Inspection system, inspection method, and inspection program

Abstract

To provide an inspection system, an inspection method, and an inspection program that can inspect for water leakage from a water area without installing new equipment in the water area. [Solution] An inspection system 1 for inspecting for water leakage from a water body having a levee includes a control unit 11. The water body is, for example, a reservoir. The control unit 11 calculates a normalized index value for each pixel constituting a multi-wavelength spectroscopic image of the reservoir photographed from above, normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band. Then, the control unit 11 determines whether the normalized index value of a pixel constituting an inspection area set as an area where water leakage from the reservoir may occur in the multi-wavelength spectroscopic image is equal to or greater than a threshold value.

Description

[Technical Field]

[0001] The present invention relates to an inspection system, an inspection method, and an inspection program for inspecting for water leakage from a water area. [Background technology]

[0002] Water bodies such as reservoirs and rivers have embankments to hold back water. Reservoirs, an example of a water body, are artificially constructed ponds that can store and draw water, primarily for agricultural use. There are approximately 150,000 reservoirs nationwide. Many of these reservoirs were constructed before the Edo period. As a result, some reservoirs do not meet safety standards for disasters such as floods and earthquakes. Therefore, in the event of a disaster, it is desirable to efficiently investigate whether reservoirs in affected areas are leaking in order to prevent secondary damage caused by reservoir collapses. For example, Patent Document 1 discloses a water leakage detection system that determines whether a reservoir is leaking based on the difference between the water level measured by a water level sensor installed in the reservoir to be monitored and a pre-calculated reference water level. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-173310 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology in Patent Document 1 requires the installation of a water level sensor in each water area to be monitored, which requires a great deal of effort for the installation and maintenance of the water level sensors. Therefore, there is a need for a technology that can determine the presence or absence of water leakage from a water area to be monitored without installing new equipment in the water area to be monitored. [Means for solving the problem]

[0005] An inspection system that solves the above problem is an inspection system for inspecting for water leakage from a water body having a levee, and a control unit provided in the inspection system calculates a normalized index value for each pixel constituting a multi-wavelength spectroscopic image of the water body photographed from above, normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band, and determines whether the normalized index value of a pixel constituting an inspection area in the multi-wavelength spectroscopic image, which has been set as an area where water leakage from the water body may occur, is greater than or equal to a threshold value. [Effects of the Invention]

[0006] According to the present invention, it is possible to check for and determine water leakage from a water area without installing new equipment in the water area. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a block diagram of the inspection system. [Figure 2] FIG. 2 is an explanatory diagram of the hardware configuration. [Figure 3] FIG. 3 is a cross-sectional view that schematically shows a reservoir. [Figure 4] FIG. 4 is a cross-sectional view that schematically shows another example of a water leakage pattern in a reservoir. [Figure 5] Figure 5 is a satellite image showing a reservoir. [Figure 6] FIG. 6 is a flowchart showing the steps of the water leakage inspection method. [Figure 7] Figure 7 is an image showing the embankment used in the test example. [Figure 8] Figure 8 shows the satellite image used in the test example. [Figure 9] Figure 9 is a heat map of the satellite image in Figure 8 normalized by NDWI. [Figure 10] FIG. 10 is an enlarged view of the heat map of FIG. 9, showing the area corresponding to the embankment slope shown in FIG. [Figure 11] FIG. 11 is a binarized version of the heat map in FIG. DETAILED DESCRIPTION OF THE INVENTION

[0008] An embodiment of an inspection system, an inspection method, and an inspection program will be described below with reference to Figs. 1 to 11. The inspection system of this embodiment determines whether or not a reservoir is leaking from an optical satellite image of a reservoir, which is an example of a water area. The water area includes a water area, which is an area where water exists, and a levee located around the water area. The water area may be a river or the like having a levee (river levee) other than a reservoir.

[0009] As shown in FIG. 1, the inspection system 1 includes a water leak inspection device 10. The water leak inspection device 10 is a computer terminal used by a user of the inspection system 1. The inspection system 1 also uses a satellite image server 20. As an example, the satellite image server 20 is an external server managed by an organization different from the organization that manages the water leak inspection device 10, but it may also be a server managed by the organization that manages the water leak inspection device 10. The water leak inspection device 10 transmits and receives data to and from the satellite image server 20 via a network line.

[0010] (Configuration example of information processing device) FIG. 2 shows an example of the hardware configuration of an information processing device H10 that functions as the water leakage inspection device 10, the satellite image server 20, and the like.

[0011] The information processing device H10 includes a communication device H11, an input device H12, a display device H13, a storage device H14, and a processor H15. Note that this hardware configuration is an example, and the information processing device H10 may include other hardware.

[0012] The communication device H11 is an interface that executes data transmission and reception by establishing a communication path with another device, and is, for example, a network interface card or a wireless interface.

