Estimation device

Abstract

An estimation device (10) according to the present disclosure comprises: a processing unit (14) that averages spectral data, acquired by a hyperspectral camera, for each prescribed wavelength, thereby generating averaged spectral data indicating the average wavelength intensity for each prescribed wavelength; and an estimation unit (16). The estimation unit: identifies a peak wavelength, at which a secondary differential spectrum obtained by secondary differentiation of the averaged spectral data has a maximum value or a minimum value, within a prescribed band in the secondary differential spectrum; obtains the relationship between a reflected light intensity at the peak wavelength and a neutralization depth on the basis of the averaged spectral data, when the neutralization depth of concrete for estimation having the same water-cement ratio as target concrete acquired in advance is zero, and a neutralization depth progression coefficient, which is a constant corresponding to the water-cement ratio; and estimates the neutralization depth of the target concrete on the basis of the obtained relationship and the reflected light intensity at the peak wavelength in the averaged spectral data of the target concrete.

Description

estimation device

[0001] This disclosure relates to an estimation device.

[0002] Many social infrastructure facilities are over 50 years old and are deteriorating. Proper maintenance is necessary to ensure the safety and security of these facilities. Furthermore, given the vast number of social infrastructure facilities, efficient inspection and maintenance work is essential.

[0003] In concrete, which is widely used as a structural material for social infrastructure facilities, cracking, spalling, and exposed reinforcement are known as typical forms of deterioration. The mechanism of these deteriorations involves a phenomenon called concrete carbonation.

[0004] The current method for inspecting concrete for carbonation involves a technique called core extraction, which involves cutting into the concrete to collect a concrete core. However, this method is costly and time-consuming because it involves localized destruction of the concrete.

[0005] Patent Document 1 describes a method for detecting the amount of salt on the surface of concrete using near-infrared spectroscopy in order to investigate salt damage to concrete.

[0006] A. Watanabe et al.:”Toward automated non-destructive diagnosis of chloride attack on concrete structures by near infrared spectroscopy, Construction and Building materials, 305(2021).

[0007] As mentioned above, destructive testing of concrete, such as core drilling, is costly and time-consuming, so there is a need for a non-destructive method to estimate the carbonation depth of concrete. As a non-destructive concrete inspection method, a method using a hyperspectral camera capable of measuring the intensity of electromagnetic waves at predetermined wavelengths is being considered. Hyperspectral cameras are easy to carry and do not require concrete drilling, so if the carbonation depth of concrete can be estimated using a hyperspectral camera, it will be possible to reduce costs and time.

[0008] However, since concrete carbonation progresses in the depth direction within the concrete, it was difficult to estimate the carbonation depth of the concrete from the results of imaging the concrete surface with a hyperspectral camera.

[0009] In view of the above-mentioned problems, the purpose of this disclosure is to provide an estimation device that can estimate the carbonation depth of concrete using a hyperspectral camera.

[0010] To solve the above problems, the estimation device according to the present disclosure is an estimation device for estimating the carbonation depth of a target concrete, comprising: a processing unit that averages spectral data showing the reflected light intensity of electromagnetic waves for each predetermined wavelength reflected from the surface of the target concrete, acquired by a hyperspectral camera, for each predetermined wavelength, and generates averaged spectral data showing the average wavelength intensity for each predetermined wavelength; and an estimation unit that identifies a peak wavelength in which the second derivative spectrum obtained by taking the second derivative of the averaged spectral data has a maximum or minimum value within a predetermined band in the second derivative spectrum, identifies an averaged spectral data obtained by taking the second derivative of the averaged spectral data, in which the carbonation depth of an estimation concrete having the same water-cement ratio as the target concrete, which has been acquired in advance, is zero, and determines the relationship between the reflected light intensity at the peak wavelength and the carbonation depth based on the relationship determined and the reflected light intensity at the peak wavelength in the averaged spectral data of the target concrete, and estimates the carbonation depth of the target concrete.

[0011] According to this disclosure, the carbonation depth of concrete can be estimated using a hyperspectral camera.

