Railway support evaluation device, railway support evaluation system, and railway support evaluation method

Abstract

To provide a railroad supporting object evaluation device, railroad supporting object evaluation system and railroad supporting object evaluation method which can improve the accuracy of evaluating the corrosion of a railroad supporting object.SOLUTION: A railroad supporting object evaluation device 1 for evaluating the corrosion of a railroad supporting object 3 comprises: an acquisition unit which acquires spectrum image data obtained by imaging the railroad supporting object 3; a selection unit which selects a specific wavelength included in a prescribed frequency band as a normalization wavelength and selects a specific wavelength included in a prescribed frequency band as an evaluation wavelength on the basis of the spectrum image data acquired by the acquisition unit; and an evaluation unit which calculates the reflectance with a difference in the spectral intensity of the normalization wavelength and the evaluation wavelength between the normalization wavelength and the evaluation wavelength selected by the selection unit, and evaluates the corrosion as an installation environment of the railroad supporting object 3 on the basis of the calculated reflectance.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a railway support evaluation device, a railway support evaluation system, and a railway support evaluation method for evaluating corrosion of railway supports. [Background technology]

[0002] Conventionally, a surface condition diagnostic device such as that disclosed in Patent Document 1 has been proposed as a technology for diagnosing the degree of deterioration of metals and structures and the state of change in various items using images of the surfaces of the target metals, structures, and various items.

[0003] The surface condition diagnostic device disclosed in Patent Document 1 includes a deterioration level determination data storage means, a deterioration level determination location designation means for designating a deterioration level determination location from surface image data, an RGB value acquisition means for acquiring the RGB values ​​of each pixel surrounding the designated location, a deterioration level data calculation means for calculating the average value and variance of the RGB values, a means for storing an average deterioration level and a variance deterioration level based on the average value and variance of the RGB values, a deterioration level comparison means for comparing whether the average deterioration level and the variance deterioration level are the same, and an operation selection means. Therefore, the operation selection means can operate a deterioration level notification means when the comparison results in the same level, and can operate an RGB value correction processing means and a feedback means when the levels are not the same. This enables a low-cost system with a short change level determination processing time and accurate and easy-to-understand notification of the change level. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6034529 Summary of the Invention [Problem to be solved by the invention]

[0005] Currently, when evaluating corrosion such as rust that occurs on the surface of railway supports, visual inspections are performed by inspectors. As a result, there is a large amount of variation in the evaluation results due to factors such as the inspector's experience and physical condition, and the weather at the time of the inspection. In particular, railway supports installed in outdoor environments are easily affected by changes in sunlight illuminance, making it difficult to separate the metal support from the background sky, which raises concerns about evaluations in outdoor environments.

[0006] In this regard, the technology disclosed in Patent Document 1 expresses photographed images of steel surfaces as data in the RGB color system and evaluates them by converting them into the UCS color system, making it difficult to evaluate corrosion taking into account changes in sunlight illuminance, the separation of the background sky from the metal support, etc. For these reasons, there is a need for improved accuracy in evaluating corrosion of outdoor railway supports.

[0007] Therefore, the present invention has been devised in consideration of the above-mentioned problems, and its purpose is to provide a railway support evaluation device, a railway support evaluation system, and a railway support evaluation method that can improve the accuracy of evaluating corrosion of railway supports outdoors. [Means for solving the problem]

[0009] A railway support evaluation device according to a first aspect of the present invention is a railway support evaluation device for evaluating corrosion of a railway support, the railway support evaluation device comprising: an acquisition unit that acquires spectral image data of an image of the railway support; and, based on the spectral image data acquired by the acquisition unit, determining a specific wavelength included in a frequency band of 500 to 600 nm as a first normalized wavelength and a specific wavelength included in a frequency band of 600 to 800 nm as a first evaluation wavelength. Further, a specific frequency in the 615 to 715 nm frequency band is selected as the second normalized wavelength, and a specific frequency in the 750 to 850 nm frequency band is selected as the second evaluation wavelength. a selector for selecting the first normalized wavelength and the first evaluation wavelength, and a selector for calculating a first reflectance based on a difference in spectral intensity between the first normalized wavelength and the first evaluation wavelength selected by the selector; and determining whether the railway support is in a sunny installation environment based on the calculated first reflectance. Further, a second reflectance is calculated based on a difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength, and the railway support is set as a shaded installation environment based on the calculated second reflectance. The corrosion detection device is characterized by comprising an evaluation unit for evaluating the corrosion.

[0011] Second InventionThe railway support evaluation device according to the present invention is First Invention In the present invention, the spectral image data acquired by the acquisition unit includes at least one of a panoramic image, a close-up image, and a mirror-illuminated image of the railway support.

[0012] Third Invention The railway support evaluation device according to the first invention or the second invention In the present invention, the acquisition unit further acquires part information that identifies the parts that make up the railway support, and the evaluation unit evaluates the state of corrosion in the installation environment of the railway support, which includes at least one of a sunny area and a shady area, based on some or all of the part information acquired by the acquisition unit.

[0013] Fourth Invention The railway support evaluation device according to the present invention is Third Invention In the present invention, the evaluation unit further includes a corrosion level assignment unit that assigns a corrosion level indicating the degree of corrosion from the distribution of pixels for each portion of the railway support structure based on the evaluation results of the spectral image data.

[0014] Fifth Invention The railway support evaluation system according to the present invention is a railway support evaluation system for evaluating corrosion of a railway support, the railway support evaluation system comprising: an acquisition means for acquiring spectral image data of an image of the railway support; and, based on the spectral image data acquired by the acquisition means, determining a specific wavelength included in a frequency band of 500 to 600 nm as a first normalized wavelength and a specific wavelength included in a frequency band of 600 to 800 nm as a first evaluation wavelength. Further, a specific frequency in the 615 to 715 nm frequency band is selected as the second normalized wavelength, and a specific frequency in the 750 to 850 nm frequency band is selected as the second evaluation wavelength. a selection means for selecting the first normalized wavelength and the first evaluation wavelength selected by the selection means, calculating a first reflectance based on a difference in spectral intensity between the first normalized wavelength and the first evaluation wavelength, and determining whether the railway support is in a sunny installation environment based on the calculated first reflectance. Further, a second reflectance is calculated based on a difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength, and the railway support is set as a shaded installation environment based on the calculated second reflectance. and an evaluation means for evaluating the corrosion.

