7 is executed by the notification unit 55. The notification is executed by using the notification device 61.

According to the above inspection method and inspection system 100, the inspection object 30 on which the identification pattern 31 is formed can be identified based on the identification pattern 31 in the image. As a result, even if images have similar compositions, the inspection object 30 can be identified, for example, to determine what the inspection object 30 is and where the inspection object 30 is located. After identifying the inspection object 30, image analysis using DIC can be performed on the identification pattern 31. As a result, for example, the displacement distribution (Figure 7) in the identified inspection object 30 can be obtained, and the strain distribution and displacement distribution occurring in the structure 1 can be quantitatively visualized.

In particular, in the above example, the inspection object 30 is formed in a portion where stress is likely to concentrate. Usually, fatigue cracks occur in portions where stress is likely to concentrate, and eventually the structure 1 is destroyed. Therefore, it is preferable to predict the life until fatigue cracks occur before the structure 1 is destroyed, or to detect and repair fatigue cracks before the structure 1 is destroyed. For this reason, by quantitatively visualizing the strain distribution and displacement distribution based on DIC for portions where stress is likely to concentrate, it becomes possible to predict the life until fatigue cracks occur and to repair at the appropriate time.

In addition to inspecting the inspection object 30 in a newly constructed structure 1, inspection can also be easily performed on an existing structure 1 that does not have an identification pattern 31. In other words, inspection can be easily performed by the inspection system 100 in an existing structure 1 by simply attaching a new identification pattern 31 to the inspection object 30.

Furthermore, by inspecting the structure 1, the tendency and frequency of occurrence of displacement and distortion can be recorded and analyzed. In particular, as described above, inspection by the inspection system 100 can be performed, for example, while the operation of the structure 1 is continued. This allows a large amount of inspection data to be acquired, and the accuracy of the analysis of the tendency and frequency of occurrence of displacement and distortion can be improved. As a result, maintenance plans can be optimized. Furthermore, by feeding back the analysis results to the designer of the structure 1, reinforcement can be implemented when designing the next model, thereby improving the reliability of the structure 1.

In the above example, the inspection target 30 in the structure 1 moves, but the inspection target 30 does not have to move. Specifically, the inspection target 30 may be provided on a fixed structure 1 such as a bridge, a road overpass, or the inner wall of a tunnel.

Figure 9 is a schematic diagram showing an inspection system 200 of the second embodiment. The structure 1 shown in Figure 9 can be inspected by the inspection system 200. The inspection system 200 is the same as the inspection system 100, except that instead of the single imaging device 3 in the inspection system 100, the inspection system 200 has two imaging devices 3a and 3b. A three-dimensional image is captured by the two imaging devices 3a and 3b. In addition, while a power generation facility is exemplified as the structure 1 in Figure 1 above, a railway vehicle is exemplified as the structure 1 in Figure 9.

The railway vehicle that constitutes the structure 1 runs on rails 41 via bogies 42 and has entrances/exits 43 and windows 44. As with the power generation equipment described above, in railway vehicles, stress is likely to concentrate, for example, at corners and at parts where the shape changes. Therefore, in the illustrated example, the imaging device 3 images the inspection object 30, which is near the window frame of the window 44 where stress is likely to concentrate (part A in FIG. 9), and the inspection is performed by the inspection system 200.

The imaging device 3 is installed on the ground. Therefore, when the structure 1 formed by the railway vehicle passes the location where the imaging device 3 is installed, the imaging device 3 captures an image of the vicinity of the window frame of the window 44. This makes it possible to simultaneously measure the amount and direction of displacement of the inspection object 30, which is the vicinity of the window frame of the window 44.

In the example shown in FIG. 9, as in the example shown in FIG. 1 above, the inspection target 30 of the structure 1 includes multiple unit parts 30a. Two imaging devices 3a, 3b capture images of each unit part 30a. Specifically, while the railway vehicle is running, the imaging devices 3a, 3b fixed to the ground continuously capture images of multiple unit parts 30a at the same height. However, even in the railway vehicle as the structure 1, there may be only one inspection target 30.

The imaging device 3 provided in the inspection system 200 includes two imaging devices 3a and 3b as described above. The image obtained by the imaging device 3 is a three-dimensional image. By obtaining a three-dimensional image, a visualized displacement distribution map of the three-dimensional surface including the in-plane direction as well as the depth direction can be obtained. Therefore, it is possible to measure the amount of displacement not only in the in-plane direction but also in the depth direction. Each of the imaging devices 3a and 3b is a monocular camera, but a single stereo camera may also be used.

Figure 10 is an enlarged view of the window 44 in Figure 9. Figure 10 illustrates the identification pattern 31 shown in Figure 3B above as an example. Part A in Figure 10 corresponds to each of the seven parts A in Figure 9. The corners 44a of the window 44 suddenly change shape discontinuously. For this reason, distortion is likely to occur at the corners 44a of the window 44. Therefore, in the example of Figure 10, an inspection target 30 is formed at each of the two lower corners 44a of the window 44. The imaging device 3 images the two corners 44a so as to include two identification patterns 31, as shown in part A.

In the example of FIG. 10, the identification pattern 31 formed on each inspection object 30 includes at least one complete identification pattern 31a (one for each corner 44a in FIG. 10) that is completely displayed to the extent that the individual identification information can be recognized. By including the complete identification pattern 31a in the identification pattern 31, the inspection object 30 can be identified by the individual identification information included in the complete identification pattern 31a. Furthermore, the inspection object 30 can be inspected based on the points that make up the complete identification pattern 31a (which may be, for example, characters 34 (FIG. 3C) that can be considered as points). It is particularly preferable that the complete identification pattern 31a does not lack any part that corresponds to the individual identification information.

