Inundation damage estimation system, inundation damage estimation program and inundation damage estimation method

Abstract

To provide an inundation damage estimation system, an inundation damage estimation method, and a program that can immediately estimate the inundation range and distribution of inundation depth with simple operation of a terminal.SOLUTION: An inundation damage estimation system 1 comprises an inundation depth information input device 2 for obtaining inundation depth input information for a reference point on map information selected by a user, an inundation depth estimation device 4 for estimating the inundation depth information for the reference point based on the inundation depth input information obtained by the inundation depth information input device 2, an inundation situation estimation device 5 for obtaining a water level altitude indicating an altitude of a water level within an object range including the reference point, and estimating the inundation depth distribution within the object range based on the inundation depth information for the reference point estimated by the inundation depth estimation device 4, the object range, and altitude information indicating the altitude of the land, and a user terminal device 8 that displays the inundation depth distribution estimated by the inundation situation estimation device 5.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a flood damage estimation system, a flood damage estimation program, and a flood damage estimation method for estimating flood damage conditions. [Background technology]

[0002] In recent years, flood damage has become more frequent due to the increase in heavy rains caused by global warming. When floods occur, it is important for local governments to grasp the extent of flooding early in order to respond to the disaster and reduce damage.

[0003] A system has been disclosed that predicts flood conditions and flood damage conditions by accurately recognizing and processing the topography and rainfall data of a target area in real time (see Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2002-298063 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology described in Patent Document 1 requires a user who owns a real flood map system to perform analysis using a processing device that stores various programs and various data required for the programs, which requires a separate system, is difficult to use, and involves a large amount of calculations.

[0006] The present invention aims to provide a flood damage estimation system, a flood damage estimation program, and a flood damage estimation method that, compared to conventional technology, can instantly estimate the flood area and flood depth distribution with simple operations on a browser. [Means for solving the problem]

[0007] The flood damage estimation system according to the present invention comprises: a flood depth information input device for acquiring flood depth input information of a reference point on map information selected by a user; A flood depth estimation device that estimates flood depth information of the reference point based on the flood depth input information acquired by the flood depth information input device; a flood situation estimation device that acquires water surface elevation indicating the elevation of the water surface within the target range based on the flood depth information of the reference point estimated by the flood depth estimation device, a target range including the reference point, and elevation information indicating the elevation of the land, and estimates the flood depth distribution within the target range; A user terminal device that displays the flood depth distribution estimated by the flood situation estimation device; and Equipped It is characterized by: Effect of the Invention

[0008] With this type of flood damage estimation system, the flooded area and flood depth distribution can be instantly estimated with simple operations on a browser. [Brief description of the drawings]

[0009] [Figure 1] 1 is an overall configuration diagram showing an example of a flood damage estimation system 1 according to the present embodiment. [Diagram 2] FIG. 2 is a block diagram showing an example of a machine learning device 3 of the present embodiment. [Diagram 3] 1 is a diagram showing an example of a learning model 10 and learning data 11 according to the present embodiment. [Figure 4] 4 is a flowchart showing an example of a machine learning method performed by the machine learning device 3 of the present embodiment. [Diagram 5] FIG. 2 is a block diagram showing an example of a flood depth estimation device 4 of the present embodiment. [Figure 6] FIG. 2 is a block diagram showing an example of a flood situation estimation device 5 of the present embodiment. [Figure 7] FIG. 2 is a block diagram showing an example of a flood damage estimation device 6 according to the present embodiment. [Figure 8]FIG. 2 is a block diagram showing an example of a route estimation device 7 according to the present embodiment. [Figure 9] FIG. 2 is a block diagram showing an example of a user terminal device 8 of the present embodiment. [Figure 10] FIG. 9 is a hardware configuration diagram showing an example of a computer 900 according to the present embodiment. [Figure 11] FIG. 2 is a flowchart showing an example of a flood damage estimation method according to the present embodiment. [Figure 12] An example of a reference point 91 used in the flood damage estimation method of this embodiment is shown. [Figure 13] An example of a flood depth acquisition method used in the flood damage estimation method of this embodiment is shown. [Figure 14] 1 shows an example of a target range 92 used in the flood damage estimation method of this embodiment. [Figure 15] 1 shows an example of a flood depth distribution 94 used in the flood damage estimation method of this embodiment. [Figure 16] 1 shows an example of a flood damage distribution 95 used in the flood damage estimation method of this embodiment. [Figure 17] 1 shows an example of an evacuation route used in the flood damage estimation method of this embodiment. [Figure 18] The following graph compares the time it takes to determine the flooding situation. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment for carrying out the present invention will be described with reference to the drawings. In the following, the scope necessary for the explanation to achieve the object of the present invention will be shown in a schematic manner, and the scope necessary for the explanation of the relevant part of the present invention will be mainly explained, and the parts that are omitted from the explanation will be based on the publicly known technology.

[0011] (Flood damage estimation system 1) FIG. 1 is an overall configuration diagram showing an example of a flood damage estimation system 1 according to the present embodiment.

