JP7621300B2 - Information processing device, information processing program, and information processing method - Google Patents

Description

The present disclosure relates to an information processing device, an information processing program, and an information processing method. In particular, the present disclosure relates to information processing for estimating the flooding status of houses or forests when a flood occurs.

In the past, ground sensors were used to grasp flood damage during water damage such as floods, heavy rains, and typhoons. Methods of grasping flood damage using ground sensors are useful for grasping the flooding situation at specific spots such as elevators or power plants. However, in situations where damage occurs over a wide area, methods of grasping flood damage using ground sensors are not suitable for obtaining information on the extent of the area, such as which areas have suffered the most damage.

As a method for grasping flood damage over a wide area, a flood depth estimation method using remote sensing data has been proposed (see Non-Patent Document 1). In Non-Patent Document 1, SAR satellite images taken at two different times, before and after a disaster, are compared to detect changed areas, and the inundated area, which is the extent of the flooding, is estimated.

"An Attempt to Rapidly Extract Flooded Area and Depth During Heavy Rain Disasters Using Satellite SAR." Honda, Kenichi et al., Journal of Remote Sensing Society of Japan, Vol. 40 No. 3 (2020), pp. 158-162.

In Non-Patent Document 1, flooded areas are calculated by comparing two SAR satellite images: a first image taken at a certain time and a second image taken at a time in the past relative to the first image. Generally, areas where there is a noticeable difference between the two images are considered to be flooded areas. In this case, there are cases where the differences are sparse due to noise in the images and seasonal fluctuations. In flooded area detection methods that rely on images like this, the image quality and imaging conditions affect the determination of flooded areas. For this reason, even if an area is actually flooded, it may be considered to be a non-flooded area because there is no difference between the images of that area. This is an issue in terms of estimation accuracy.

The purpose of this disclosure is to estimate flooded areas with high accuracy, regardless of the image quality and imaging conditions of the remote sensing data used.

The information processing device of the present disclosure includes:
an estimation unit that estimates an initially flooded area that is an area that exists in both the first image and the second image and indicates a flooded area by comparing a first image with a second image that was taken at an earlier date and time than the first image;
a flooded area filling unit which, when at least a portion of the initially flooded area is included in a depression area obtained by analyzing the terrain gradient for elevation information indicating the elevation of the terrain, calculates an index value indicating the degree of flooding of the depression area, and, when the index value is equal to or greater than a threshold value, determines that the depression area is flooded, and fills the depression area with a flooded area based on the determination result indicating flooding;
Equipped with.

The information processing device disclosed herein can estimate flooded areas with high accuracy, regardless of the image quality or imaging conditions of the remote sensing data used.

FIG. 1 is a block diagram of an information processing device 101 according to a first embodiment. FIG. 1 is a diagram showing a hardware configuration of an information processing device 101 according to a first embodiment. FIG. 1 is a flowchart showing the operation of an information processing method according to the first embodiment. A diagram of the first embodiment showing a potential flooded depression 131A and an initial flooded area 11A. FIG. 11 is a block diagram of an information processing device 20 according to a modified example of the first embodiment. FIG. 2 is a diagram supplementing the hardware configuration of information processing devices 10 and 20 in the first embodiment.

In the description of the embodiments and drawings, the same elements and corresponding elements are given the same reference numerals. Descriptions of elements given the same reference numerals are omitted or simplified as appropriate. In the following embodiments, "part" may be read as "circuit," "process," "procedure," "process," or "circuitry" as appropriate.

Embodiment 1.
FIG. 1 is a functional configuration diagram of an information processing device 10.
FIG. 2 shows the hardware configuration of the information processing device 101 .

***Configuration Description***
1 , the information processing device 101 includes an initial flooded area estimation unit 11, a depression detection unit 12, and a flooded area filling unit 13. The flooded area filling unit 13 includes a candidate detection unit 131, a flood determination unit 132, and a flooded depression filling unit 133.
The initial flooded area estimation unit 11 is an estimation unit.