[0016] Sixth InventionThe railway support evaluation method according to the present invention is a railway support evaluation method for evaluating corrosion of a railway support, the method comprising: acquiring spectral image data of an image of the railway support; and determining, based on the spectral image data acquired by the acquiring step, a specific wavelength included in a frequency band of 500 to 600 nm as a first normalized wavelength and a specific wavelength included in a frequency band of 600 to 800 nm as a first evaluation wavelength. Further, a specific frequency in the 615 to 715 nm frequency band is selected as the second normalized wavelength, and a specific frequency in the 750 to 850 nm frequency band is selected as the second evaluation wavelength. a selecting step of selecting a first normalized wavelength and a first evaluation wavelength, the first normalized wavelength being selected by the selecting step, the first normalized wavelength being the first evaluation wavelength, and a first reflectance being calculated based on the calculated first reflectance, the railway support being set as an installation environment in a sunny spot, Further, a second reflectance is calculated based on a difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength, and the railway support is set as a shaded installation environment based on the calculated second reflectance. and an evaluation step of evaluating the corrosion. [Effects of the Invention]

[0018] First Invention According to the present invention, the acquisition unit acquires spectral image data of railway supports. This allows spectral information from a wide frequency band to be selected, thereby improving the accuracy of evaluating corrosion of outdoor railway supports.

[0019] Also, the first invention According to the method, the selector selects a specific wavelength included in a predetermined frequency band as a normalized wavelength and a specific wavelength included in the predetermined frequency band as an evaluation wavelength. This allows the characteristics of the captured spectral image data to be separated. This improves the accuracy of evaluating corrosion of outdoor railway supports.

[0020] especially, First Invention According to the method, the selector selects a specific wavelength in the 500-600 nm frequency band as the first normalized wavelength and a specific wavelength in the 600-800 nm frequency band as the first evaluation wavelength. This allows the characteristics of sunny conditions to be separated from the captured spectral image data. This improves the accuracy of evaluating corrosion of railway supports in sunny outdoor locations.

[0021] Also First Invention According to the method, the evaluation unit calculates the first reflectance based on the difference in spectral intensity between the first normalized wavelength and the first evaluation wavelength. Therefore, based on the calculated first reflectance, corrosion of the railway support can be evaluated assuming that the railway support is installed in a sunny environment. This makes it possible to improve the accuracy of evaluating corrosion of railway supports installed in a sunny environment outdoors.

[0022] First Invention According to the method, the selector selects a specific frequency in the 615 to 715 nm frequency band as the second normalized wavelength and a specific frequency in the 750 to 850 nm frequency band as the second evaluation wavelength. This allows the characteristics of shaded conditions to be separated from the captured spectral image data. This improves the accuracy of evaluating corrosion of shaded railway supports outdoors.

[0023] Also First Invention According to the method, the evaluation unit calculates the second reflectance based on the difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength. Therefore, based on the calculated second reflectance, corrosion of the railway support can be evaluated assuming that the support is installed in a shaded environment. This makes it possible to improve the accuracy of evaluating corrosion of a shaded railway support outdoors.

[0024] Second Invention According to the method, the acquisition unit includes at least one of a panoramic image, a close-up image, and a mirror-illuminated image of the railway support. This allows for evaluation that takes into account the characteristics that vary depending on the installation location and scale in the outdoor environment, the structure of the railway support, and the weather and solar radiation conditions. This further improves the accuracy of evaluation of corrosion of railway supports.

[0025] Third Invention According to the method, the acquisition unit further acquires part information that identifies parts that make up the railway support. Therefore, the evaluation unit can evaluate the state of corrosion in the installation environment of the railway support, including at least one of a sunny area and a shaded area, based on part or all of the part information. This makes it possible to further improve the accuracy of the evaluation of corrosion of the railway support.

[0026] Fourth Invention According to the present invention, the evaluation unit further includes a corrosion level assigning unit. Therefore, based on the evaluation results of the spectral image data, a corrosion level indicating the degree of corrosion can be assigned from the pixel distribution for each part of the railway support. This makes it possible to further improve the accuracy of evaluating the corrosion of the railway support.

[0027] Fifth Invention According to the present invention, the acquisition means acquires spectral image data of an image of a railway support. This allows spectral information of a wide frequency band to be selected. This improves the accuracy of evaluating corrosion of outdoor railway supports.

[0028] especially, Fifth Invention According to the method, the selection means selects a specific wavelength in the 500-600 nm frequency band as the first normalized wavelength and a specific wavelength in the 600-800 nm frequency band as the first evaluation wavelength. This allows the characteristics of sunny conditions to be separated from the captured spectral image data. This makes it possible to improve the accuracy of evaluating corrosion of railway supports in sunny outdoor locations.

[0029] especially, Fifth Invention According to the method, the selecting means selects a specific frequency in the 615 to 715 nm frequency band as the second normalized wavelength and a specific frequency in the 750 to 850 nm frequency band as the second evaluation wavelength. This allows the characteristics of shaded conditions to be separated from the captured spectral image data. This makes it possible to improve the accuracy of evaluating corrosion of shaded railway supports outdoors.

[0030] Sixth Invention According to the method, the acquisition step acquires spectral image data of an image of a railway support. This allows spectral information of a wide frequency band to be selected. This improves the accuracy of evaluating corrosion of outdoor railway supports.

[0031] especially, Sixth Invention According to the method, the selection step selects a specific wavelength in the 500-600 nm frequency band as the first normalized wavelength and a specific wavelength in the 600-800 nm frequency band as the first evaluation wavelength. This allows the characteristics of sunny conditions to be separated from the captured spectral image data. This makes it possible to improve the accuracy of evaluating corrosion of railway supports in sunny outdoor locations.