[0013] The input device H12 is a device that accepts input from a user, etc. The input device H12 is, for example, a mouse, a keyboard, etc. The display device H13 is a display, a touch panel, etc. that displays various information.

[0014] The storage device H14 is a storage device that stores data and various programs for executing various functions of the water leakage inspection device 10 and the satellite image server 20. Examples of the storage device H14 include a ROM, a RAM, and a hard disk.

[0015] The processor H15 uses programs and data stored in the storage device H14 to control each process in the information processing device H10, which functions as the water leak inspection device 10, the satellite image server 20, etc. Examples of the processor H15 include a CPU and an MPU. The processor H15 executes various processes corresponding to various processes by expanding programs stored in a ROM or the like into a RAM. For example, when an application program for the water leak inspection device 10 or the satellite image server 20 is started, the processor H15 operates a process that executes each process described below.

[0016] The processor H15 is not limited to a processor that performs all of its processing using software. For example, the processor H15 may include a dedicated hardware circuit (e.g., an application-specific integrated circuit (ASIC)) that performs hardware processing for at least some of the processing it performs. That is, the processor H15 may be configured as follows:

[0017] [1] One or more processors that operate according to a computer program (software). [2] One or more dedicated hardware circuits that perform at least some of the various processes; or [3] Circuits containing combinations of these The processor includes a CPU and memory such as RAM and ROM. The memory stores program code or instructions configured to cause the CPU to perform processes. Memory, or computer-readable media, includes any available media that can be accessed by a general-purpose or special-purpose computer.

[0018] (Configuration of 100 reservoirs) The configuration of the reservoir 100 in which leakage inspection is performed by the inspection system 1 will be described with reference to FIG.

[0019] As shown in Fig. 3, the reservoir 100 comprises a bank body 101 and a water storage section 102 in which a stream or small river is dammed by the bank body 101. Water for use is stored in the water storage section 102. The bank body 101 is equipped with intake facilities (not shown) for taking in the water for use. Note that the reservoir 100 refers to a structure composed of the bank body 101 and intake facilities as artificially constructed facilities.

[0020] A common cause of water leakage from the reservoir 100 is a phenomenon in which the groundwater level of the seepage water in the embankment 101 rises, causing water to seep out from the slope 101S located downstream of the reservoir 100. Figure 3 shows an infiltration line 103 in the case where the groundwater level of the seepage water in the embankment 101 rises, causing water to seep out from the slope 101S, and also shows a schematic diagram of the area into which water seeps out in this case as the leakage area A1.

[0021] Furthermore, as shown in Figure 4, a form known as seepage failure is known as an example of damage caused by typhoons and heavy rain. In seepage failure, when the water-blocking function of the embankment body 101 is reduced due to internal deterioration of the embankment body 101, the embankment body 101 is locally destroyed, resulting in the formation of a water path 104 called a piping hole that penetrates the embankment body 101. In this case, water stored in the water storage section 102 leaks from a slope 101S located downstream of the embankment body 101. Water leaking from the reservoir 100 spreads from the outlet of the water path 104 on the slope 101S to the surrounding area. Note that in Figure 4, the range into which water leaks from the water path 104 spreads is schematically illustrated as a water leakage range A2.

[0022] (Inspection System 1 functions) Next, each function of the inspection system 1 will be described. In the inspection system 1, the water leakage inspection device 10 acquires a satellite image of a reservoir 100 located in a remote location from a satellite image server 20, and uses the optical satellite image acquired from the satellite image server 20 to inspect the presence or absence of water leakage in the reservoir 100 shown in the satellite image.

[0023] As shown in FIG. 1, the water leakage inspection device 10 includes a control unit 11, a storage unit 12, an input unit 13, an output unit 14, and a communication unit 15. The control unit 11 performs various processes including the water leakage inspection process described below. The control unit 11 executes an inspection program for performing the water leakage inspection process, thereby functioning as a management unit 11A, an area setting unit 11B, a normalization calculation unit 11C, a threshold setting unit 11D, a determination unit 11E, and the like.

[0024] The management unit 11A acquires a satellite image of the reservoir 100 to be inspected for leaks from the satellite image server 20. The satellite image preferably has a photographing accuracy (resolution) of at least 1 m x 1 m. Furthermore, the management unit 11A controls the area setting unit 11B, the normalization calculation unit 11C, the threshold setting unit 11D, and the judgment unit 11E to perform the leak inspection process.

[0025] The area setting unit 11B executes an area setting process to set an area corresponding to the water surface 102S (see Figure 3) of the water stored in the reservoir 100 to be inspected and an area to be inspected for the presence or absence of water leakage in the satellite image obtained from the satellite image server 20.

