Corrosion depth estimation device and corrosion depth estimation method

Abstract

To estimate corrosion depth simply and with high accuracy. [Solution] The corrosion depth estimation device comprises an extraction unit that extracts a corrosion area from an image of the surface of a metal structure, a diameter calculation unit that determines the diameter of the largest inscribed circle of the corrosion area, and a depth calculation unit that calculates the depth of the corrosion area based on the diameter of the largest inscribed circle.

Description

Technical Field

[0001] The present invention relates to a corrosion depth estimation device and a corrosion depth estimation method for estimating the depth of corrosion on a metal surface.

Background Art

[0002] Conventionally, in the evaluation of the depth of corrosion (corrosion depth) occurring in an aircraft structure, the corroded part was actually removed and its depth was measured. Therefore, a lot of man-hours were required for this actual measurement work.

[0003] In fields other than aircraft structures, there are technologies for detecting abnormal parts of metal materials using ultrasonic waves, ultraviolet rays, X-rays, photothermal, etc. However, these technologies are exclusively applied to those with standardized shapes, such as the measurement of the remaining wall thickness of an iron pipe and the detection of damage to an aluminum conductor steel-reinforced (ACSR) wire. Therefore, it is difficult to apply such technologies to the evaluation of corrosion occurring in aluminum alloys frequently used in aircraft or the evaluation of corrosion in various component shapes and complex locations.

[0004] In this regard, for example, in the technologies described in Patent Documents 1 and 2, based on the perimeter or area of the corroded part calculated from image data, the corrosion depth or the degree of its damage is estimated without actually measuring the corroded part.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] It would be useful if the corrosion depth could be estimated with higher accuracy. This invention has been made in view of the above circumstances, and aims to estimate corrosion depth simply and with high accuracy. [Means for solving the problem]

[0007] To achieve the above objective, one embodiment of the present invention is: An extraction unit that extracts corroded areas from images of the surface of metal structures, A diameter calculation unit for determining the diameter of the largest inscribed circle of the corrosion region, A depth calculation unit that calculates the depth of the corrosion region based on the diameter of the maximum inscribed circle, It is equipped with. [Effects of the Invention]

[0008] According to this invention, a corroded area is extracted from an image of the surface of a metal structure, and the diameter of its maximum inscribed circle is determined. Then, the depth of the corroded area is calculated based on the diameter of the maximum inscribed circle. This makes it possible to estimate the corrosion depth simply and with high accuracy. [Brief explanation of the drawing]

[0009] [Figure 1] This is a block diagram showing the schematic configuration of a corrosion depth estimation device according to an embodiment. [Figure 2] This flowchart shows the procedure for estimating corrosion depth according to the embodiment. [Figure 3] This is an example of image data of a corroded area. [Figure 4] This is an example of image data of a corroded area. [Figure 5] This is a diagram illustrating the corrosion depth estimation process according to the embodiment. [Figure 6] This graph shows an example of measured maximum inscribed circle diameter and corrosion depth in a corrosive region, and illustrates an example of depth conversion data according to the embodiment. [Figure 7] This graph shows an example of actual measurements of the perimeter and corrosion depth of a corroded area. [Figure 8] This graph shows an example of actual measurements of the area and depth of the corroded region. [Figure 9]It is a graph showing an example of depth conversion data according to a modification of the embodiment. [Figure 10] It is a diagram for explaining the aspect ratio of the acicular shape in the corrosion area. [Figure 11] It is a flowchart showing the procedure of the corrosion depth estimation process according to a modification of the embodiment. [Figure 12] It is a diagram for explaining how to obtain the aspect ratio of the acicular shape in the corrosion area. [Figure 13] It is a graph showing that there are multiple tendencies in the relationship between the maximum inscribed circle diameter and the corrosion depth. [Figure 14] It is a diagram illustrating a corrosion area with a small maximum inscribed circle diameter but deep corrosion. [Figure 15] It is a diagram illustrating a corrosion area with a large maximum inscribed circle diameter but shallow corrosion.

Mode for Carrying Out the Invention

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0011] [Configuration of Corrosion Depth Estimation Device] FIG. 1 is a block diagram showing a schematic configuration of a corrosion depth estimation device 1 according to the present embodiment. As shown in this figure, the corrosion depth estimation device 1 is a computer that evaluates the damage level due to corrosion of an aircraft structure. Specifically, the corrosion depth estimation device 1 includes an operation unit 12, a display unit 13, a storage unit 16, and a control unit 17.

[0012] The operation unit 12 is an operation means for a user to perform various operations for operating the corrosion depth estimation device 1, and includes, for example, a pointing device such as a mouse and a keyboard. The display unit 13 is composed of, for example, a liquid crystal display, an organic EL display, or other displays, and displays various information based on a display signal from the control unit 17. Note that the display unit 13 may be a touch panel that also serves as a part of the operation unit 12, or may perform voice output.