[0032] especially, Sixth Invention According to the method, the selecting step selects a specific frequency in the 615 to 715 nm frequency band as the second normalized wavelength and a specific frequency in the 750 to 850 nm frequency band as the second evaluation wavelength. This allows the characteristics of shaded conditions to be separated from the captured spectral image data. This makes it possible to improve the accuracy of evaluating corrosion of shaded railway supports outdoors. [Brief explanation of the drawings]

[0033] [Figure 1] FIG. 1 is a schematic diagram showing an example of a railway support structure evaluation system according to this embodiment. [Figure 2] FIG. 2 is a schematic diagram showing an example of the operation of the railway support structure evaluation system in this embodiment. [Figure 3] FIG. 3(a) is a schematic diagram showing an example of the configuration of a railway support structure evaluation device in this embodiment, and FIG. 3(b) is a schematic diagram showing an example of the function of the railway support structure evaluation device in this embodiment. [Figure 4] FIG. 4(a) is a schematic diagram showing an example of spectral image data in this embodiment, and FIG. 4(b) is a schematic diagram showing an example of a normalized first spectrograph in this embodiment. [Figure 5] FIG. 5(a) is a schematic diagram showing an example of spectral image data in this embodiment, and FIG. 5(b) is a schematic diagram showing an example of a normalized second spectrograph in this embodiment. [Figure 6] FIG. 6 is a flowchart showing an example of the operation of the railway support structure evaluation system in this embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0034] Hereinafter, an example of a railway support evaluation device, a railway support evaluation system, and a railway support evaluation method according to an embodiment of the present invention will be described with reference to the drawings.

[0035] An example of a railway support structure evaluation system 100 according to this embodiment will be described with reference to FIG. (First embodiment: railway support evaluation system 100) An example of the configuration of a railway support structure evaluation system 100 in the first embodiment will be described with reference to Figures 1 to 6. Figure 1 is a schematic diagram showing an example of the configuration of a railway support structure evaluation system 100 in the first embodiment.

[0036] 1, the railway support evaluation system 100 in this embodiment includes a railway support evaluation device 1. The railway support evaluation device 1 is connected to, for example, an imaging device 2 (imaging devices 2a, 2b, imaging device 2c, etc.), and may also be connected to another terminal 5 or a server 6 via, for example, a communication network 4.

[0037] The railway support evaluation system 100 evaluates corrosion of a railway support 3 installed, for example, at an outdoor installation location 40 (for example, XX Line, XX Town, Main Route 10). The railway support evaluation system 100 can be used, for example, for post-work confirmation evaluation performed after inspecting the railway support 3 (for example, a "construction confirmation inspection" performed by an inspector), as well as for pre-evaluation performed before inspecting the railway support 3.

[0038] In the railway support evaluation system 100, as shown in FIG. 2, an image of a railway support 3 installed at an outdoor installation location 40 is captured using an imaging device 2a or the like. The railway support evaluation system 100 acquires, for example, image data generated by the imaging device 2a or spectral image data formed of numerous continuous wavelength elements via the railway support evaluation device 1. The railway support evaluation system 100 acquires, for example, spectral image data of the railway support 3 captured by the imaging device 2 using the railway support evaluation device 1, and stores the acquired spectral image data in the server 6, the railway support evaluation device 1, or the like.

[0039] The railway support evaluation device 1 refers to the acquired spectral image data, selects a normalized wavelength and an evaluation wavelength included in the frequency band of the spectral image data, and calculates the reflectance based on the difference in spectral intensity between the selected normalized wavelength and the evaluation wavelength. The railway support evaluation device 1 evaluates the corrosion state of an outdoor railway support 3 in both sunny and shaded areas based on the calculated reflectance, for example. The railway support evaluation device 1 outputs the evaluation results, etc. to a display unit 109, for example.

[0040] The railway support evaluation device 1 displays, for example, the contents of the evaluation results on a display screen 1a. The display screen 1a displays, for example, the location of the installation location 40, the name of the line, and the like, as well as, for example, a captured image of the railway support 3 and an image showing the state of corrosion. The railway support evaluation device 1 displays, for example, countermeasure information 1c for the state of corrosion selected by a designation unit 1b displayed on the display screen 1a. The display screen 1a may be configured to display, for example, the captured image of the railway support 3, a corresponding image showing the evaluation result of corrosion, and various other related information, each associated with the location where the inspection was performed, such as "XX Line, XX Town, Main No. 10."

[0041] The railway support evaluation device 1 may, for example, associate the parts that make up the railway support 3 with the degree of surface corrosion, and display measures to be taken in response to the evaluation results of corrosion on the railway support 3, such as "Part A: No abnormalities" or "Part B: Painting required." The railway support evaluation device 1 may also display other evaluation results, such as the degree of corrosion occurring on the surface of the railway support 3, such as "corrosion level 1" or "corrosion level 1 80%," or the probability that corrosion factors have occurred.

[0042] In the railway support evaluation system 100, for example, an evaluator may directly input data into the railway support evaluation device 1, or spectral image data to be evaluated may be transmitted from another terminal 5, in addition to the railway support evaluation device 1 and the imaging device 2.

[0043] As the railway support evaluation device 1, electronic devices such as a personal computer (PC) can be used, as well as electronic devices such as smartphones, tablet devices, wearable devices, IoT (Internet of Things) devices, and single-board computers such as Raspberry Pi (registered trademark), and may have an imaging device 2 built in.

[0044] The railway support evaluation system 100 may also use a head mounted display (HMD) with a built-in imaging device 2. In this case, an evaluator or the like can visually check the railway support 3 at the installation location 40 through the display and recognize the evaluation results of the railway support 3. This reduces the difficulty of the corrosion evaluation work for the railway support 3 and also shortens the evaluation work time.

[0045] For example, when one railway support 3 is formed of multiple members or structures, the railway support evaluation device 1 may separately generate spectral image data formed with numerous wavelength elements. For example, the railway support evaluation device 1 may display evaluation results for each member, structure, or other component that constitutes the railway support 3 by specifying some or all of the components that constitute the railway support 3. Furthermore, when generating evaluation results for one railway support 3, for example, the railway support evaluation device 1 may use multiple spectral image data formed with numerous wavelength elements.

[0046] The railway support evaluation device 1 may acquire spectral image data including at least one of a panoramic image of the railway support 3 captured by the imaging device 2, a close-up image, and a mirror image illuminated by a mirror. The railway support evaluation device 1 can evaluate the corrosion status of the railway support 3 outdoors in both sunny and shady locations based on the acquired various spectral image data such as the panoramic image, close-up image, and mirror image. This enables the railway support evaluation device 1 to improve evaluation accuracy.