[0026] The region setting process by the region setting unit 11B will be described with reference to Fig. 5. Image data 30 shown in Fig. 5 is an example of a satellite image of the reservoir 100 taken from above. The area setting unit 11B sets, in the image data 30, a water storage area 31 corresponding to the water surface 102S of the reservoir 100 and an inspection area 32 where there is a possibility of water leakage from the reservoir 100. For example, the area setting unit 11B sets, in the reservoir 100, an area corresponding to the entire embankment 101 including the slope 101S as the inspection area 32.

[0027] As an example, the region setting unit 11B may set the water storage region 31 and the inspection region 32 in the image data 30 based on an input such as a range specification by the user. As another example, the area setting unit 11B may identify the position of the embankment 101 in the image data 30 based on various types of geographic information associated with the position information of the reservoir 100 to be inspected. Examples of the geographic information include a topographical map, elevation data, or a land use map. The geographic information may also be information collected as a GIS (Geographic Information System). The area setting unit 11B may then set the inspection area 32 at the identified position of the embankment 101. In this case, the area setting unit 11B may set the inspection area 32 by combining image processing such as edge detection on the image data 30.

[0028] 5, the outlines of the water storage area 31 and the inspection area 32 are indicated by dashed lines. Note that in FIG. 5, the inspection area 32 is an area surrounded by the dashed line representing the outline of the water storage area 31 and the dashed line representing the outline of the inspection area 32.

[0029] 1, the normalization calculation unit 11C performs normalization processing to calculate a normalized index value for each pixel constituting the satellite image, normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band. The normalized index value is a numerical value obtained by normalizing each pixel using the spectral radiance values ​​in the two different wavelength bands in order to distinguish the satellite image into areas with a relatively large amount of water and areas with a relatively small amount of water.

[0030] Note that the wavelength here refers to the wavelength of light observed by an optical sensor onboard a satellite, i.e., the wavelength of sunlight reflected on Earth. For example, Stellogic uses an optical sensor that observes four wavelength bands: blue, green, red, and NIR (near-infrared) to capture satellite images. Blue is the wavelength band from 450 nm to 510 nm. Green is the wavelength band from 510 nm to 580 nm. Red is the wavelength band from 590 nm to 690 nm. NIR is the wavelength band from 750 nm to 900 nm. In other words, optical satellite images are an example of multi-wavelength spectroscopic images that observe electromagnetic waves in different wavelength bands. Note that the wavelengths that an optical sensor can observe vary depending on the optical sensor.

[0031] For example, the normalization calculation unit 11C may calculate an NDWI (Normalized Difference Water Index), which is an example of a normalization index value. The formula for calculating the NDWI is given as NDWI=(spectral radiance value of short wavelength band−spectral radiance value of long wavelength band) / (spectral radiance value of short wavelength band+spectral radiance value of long wavelength band). In this case, the normalization calculation unit 11C substitutes the spectral radiance value of the green wavelength band as the spectral radiance value of the short wavelength band, and substitutes the spectral radiance value of the NIR wavelength band as the spectral radiance value of the long wavelength band in the above formula. That is, in this case, the normalization calculation unit 11C normalizes each pixel constituting the satellite image based on the spectral radiance value of the green wavelength band and the spectral radiance value of the NIR wavelength band.

[0032] The normalized index value is not limited to NDWI and may be another index value. For example, the normalization calculation unit 11C may calculate LSWI (Land Surface Water Index). For example, the formula for calculating LSWI is given as LSWI = (reflectance in near-infrared region - reflectance in short-wavelength infrared region) / (reflectance in near-infrared region + reflectance in short-wavelength infrared region). Note that the near-infrared region in the above formula is, for example, a wavelength region from 750 nm to 1000 nm. The short-wavelength infrared region in the above formula is, for example, a wavelength region from 1000 nm to 2500 nm. The reflectance can be calculated by correcting the spectral reflection luminance value using parameters such as atmospheric correction and the angle of sunlight incidence.

[0033] For example, the normalization calculation unit 11C may calculate a Modified Land Surface Water Index (MLSWI). For example, the formula for calculating the MLSWI is given as LSWI=(absorbance in near-infrared region−reflectance in short-wavelength infrared region) / (absorbance in near-infrared region+reflectance in short-wavelength infrared region). Note that the absorptance in the near-infrared region in the above formula is given as absorptance=(1−reflectance in near-infrared region). The above-mentioned LSWI and MLSWI are examples of normalized index values ​​normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band.

[0034] In a satellite image normalized by one of the above normalized index values, a larger normalized index value indicates a larger amount of water. For example, the normalized index value is large at the water surface 102S of the water stored in the reservoir 100 or at a location where a water leak has occurred.

[0035] The threshold setting unit 11D executes a threshold setting process to set a threshold for the normalized index value for determining whether or not there is water leakage from the reservoir 100 in the inspection area 32. As an example, the threshold setting unit 11D may set a threshold for each satellite image based on an input by a user.