[0012] This figure shows an example of the configuration of an estimation device according to one embodiment of the present disclosure. This figure shows an example of the configuration of the extraction unit shown in Figure 1. This figure shows an example of an image captured by a hyperspectral camera. This figure is for explaining the operation of the binarization unit shown in Figure 2. This figure is for explaining the operation of the rectangle setting unit shown in Figure 2. This figure is for explaining the operation of the rectangle setting unit shown in Figure 2. This figure is for explaining the operation of the deletion unit shown in Figure 2. This figure shows an example of the configuration of the processing unit shown in Figure 1. This figure shows an example of averaged spectral data. This figure shows an example of the configuration of the estimation unit shown in Figure 2. This figure shows an example of a second derivative spectrum. This figure is for explaining the operation of the graph creation unit shown in Figure 6. This figure is for explaining the operation of the neutralization depth estimation unit shown in Figure 6. This is a flowchart showing an example of the operation of the estimation device shown in Figure 1. This figure shows an example of the configuration of a computer that functions as an estimation device according to the present disclosure.

[0013] Embodiments of this disclosure will be described below with reference to the drawings.

[0014] Figure 1 shows an example of the configuration of an estimation device 10 according to one embodiment of the present disclosure. The estimation device 10 according to the present disclosure estimates the carbonation depth of concrete (target concrete) that is the target of carbonation depth estimation.

[0015] As shown in Figure 1, the estimation device 10 according to this embodiment includes a first input unit 11, an extraction unit 12, a second input unit 13, a processing unit 14, a third input unit 15, an estimation unit 16, and an output unit 17.

[0016] The first input unit 11 receives an image of the surface of the target concrete, captured by a hyperspectral camera capable of measuring the intensity of electromagnetic waves for each predetermined wavelength. Using the hyperspectral camera, the wavelength λ α wavelengths from nm to λ β When imaging is performed at tnm intervals up to nm, the captured image is [{(λ β -λ α) / t} + 1] sheets. The first input unit 11 inputs all the captured images of [{(λ β -λ α ) / t} + 1] sheets. The first input unit 11 outputs the input captured images of the hyperspectral camera to the extraction unit 12.

[0017] The extraction unit 12 extracts the central part excluding the peripheral part of the target concrete from the captured images of the hyperspectral camera output from the first input unit 11. FIG. 2 is a diagram showing an example of the configuration of the extraction unit 12.

[0018] As shown in FIG. 2, the extraction unit 12 includes a grayscale conversion unit 121, a binarization unit 122, a rectangle setting unit 123, and a deletion unit 124.

[0019] The grayscale conversion unit 121 converts each of the captured images of the hyperspectral camera for each predetermined wavelength ([{(λ β -λ α ) / t} + 1] sheets of captured images) into a grayscale image. The grayscale conversion unit 121 extracts the highest pixel value (maximum pixel value) within each grayscale image.

[0020] The binarization unit 122 performs binarization on the image with the largest maximum pixel value among the [{(λ β -λ α ) / t} + 1] sheets of grayscale images. FIG. 3A is a diagram showing an example of a captured image of the hyperspectral camera. As shown in FIG. 3A, the target concrete and its surroundings are imaged by the hyperspectral camera. Also, FIG. 3B is a diagram showing an example of a binary image generated by the binarization unit 122. As shown in FIGS. 3A and 3B, the white area of the binary image corresponds to the concrete part.

[0021] Depending on the imaging wavelength of the hyperspectral camera, the entire captured image may become dark, and it may be difficult to set an optimal threshold when converting from a grayscale image to a binary image. In this embodiment, by generating grayscale images for all the captured images for each predetermined wavelength and generating a binary image from the grayscale image having the maximum pixel value, a clear binary image can be obtained.

[0022] The rectangle setting unit 123 determines the maximum and minimum x and y coordinates of the white region in the binary image. As shown in Figure 3C, the rectangle setting unit 123 sets a rectangle with the determined maximum and minimum x and y coordinates as its vertices. Hereinafter, we assume that the minimum x coordinate of the white region is A and the maximum x coordinate is C. Also, we assume that the minimum y coordinate of the white region is B and the maximum y coordinate is D. In this case, as shown in Figure 3D, the rectangle setting unit 123 sets a rectangle on the binary image with the four points (A, B), (C, B), (A, D), and (C, D) as its vertices.