[0047] The railway support evaluation device 1 may acquire, in addition to spectral image data relating to the railway support 3, information such as the installation location 40, setting information relating to the captured images and corrosion evaluation, location information of the evaluation range, and the photographer in text format or other formats.

[0048] The image data acquired by the railway support evaluation device 1 may be, for example, spectral image data, hyperspectral data, multispectral image data with any wavelength selected, or target spectral image data limited to a specific wavelength.

[0049] The railway support structure evaluation device 1 may also be configured to acquire various types of information related to, for example, spectral image data, etc. The railway support structure evaluation device 1 may also be configured to acquire, as installation information, for example, solar radiation conditions such as sunny or shaded in the area or location where the evaluation of the railway support structure 3 where the spectral image data was captured is carried out.

[0050] The installation information may include, for example, weather information related to the weather in the area where the evaluation is performed, such as sunny, cloudy, or rainy, as well as solar radiation information such as direct sunlight, skylight, ground reflected light, and mirror irradiation. By acquiring installation information including various information related to, for example, spectral image data, the railway support evaluation device 1 can achieve highly accurate corrosion evaluation that takes into account, for example, solar radiation and seasonal characteristics.

[0051] The railway support evaluation device 1 may include inspection information in the installation information, such as the date, temperature, inspection target area, and inspector when the railway support 3 imaged in the spectral image data was inspected, as well as the planned date for inspecting the railway support 3.

[0052] The inspection information is, for example, information about the internal materials and installation of the railway support structure 3, such as the material and the number of years since installation.

[0053] (Railway support evaluation device 1) Next, an example of the railway support structure evaluation device 1 in this embodiment will be described with reference to Fig. 3. Fig. 3(a) is a schematic diagram showing an example of the configuration of the railway support structure evaluation device 1 in this embodiment, and Fig. 3(b) is a schematic diagram showing an example of the function of the railway support structure evaluation device 1 in this embodiment.

[0054] 3(a), the railway support structure evaluation device 1 includes a housing 10, a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, a storage unit 104, and I / Fs 105 to 107. The components 101 to 107 are connected by an internal bus 110.

[0055] The CPU 101 controls the entire railway support structure evaluation device 1. The ROM 102 stores operation code for the CPU 101. The RAM 103 is a work area used when the CPU 101 is operating. The storage unit 104 stores various information such as various installation information, information on the evaluation target, and a reference database. As the storage unit 104, for example, a data storage device such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), an SD card, or a miniSD card is used. Note that the railway support structure evaluation device 1 may also have a GPU (Graphics Processing Unit) (not shown). The inclusion of a GPU enables faster calculation processing than usual.

[0056] The I / F 105 is an interface for transmitting and receiving various information to and from the imaging device 2, and may also be an interface for transmitting and receiving various information to and from other terminals 5, servers 6, etc. via a communication network 4 such as the Internet. The I / F 106 is an interface for transmitting and receiving information to and from the input unit 108. A keyboard, for example, is used as the input unit 108, and an evaluator using the railway support structure evaluation device 1 inputs various information or control commands for the railway support structure evaluation device 1 via the input unit 108. The I / F 107 is an interface for transmitting and receiving various information to and from the display unit 109. The display unit 109 outputs various information such as evaluation results stored in the storage unit 104, or the processing status of the railway support structure evaluation device 1. A display, for example, a touch panel type, is used as the display unit 109.

[0057] <Reference database> The reference database stored in the storage unit 104 stores, for example, spectral image data (e.g., still images, videos, etc.) of railway support structures 3 acquired from an imaging device 2, as well as installation information including inspection information related to the spectral image data, corrosion level information, evaluation results, evaluation algorithms, etc.

[0058] Here, an example of the spectral image data and an example of a normalized spectrum graph in this embodiment will be described with reference to FIGS.

[0059] Fig. 4(a) is a schematic diagram showing an example of spectral image data in this embodiment, Fig. 4(b) is a schematic diagram showing an example of a normalized first spectrograph in this embodiment, Fig. 5(a) is a schematic diagram showing an example of spectral image data in this embodiment, and Fig. 5(b) is a schematic diagram showing an example of a normalized second spectrograph in this embodiment.

[0060] The imaging device 2 captures spectral image data of multiple railway supports 3, for example, at an installation location. Figures 4(a) and 5(a) show multiple spectrographs of the spectral image data of multiple railway supports 3 captured by the imaging device 2. Each of the solid and dashed lines in the multiple spectrographs corresponds to, for example, the spectral images of the captured multiple railway supports 3. The spectrographs show graphs of the spectral image data acquired by, for example, the railway support evaluation device 1 before being normalized.

[0061] For example, before normalization, spectral graphs tend to have little difference between each other, making it difficult to extract differences (features) between sunny and shaded areas.

[0062] For this reason, the railway support evaluation device 1 selects a normalized wavelength and an evaluation wavelength from the frequency band of the spectrograph based on the acquired spectral image data, calculates the reflectance from the difference in spectral intensity between the normalized wavelength and the evaluation wavelength, and evaluates corrosion of the railway support 3 in a sunny or shady installation environment based on the calculated reflectance. This makes it possible to distinguish between the corrosion conditions in sunny and shady areas, which are difficult to distinguish outdoors, and to clarify the differences in corrosion.

[0063] Here, for example, the frequency pair to be selected is calculated based on a certain tendency of characteristic difference. The frequency pair to be selected is calculated based on the spectral change due to the difference (slope) of the spectral intensity (spectral intensity) between, for example, two wavelengths (λ1, λ2). In this case, by normalizing (dividing) by the sum of the two wavelengths, even if the conditions are different, for example, uneven lighting or shadows, the influence of these can be reduced and the characteristics can be compared. These calculations are obtained by the following formula using, for example, a known spectrum measurement method or spectrum analysis technique (for example, "NDSI: normalized difference spectral index"): "I λ " is, for example, "λ nm ", the normalized wavelength is "λ2" and the evaluation wavelength is "λ1".