[0036] As another example, the threshold setting unit 11D may set an arbitrary pre-stored value as the threshold. For example, when NDWI is used as the normalized index value, the NDWI value varies depending on the weather and the incident angle of sunlight. Therefore, when the threshold setting unit 11D sets an arbitrary pre-stored value as the threshold, the threshold setting unit 11D may set a threshold corrected according to the conditions at the time of capturing the satellite image. The conditions at the time of capturing the satellite image include, for example, information for calculating the incident angle of sunlight, such as the time of day and season, and weather information.

[0037] As another example, the threshold setting unit 11D may set a threshold so that the outline of a portion of the image data 30 where the normalized index value is equal to or greater than the threshold and overlaps with the water storage area 31 corresponds to the outline of the water storage area 31. For example, the threshold setting unit 11D may calculate a threshold for each satellite image such that the outline of a portion of the image data 30 where the normalized index value is equal to or greater than the threshold and overlaps with the water storage area 31 most closely resembles the outline of the water storage area 31. For example, a discriminant analysis method called Otsu's binarization may be used to set a threshold for the normalized index value to distinguish between the normalized index value of the water surface 102S reflected in the water storage area 31 and the normalized index value of an area where the amount of water is relatively small, such as the levee body 101, reflected in the inspection area 32.

[0038] The determination unit 11E executes a determination process to determine whether or not the normalized index values ​​of the pixels that make up the inspection area 32 are equal to or greater than the threshold value set by the threshold setting unit 11D. If the normalized index values ​​of the pixels that make up the inspection area 32 are equal to or greater than the threshold value set by the threshold setting unit 11D, the determination unit 11E determines that water leakage from the reservoir 100 is occurring in the inspection area 32. If the normalized index values ​​of the pixels that make up the inspection area 32 are less than the threshold value set by the threshold setting unit 11D, the determination unit 11E determines that water leakage from the reservoir 100 is not occurring.

[0039] The storage unit 12 stores programs and data for the control unit 11 to execute various processes. For example, the storage unit 12 stores an inspection program for performing a water leakage inspection process. The storage unit 12 also includes an image storage unit 12A, a reservoir information storage unit 12B, an arithmetic formula storage unit 12C, and a threshold information storage unit 12D.

[0040] The image storage unit 12A stores satellite images acquired from the satellite image server 20. The image storage unit 12A stores the results of each process, such as the area setting process by the area setting unit 11B, the normalization process by the normalization calculation unit 11C, the threshold setting process by the threshold setting unit 11D, and the judgment process by the judgment unit 11E, as the results of the leak inspection process on the satellite image. The image storage unit 12A stores the results of the leak inspection process on the satellite image in a state where no water is leaking from the reservoir 100.

[0041] The reservoir information storage unit 12B stores various types of geographic information associated with the location information of the reservoir 100 to be inspected. The geographic information stored in the reservoir information storage unit 12B is used in the area setting process by the area setting unit 11B.

[0042] The arithmetic formula storage unit 12C stores an arithmetic formula, parameters, etc., used by the normalization calculation unit 11C to perform the normalization process. The arithmetic formula used to perform the normalization process is, for example, at least one of the above-mentioned NDWI, LSWI, and MLSWI calculation formulas.

[0043] The threshold information storage unit 12D stores various information for threshold setting processing by the threshold setting unit 11D. For example, when the threshold setting unit 11D automatically sets a threshold according to the outer shape of the water storage area 31, the threshold information storage unit 12D may store an arithmetic expression for performing discriminant analysis. Furthermore, when the threshold setting unit 11D sets a pre-stored arbitrary value as the threshold, the threshold information storage unit 12D may store a function or table for correcting the threshold according to the conditions at the time of capturing the satellite image.

[0044] The input unit 13 is a device that allows a user to input various instructions and information to the water leak inspection device 10. The input unit 13 includes, for example, a keyboard and a pointing device. The output unit 14 is a device that outputs various information handled by the water leak inspection device 10. The output unit 14 is, for example, a display. The input unit 13 and the output unit 14 may be a touch panel display or the like. The communication unit 15 is a device that transmits and receives data to and from the satellite image server 20 via a network line.

[0045] (Leak inspection method) As shown in FIG. 6, the water leakage inspection method includes steps S1 to S5. First, the management unit 11A acquires, from the satellite image server 20, a satellite image showing the reservoir 100 to be inspected for the presence or absence of water leakage (step S1).

[0046] Next, the area setting unit 11B performs an area setting process to set a water storage area 31 corresponding to the water surface 102S of the reservoir 100 to be inspected and an inspection area 32 corresponding to the embankment body 101 in the image data 30 (step S2).

[0047] Next, the normalization calculation unit 11C performs a normalization process to calculate a normalized normalized index value for each pixel constituting the image data 30 according to the first spectral radiance value in the first wavelength band and the second spectral radiance value in the second wavelength band (step S3).