In the example of FIG. 10, the identification pattern 31 formed on each inspection target 30 further includes an incomplete identification pattern 31b in which the individual identification information cannot be identified because, for example, a part of the complete identification pattern is missing. The incomplete identification pattern 31b allows inspection, for example, by DIC.

For example, when applying an identification pattern 31 to a corner 44a, the shape of the corner 44a changes as described above, and depending on the shape of the identification pattern 31, it may be difficult to apply the identification pattern 31 so as to cover the entire corner 44a. Therefore, for example, it is preferable to apply the identification pattern 31 multiple times while overlapping the identification pattern 31 so as to cover the entire corner 44a. In this case, another identification pattern 31 is placed so as to overlap the identification pattern 31 that has already been placed. In this way, the identification pattern 31 can be placed so as to cover the entire corner 44a, and the entire corner 44a can be inspected as the inspection target 30.

The complete identification pattern 31a is preferably placed at least in one position on the inspection object 30 that is included in the imaging range of the imaging device 3 (part A in the example of FIG. 10). By placing the complete identification pattern 31a in such a position, the inspection object 30 can be identified. The complete identification pattern 31a can be placed, for example, by placing the identification pattern 31 so as to cover the entire corner 44a, and then placing the complete identification pattern 31a from above the placed identification pattern 31.

The identification patterns 31 for image analysis may be arranged in an orderly manner in the vertical and horizontal directions, or may be arranged randomly regardless of direction. To stabilize the measurement accuracy, it is preferable that the dimensions of the identification patterns 31 are approximately the same (or may be the same) for all of the identification patterns 31.

In addition to the corners 44a, other areas where stress is likely to concentrate include, for example, notches, holes, and holes for fastening or joining. Therefore, these areas may be subject to inspection 30.

Figure 11 is a schematic diagram showing an inspection system 300 of the third embodiment. In the illustrated example, the structure 1 inspected by the inspection system 300 is a construction machine, more specifically, a hydraulic excavator. The structure 1 shown in Figure 11 can be inspected by an inspection system 300 similar to the inspection system 100 described above, for example.

In the structure 1 formed by a hydraulic excavator, an upper rotating body 82 is mounted on a lower traveling body 81. The upper rotating body 82 includes a boom 83, an arm 84, a bucket 85, a boom cylinder 86, an arm cylinder 87, and a bucket cylinder 88. The imaging device 3 is installed on the side of the boom 83 and is capable of imaging the corners of the side of the boom 83. This allows the displacement and displacement direction at the corners of the side of the boom 83 to be measured simultaneously. However, the installation location and imaging portion of the imaging device 3 are not limited to this, and it may be installed on the upper rotating body 82, for example, to image the upper rotating body 82.

In this specification, the inspection systems 100, 200 according to the present invention have been described using wind power generation equipment, railway vehicles, and hydraulic excavators as examples, but the type and shape of the structure 1 to be measured are not limited.

REFERENCE SIGNS LIST 1 Structure 100, 200, 300 Inspection system 20 Blade 21 Tower 22 Rotating shaft 23 Nacelle 3 Imaging device 30 Inspection target 30a Unit portion 31 Discrimination pattern 31a Complete discrimination pattern 31b Incomplete discrimination pattern 32 Dot portion 32a Unit dot 33 Information portion 33a Dot 34 Character 3a. 3b Imaging device 41 Rail 42 Cart 43 Entrance/exit 44 Window 44a Corner 50 Processing device 51 Image acquisition section 52 Recognition section 53 Image analysis section 54 Display section 55 Notification section 60 Display device 61 Notification device 81 Lower travel structure 82 Upper rotating structure 83 Boom 84 Arm 85 Bucket 56 Boom cylinder 87 Arm cylinder 88 Bucket cylinder Q Points R, S Area S1 Formation process S2 Imaging process S3 Image acquisition process S4 Recognition process S5 Image analysis process S6 Display process S7 Notification process U Center point

Claims (10)

an imaging device for imaging an identification pattern formed on an object to be inspected, the identification pattern including individual identification information;
A processor,
The arithmetic processing device includes:
an identification unit that identifies the imaged inspection object based on the identification pattern in the image captured by the imaging device;
an image analysis unit that performs image analysis on the identification pattern in the image based on a digital image correlation method,
The identification pattern further includes characters. The inspection system according to claim 1 , wherein the image analysis unit analyzes at least one of a displacement amount and a displacement direction of an arbitrary point on the inspection object caused by a deformation of the inspection object. The inspection system according to claim 1 , wherein the identification patterns include at least one complete identification pattern in which the individual identification information is completely displayed to a degree that allows the individual identification information to be recognized. The test object includes a plurality of unit sites,
The inspection system according to claim 1 , wherein the imaging device images each of the unit parts. The inspection system according to claim 4 , wherein the imaging device successively images the plurality of unit sites. The inspection system of claim 1 , wherein the image obtained by the imaging device is a three-dimensional image. The inspection system according to claim 1 , wherein the arithmetic processing device further comprises a display unit that displays a result of the image analysis by the image analysis unit on a display device. The inspection system according to claim 1 , wherein the arithmetic processing device further comprises a notification unit that issues an alarm to a user when a result of the image analysis by the image analysis unit deviates from a predetermined standard. The inspection system according to claim 1 , wherein the inspection object is a surface of at least one of a power generation facility, a railway vehicle, and a construction machine. an imaging step of imaging an identification pattern formed on an object to be inspected, the identification pattern including computer-readable individual identification information and characters , which is formed based on a predetermined rule, by an imaging device;
an identification step of identifying the imaged inspection object based on the identification pattern in an image captured by an imaging device;
and an image analysis step of performing image analysis based on a digital image correlation method on the identification pattern in the image.

Images