[0064]

number

[0065] Figures 4(b) and 5(b) show normalized spectral graphs. Figure 4(b) is a normalized spectral graph of, for example, a railway support 3 in the sun and a railway support 3 illuminated by a mirror in the sun, while Figure 5(b) shows normalized spectral data of, for example, a railway support 3 in the shade at two locations. In other words, by identifying each normalized wavelength and evaluation wavelength in the normalized spectral graphs and creating a normalized spectral graph (spectral intensity) within that range, it is possible to evaluate corrosion on other railway supports 3.

[0066] Any frequency in the frequency band included in the spectrograph that can evaluate the corrosion of such railway support structure 3 is selected as a specific wavelength, and is then selected as a normalized wavelength (e.g., first normalized wavelength, second normalized wavelength) and an evaluation wavelength (e.g., first evaluation wavelength, second evaluation wavelength).

[0067] The normalized wavelength and evaluation wavelength are selected based on the difference in each frequency in the normalized spectrum graph shown in Figure 4(b), for example, 500 to 600 nm is selected as the target of the first normalized wavelength of reflected light in sunlight, and a clearly different evaluation wavelength of 600 to 800 nm is selected.Furthermore, based on the difference in each frequency in the normalized spectrum graph shown in Figure 5(b), for example, 615 to 715 nm is selected as the target of the second normalized wavelength of reflected light in the shade, and a second evaluation wavelength of 750 to 850 nm is selected.

[0068] By focusing on the normalized wavelengths (e.g., first normalized wavelength, second normalized wavelength) and evaluation wavelengths (e.g., first evaluation wavelength, second evaluation wavelength), it is possible to evaluate the corrosion levels in sunny and shady areas separately, enabling more accurate analysis.

[0069] In this embodiment, the first normalized wavelength (500 to 600 nm) and evaluation wavelength (600 to 800 nm) are selected based on the reflected light in sunlight shown in FIG. 4(b), and the second normalized wavelength (615 to 715 nm) and second evaluation wavelength (750 to 850 nm) are selected as specific wavelengths for the reflected light in the shade shown in FIG. 5(b). Here, for example, the specific wavelength may be specified as one point or multiple points. The specific wavelength may be determined, for example, as the center wavelength of the wavelength range of each frequency band, or as the wavelength at which the difference in spectral intensity between each spectrograph is largest.

[0070] Alternatively, a specific wavelength may be identified as a specific point where a convex peak is formed on each normalized spectrum graph. The selection of this specific wavelength may vary for each inspection target performed by the evaluator, depending on, for example, the location 40 of the railway support 3, the date and time, the weather, the season, the part, the material or paint of the part, etc.

[0071] Alternatively, a specific wavelength range may be set around these specific wavelengths. The specific wavelength range may be a predetermined wavelength range set in advance, such as ±10 nm. For example, if the specific wavelength is 550 nm and the specific wavelength range is ±10 nm, the range in which the spectrum data is actually detected will be 540 to 560 nm. This specific wavelength range may differ depending on the purpose of evaluation of the railway support 3.

[0072] The reference database stores such specified normalized wavelengths (e.g., first normalized wavelength, second normalized wavelength) and specified evaluation wavelengths (e.g., first evaluation wavelength, second evaluation wavelength) as, for example, a detection algorithm. Furthermore, other specified wavelengths, ranges of specified wavelengths, and in some cases, calculation methods and the calculation formulas themselves that define them, are stored in association with each evaluation purpose of the railway support 3.

[0073] Furthermore, for example, an evaluation based on a detection algorithm using a normalized wavelength of 500 to 600 nm and an evaluation wavelength of 600 to 800 nm may be referred to as a "sunny level," and an evaluation based on a detection algorithm using a normalized wavelength of 615 to 715 nm and an evaluation wavelength of 750 to 850 nm may be referred to as a "shade level," and evaluation results may be assigned in stages.

[0074] Furthermore, the corrosion level indicating the degree of corrosion of the railway support 3 may be expressed as a specific level of corrosion such as levels 1 to 5, depending on the progress and extent of corrosion of the zinc plating per area, peeling of the surface paint, discoloration, etc. The corrosion level may be calculated from the distribution of pixels for each part of the railway support 3 based on the evaluation results of the spectral image data, for example, and may be assigned as a corrosion level indicating the degree of corrosion.

[0075] As a criterion for determining the corrosion level, for example, based on the evaluation results of the spectral image data, the pixels of each evaluation target constituting the railway support 3 may be classified by each corrosion level information and the ratio may be calculated. Note that in this evaluation, in order to take into account the difference in level for each part within the support, the support may be divided into three areas and the number of pixels of each level may be calculated in each area.

[0076] Fig. 3(b) is a schematic diagram showing an example of the functions of the railway support structure evaluation device 1. The railway support structure evaluation device 1 includes an acquisition unit 11, a selection unit 12, an evaluation unit 13, a storage unit 14, and an output unit 15. Note that each function shown in Fig. 3(b) is realized by the CPU 101 using the RAM 103 as a work area to execute a program stored in the storage unit 104 or the like, and may be controlled by, for example, artificial intelligence.

[0077] <Acquisition part 11> The acquisition unit 11 acquires spectral image data of an image of the railway support 3. The acquisition unit 11 acquires spectral image data from the imaging device 2 or the like, and may also acquire spectral image data, for example, from the imaging device 2 built into the railway support evaluation device 1. The acquisition unit 11 acquires installation information, etc. input in advance by, for example, an evaluator (or inspector), and also acquires installation information, for example, from the imaging device 2 or the like, including inspection information regarding the evaluation location to be evaluated of the railway support 3 corresponding to the spectral image data or the installation environment. The frequency and period at which the acquisition unit 11 acquires the spectral image data and various related information, etc., are arbitrary.

[0078] The acquisition unit 11 also acquires spectral image data including at least one of a panoramic image, a close-up image, and a mirror-illuminated image of the railway support 3 captured by the imaging device 2. The acquisition unit 11 may acquire information such as model information of the imaging device 2 that captured the spectral image data, the imaging date and time, location information (latitude and longitude), and imaging setting information of the imaging device 2.

[0079] The acquisition unit 11 may also acquire part information that identifies parts that make up the railway support 3. The part information may be acquired, for example, before or after capturing an image by the imaging device 2.

[0080] The acquisition unit 11 may also acquire, for example, various types of information transmitted to the railway support evaluation device 1. The acquisition unit 11 may also receive, for example, various types of installation information including spectral image data, inspection information, etc. transmitted from an external terminal such as another terminal 5 via the communication network 4 and the I / F 105.