[0023] Referring again to Figure 2, the deletion unit 124 shrinks the rectangle set on the binary image by a pixels from the top edge toward the interior of the rectangle, a' pixels from the bottom edge toward the interior of the rectangle, b pixels from the right edge toward the interior of the rectangle, and b' pixels from the left edge toward the interior of the rectangle, and deletes the part other than the shrunken rectangle. That is, as shown in Figure 3E, the deletion unit 124 deletes pixels in the part other than the rectangle with the four points (A+b', B-a), (C-b, B-a), (A+b', D+a'), and (C-b, D+a') as vertices. Note that a and a' may be the same value or different values. Also, b and b' may be the same value or different values.

[0024] Through the process described above, the extraction unit 12 can extract only the central portion of the binary image, that is, the central part of the surface shape of the target concrete excluding the peripheral portion. The extraction unit 12 outputs the binary image data of the central portion to the processing unit 14.

[0025] Generally, concrete samples do not have a uniform surface shape, and the edges of concrete samples, in particular, may be chipped. Concrete contains cement, coarse aggregate, and fine aggregate, but when concrete chips due to external impact, the lower-strength cement chips first, exposing the coarse aggregate on the surface. Therefore, using data from the peripheral edge of the target concrete will result in a greater influence from the aggregate than from a normal concrete surface. In this embodiment, by extracting the central part of the target concrete, excluding the peripheral edge, from the image captured by the hyperspectral camera, the influence of exposed aggregate can be reduced.

[0026] Referring again to Figure 1, the second input unit 13 receives spectral data acquired by imaging the target concrete with a hyperspectral camera. The spectral data is data that shows the reflected light intensity of electromagnetic waves for each predetermined wavelength reflected from the surface of the target concrete. The second input unit 13 outputs the input spectral data as a three-dimensional data cube to the processing unit 14.

[0027] For example, if the surface area of ​​the target concrete is width x pixels and length y pixels (e.g., width 191 pixels, length 300 pixels), and a hyperspectral camera measures wavelength λ α wavelengths from nm to λ β Assume that imaging is performed at tnm intervals up to nm (for example, at 9nm intervals from 1295nm to 2218nm). In this case, the second input unit 13 captures a two-dimensional image with width x pixels × length y pixels that shows the intensity of the reflected light (reflected light intensity) reflected at each position of the target concrete from electromagnetic waves with wavelengths at 9nm intervals from 1295nm to 2218nm, using the number of imaging wavelengths [{(λ β -λ α A three-dimensional data cube stacked by [) / t] + 1] is output to the processing unit 14.

[0028] There are two imaging methods using hyperspectral cameras: line-type and area-type. In the line-type method, the target concrete is moved in one direction relative to the hyperspectral camera, and images are taken with the hyperspectral camera line by line in a direction intersecting that direction (measuring the wavelength intensity of electromagnetic waves reflected by the target concrete). In the area-type method, the hyperspectral camera images a certain area of ​​the target concrete at once. In the area-type method, only the intensity of one wavelength can be measured, so the same area is imaged multiple times with different wavelengths. Regardless of which method is used for imaging, the second input unit 13 outputs a 3D data cube to the processing unit 14.

[0029] The processing unit 14 averages the spectral data, which shows the reflected light intensity of electromagnetic waves at predetermined wavelengths reflected from the surface of the target concrete and acquired by the hyperspectral camera, for each predetermined wavelength. The processing unit 14 then generates averaged spectral data showing the average wavelength intensity for each predetermined wavelength. Figure 4 shows an example of the configuration of the processing unit 14.

[0030] As shown in Figure 4, the processing unit 14 includes a spectral data extraction unit 141 and a spectral data averaging unit 142.

[0031] The spectral data extraction unit 141 receives spectral data (3D data cube) output from the second input unit 13 and a binary image of the central part output from the extraction unit 12. The spectral data extraction unit 141 associates the pixels of the 3D data cube with the pixels of the binary image of the central part and extracts the pixels corresponding to the binary image of the central part from the 3D data cube. In other words, similar to the extraction unit 12, the spectral data extraction unit 141 removes the peripheral part of the target concrete from the 3D data cube.

[0032] The spectral data averaging unit 142 averages the spectral data of the central part of the target concrete extracted by the spectral data extraction unit 141 for each wavelength, and generates averaged spectral data showing the average wavelength intensity for each predetermined wavelength, as shown in Figure 5.

[0033] In this way, the processing unit 14 generates averaged spectral data showing the average wavelength intensity for each predetermined wavelength as shown in FIG. 5, using the spectral data corresponding to the central portion of the target concrete. The surface of the concrete has different distributions of cement, coarse aggregate, and fine aggregate. By averaging the spectral data of the central portion of the target concrete, the characteristics of the entire surface of the target concrete can be grasped.