[0048] Next, the threshold setting unit 11D executes a threshold setting process to set a threshold for the normalized index value for determining whether or not there is leakage from the reservoir 100 in the inspection area 32 (step S4). After setting the threshold, the threshold setting unit 11D may output to the output unit 14 a binarized image obtained by extracting only those portions of the pixels constituting the image data 30 whose normalized index values ​​are equal to or greater than the threshold.

[0049] Finally, the determination unit 11E executes a determination process to determine whether or not the normalized index values ​​of the pixels that make up the inspection area 32 are equal to or greater than the threshold value set by the threshold setting unit 11D in step S4 (step S5). Note that the determination unit 11E may display the determination result, whether or not the normalized index values ​​of the pixels that make up the inspection area 32 are equal to or greater than the threshold value set by the threshold setting unit 11D, on the output unit 14, depending on the determination result of step S5.

[0050] (Test example) Test examples of this embodiment will be described with reference to Figures 7 to 11. The following test examples were carried out during the day on a clear day.

[0051] As shown in Fig. 7, in a test example of this embodiment, an embankment 200 simulating the embankment body 101 was formed in a test site set up near the reservoir 100. Then, tap water was sprinkled on the embankment slope 201 of the embankment 200 to reproduce water leakage from the reservoir 100.

[0052] Water was sprayed at three points on the embankment slope 201: the first point P1, the second point P2, and the third point P3. At the first point P1, water was sprayed in an area of ​​2m x 2m before satellite images were taken until water floated on the surface. Water was then continued to be sprayed while the satellite images were being taken so that the water remained floating on the surface. The first point P1 was at a level that reproduced the condition in which water leakage from the reservoir 100 had occurred.

[0053] At the second point P2, water was sprayed over an area of ​​2m x 2m before satellite imagery was taken. At the third point P3, water was sprayed over an area of ​​1m x 1m before satellite imagery was taken. At the second point P2 and the third point P3, there was no water floating on the surface and the surface was wet when the satellite imagery was taken. Note that the second point P2 and the third point P3 are comparison levels that reproduce a condition where there is no leakage but the ground is wet after rainfall, etc. Under the above conditions, satellite imagery was taken so that the reservoir 100 and the embankment 200 were included in the shooting range.

[0054] In step S1, the management unit 11A obtained a satellite image showing the reservoir 100 and the embankment 200 from the satellite image server 20. In step S1, test image data 40 shown in FIG. 8 was obtained from the satellite image server 20 as the satellite image.

[0055] Satellite images taken by Stellogic were used as the test image data 40. The test image data 40 was taken using an optical sensor that observes the four wavelength bands of blue, green, red, and NIR, as described above. The test image data 40 was also taken at a resolution of 1 m x 1 m.

[0056] 8, in step S2, the area setting unit 11B sets a water storage area 41 in the test image data 40 at a position corresponding to the water surface 102S of the reservoir 100, based on input by the user. In addition, the area setting unit 11B sets an inspection area 42 in the test image data 40 at a position corresponding to the embankment slope 201 of the embankment 200, based on input by the user.

[0057] In step S3, the normalization calculation unit 11C calculated NDWI, which is a normalization index value, for each pixel constituting the test image data 40. NDWI was calculated as NDWI = (Green spectral radiance value - NIR spectral radiance value) / (Green spectral radiance value + NIR spectral radiance value).

[0058] 9 shows the magnitude of the NDWI of each pixel constituting the test image data 40. Note that in the heat map 50, the darker the color, the larger the NDWI.

[0059] 9, in the heat map 50, it is confirmed that the NDWI is large in the water storage area 41, the inspection area 42, etc. in the test image data 40. In addition, it is confirmed that the NDWI is also large in the shaded areas in the test image data 40 other than the water storage area 41 and the inspection area 42.

[0060] FIG. 10 shows an enlarged image 51 obtained by enlarging the inspection area 42 in the heat map 50. Note that the color scale in the enlarged image 51 is reconfigured based on the minimum and maximum NDWI values ​​within the inspection area 42. As shown in FIG. 10, in the enlarged image 51, the NDWI values ​​are relatively large throughout the entire portion of the embankment slope 201 corresponding to the first point P1. In the portion corresponding to the second point P2, the NDWI values ​​are relatively large in the portion close to the first point P1, but are relatively small in the portion far from the first point P1. In the portion corresponding to the third point P3, the NDWI values ​​are relatively small throughout the entire portion. Therefore, by normalizing each pixel constituting the satellite image with a normalized index value such as NDWI, it is possible to determine whether water is present in the satellite image based on the magnitude of the normalized index value.

[0061] In step S4, the threshold value setting unit 11D set the NDWI threshold value so that the outline of the portion of the test image data 40 where the NDWI is equal to or greater than the threshold value and overlaps with the water storage area 41 corresponds to the outline of the water storage area 41.