The information processing device 10 is a computer. The information processing device 10 includes a processor 510. In addition to the processor 510, the information processing device 10 includes multiple pieces of hardware. The multiple pieces of hardware are a main memory device 520, an auxiliary memory device 530, an input interface 540, an output interface 550, and a communication interface 560. The processor 510 is connected to the other hardware via a signal line 570 and controls the other hardware.

The information processing device 10 has, as functional elements, an initial flooded area estimation unit 11, a depression detection unit 12, and a flooded area filling unit 13.
The functions of the initial flooded area estimation unit 11, depression detection unit 12, and flooded area filling unit 13 are realized by the information processing program 531.

The processor 510 is a device that executes the information processing program 531. When the processor 510 executes the information processing program 531, the functions of the initial flooded area estimation unit 11, the depression detection unit 12, and the flooded area filling unit 13 are realized. The processor 510 is an IC (Integrated Circuit) that performs arithmetic processing. Specific examples of the processor 510 are a CPU (Central Processing Unit), a DSP (Digital Signal Processor), and a GPU (Graphics Processing Unit).

The main memory 520 is a memory device. Specific examples of the main memory 520 are SRAM (Static Random Access Memory) and DRAM (Dynamic Random Access Memory). The main memory 520 holds the results of calculations by the processor 510.

The auxiliary storage device 530 is a storage device that stores data in a non-volatile manner. A specific example of the auxiliary storage device 530 is a HDD (Hard Disk Drive). The auxiliary storage device 530 may also be a portable recording medium. Examples of portable recording media include an SD (registered trademark) (Secure Digital) memory card, a NAND flash, a flexible disk, an optical disk, a compact disk, a Blu-ray (registered trademark) disk, and a DVD (Digital Versatile Disk). The auxiliary storage device 530 stores an information processing program 531.

The input interface 540 is a port through which data is input from each device. The output interface 550 is connected to various devices. The output interface 550 is a port through which the processor 510 outputs data to various devices. The communication interface 560 is a communication port through which the processor communicates with other devices.

The processor 510 loads the information processing program 531 from the auxiliary storage device 530 to the main storage device 520. The processor 510 reads the loaded information processing program 531 from the main storage device 520 and executes it. In addition to the information processing program 531, an OS (Operating System) is also stored in the main storage device 520. The processor 510 executes the information processing program 531 while executing the OS. The information processing device 10 may include multiple processors that replace the processor 510. These multiple processors share the execution of the information processing program 531. Each processor is a device that executes the information processing program 531, just like the processor 510. Data, information, signal values, and variable values used, processed, or output by the information processing program 531 are stored in the main storage device 520, the auxiliary storage device 530, or a register or cache memory in the processor 510.

The information processing program 531 is a program that causes a computer to execute each process, procedure, or step of the initial flooded area estimation unit 11, depression detection unit 12, flooded area filling unit 13, candidate detection unit 131, flooding assessment unit 132, and flooded depression filling unit 133, where "part" is replaced with "process," "procedure," or "step."

The method is an information processing method performed by an information processing device 10, which is a computer, executing an information processing program 531. The information processing program 531 may be provided by being stored in a computer-readable recording medium, or may be provided as a program product.

*** Operation Description ***
The operation of the information processing device 10 will be described with reference to FIG.
3 is a flowchart illustrating the operation of the information processing device 10. The operation of the information processing device 10 corresponds to an information processing method. Also, the operation of the information processing device 10 corresponds to the processing of an information processing program.