[0013] The memory unit 16 is a memory composed of RAM (Random Access Memory), ROM (Read Only Memory), etc. The memory unit 16 stores various programs and data, and also functions as a work area for the control unit 17. The memory unit 16 has a program for executing the corrosion depth estimation process described later stored in it, as well as image data 21 of the corroded area. Furthermore, the memory unit 16 has depth conversion data 22 pre-stored for determining the corrosion depth P from the image data 21 (see Figure 6). Details of the depth conversion data 22 will be described later.

[0014] The control unit 17 is composed of, for example, a CPU (Central Processing Unit) and controls the operation of each part of the corrosion depth estimation device 1. Specifically, the control unit 17 operates each part of the corrosion depth estimation device 1 based on the operation content of the operation unit 12, and also loads programs pre-stored in the memory unit 16 and executes various processes in cooperation with the loaded programs.

[0015] [Corrosion depth estimation process] Next, we will explain the corrosion depth estimation process used to estimate the corrosion depth of aircraft structures. Figure 2 is a flowchart showing the procedure for estimating corrosion depth, and Figures 3 and 4 are examples of image data of the corroded area. Figure 5 is a diagram to explain the corrosion depth estimation process, and Figure 6 is a graph showing an example of depth conversion data 22. In the corrosion depth estimation process, the corrosion depth (depth of corrosion) is estimated based on images of the corroded area. Then, the damage level (degree of damage) of the structure is determined based on the estimated corrosion depth. The corrosion depth estimation process is performed when the control unit 17 of the corrosion depth estimation device 1 reads a predetermined program from the storage unit 16 and loads it.

[0016] As shown in Figure 2, first, the corroded area to be evaluated is pre-treated (Step S1). Here, a worker exposes the bare metal surface within a predetermined range, including the black corroded area (corrosion region), by, for example, removing the surface coating with sandpaper from the corroded area to be evaluated. The structure to be evaluated is not particularly limited, but in this embodiment it is an aircraft structure made of aluminum alloy.

[0017] Next, an image of the corroded area is obtained (step S2). Here, the surface of the structure, including the corroded area, is photographed with a camera at the required resolution, and the image is acquired. This yields two-dimensional image data 21 including the corroded area 30, as shown in Figure 3, for example. The acquired image data 21 is stored in the storage unit 16. Furthermore, the data format of the image data 21 is not particularly limited, as long as it contains information that allows for the extraction of the corroded area in step S3 described later.

[0018] Next, the control unit 17 of the corrosion depth estimation device 1 extracts the corrosion region 31 from the corrosion region 30 based on the image data 21 of the corrosion region 30 (step S3). The corrosion region 31 refers to the portion of the corroded part 30 where metal corrosion has progressed to a predetermined level or more. Here, for example, based on the color information of the image data 21, the portion of the corroded part 30 that is darker in color (closer to black) than a predetermined threshold is defined as the corrosion region 31. As a result, multiple corrosion regions 31 are extracted from the corroded part 30, as shown in Figure 4, for example. The image information used to extract the corroded region 31 is not particularly limited; for example, information such as changes in brightness and contrast may be used. Also, it is sufficient for at least one corroded region 31 to be extracted in this step.

[0019] Next, the control unit 17 determines the maximum inscribed circle diameter for each of the corrosion regions 31 extracted in step S3 (step S4). Here, the control unit 17 sets the maximum inscribed circle S contained within the corrosion region 31 based on the shape of the corrosion region 31 in the image data 21, as shown in Figure 5, for example, and determines the maximum inscribed circle diameter D as its diameter. Furthermore, the calculation of the maximum inscribed circle diameter D does not need to be performed for all corrosion regions 31; for example, it may be performed only for corrosion regions 31 of a certain size or larger.

[0020] Next, the control unit 17 converts the maximum inscribed circle diameter D into corrosion depth P using the depth conversion data 22 (step S5). Here, the control unit 17 calculates the corrosion depth P by inputting the maximum inscribed circle diameter D into the depth conversion data 22 read from the storage unit 16. The depth conversion data 22 is correlation data that represents the correlation between the maximum inscribed circle diameter D and the corrosion depth P, as shown in Figure 6, for example, and is pre-set based on actual measurements. In the example in Figure 6, the plots in the figure are the measured values ​​of the maximum inscribed circle diameter D and the corrosion depth P, and the solid line (thick line) is the linear approximation formula for these values, which is the depth conversion data 22 in this embodiment. The depth conversion data 22 is not limited in format as long as it can convert the maximum inscribed circle diameter D to corrosion depth P; for example, it may be a conversion table.