[0062] 11 shows a binarized image 60 obtained by binarizing each pixel constituting the test image data 40 based on the NDWI threshold value. The binarized image 60 is obtained by extracting only those pixels constituting the test image data 40 whose NDWI values ​​are equal to or greater than the threshold value set in step S4.

[0063] 11, it was confirmed that the NDWI was equal to or greater than the threshold set in step S4 in the inspection area 42 in the binarized image 60 in the portion corresponding to the first point P1 that reproduced the state in which water leakage from the reservoir 100 occurred. Therefore, it was confirmed that the presence or absence of water leakage in the inspection area 42 can be determined by determining in the determination process of step S5 whether the NDWI of the pixels that make up the inspection area 42 is equal to or greater than the threshold set by the threshold setting unit 11D in step S4.

[0064] It was confirmed that the NDWI was equal to or greater than the threshold set in step S4 on the side of the second point P2 closer to the first point P1 in the inspection area 42 in the binarized image 60. This is thought to be because the NDWI value on the side of the second point P2 closer to the first point P1 was large due to the influence of a large amount of water present at the first point P1.

[0065] Furthermore, in the binarized image 60, in addition to the reservoir area 41 and the inspection area 42, there are also some areas where the NDWI is equal to or greater than the threshold value set in step S4. These are areas that are shaded in the test image data 40. In other words, when NDWI is used as the normalized index value, it has been confirmed that the NDWI value is affected by shadows. Therefore, it is preferable that the satellite image used for the leak inspection process does not have a shadow on the embankment 101 to be inspected for leaks. As an alternative method, the leak inspection process may be performed on multiple satellite images of the same reservoir 100 taken at different times, such as in the morning and afternoon, with different shadow positions. This can further improve the accuracy of determining whether or not there is a leak.

[0066] (Effects of the embodiment) (1) In the inspection system 1, the normalization calculation unit 11C calculates a normalized index value normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band for each pixel constituting image data 30, which is a satellite image of the reservoir 100. Then, the determination unit 11E determines whether the normalized index value of a pixel constituting an inspection area 32 corresponding to the embankment 101 in the image data 30 is equal to or greater than a threshold value. This makes it possible to confirm the presence or absence of water leakage from the embankment 101 corresponding to the inspection area 32. Therefore, the presence or absence of water leakage from the reservoir 100 can be determined without installing any new equipment in the reservoir 100.

[0067] (2) The threshold setting unit 11D sets the threshold so that the outline of the portion of the image data 30 where the normalized index value is equal to or greater than the threshold and that overlaps with the water storage area 31 corresponds to the outline of the water storage area 31. For example, when the NDWI is used as the normalized index value, the NDWI varies depending on the weather and the angle of incidence of sunlight. However, by setting the threshold according to the shape of the water storage area 31 as described above, the threshold for the normalized index value can be optimized for each satellite image.

[0068] (3) The area setting unit 11B may identify the position of the embankment 101 in the image data 30 based on various geographic information associated with the location information of the reservoir 100 to be inspected, and set the inspection area 32 at the identified position of the embankment 101. This process makes it possible to automatically extract an area in the image data 30 where there is a possibility of water leakage from the reservoir 100 to be inspected.

[0069] (Example of change) This embodiment can be modified as follows: This embodiment and the following modifications can be combined and implemented within the scope of technical compatibility.

[0070] The water leak detection device 10 may be realized as a single device, or may be distributed across multiple devices or subsystems that cooperate to execute programs. The water leak detection device 10 and the satellite image server 20 may be realized as a single device.

[0071] In the above embodiment, an example was given of a configuration in which an area corresponding to the entire embankment body 101 in the image data 30 is set as the inspection area 32. However, an area of ​​the embankment body 101 corresponding to the slope 101S may also be set as the inspection area 32. In this case, the inspection area 32 is made up of only slopes where puddles are unlikely to form after rainfall. This prevents puddles and other water puddles from being erroneously detected as leak locations in the inspection area 32.

[0072] In the NDWI calculation formula, the short wavelength band is Green and the long wavelength band is NIR, but this is not limited to this and the spectral radiance value of any wavelength band may be substituted into the NDWI calculation formula. Also, if the optical sensor capturing the satellite imagery can observe the short wavelength infrared (SWIR) range, the spectral radiance value of the short wavelength infrared range may be substituted into the NDWI calculation formula.

[0073] In the above embodiment, an example has been given of a configuration in which the presence or absence of water leakage in the reservoir 100 is inspected from an optical satellite image of the reservoir 100. The multi-wavelength spectroscopic images handled by the inspection system 1 are not limited to optical satellite images, and may be, for example, multi-spectral images or hyperspectral images of the reservoir 100 including the embankment 101 taken from above using an airplane, helicopter, balloon, drone, or the like.

[0074] In step S5, the determination unit 11E may perform a process of comparing a reference satellite image in a state where no water leakage occurs with a satellite image that is the target of the water leakage inspection process, in addition to the process of determining whether the normalized index values ​​of the pixels that make up the inspection area 32 are equal to or greater than a threshold value.