[0081] <Selection section 12> The selector 12 sets a specific wavelength included in a predetermined frequency band as a normalized wavelength and selects a characteristic wavelength included in the predetermined frequency band as an evaluation wavelength based on the spectral image data acquired by the acquirer 11. For example, the selector 12 sets a specific wavelength included in a frequency band of 500 to 600 nm as a first normalized wavelength and selects a characteristic wavelength included in a frequency band of 600 to 800 nm as a first evaluation wavelength.

[0082] Selection section 12 Alternatively, the selected first normalized wavelength (500 to 600 nm) and the first evaluation wavelength (600 to 800 nm) may be used as a detection algorithm (known specific frequency band) for evaluating corrosion in a sunny installation environment of the railway support structure 3, for example.

[0083] Furthermore, based on the spectral image data acquired by the acquisition unit 11, the selection unit 12 selects, for example, a specific wavelength included in the frequency band of 615 to 715 nm as the second normalized wavelength, and a characteristic wavelength included in the frequency band of 750 to 850 nm as the second evaluation wavelength.

[0084] Selection section 12 Alternatively, the selected second normalized wavelength (615 to 715 nm) and the first evaluation wavelength (750 to 850 nm) may be used as a detection algorithm (known specific frequency band) for evaluating corrosion in a shaded installation environment of, for example, a railway support structure 3.

[0085] <Evaluation Section 13> The evaluation unit 13 evaluates the spectral image data acquired by the acquisition unit 11 by calculating a first reflectance between the first normalized wavelength selected by the selection unit 12 and the first evaluation wavelength based on the difference in spectral intensity between the first normalized wavelength and the first evaluation wavelength, and evaluates corrosion of the railway support 3 in a sunny installation environment based on the calculated first reflectance.

[0086] In addition, the evaluation unit 13 calculates a second reflectance between the second normalized wavelength selected by the selection unit 12 and the second evaluation wavelength based on the difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength, and evaluates corrosion of the railway support 3 in a shady installation environment based on the calculated second reflectance.

[0087] The evaluation unit 13 may evaluate the spectral image data acquired by the acquisition unit 11 based on, for example, known normalized wavelengths and evaluation wavelengths stored as a detection algorithm, by referring to a reference database or the storage unit 104. The evaluation unit 13 may perform evaluation using, for example, a search algorithm, data such as evaluation parameters and setting information, or known techniques.

[0088] <Corrosion level assignment section> The evaluation unit 13 further includes, for example, a corrosion level assigning unit that assigns a corrosion level indicating the degree of corrosion from the distribution of pixels for each portion of the railway support 3 based on, for example, the evaluation result of the spectral image data.

[0089] <Storage section 14> The storage unit 14 retrieves, as necessary, various pieces of information such as the reference database stored in the storage unit 104. The storage unit 12 stores in the storage unit 104 various pieces of information acquired or generated by the components 11, 13 to 15.

[0090] <Output section 15> The output unit 15 outputs the evaluation result. The output unit 15 transmits the evaluation result to the display unit 109 via the I / F 107, and also transmits the evaluation result to another terminal 5 or the like via, for example, the I / F 105. The output unit 15 outputs data displaying an image that reproduces the surface condition of the railway support 3 shown in FIG. 2 to the display unit 109 or the like.

[0091] <Display section 109> The display unit 109 displays the evaluation results. For example, as shown in FIG. 2, the display unit 109 displays a display image 1a, a designation unit 1b, and the evaluation results. The display image 1a shows an image of the railway support structure 3 based on the spectral image data acquired by the acquisition unit 11. The designation unit 1b designates an evaluation location of the spectral image data of the railway support structure 3 to be evaluated in the display image 1a (for example, the right side of the frame in FIG. 2 is shown).

[0092] The evaluation results are displayed via the designation unit 1b, for example, for the railway support structure 3 to be evaluated. This makes it possible to display the evaluation results for each railway support structure 3, even when evaluating multiple locations that make up the railway support structure 3 at once. Note that the display unit 109 may display the evaluation results using, for example, only a list or character strings. Known technologies can be used for the above display method.

[0093] For example, when an HMD is used as the railway support structure evaluation system 1, a transmissive display is used as the display unit 109. In this case, the display unit 109 can display the designation unit 1b and the evaluation result on the railway support structure 3 that is visually recognized by, for example, an evaluator through the display unit 109.

[0094] <Imaging device 2> The imaging device 2 is a known camera that captures an image of the railway support 3 and generates spectral image data. For example, an RGB camera, a multispectral camera, a target spectral camera, a hyperspectral camera, or the like may be used as the imaging device 2, and multiple cameras may be used. The imaging device 2 may be, for example, a video camera that captures video, and may be built into the railway support evaluation device 1. If the imaging device 2 is a spectral video camera, for example, the spectral image data may be extracted from a portion of the captured video.

[0095] <Communication Network 4> The communication network 4 is, for example, an internet network to which the railway support structure evaluation system 100 and the like are connected via communication circuits. The communication network 4 may be configured as a so-called optical fiber communication network. Furthermore, the communication network 4 may be realized by a known communication network such as a wired communication network or a wireless communication network.

[0096] <Other terminal 5> The other terminal 5 may be, for example, an electronic device embodied in the same way as the railway support evaluation device 1. The other terminal 5 may, for example, represent a central control device or the like that is capable of communicating with the multiple railway support evaluation devices 1 that make up the railway support evaluation system 100. The other terminal 5 may, for example, be connectable to multiple railway support evaluation devices 1, and may acquire the evaluation results generated by each of the railway support evaluation devices 1. This makes it possible to analyze the evaluation results of railway supports 3 produced at multiple locations, for example, and to improve corrosion of the railway supports 3.

[0097] <Server 6> The server 6 stores, for example, the various types of information described above. The server 6 accumulates, for example, various types of information sent via the communication network 4. The server 6 may store, for example, information similar to that stored in the storage unit 104, and may transmit and receive various types of information to and from one or more railway support structure evaluation devices 1 via the communication network 4. That is, the railway support structure evaluation device 1 may use the server 6 instead of the storage unit 104.