[0021] Next, the control unit 17 determines the damage level (degree of damage) of the structure due to the corrosion based on the corrosion depth P obtained in step S5 (step S6). Specifically, the control unit 17 determines that if the maximum corrosion depth P obtained in step S5 is less than a predetermined A value, no repair is necessary; if it is greater than or equal to A value and less than B value (> A value), repair is necessary (repairable); and if it is greater than or equal to B value, replacement is necessary (not repairable). However, this determination method is not particularly limited. For example, the distribution and size of the corrosion depth P in the corroded area 30, the location on the structure, etc., may also be taken into consideration.

[0022] [Technical effects of this embodiment] As described above, according to this embodiment, the corroded region 31 is extracted from the image data 21 of the corroded portion 30, and its maximum inscribed circle diameter D is determined. Then, the corrosion depth P is calculated based on this maximum inscribed circle diameter D. This makes it possible to easily estimate the corrosion depth P using only image analysis, without the need to perform actual measurements of the corroded area 30 (corrosion region 31). Furthermore, since it is sufficient to obtain image data 21, it can be suitably applied to the evaluation of areas where actual measurement is difficult, such as complex part shapes or recessed areas of structures. Furthermore, instead of the perimeter or area of ​​the corroded region 31, the maximum inscribed circle diameter D is used as a feature quantity for the image data 21 used to determine the corrosion depth P. This allows for highly accurate determination of the corrosion depth P.

[0023] In other words, the inventors have found through actual measurements that the maximum inscribed circle diameter D of the corrosion region 31 has a higher degree of correlation with the corrosion depth P than with the perimeter length or area of ​​the corrosion region 31. Here, Figure 7 shows an example of the measured perimeter and corrosion depth of the corrosion region 31, and Figure 8 shows an example of the measured area and corrosion depth P of the corrosion region 31. As shown in these figures, no good correlation was found between the perimeter and area of ​​the corrosion region 31 and the corrosion depth P. In contrast, a good correlation was found between the maximum inscribed circle diameter D of the corrosion region 31 and the corrosion depth P, as shown in Figure 6. This is thought to be because corrosion on the metal surface and corrosion in the depth direction proceed at roughly the same erosion rate. In other words, the corrosion region with the largest inscribed circle diameter is considered to be the deepest part of the corrosion within its corrosion occurrence range. Therefore, by using the maximum inscribed circle diameter D of the corrosion region 31, the corrosion depth P can be estimated with higher accuracy than by using the perimeter or area.

[0024] Furthermore, according to this embodiment, the damage level (degree of damage) of the aircraft structure (metal structure) is determined based on the calculated corrosion depth P. This allows, for example, if the corrosion depth P is determined to be irreparable, corrosion removal work can be immediately stopped, reducing the man-hours spent on unnecessary corrosion removal, inspection, and measurement. In turn, this can lead to shorter lead times for regularly maintained machines, an increase in the number of machines that can be maintained, improved machine utilization rates, and reduced repair costs.

[0025] [Differentiation] In the above embodiment, in step S5 of the corrosion depth estimation process, the corrosion depth P is calculated using the depth conversion data 22. However, in this step, the corrosion depth P may be calculated using depth conversion data 22A instead of depth conversion data 22.

[0026] The depth conversion data 22A is correlation data that represents the correlation between the product of the maximum inscribed circle diameter D and the needle-shape ratio R and the corrosion depth P, as shown in Figure 9, for example, and is pre-set based on actual measurements. In the example in Figure 9, the plots in the figure are actual measured values, and the thick lines are their approximation formulas, which are the depth conversion data 22A of this embodiment. The aspect ratio R is the ratio of the maximum length to the maximum width of the target corrosion region 31, as shown in Figure 10. The specific method for calculating the aspect ratio R will be described later. The data format of the depth conversion data 22A is not particularly limited and may be, for example, a conversion table.

[0027] Specifically, in the corrosion depth estimation process in this modified example, as shown in Figure 11, steps S4a and S5a are executed instead of steps S4 and S5 in the above embodiment.

[0028] In step S4a, the control unit 17 determines the maximum inscribed circle diameter D and the needle-shape ratio R of the corrosion region 31. The maximum inscribed circle diameter D is determined in the same way as in step S4 of the above embodiment. In calculating the needle-like ratio R, as shown in Figure 12, the control unit 17 first determines the maximum length L of the corroded region 31. Next, the control unit 17 determines two circumtangential lines 32 that are parallel to the direction of the maximum length L and circumtangential to the corrosion region 31. Next, the control unit 17 determines the distance W between the two external tangent lines 32 (the maximum width of the corrosion region 31). Next, the control unit 17 calculates the needle-shape ratio R by dividing the maximum length L by the distance W, which is L / W.