<Step S1>
In step S1, the initial flooded area estimation unit 11 estimates the initial flooded area 11A from the first image #1 and the second image #2.
The initial flooded area estimation unit 11 estimates an initial flooded area 11A, which indicates an area that exists in both the first image #1 and the second image #2 and is flooded, by comparing the first image #1 with the second image #2 that was taken on an earlier date and time than the first image #1. This will be described in detail below.
The first image #1 and the second image #2 are remote sensing data captured by a photographing device mounted on an air vehicle. The air vehicle may be an aircraft, a UAV, or an artificial satellite. The initial flooded area estimation unit 11 receives the first image #1 and the second image #2 as input data. The first image #1 is remote sensing data acquired at a time T1 indicating a certain date and time. The second image #2 is remote sensing data acquired at a time T2 indicating a certain date and time older than the certain time T1. The second image #2 and the first image #1 include the same area. The second image #2 and the first image #1 are compared for the difference between the same area. The initial flooded area estimation unit 11 compares the first image #1 and the second image #2, and estimates the initial flooded area 11A from the comparison result.
Here, the "initial" in "initial flooded area 11A" refers to the "flooded area" before it is input to the flooded area filling unit 13. In comparison with the "final flooded area 133A" after correction processing has been performed by the flooded area filling unit 13, the "initial" is added to indicate the area before correction processing has been performed. The initial flooded area estimation unit 11 outputs the initial flooded area 11A, which is the estimation result, to the candidate detection unit 131 of the flooded area filling unit 13. The initial flooded area estimation unit 11 can estimate the presence or absence of flooding from, for example, (1) optical images or (2) SAR images, as follows.

(1) Optical Image When estimating the initial flooded area 11A from an optical image, the initial flooded area estimation unit 11 can use NDWI (Normalized Difference Water Index), which is a water index that uses an image in the near-infrared band. ρ _NIR and ρ _SWIR are measurement values of the NIR (Near Inrealed) and SWIR (Short-Wave Infrared) bands at a certain point, respectively. In this case, the NDWI is expressed by the following formula (1). The initial flooded area estimation unit 1 can estimate the initial flooded area 11A by detecting a change in NDWI between the first image #1 and the second image #2.
NDWI=[ρ _NIR −ρ _SWIR ]/[ρ _NIR +ρ _SWIR ] Formula (1)
In recent years, methods for estimating flooded areas using visible bands instead of infrared bands have also been proposed.
The initial flooded area estimation unit 11 can also estimate the initial flooded area 11A using a method that utilizes the visible band.
(2) SAR Images When estimating the initial flooded area 11A from SAR images, the method of Non-Patent Document 1 compares SAR images before and after the occurrence of a flood disaster, and extracts areas with reduced backscattering intensity as flooded areas. The initial flooded area estimation unit 11 may use the method of Non-Patent Document 1.

The above-described (1) optical image and (2) SAR image techniques are examples of remote sensing data. Various remote sensing data can be input to the information processing device of this specification by combining sensors and measurement devices.

<Step S2>
In step S2, the depression detection unit 12 acquires the elevation information 30 and detects the depression region 12A from the acquired elevation information 30. The elevation information 30 described below is input to the depression detection unit 12. The depression detection unit 12 analyzes the elevation information 30 and detects the depression region 12A within the analysis range.

<Altitude information 30>
As the elevation information 30, DEM (Digital Elevation Model) information created by remote sensing technology or surveying can be used. DEM is created by methods such as laser surveying or photogrammetry. DEM is open data that can be acquired by general users. An example of open data that can be acquired by general users is "Geospatial Information Authority of Japan: Basic Map Information https://fgd.gsi.go.jp/download/menu.php". The Geospatial Information Authority of Japan has released 5m mesh DEM5A. SRTM-3 (Shuttle Rador Topography Mission) is released worldwide, and DEM data with high vertical resolution is also available.

<Depression area 12A>
The depression detection unit 12 regards depressions as spatially connected regions with no gradient. In this way, the depression detection unit 12 detects depression regions 12A from the elevation information 30. The depression detection unit 12 calculates a two-dimensional numerical gradient of the elevation information 30, and calculates coordinates where the absolute value of the gradient is equal to or less than a preset threshold. Of the connected elements formed by the calculated coordinates, the depression detection unit 12 detects, as depression regions 12A, regions whose area is equal to or greater than a first threshold and less than a second threshold.

Alternatively, as another implementation example, the depression detection unit 12 detects depression regions 12A by the following process. The depression detection unit 12 calculates the flow direction of each pixel from the two-dimensional numerical gradient of the elevation information 30, and calculates coordinates where the flow direction does not apply to any direction. Among the connected elements formed by the calculated coordinates, the depression detection unit 12 detects an area whose area is equal to or greater than a first threshold value and less than a second threshold value as depression regions 12A.