[0034] Referring to FIG. 1 again, the processing unit 14 outputs the generated averaged spectral data to the estimation unit 16.

[0035] The third input unit 13 receives the water-cement ratio w% of the target concrete, the spectral data when the carbonation depth of the concrete with a water-cement ratio of w% prepared in advance (hereinafter referred to as "reference concrete") is zero (no carbonation), and the carbonation depth progression coefficient γ. The spectral data input to the third input unit 13 is data (averaged spectral data) obtained by performing the same processing as the processing by the above-described processing unit 14 on the spectral data of the reference concrete previously acquired by the hyperspectral camera. The carbonation depth progression coefficient γ is a constant corresponding to the water-cement ratio w%, and is given by, for example, the following formula: γ = 0.0014ε where ε is a constant set according to environmental factors such as temperature and humidity.

[0036] Most of the range of the water-cement ratio of the concrete used in concrete structures is 40 - 60%. That is, the water-cement ratio w% is mostly in the range of 40% to 60%. Therefore, the spectral data of the concrete for estimation may also be prepared for concrete with a water-cement ratio of 40% to 60%. By doing so, the spectral data of the concrete for estimation that should be prepared in advance can be reduced. The third input unit 13 may acquire the spectral data of the concrete for estimation with the input water-cement ratio w from among the spectral data of a plurality of concretes for estimation with different water-cement ratios w prepared in advance. Also, the third input unit 13 may acquire the neutralization depth progression coefficient γ corresponding to the input water-cement ratio w from among the neutralization depth progression coefficients γ corresponding to different water-cement ratios w prepared in advance.

[0037] The third input unit 15 outputs the input spectral data of the concrete for estimation and the neutralization depth progression coefficient γ to the estimation unit 16.

[0038] The estimation unit 16 receives the averaged spectral data of the target concrete output from the processing unit 14. Also, the estimation unit 146 receives the spectral data (averaged spectral data) of the concrete for estimation and the neutralization depth progression coefficient γ output from the third input unit 15.

[0039] The estimation unit 16 estimates the neutralization depth of the target concrete. FIG. 6 is a diagram showing an example of the configuration of the estimation unit 16.

[0040] As shown in FIG. 6, the estimation unit 16 includes a second derivative spectrum calculation unit 161, a graph creation unit 162, and a neutralization depth estimation unit 163.

[0041] The second derivative spectrum calculation unit 161 calculates the second derivative spectrum by second differentiating the averaged spectral data of the target concrete. As a method for calculating the second partial spectrum, for example, the Savitzky-Golay method can be used, but it is not limited to this, and any method can be used.

[0042] As shown in Figure 7, the second derivative spectrum calculation unit 161 identifies a peak wavelength u [nm] in which the second derivative spectrum obtained by second derivative of the averaged spectral data has a maximum or minimum value within a predetermined bandwidth. The predetermined wavelength bandwidth is, for example, the wavelength range of 2000-2200 nm. Therefore, the second derivative spectrum calculation unit 161 identifies a peak wavelength u in which the second derivative spectrum has a maximum or minimum value within the wavelength range of 2000-2200 nm.

[0043] Concrete carbonation is represented by the following chemical formula: Ca(OH) ２ +CO ２ →CCO ３ +H ２ 0 In other words, due to the neutralization of concrete, calcium hydroxide (Ca(OH) ２ ) decreases, and carbonate ions (CaCO3) decrease. ３ ２－ The absorption bands for calcium hydroxide are expected to increase, based on the typical absorption wavelength ranges of OH bonds and carboxylic acids, around 1420 nm and 2150 nm. The absorption band for carbonate ions is expected to be 1930–1950 nm. Furthermore, spectral data obtained by hyperspectral cameras are also affected by water in the air. The absorption bands for water are expected to be 1450–1460 nm and 1930–1940 nm.