[0075] In this case, the determination unit 11E first acquires the results of the water leak inspection process for the reference satellite image stored in advance in the image storage unit 12A. Then, for example, the determination unit 11E may compare the reference satellite image and the satellite image to be subjected to the water leak inspection process for areas in the inspection area 32 where the normalized index value is equal to or greater than a threshold. Note that the threshold value for the normalized index value is set separately for the reference satellite image and the satellite image to be subjected to the water leak inspection process.

[0076] The comparison of the regions where the normalized index value is equal to or greater than the threshold may be performed, for example, by quantitatively comparing the number of pixels or the area. The comparison of the regions where the normalized index value is equal to or greater than the threshold may also be performed visually by displaying the reference satellite image and the satellite image to be inspected for water leakage side by side or overlaid on each other, the images being binarized according to the normalized index value.

[0077] According to the above process, if the area in the inspection area 32 where the normalized index value is equal to or greater than the threshold is larger in the satellite image being subjected to the leak inspection process than in the reference satellite image, it can be determined that there is a high possibility of a leak. On the other hand, even if there are pixels in the inspection area 32 where the normalized index value is equal to or greater than the threshold, if the area where the normalized index value is equal to or greater than the threshold is equal to or smaller than the reference satellite image, it can be determined that the increase in the normalized index value is due to other factors such as shadows, rather than a leak. Therefore, by performing a comparison process with the reference satellite image, it is possible to improve the accuracy of determining the presence or absence of a leak in the leak inspection process.

[0078] When performing a water leak inspection process on multiple satellite images of the same reservoir 100 taken at different times, the determination unit 11E may calculate the amount of change over time in the area (e.g., number of pixels) of the area in the inspection area 32 where the normalized index value is equal to or greater than the threshold. In this case, if the area of ​​the area in the inspection area 32 where the normalized index value is equal to or greater than the threshold is larger in the satellite image taken later than in the satellite image taken earlier, it can be determined that there is a high possibility of a water leak. This can further improve the accuracy of determining whether or not there is a water leak.

[0079] Furthermore, when performing a water leak inspection process on multiple satellite images of the same reservoir 100 taken at different times, the determination unit 11E may calculate the amount of change over time in the area (e.g., number of pixels) of the water storage area 31. For example, if the area of ​​the water storage area 31 in a satellite image taken later is significantly smaller than in a satellite image taken earlier, even though no water intake work is being carried out, it can be determined that there is a high possibility of a water leak. This can further improve the accuracy of determining whether or not there is a water leak.

[0080] The area setting unit 11B may adjust the size of the inspection area 32 depending on the size of the water storage area 31. For example, the area setting unit 11B may estimate the water level of the reservoir 100 depending on the size of the water storage area 31, and set a portion of the embankment 101 shown in the image data 30 that is lower in elevation than the estimated water level as the inspection area 32.

[0081] Once the reservoir area 31 corresponding to the water surface 102S of the reservoir 100 to be inspected and the inspection area 32 corresponding to the embankment 101 are set in the image data 30, the results of the area setting process may be reused the next time the same reservoir 100 is inspected. That is, in the image data 30 depicting the reservoir 100 to be inspected, the positions of the reservoir area 31 and the inspection area 32, for example, polygons corresponding to the reservoir area 31 and the inspection area 32, are stored in the memory unit 12. Then, the next time the same reservoir 100 is inspected, the reservoir 100 may be inspected based on the positions of the reservoir area 31 and the inspection area 32 stored in the memory unit 12. This makes it possible to monitor the presence or absence of water leakage at any time for a reservoir 100 for which the area setting process has been performed once, without having to perform the area setting process each time the image data 30 is updated. The polygons corresponding to the dam body 101 and the water storage section 102, that is, the polygons corresponding to the water storage area 31 and the inspection area 32, may be stored in the storage section 12 in advance.

[0082] The inspection system can also detect water leakage in river levees. For example, water leakage in river levees occurs when a seepage surface forms within the levee as the river water level rises. For example, the inspection system first calculates the NDWI of the toe of the river levee in a multi-wavelength spectroscopic image that includes the river levee to be inspected for water leakage. Next, a threshold NDWI is set based on the boundary between the river water body and the river levee. Finally, any part of the toe of the river levee where the NDWI is above the threshold is determined to be a water leak.

[0083] (Addendum) The technical ideas that can be understood from the above-described embodiment and modified examples will be described. (Appendix 1) An inspection system for inspecting for water leakage from a water area having a levee, The control unit provided in the inspection system calculating a normalized index value normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band for each pixel constituting the multi-wavelength spectral image obtained by photographing the water area from above; identifying a position corresponding to the bank body in the multi-wavelength spectral image based on geographic information associated with the position information of the water area, and setting an inspection area at the identified position of the bank body; In the multi-wavelength spectroscopic image, it is determined whether the normalized index value of the pixels constituting the inspection area is equal to or greater than a threshold value. Inspection system.