[0098] (An example of the operation of the railway support evaluation device 1) Next, a description will be given of an example of the operation of the railway support evaluation device 1 in this embodiment. Fig. 6 is a flowchart showing an example of the operation of the railway support evaluation device 1 in this embodiment.

[0099] <Acquisition means S110> 6, the acquisition unit 11 acquires spectral image data of the railway support 3 (acquisition means S110). The acquisition unit 11 acquires, for example, spectral image data of the railway support 3 captured by the imaging device 2 as information to be evaluated. The acquisition unit 11 stores the spectral image data in the storage unit 104 via, for example, the memory unit 12.

[0100] The acquisition unit 11 acquires, for example, information about the installation location 40, installation information including various inspection information related to the evaluation and inspection of the railway support 3, evaluation target information related to the spectral image data, etc. The various pieces of information acquired by the acquisition unit 11 may be input to the railway support evaluation device 1 by an evaluator, inspector, etc. in advance so as to be linked to the spectral image data, or the acquisition unit 11 may select each piece of information suitable for the spectral image data based on the spectral image data, for example. In this case, the acquisition unit 11 acquires each piece of information suitable for the spectral image data from the installation information including various inspection information, etc., stored in advance in the storage unit 104.

[0101] For example, when one railway support 3 is imaged using multiple imaging devices 2, the acquisition unit 11 acquires multiple spectral image data generated by the multiple imaging devices 2 as one piece of evaluation target information. The acquisition unit 11 may acquire, for example, related evaluation target information, etc., each time spectral image data is generated, or may receive spectral image data generated during an arbitrary period at once to acquire the spectral image data.

[0102] The acquisition unit 11 also acquires spectral image data such as a panoramic image, a close-up image, or a mirror-illuminated image of the railway support 3. The acquisition unit 11 may acquire multiple pieces of spectral image data of the railway support 3 captured by the imaging device 2 at once.

[0103] The acquisition unit 11 also acquires part information that identifies parts that make up the railway support 3. The spectral image data and various related information, data, and images acquired by the acquisition unit 11 are stored in the storage unit 104, for example.

[0104] <Selection Means S120> Next, based on the spectral image data acquired by the acquisition unit 11, a normalized wavelength and an evaluation wavelength included in the frequency band of the spectrograph are selected (selection means S120).

[0105] The selector 12 sets a specific wavelength included in a predetermined frequency band as a normalized wavelength and selects a characteristic wavelength included in the predetermined frequency band as an evaluation wavelength based on the spectral image data acquired by the acquirer 11. Note that the selector 12 may set a specific wavelength included in a frequency band of, for example, 500 to 600 nm as a first normalized wavelength and select a characteristic wavelength included in a frequency band of, for example, 600 to 800 nm as the first evaluation wavelength.

[0106] Furthermore, based on the spectral image data acquired by the acquisition unit 11, the selection unit 12 may select, for example, a specific frequency included in a frequency band of 615 to 715 nm as the second normalized wavelength, and a specific frequency included in a frequency band of 750 to 850 nm as the second evaluation wavelength.

[0107] The normalized wavelength selected by the selection unit 12 may be selected in advance as, for example, a known normalized wavelength, a detection algorithm, etc. The selection unit 12 selects an evaluation wavelength to be evaluated from a range of frequency bands based on, for example, the known normalized wavelength, the detection algorithm, etc. <Evaluation Means S130> Next, the reflectance is calculated based on the difference in spectral intensity between the normalized wavelength selected by the selection unit and the evaluation wavelength, and corrosion of the railway support 3 is evaluated (evaluation means S130). The evaluation unit 13 refers to a reference database, for example, and evaluates corrosion of the railway support 3 in a sunny or shady installation environment.

[0108] The evaluation unit 13 calculates a first reflectance based on the difference in spectral intensity between the first normalized wavelength and the first evaluation wavelength selected by the selection unit 12. The evaluation unit 13 evaluates corrosion of the railway support 3 in a sunny installation environment based on the calculated first reflectance.

[0109] Furthermore, the evaluation unit 13 calculates a second reflectance based on the difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength, for example, between the second normalized wavelength and the second evaluation wavelength selected by the selection unit 12. Based on the calculated second reflectance, the evaluation unit 13 evaluates corrosion of the railway support 3 in a shaded installation environment.

[0110] The evaluation unit 13 also evaluates the state of corrosion for each range (area) constituting the railway support 3 based on the part information that identifies the parts constituting the railway support 3 acquired by the acquisition unit 11. The evaluation unit 13 evaluates the state of corrosion in the installation environment of the railway support 3, which includes at least one of a sunny area and a shaded area, based on some or all of the part information of the railway support 3.

[0111] Furthermore, the evaluation unit 13 further includes a corrosion level assigning unit that assigns a corrosion level indicating the degree of corrosion from the distribution of pixels for each part of the railway support 3, for example, based on the evaluation result of the spectral image data.

[0112] The evaluation unit 13 may generate one evaluation result for, for example, one railway support 3 as the evaluation target, or may generate one evaluation result for, for example, multiple railway supports 3 or the parts or range that make up a railway support 3 as the evaluation target. The evaluation unit 13 generates the evaluation result using format data such as an output format stored in the storage unit 104. The evaluation unit 13 stores the evaluation result in the storage unit 104, for example, via the memory unit 12.

[0113] <Output method> Next, the evaluation result is output by the output unit 15 (output means). The output unit 15 outputs the evaluation result to the display unit 109 or the like. The output unit 15 may output the evaluation result to another terminal 5 or a server 6 via the communication network 4, for example.

[0114] The output unit 15 outputs to the display unit 109, for example, a display image 1a showing a railway support structure 3 based on spectral image data, a designation unit 1b that designates the railway support structure 3 to be evaluated in the display image 1a, and information for displaying the evaluation result for the railway support structure 3 to be evaluated via the designation unit 1b. As a result, the display unit 109 displays the display image 1a, the designation unit 1b, and the evaluation result.

[0115] This completes the operation of the railway support evaluation device 1 in this embodiment. Note that the timing for performing the update means S140 is arbitrary.