[0044] In this embodiment, the wavelength range for searching for the peak wavelength at which the second derivative spectrum reaches a maximum or minimum value is limited to 2000-2200 nm. That is, in this embodiment, the carbonation depth of the concrete is estimated by focusing on the absorption band of calcium hydroxide. In the near-infrared or short-wave infrared region, in the wavelength range below 2000 nm or above 2200 nm, absorption of light of that wavelength occurs due to the presence of water molecules. Therefore, when moisture is present in the air (for example, in humid conditions), there is a problem in that it is difficult to distinguish whether the spectral peak is due to concrete carbonation or a peak due to the presence of water molecules. In this embodiment, the above problem can be prevented by limiting the wavelength range for searching for the peak wavelength u at which the second derivative spectrum reaches a maximum or minimum value to a predetermined wavelength range (for example, 2000-2200 nm) that includes peaks of substances related to concrete carbonation and is less affected by peaks of substances unrelated to carbonation.

[0045] In this embodiment, the peak wavelength u is searched for using the second derivative spectrum obtained by taking the second derivative of the averaged spectral data of the target concrete. Here, since there are many absorption bands in the infrared spectroscopy spectrum, peaks may overlap. As mentioned above, calcium hydroxide, hydroxide ions, and carbonate ions are involved in the carbonation of concrete, and the peaks of these substances are included in the spectral data. Peaks due to moisture in the air are also included in the spectral data. In this embodiment, we focus on the wavelength band of 2000-2200 nm and estimate the carbonation depth from the calcium hydroxide peak. However, even peaks in the wavelength band of 2000-2200 nm may be affected by the presence of absorption bands of hydroxide ions, carbonate ions, or other substances.

[0046] By differentiating a spectrum, the peaks contained within that spectrum can be separated (see Reference 1). In this embodiment, by focusing on the maximum and minimum values ​​of the second derivative spectrum obtained by differentiating the averaged spectral data, overlapping peaks can be separated and the peak tops can be clearly identified. [Reference 1] "Let's separate peaks by differentiating a spectrum", FTIR Blog - PerkinElmer Japan, [Retrieved October 18, 2024], Internet<URL:https: / / www.perkinelmer.co.jp / tabid / 2470 / Default.aspx>

[0047] The graph generation unit 162 receives the averaged spectral data for the estimated concrete when the carbonation depth is zero and the carbonation depth progression coefficient γ corresponding to the water-cement ratio w%, which are output from the third input unit 15. Based on the averaged spectral data for the estimated concrete when the carbonation depth is zero and the carbonation depth progression coefficient γ, the graph generation unit 162 creates a graph showing the relationship between the reflected light intensity at the peak wavelength u and the carbonation depth. Specifically, the graph generation unit 162 obtains a straight line (graph) as the relationship between the reflected light intensity at the peak wavelength u and the carbonation depth, which passes through the average wavelength intensity at the peak wavelength u in the averaged spectral data of the estimated concrete and has a slope equal to the carbonation depth progression coefficient γ, as shown by the dashed line in Figure 8.

[0048] Referring again to Figure 6, the carbonation depth estimation unit 163 estimates the carbonation depth of the target concrete based on the graph created by the graph creation unit 162, which shows the relationship between reflected light intensity at peak wavelength u and carbonation depth, and the reflected light intensity at peak wavelength u in the averaged spectral data of the target concrete. The carbonation depth estimation unit 163 applies the relationship between reflected light intensity at peak wavelength and carbonation depth obtained for the estimation concrete to the target concrete to estimate the carbonation depth of the target concrete. Specifically, as shown in Figure 9, the carbonation depth estimation unit 163 inputs the reflected light intensity at peak wavelength u in the averaged spectral data of the target concrete to the relationship between reflected light intensity at peak wavelength u and carbonation depth to estimate the carbonation depth.

[0049] In this way, the estimation unit 16 identifies a peak wavelength u in the second derivative spectrum obtained by taking the second derivative of the averaged spectral data of the target concrete, within a predetermined band (2000-2200 nm), where the second derivative spectrum has a maximum or minimum value. The estimation unit 16 also determines the relationship between the reflected light intensity at the peak wavelength u and the carbonation depth based on the averaged spectral data of the estimation concrete, which has the same water-cement ratio w as the target concrete and has zero carbonation depth, and the carbonation depth progression coefficient γ, which is a constant corresponding to the water-cement ratio w. Then, the estimation unit 16 estimates the carbonation depth of the target concrete based on the determined relationship and the reflected light intensity at the peak wavelength u in the averaged spectral data of the target concrete.