<Step S3>
In step S3, the flooded area filling unit 13 calculates an index value indicating the degree of flooding of a depression area when at least a part of the initial flooded area 11A is included in a depression area obtained by analyzing the terrain gradient for the elevation information 30 indicating the elevation of the terrain. This depression area is a flooded depression candidate, as described below. Examples of the index value are the ratio R or the area A, as described below. If the index value is equal to or greater than a threshold value, the flooded area filling unit 13 determines that the depression area is flooded, and fills the depression area with a flooded area based on the determination result indicating flooding. This is described in detail below.
The flooded area filling unit 13 determines which depression areas are flooded from the initial flooded area 11A and the depression area 12A. The flooded area filling unit 13 fills the initial flooded area 11A with the depression area determined to be flooded, thereby calculating a final flooded area 133A that is a more accurate version of the initial flooded area 11A. If a certain percentage or more of the initial flooded area 11A is present in the depression area 12A detected by the depression detection unit 12, the flooded area filling unit 13 fills the entire corresponding depression area into the initial flooded area and outputs it as the final flooded area 133A.
The operation of the candidate detection unit 131, the flooding determination unit 132, and the flooded depression filling unit 133 provided in the flooded area filling unit 13 will be described below.

<Candidate detection unit 131>
The candidate detection unit 131 detects a depression area that contains even a small amount of the initial flooded area 11A as a flooded depression candidate 131A.

<Flood determination unit 132>
The flooding determination unit 132 determines whether the flooded depression candidate 131A is flooded or not, and outputs the determination result, that is, a flooding determination result 132A, to the flooded depression filling unit 133.
The flooding determination result 132A indicates either "flooded" or "not flooded."
The specific details are explained below.
Fig. 4 shows a flooded depression candidate 131A and an initially flooded area 11A. In Fig. 4, the flooded depression candidate 131A includes at least a part of the initially flooded area 11A.
The flooding assessment unit 132 calculates the percentage or area of the initially flooded area 11A included in the area of the potential flooded depression 131A. 2) or (3), it can be determined that "the flooded depression candidate 131A is flooded."
(1) When the calculated ratio is equal to or greater than the ratio threshold value THr, the flooding determination unit 132 determines that the floodable depression candidate 131A is flooded. If the ratio is R, then
Ratio R = [area of initial flooded area 11A] / [area of flooded depression candidate 131A]
In
R≧THr
The flooding determination unit 132 then determines that there is flooding.
(2) If the calculated area is equal to or greater than the area threshold THa, the flooding assessment unit 132 determines that the flooded depression candidate 131A is flooded.
The calculated area is the area of the initial flooded area 11A within the area of the flooded depression candidate 131A.
If the area of the initial flooded area 11A is area A, then
A ≧ THa
The flooding determination unit 132 then determines that there is flooding.
(3) The flooding assessment unit 132 assesses that the flooded depression candidate 131A is flooded when the calculated ratio R is greater than or equal to a threshold value THr and the calculated area A is greater than or equal to a threshold value THa.
R ≧ THr and A ≧ THa
The flooding determination unit 132 then determines that there is flooding.
In addition, when the candidate detection unit 131 determines that the flooded depression candidate 131A that does not meet any of the criteria (1), (2), or (3) is “not flooded,” the candidate detection unit 131 determines that the flooded depression candidate 131A is “not flooded.”

Furthermore, if the initial flooded area 11A is contained in a proportion or area of the flooded depression candidate 131A that is less than a certain percentage, it may be determined that the "flooded depression candidate 131A is not flooded." That is, the flooding determination unit 132 may determine that the "flooded depression candidate 131A is not flooded" based on any of the following (4), (5), or (6). Conversely, if the flooding determination unit 132 determines that any of the following (4), (5), or (6) is true, the flooded depression candidate 131A that does not fall under that condition is determined to be "flooded."
(4) If the calculated ratio R is less than the threshold value THr, the flooding assessment unit 132 determines that the flooded depression candidate 131A is not flooded.
R < T Hr
The flooding determination unit 132 then determines that the area is not flooded.
(5) If the calculated area A is less than the area threshold THa, the flooding assessment unit 132 determines that the flooded depression candidate 131A is not flooded.
A < THa
The flooding determination unit 132 then determines that the area is not flooded.
(6) If the calculated ratio R is less than the threshold value THr and the calculated area A is less than the threshold value THa, the flooding assessment unit 132 determines that the flooded depression candidate 131A is not flooded.
R<THr and A<THa
The flooding determination unit 132 then determines that the area is not flooded.