[0029] In step S5a, the control unit 17 calculates the corrosion depth P based on the maximum inscribed circle diameter D and needle-shape ratio R calculated in step S4a. Specifically, the control unit 17 converts the product of the maximum inscribed circle diameter D and the needle-shape ratio R into the corrosion depth P using the depth conversion data 22A. Note that the calculation of the needle-shape ratio R does not have to be performed for all corrosion regions 31; for example, it may be performed only for corrosion regions 31 of a certain size or larger.

[0030] As described above, according to this modified example, the corrosion depth P is calculated based on the maximum inscribed circle diameter D and the needle-shape ratio R of the corrosion region 31. This allows the same effects as in the above embodiment to be obtained.

[0031] Furthermore, in this modified example, the product of the maximum inscribed circle diameter D and the needle-shape ratio R of the corroded region 31 is used as a feature quantity for determining the corrosion depth P in the image data 21. This allows for highly accurate determination of the corrosion depth P. In other words, the inventors found that by using a parameter obtained by correcting the maximum inscribed circle diameter D with the needle-shape ratio R, the corrosion depth P can be estimated with greater accuracy. Figure 13 shows the relationship between the maximum inscribed circle diameter D and the corrosion depth P when the number of plots is increased. As shown in this figure, when the range of plots is expanded, there are multiple trends in the correlation between the maximum inscribed circle diameter D and the corrosion depth P. This is presumed to be because the trends differ between, for example, the "corrosion region 31 (corrosion group) with a small maximum inscribed circle diameter but deep corrosion" as shown in Figure 14, and the "corrosion region 31 (corrosion group) with a large maximum inscribed circle diameter but shallow corrosion" as shown in Figure 15. Therefore, if only the maximum inscribed circle diameter D is used as a parameter, the corrosion depth P may not be uniquely determined. In contrast, by using the product of the maximum inscribed circle diameter D and the needle-shape ratio R as a parameter, the aforementioned trends converged, as shown in Figure 9, and a good correlation with the corrosion depth P was found. Therefore, by using a parameter in which the maximum inscribed circle diameter D of the corrosion region 31 is corrected by the needle-shape ratio R, the corrosion depth P can be estimated with higher accuracy.

[0032] [others] Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above. For example, in the above embodiment, an aluminum aircraft structure was given as the structure to be evaluated (metal structure). However, the material of the metal structure according to the present invention is not limited to aluminum, nor is its application limited to aircraft applications. However, depending on the material, it will be necessary to acquire depth transformation data 22. In addition, the image analysis method (for example, the method for extracting the corrosion region 31) may need to be changed depending on the corrosion mechanism in the material.

[0033] Furthermore, in the analysis of image data 21, machine learning or other methods may be used to improve accuracy. Furthermore, details shown in the above embodiments can be modified as appropriate without departing from the spirit of the invention. [Explanation of Symbols]

[0034] 1. Corrosion depth estimation device 16 Memory section 17 Control Unit (Extraction Unit, Diameter Calculation Unit, Depth Calculation Unit) 21 Image data 22. Depth-transformed data (correlation data) 30 Corroded parts 31 Corrosion Area 32 External tangent S is the largest inscribed circle D: Maximum inscribed circle diameter (diameter of the largest inscribed circle) P Corrosion depth (depth of the corrosive area) L Maximum length W distance R Acicular ratio

Claims

1. An extraction unit that extracts corroded areas from images of the surface of metal structures, A diameter calculation unit for determining the diameter of the largest inscribed circle of the corrosion region, A depth calculation unit that calculates the depth of the corrosion region based on the diameter of the maximum inscribed circle, A corrosion depth estimation device equipped with the following features.

2. The system includes a storage unit that pre-stores correlation data representing the correlation between the diameter of the largest inscribed circle and the depth of the corrosion region. The depth calculation unit calculates the depth of the corrosion region from the diameter of the maximum inscribed circle based on the correlation data stored in the storage unit. The corrosion depth estimation device according to claim 1.

3. The system includes a determination unit that determines the degree of damage to the metal structure based on the depth of the corrosion region. The corrosion depth estimation device according to claim 1.

4. The aforementioned metal structure is made of aluminum. The corrosion depth estimation device according to claim 1.

5. The depth calculation unit is, The depth of the corrosion region is calculated based on the diameter of the largest inscribed circle and the needle-shape ratio of the corrosion region. The corrosion depth estimation device according to claim 1.

6. The acquisition process involves obtaining an image of the surface of a metal structure, An extraction step for extracting the corroded area from the image acquired in the acquisition step, A diameter calculation step to determine the diameter of the largest inscribed circle of the corrosion region, A depth calculation step, which calculates the depth of the corrosion region based on the diameter of the maximum inscribed circle, A method for estimating corrosion depth, including the method described above.

Images