[0050] By estimating the carbonation depth by considering the change in wavelength intensity from a point where the carbonation depth is zero, it becomes possible to estimate the carbonation depth more accurately than by using a carbonation depth progression curve. Furthermore, by incorporating a parameter ε that takes into account environmental factors such as temperature and humidity into the carbonation depth progression coefficient γ, it becomes possible to estimate the carbonation depth while reducing the influence of the installation environment of the target concrete and the imaging environment of the hyperspectral camera.

[0051] Referring again to Figure 1, the estimation unit 16 outputs the estimated result of the carbonation depth of the target concrete to the output unit 17.

[0052] The output unit 17 outputs the estimated result of the carbonation depth of the target concrete, which was output from the estimation unit 16, and displays the result.

[0053] The carbonation depth progression coefficient γ can be calculated experimentally, for example. The carbonation depth progression coefficient γ will be explained below.

[0054] Concrete samples with a water-cement ratio of w% were prepared. w is a water-cement ratio commonly used in concrete structures, for example, in the range of 40-60. The prepared concrete samples were subjected to carbonation acceleration using a carbonation accelerator, and multiple concrete samples with target carbonation depths in the range of 0-10 mm were prepared. Next, phenolphthalein solution was sprayed onto the cross-sections of the multiple concrete samples, and the carbonation depth of each concrete sample was measured.

[0055] Next, the carbonation-accelerating surface of concrete samples, whose carbonation depth was measured, was imaged with a hyperspectral camera, and spectral data was acquired. From the acquired spectral data, wavelength intensity data at each wavelength in the near-infrared or short-wave infrared region was obtained.

[0056] Next, the spectral data from the central part of the concrete sample was extracted from the acquired spectral data, similar to the processing section 14, and average spectral data was calculated. Furthermore, the second derivative of the calculated average spectral data was obtained by taking the second derivative. The wavelength intensity at the wavelength where the second derivative spectrum takes a minimum value was plotted against the measured value of the carbonation depth, and the slope of the approximate straight line of the plotted points was calculated as the carbonation depth progression coefficient γ.

[0057] Next, the operation of the estimation device 10 according to this embodiment will be described. Figure 10 is a flowchart showing an example of the operation of the estimation device 10 according to this embodiment, and is a diagram for explaining the estimation method performed by the estimation device 10 according to this embodiment.

[0058] The extraction unit 12 extracts the central part of the target concrete, excluding the peripheral part, from the image captured by the hyperspectral camera (step S11).

[0059] The processing unit 14 averages the spectral data, which shows the reflected light intensity of electromagnetic waves at predetermined wavelengths reflected from the surface of the target concrete and is acquired by the hyperspectral camera, for each predetermined wavelength, and generates averaged spectral data showing the average wavelength intensity for each predetermined wavelength (step S12). Specifically, the processing unit 14 generates averaged spectral data using spectral data corresponding to the central part of the target concrete extracted by the extraction unit 12.

[0060] The processing unit 14 identifies a peak wavelength u within a predetermined bandwidth in the second derivative spectrum obtained by taking the second derivative of the generated averaged spectral data (step S13).

[0061] The estimation unit 16 determines the relationship between the reflected light intensity at the peak wavelength u and the carbonation depth based on the averaged spectral data of the estimation concrete having the same water-cement ratio w as the target concrete obtained in advance, when the carbonation depth is zero, and the carbonation depth progression coefficient γ, which is a constant corresponding to the water-cement ratio w (step S14). Specifically, the estimation unit 16 determines the relationship between the reflected light intensity at the peak wavelength u and the carbonation depth as a straight line that passes through the average wavelength intensity at the peak wavelength u in the averaged spectral data of the estimation concrete when the carbonation depth is zero, and whose slope is the carbonation progression coefficient γ.

[0062] The estimation unit 16 estimates the carbonation depth of the target concrete based on the relationship obtained and the reflected light intensity at the peak wavelength u in the averaged spectral data of the target concrete (step S15). Specifically, the estimation unit 16 inputs the reflected light intensity at the peak wavelength u in the averaged spectral data of the target concrete into the relationship obtained to estimate the carbonation depth of the target concrete.