<Flood depression filling part 133>
The flooded depression filling unit 133 fills in the flooded depression candidates, which are depression areas, with flooded areas based on the determination result of the flood determination unit 132. This will be described below.
(1) The flooded depression filling unit 133 of the flooded area filling unit 13 converts a part of the initial flooded area 11A included in the flooded depression candidate 131A, which is a depression area, into a flooded area based on the flooding determination result 132A indicating “flooding.” The flooded depression candidate 131A is filled with a flooded area by using the above-mentioned formula (132). The flooded depression filling unit 133 outputs the filling result as a final flooded area 133A.
(2) Alternatively, the flooded area filling unit 13 may determine, based on the flooding determination result 132A indicating “flooding,”
The flooded depression candidate 131A is filled with a flooded area by using the entirety of the flooded depression candidate 131A as a flooded area. The flooded depression filling unit 133 outputs the filling result as a final flooded area 133A.
(3) In addition, when the index value is less than the threshold value, the flooded area filling unit 13 judges the flooded depression candidate 131A to be “non-flooded” and, based on the flooding judgment result 132A indicating “non-flooded”, A part of the initially flooded area 11A included in the candidate 131A is excluded from the initially flooded area 11A. The flooded depression filling unit 133 outputs the initially flooded area 11A reflecting the filling result as the final flooded area 133A.

***Description of Effect of First Embodiment***
As described above, the information processing device 10 can correct the sparse initial flooded area 11A that occurs depending on at least one of the image quality and imaging conditions of the input images, the first image #1 and the second image #2, and estimate the flooded area with high accuracy. In other words, the flooded area can be estimated with high accuracy regardless of the image quality and imaging conditions of the remote sensing data used.

In the information processing device 10 of the first embodiment, the depression area 12A detected by the depression detection unit 12 is used to perform improvement processing of the initial flooded area 11A, but this is not limited to an example. For example, the information processing device 10 may acquire the depression area 12A from an external system, such as another system or device, different from the information processing device 10, and the flooded area filling unit 13 may use the acquired depression area 12A to calculate the final flooded area 133A. This configuration will be described below as a modified example.

<Modification>
Fig. 5 shows a functional configuration diagram of an information processing device 20 which is a modified example of the information processing device 10. Fig. 5 corresponds to Fig. 1. The hardware configuration of the information processing device 20 is similar to the hardware configuration of the information processing device 10 in Fig. 2. The information processing device 20 has the same configuration as the information processing device 10, except that it acquires depression areas from an external system and inputs them to the flooded area filling unit 13.

For example, depression information published by a local government can be used for the depression area 12A of the modified example. Some depression information published by local governments is obtained by manually detecting areas where the ground level is lower than the surrounding area from topographical information. Compared to a case where the depression detection unit 12 of the information processing device 10 automatically calculates the depression area 12A from the elevation information 30, manual detection may enable depressions to be detected with high accuracy.

***Description of Effects of Modifications***
The information processing device 20 can estimate flooded areas with high accuracy, thereby improving the speed of analysis, without the need to use the elevation information 30. Furthermore, the information processing device 20 can estimate flooded areas with even higher accuracy depending on the accuracy of the depression information used.

<Hardware configuration supplement>
2, the functions of the information processing apparatus 10 are implemented by software. However, the functions of the information processing apparatuses 10 and 20 may be implemented by hardware.
FIG. 6 shows a configuration in which the functions of the information processing devices 10 and 20 are realized by hardware. The information processing device 10 will be described below as an example. The electronic circuit 90 in FIG. 6 is a dedicated electronic circuit that realizes the functions of the initial flooded area estimation unit 11, the depression detection unit 12, and the flooded area filling unit 13 of the information processing device 10. The electronic circuit 90 is connected to a signal line 91. Specifically, the electronic circuit 90 is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA, an ASIC, or an FPGA. GA is an abbreviation for Gate Array. ASIC is an abbreviation for Application Specific Integrated Circuit. FPGA is an abbreviation for Field-Programmable Gate Array. The functions of the components of the information processing device 10 may be realized by one electronic circuit, or may be distributed and realized by multiple electronic circuits. Also, some of the functions of the components of the information processing device 10 may be realized by electronic circuits, and the remaining functions may be realized by software.