[0084] (Appendix 2) An inspection system for inspecting for water leakage from a water area having a levee, a control unit provided in the inspection system, which executes a water leak inspection process on a plurality of multi-wavelength spectroscopic images obtained by photographing the water area from above at different time periods; The water leak inspection process calculates a normalized index value normalized for each pixel constituting each multi-wavelength spectral image according to a first spectral radiance value of a first wavelength band and a second spectral radiance value of a second wavelength band different from the first wavelength band; and determining whether or not the normalized index value of pixels constituting an inspection area set as an area where water leakage from the water area may occur is equal to or greater than a threshold value in each multi-wavelength spectroscopic image. Inspection system.

[0085] (Appendix 3) An inspection system for inspecting for water leakage from a water area having a levee, The control unit provided in the inspection system calculating a normalized index value normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band for each pixel constituting the multi-wavelength spectral image obtained by photographing the water area from above; determining whether or not the normalized index value of a pixel constituting an inspection area set as an area where water leakage from the water area may occur is equal to or greater than a first threshold value in the multi-wavelength spectroscopic image; execute a process of comparing a region in the inspection area where the normalized index value is equal to or greater than the first threshold value with a region in the inspection area in a reference multi-wavelength spectroscopic image including the water body portion in a state where no water leakage occurs where the normalized index value is equal to or greater than a second threshold value; the first threshold is set so that an outline of a portion of the multi-wavelength spectroscopic image where the normalized index value is equal to or greater than the first threshold and overlaps with a first water storage area set to correspond to the water surface of the water area corresponds to an outline of the first water storage area; The second threshold value is set so that the outline of a portion of the reference multi-wavelength spectroscopic image where the normalized index value is equal to or greater than the second threshold value and overlaps with a second water storage area set to correspond to the water surface of the water area corresponds to the outline of the second water storage area. Inspection system. [Explanation of symbols]

[0086] A1, A2... water leakage range, H10... information processing device, H11... communication device, H12... input device, H13... display device, H14... storage device, H15... processor, P1... first location, P2... second location, P3... third location, S1 to S5... steps, 1... inspection system, 10... water leakage inspection device, 11... control unit, 11A... management unit, 11B... area setting unit, 11C... normalization calculation unit, 11D... threshold setting unit, 11E... judgment unit, 12... memory unit, 12A... image memory unit, 12B... reservoir Information storage unit, 12C...arithmetic formula storage unit, 12D...threshold information storage unit, 13...input unit, 14...output unit, 15...communication unit, 20...satellite image server, 30...image data, 31, 41...water storage area, 32, 42...inspection area, 40...test image data, 50...heat map, 51...enlarged image, 60...binarized image, 100...reservoir, 101...embankment body, 101S...slope, 102...water storage area, 102S...water surface, 103...infiltration line, 104...water path, 200...embankment, 201...embankment slope.

Claims

1. An inspection system for inspecting water leakage from a water area having a dam body, The control unit provided in the inspection system calculating a normalized index value normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band for each pixel constituting the multi-wavelength spectral image obtained by photographing the water area from above; In the multi-wavelength spectroscopic image, it is determined whether the normalized index value of pixels constituting an inspection area set as an area where water leakage from the water area may occur is equal to or greater than a threshold value. Inspection system.

2. The control unit sets the threshold value so that an outline of a portion of the multi-wavelength spectroscopic image where the normalized index value is equal to or greater than the threshold value and overlaps with a water storage area set to correspond to the water surface of the water area corresponds to the outline of the water storage area. The inspection system of claim 1 .

3. In the multi-wavelength spectral image, a region corresponding to a slope of the bank body is set as the inspection region.

3. An inspection system according to claim 1 or 2.

4. An inspection method for inspecting for the presence or absence of water leakage from a water area having a levee using an inspection system including a control unit, comprising: The control unit calculating a normalized index value normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band for each pixel constituting the multi-wavelength spectral image obtained by photographing the water area from above; In the multi-wavelength spectroscopic image, it is determined whether the normalized index value of pixels constituting an inspection area set as an area where water leakage from the water area may occur is equal to or greater than a threshold value. Inspection method.

5. An inspection program for inspecting for the presence or absence of water leakage from a water area having a levee using an inspection system including a control unit, The control unit calculating a normalized index value normalized according to a first spectral radiance value in a first wavelength band and a second spectral radiance value in a second wavelength band different from the first wavelength band for each pixel constituting the multi-wavelength spectral image obtained by photographing the water area from above; The multi-wavelength spectroscopic image is configured to function as a means for determining whether or not the normalized index value of pixels constituting an inspection area set as an area where water leakage from the water area may occur is equal to or greater than a threshold value. Inspection program.

Images