[0063] As described above, the estimation device 10 according to this embodiment comprises a processing unit 14 and an estimation unit 16. The processing unit 14 averages spectral data, which shows the reflected light intensity of electromagnetic waves reflected from the surface of the target concrete at predetermined wavelengths and is acquired by a hyperspectral camera, for each predetermined wavelength to generate averaged spectral data. The estimation unit 16 identifies the peak wavelength u at which the second derivative spectrum has a maximum or minimum value within a predetermined band in the second derivative spectrum obtained by taking the second derivative of the averaged spectral data. The estimation unit 16 also determines the relationship between the reflected light intensity at the peak wavelength u and the carbonation depth based on the averaged spectral data of an estimation concrete having the same water-cement ratio as the target concrete, which has been acquired in advance, when the carbonation depth is zero, and the carbonation depth progression coefficient γ. Then, the estimation unit 16 estimates the carbonation depth of the target concrete based on the determined relationship and the reflected light intensity at the peak wavelength u in the averaged spectral data of the target concrete.

[0064] By doing this, it is possible to estimate the carbonation depth of concrete using a hyperspectral camera.

[0065] The estimation device 10 described above can be implemented by the computer 20 shown in Figure 11. A program for causing the computer 20 to function as the estimation device 10 may be provided. This program may be stored on a storage medium or provided via a network. Figure 11 is a block diagram illustrating the schematic configuration of the computer 20 functioning as the estimation device 10. The computer 20 may be a general-purpose computer, a dedicated computer, a workstation, a PC (Personal Computer), an electronic notepad, etc. Program instructions may be program code, code segments, etc., for executing the required tasks.

[0066] As shown in Figure 11, the computer 20 includes a processor 21, ROM (Read Only Memory) 22, RAM (Random Access Memory) 23, storage 24, input unit 25, display unit 26, and communication interface (I / F) 27. Each component is connected to each other via a bus 29 so as to be able to communicate with each other. The processor 21 is specifically a CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), SoC (System on a Chip), etc., and may be composed of multiple processors of the same or different types.

[0067] The processor 21 is a control unit that controls each component and performs various calculations. Specifically, the processor 21 reads a program from the ROM 22 or storage 24 and executes the program using the RAM 23 as a working area. The processor 21 controls each component and performs various calculations according to the program stored in the ROM 22 or storage 24. In this embodiment, the ROM 22 or storage 24 stores a program for operating the computer 20 as the estimation device 10 according to this disclosure. When this program is read and executed by the processor 21, each component of the estimation device 10, specifically the extraction unit 12, the processing unit 14, and the estimation unit 16, is realized.

[0068] The program may be provided in a form stored on a non-transitory storage medium such as a CD-ROM (Compact Disk Read Only Memory), DVD-ROM (Digital Versatile Disk Read Only Memory), or USB (Universal Serial Bus) memory. Alternatively, the program may be provided in a form that can be downloaded from an external device via a network.

[0069] ROM 22 stores various programs and data. RAM 23 temporarily stores programs or data as a working area. Storage 24 consists of an HDD (Hard Disk Drive) or SSD (Solid State Drive) and stores various programs and data, including the operating system.

[0070] The input unit 25 includes a pointing device such as a mouse and a keyboard, and is used for various types of input. The input unit 25 functions as a first input unit 11, a second input unit 13, and a third input unit 15.

[0071] The display unit 26 is, for example, a liquid crystal display and displays various information. The display unit 26 may also function as an input unit 25 by employing a touch panel system. The display unit 26 is an example of an output unit 17.

[0072] The communication interface 27 is an interface for communicating with other devices, for example, a LAN interface.

[0073] The following additional information is disclosed regarding the embodiments described above.

[0074] [Note 1] An estimation device for estimating the carbonation depth of a target concrete, comprising a control unit, wherein the control unit averages spectral data showing the reflected light intensity of electromagnetic waves reflected from the surface of the target concrete at predetermined wavelengths, acquired by a hyperspectral camera, for each predetermined wavelength to generate averaged spectral data showing the average wavelength intensity for each predetermined wavelength, identifies a peak wavelength at which the second derivative spectrum has a maximum or minimum value within a predetermined band in the second derivative spectrum obtained by taking the second derivative of the averaged spectral data, determines the relationship between the reflected light intensity at the peak wavelength and the carbonation depth based on the averaged spectral data of an estimation concrete having the same water-cement ratio as the target concrete, which has been acquired in advance, when the carbonation depth is zero, and a carbonation depth progression coefficient which is a constant corresponding to the water-cement ratio, and estimates the carbonation depth of the target concrete based on the relationship obtained and the reflected light intensity at the peak wavelength in the averaged spectral data of the target concrete.