Each of the processor 510 and the electronic circuit 90 is also called a processing circuit or a circuitry. In the information processing device 10, the functions of the initial flooded area estimation unit 11, the depression detection unit 12, and the flooded area filling unit 13 may be realized by the circuitry.

The above is a description of the first embodiment. Of the multiple technical matters included in the first embodiment, one technical matter may be partially implemented. Alternatively, the technical matters included in the first embodiment may be partially implemented in combination.

10, 20 Information processing device, 11 Initial flooded area estimation unit, 11A Initial flooded area, 12 Depression detection unit, 12A Depression area, 13 Flooded area filling unit, 131 Candidate detection unit, 131A Flooded depression candidate, 132 Flooding determination unit, 132A Flooding determination result, 133 Flooded depression filling unit, 133A Final flooded area, 30 Elevation information, 510 Processor, 520 Main memory device, 530 Auxiliary memory device, 540 Input interface, 550 Output interface, 560 Communication interface, 570 Signal line, 90 Electronic circuit, 91 Signal line.

Claims (8)

an estimation unit that estimates an initially flooded area that is an area that exists in both the first image and the second image and indicates a flooded area by comparing a first image with a second image that was taken at an earlier date and time than the first image;
a flooded area filling unit which, when at least a portion of the initially flooded area is included in a depression area obtained by analyzing the terrain gradient for elevation information indicating the elevation of the terrain, calculates an index value indicating the degree of flooding of the depression area, and, when the index value is equal to or greater than a threshold value, determines that the depression area is flooded, and fills the depression area with a flooded area based on the determination result indicating flooding;
An information processing device comprising: The flooded area filling portion is
The information processing device according to claim 1 , wherein the depression region is filled by using a part of the initial flooded area included in the depression region as the flooded area. The flooded area filling portion is
The information processing apparatus according to claim 1 , wherein the depression area is filled by using the entire depression area as the flooded area. The flooded area filling portion is
An information processing device as described in any one of claims 1 to 3, wherein when the index value is less than a threshold value, the depression area is judged to be non-flooded, and a portion of the initial flooded area contained in the depression area is excluded from the initial flooded area based on the judgment result indicating non-flooded. The information processing device further includes:
The information processing apparatus according to claim 1 , further comprising a depression detection unit that acquires the altitude information and detects a depression region from the acquired altitude information. The first image and the second image are
6. The information processing apparatus according to claim 1, wherein the data is remote sensing data photographed by an imaging device mounted on a flying object. On the computer,
an estimation process for estimating an initially flooded area that is an area that exists in both the first image and the second image and indicates a flooded area by comparing a first image with a second image that was taken at an earlier date and time than the first image;
a flooded area filling process for, when at least a portion of the initially flooded area is included in a depression area obtained by analyzing the terrain gradient for elevation information indicating the elevation of the terrain, calculating an index value indicating the degree of flooding of the depression area, and judging the depression area to be flooded if the index value is equal to or greater than a threshold value, and filling the depression area with a flooded area based on the judgment result indicating flooding;
An information processing program that executes the above. The computer
By comparing a first image with a second image taken at an earlier date and time than the first image, an initial flooded area indicating a flooded area that exists in both the first image and the second image is estimated;
When at least a portion of the initially flooded area is included in a depression area obtained by analyzing the terrain gradient for elevation information indicating the elevation of the terrain, an index value indicating the degree of flooding of the depression area is calculated, and when the index value is equal to or greater than a threshold value, the depression area is determined to be flooded, and the depression area is filled with a flooded area based on the determination result indicating flooding.
Information processing methods.

Images