[0116] According to this embodiment, the evaluation unit 13 refers to a reference database and generates evaluation results for the installation information and evaluation target information. The reference information includes corrosion level information. Therefore, it is possible to generate evaluation results based on the results of past evaluations of the corrosion of the railway support 3. This makes it possible to improve the accuracy of evaluating the corrosion of the railway support 3.

[0117] Furthermore, according to this embodiment, the installation information includes various types of inspection information. Therefore, it is possible to realize an evaluation that takes into account the surface condition of the railway support 3, which varies depending on the inspection conditions. This makes it possible to further improve the accuracy of evaluating the corrosion of the railway support 3. Furthermore, since the evaluation results are linked to the inspection conditions, it is possible to identify the conditions that cause corrosion deterioration and understand changes in corrosion that occur when the inspection conditions are changed. This makes it possible to improve the corrosion of the railway support 3.

[0118] Furthermore, according to this embodiment, the display unit 109 displays a designation section 1b for designating the railway support structure 3 to be evaluated. Therefore, even when multiple railway support structures 3 are imaged as a single set of spectral image data, the evaluation results for each railway support structure 3 can be easily known. This makes it possible to easily implement countermeasures and repair plans for the railway support structures 3 based on the evaluation results.

[0119] Furthermore, according to this embodiment, a railway support evaluation method for evaluating corrosion of a railway support can be provided by an acquisition step of performing an acquisition means S110, a selection step of performing a selection means S120, and an evaluation step of performing an evaluation means S130.

[0120] Although the embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the inventions and their equivalents as defined in the claims. [Explanation of symbols]

[0121] 1: Railway support evaluation device 1a:Display screen 1b:Specified part 1c: Troubleshooting information 2: Imaging device 2a: Imaging device 2b: Imaging device 2c: Imaging device 3:Railway support 4: Communication network 5: Other devices 6: Server 10: Housing 11: Acquisition part 12: Selection section 13: Evaluation section 14: Storage section 15: Output section 40: Installation location 100:Railway support evaluation system 101: CPU 102:ROM 103:RAM 104: Preservation Department 105: Interface 106: Interface 107: Interface 108: Input section 109:Display section 110: Internal bus S110: Acquisition means S120: Selection means S130: Evaluation method

Claims

1. A railway support evaluation device for evaluating corrosion of a railway support, an acquisition unit that acquires spectral image data of the railway support; a selector that, based on the spectral image data acquired by the acquirer, selects a specific wavelength included in a frequency band of 500 to 600 nm as a first normalized wavelength, a specific wavelength included in a frequency band of 600 to 800 nm as a first evaluation wavelength, and further selects a specific frequency included in a frequency band of 615 to 715 nm as a second normalized wavelength, and a specific frequency included in a frequency band of 750 to 850 nm as a second evaluation wavelength; an evaluation unit that calculates a first reflectance between the first normalized wavelength and the first evaluation wavelength selected by the selection unit, based on a difference in spectral intensity between the first normalized wavelength and the first evaluation wavelength, and evaluates the corrosion of the railway support as being installed in a sunny environment based on the calculated first reflectance, and calculates a second reflectance between the second normalized wavelength and the second evaluation wavelength, based on a difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength, and evaluates the corrosion of the railway support as being installed in a shaded environment based on the calculated second reflectance; To be prepared A railway support evaluation device characterized by the above.

2. The spectral image data acquired by the acquisition unit includes: The image includes at least one of a panoramic image, a close-up image, and a mirror-illuminated image of the railway support.

2. The railway support evaluation device according to claim 1, wherein:

3. The acquisition unit Further, part information for identifying parts constituting the railway support is acquired; The evaluation unit Evaluating the state of corrosion in an installation environment including at least one of a sunny area and a shaded area of ​​the railway support based on a part or all of the part information acquired by the acquisition unit.

3. The railway support evaluation device according to claim 1 or 2,

4. The evaluation unit The method further comprises a corrosion level assigning unit that assigns a corrosion level indicating the degree of corrosion from the distribution of pixels for each portion of the railway support based on the evaluation result of the spectral image data.

4. The railway support evaluation device according to claim 3, wherein:

5. A railway support evaluation system for evaluating corrosion of a railway support, comprising: an acquisition means for acquiring spectral image data of the railway support; a selecting means for selecting, based on the spectral image data acquired by the acquiring means, a specific wavelength included in a frequency band of 500 to 600 nm as a first normalized wavelength, a specific wavelength included in a frequency band of 600 to 800 nm as a first evaluation wavelength, and further selecting, based on the spectral image data acquired by the acquiring means, a specific frequency included in a frequency band of 615 to 715 nm as a second normalized wavelength, and a specific frequency included in a frequency band of 750 to 850 nm as a second evaluation wavelength; an evaluation means for calculating a first reflectance between the first normalized wavelength selected by the selection means and the first evaluation wavelength based on a difference in spectral intensity between the first normalized wavelength and the first evaluation wavelength, and evaluating the corrosion of the railway support in a sunny installation environment based on the calculated first reflectance, and for calculating a second reflectance between the second normalized wavelength and the second evaluation wavelength based on a difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength, and evaluating the corrosion of the railway support in a shaded installation environment based on the calculated second reflectance; To be prepared A railway support evaluation system characterized by:

6. A railway support evaluation method for evaluating corrosion of a railway support, comprising: an acquisition step of acquiring spectral image data of the railway support; a selection step of selecting, based on the spectral image data acquired by the acquisition step, a specific wavelength included in a frequency band of 500 to 600 nm as a first normalized wavelength, a specific wavelength included in a frequency band of 600 to 800 nm as a first evaluation wavelength, a specific frequency included in a frequency band of 615 to 715 nm as a second normalized wavelength, and a specific frequency included in a frequency band of 750 to 850 nm as a second evaluation wavelength; an evaluation step of calculating a first reflectance between the first normalized wavelength and the first evaluation wavelength selected in the selection step, based on the difference in spectral intensity between the first normalized wavelength and the first evaluation wavelength, and assuming that the railway support is installed in a sunny environment based on the calculated first reflectance; and calculating a second reflectance between the second normalized wavelength and the second evaluation wavelength, based on the difference in spectral intensity between the second normalized wavelength and the second evaluation wavelength, and evaluating the corrosion of the railway support in a shaded installation environment based on the calculated second reflectance. Having A railway support evaluation method characterized by:

Images