[0075] [Addendum 2] Estimation device as described in Addendum 1, wherein the control unit extracts the central portion of the target concrete from the image captured by the hyperspectral camera, excluding the peripheral portion, and generates the averaged spectral data using the spectral data corresponding to the central portion.

[0076] [Appendix 3] Estimation device according to Appendix 1 or 2, wherein the control unit determines the relationship as a straight line passing through the average wavelength intensity at the peak wavelength in the averaged spectral data when the carbonation depth of the concrete for estimation is zero, and with the carbonation depth progression coefficient as its slope.

[0077] [Appendix 4] An estimation method performed by an estimation device for estimating the carbonation depth of a target concrete, comprising: averaging spectral data showing the reflected light intensity of electromagnetic waves reflected from the surface of the target concrete at predetermined wavelengths, acquired by a hyperspectral camera, for each predetermined wavelength to generate averaged spectral data showing the average wavelength intensity for each predetermined wavelength; identifying a peak wavelength at which the second derivative spectrum has a maximum or minimum value within a predetermined band in the second derivative spectrum obtained by taking the second derivative of the averaged spectral data; determining the relationship between the reflected light intensity at the peak wavelength and the carbonation depth based on the averaged spectral data of an estimation concrete having the same water-cement ratio as the target concrete, which has been acquired in advance, when the carbonation depth is zero, and a carbonation depth progression coefficient which is a constant corresponding to the water-cement ratio; and estimating the carbonation depth of the target concrete based on the relationship obtained and the reflected light intensity at the peak wavelength in the averaged spectral data of the target concrete.

[0078] [Appendix 5] A non-temporary storage medium storing a program executable by a computer, the non-temporary storage medium storing a program that causes the computer to operate as an estimation device described in any one of the appendix items 1-3.

[0079] Although the embodiments described above are representative examples, it will be apparent to those skilled in the art that many modifications and substitutions are possible within the spirit and scope of this disclosure. Therefore, the present invention should not be construed as being limited by the embodiments described above, and various modifications or changes are possible without departing from the claims. For example, it is possible to combine multiple component blocks shown in the configuration diagram of the embodiments into one, or to divide one component block.

[0080] 10 Estimation device 11 First input unit 12 Extraction unit 13 Second input unit 14 Processing unit 15 Third input unit 16 Estimation unit 17 Output unit 121 Grayscale conversion unit 122 Binarization unit 123 Rectangle setting unit 124 Deletion unit 141 Spectrum data extraction unit 142 Spectrum data averaging unit 161 Second derivative spectrum calculation unit 162 Graph creation unit 163 Neutralization depth estimation unit 20 Computer 21 Processor 22 ROM 23 RAM 24 Storage 25 Input unit 26 Display unit 27 Communication I / F 29 Path

Claims

1. An estimation device for estimating the carbonation depth of a target concrete, comprising: a processing unit that averages spectral data showing the reflected light intensity of electromagnetic waves reflected from the surface of the target concrete at predetermined wavelengths, acquired by a hyperspectral camera, for each predetermined wavelength, and generates averaged spectral data showing the average wavelength intensity for each predetermined wavelength; and an estimation unit that identifies a peak wavelength at which the second derivative spectrum has a maximum or minimum value within a predetermined band in the second derivative spectrum obtained by taking the second derivative of the averaged spectral data, and determines the relationship between the reflected light intensity at the peak wavelength and the carbonation depth based on averaged spectral data of an estimation concrete having the same water-cement ratio as the target concrete, which has been acquired in advance, when the carbonation depth is zero, and a carbonation depth progression coefficient which is a constant corresponding to the water-cement ratio, and estimates the carbonation depth of the target concrete based on the relationship obtained and the reflected light intensity at the peak wavelength in the averaged spectral data of the target concrete.

2. The estimation device according to claim 1, further comprising an extraction unit for extracting the central portion of the target concrete excluding the peripheral portion from the image captured by the hyperspectral camera, wherein the processing unit generates the averaged spectral data using the spectral data corresponding to the central portion.

3. The estimation device according to claim 1, wherein the estimation unit determines the relationship as a straight line passing through the average wavelength intensity at the peak wavelength in the averaged spectral data when the carbonation depth of the concrete for estimation is zero, and with the carbonation depth progression coefficient as its slope.

Images