[0012] The flood damage estimation system 1 of this embodiment comprises a flood depth information input device 2 for acquiring flood depth input information for a reference point 91 on map information selected by a user, a flood depth estimation device 4 that estimates the flood depth of the reference point 91 based on the flood depth input information input by the flood depth information input device 2, a flood situation estimation device 5 that acquires a water surface elevation 93 indicating the elevation of the water surface within the target range 92 based on flood depth information 112 for the reference point 91 estimated by the flood depth estimation device 4, a target range 92 including the reference point 91, and elevation information 522 indicating the elevation of the land, and estimates a flood depth distribution 94 within the target range 92, and a user terminal device 8 that displays the flood depth distribution 94 estimated by the flood situation estimation device 5.

[0013] Furthermore, the flood damage estimation device 6 estimates a flood damage distribution 95 including at least one of road flooding information and the number of flooded buildings within the target range 92, based on the flood depth distribution 94 within the target range 92 estimated by the flood situation estimation device 5, and social information 623 of roads, buildings, etc. Furthermore, the route estimation device 7 estimates at least one of an evacuation route and a rescue route from a starting point on the map information selected by the user to a destination on the map information, based on the flood damage distribution 95 and the social information 623. Furthermore, the machine learning device 3 operates as a subject of the machine learning learning phase for the flood depth input information input by the flood depth information input device 2.

[0014] The flood depth information input device 2, machine learning device 3, flood depth estimation device 4, flood situation estimation device 5, flood damage estimation device 6, route estimation device 7, and user terminal device 8 are configured, for example, as a general-purpose or dedicated computer (see FIG. 8 described later) and are connected to a wired or wireless network 9 so as to be able to transmit and receive various data to and from each other. Note that the number of flood depth information input devices 2, machine learning devices 3, flood depth estimation devices 4, flood situation estimation devices 5, flood damage estimation devices 6, route estimation devices 7, and user terminal devices 8, as well as the number of connections or connection configuration of the network 9 are not limited to the example of FIG. 1 and may be changed as appropriate. For example, each device may be configured using one or more servers in common or in combination.

[0015] (Flood depth information input device 2) The flood depth information input device 2 acquires flood depth input information for a reference point 91 selected by the user. The flood depth information input device 2 in this embodiment is an imaging device such as a camera or video camera, and may be a standalone camera or a camera attached to a smartphone or the like. The flood depth input information in this embodiment may be image data 111 captured by a camera or the like. The flood depth information input device 2 transmits the acquired flood depth input information for the reference point 91 to the machine learning device 3. It is preferable that the flood depth information input device 2 also transmits the time of acquisition.

[0016] (Machine learning device 3) Fig. 2 is a block diagram showing an example of a machine learning device 3 of this embodiment. Fig. 3 is a diagram showing an example of a learning model 10 and learning data 11 of this embodiment. The machine learning device 3 includes a control unit 30, a communication unit 31, a learning data storage unit 32, a trained model storage unit 33, an input unit 34, and an output unit 35.

[0017] The control unit 30 functions as a learning data acquisition unit 301 and a machine learning unit 302. The communication unit 31 is connected to external devices (e.g., the flood depth information input device 2, the flood depth estimation device 4, the flood situation estimation device 5, the flood damage estimation device 6, the route estimation device 7, and the user terminal device 8) via the network 9, and functions as a communication interface for transmitting and receiving various data. The input unit 34 accepts various input operations, and the output unit 35 functions as a user interface by outputting various information via a display screen or voice.

[0018] The learning data acquisition unit 301 acquires learning data 11 consisting of image data 111 as input data and flood depth information 112 as output data. The learning data acquisition unit 301 may acquire the learning data 11, for example, in cooperation with an external device connected via the communication unit 31 and the network 9, or may acquire the learning data 11 by accepting an input operation via the input unit 34 and the output unit 35. The learning data 11 is data used as teacher data (training data), verification data, and test data in supervised learning. Moreover, the flood depth information 112 is data used as a correct answer label in supervised learning.

[0019] The learning data storage unit 32 is a database that stores a plurality of sets of learning data 11 acquired by the learning data acquisition unit 301. The specific configuration of the database that constitutes the learning data storage unit 32 may be designed as appropriate.

[0020] The machine learning unit 302 performs machine learning using multiple sets of learning data 11 stored in the learning data storage unit 32. That is, the machine learning unit 302 inputs multiple sets of learning data 11 to the learning model 10 and generates a trained learning model 10 by having the learning model 10 learn the correlation between image data 111 and flood depth information 112 contained in the learning data 11.

[0021] The trained model storage unit 33 is a database that stores the trained learning model 10 (specifically, a group of adjusted weight parameters) generated by the machine learning unit 302. The trained learning model 10 stored in the trained model storage unit 33 is provided to an actual system (e.g., the inundation depth estimation device 4) via the network 9, a recording medium, etc. Note that, although the learning data storage unit 32 and the trained model storage unit 33 are shown as separate storage units in FIG. 2, they may be configured as a single storage unit.

[0022] As shown in FIG. 3, each of the multiple learning data 11 used for machine learning of the learning model 10 is composed of image data 111 and flood depth information 112.

[0023] The image data 111 constituting the learning data 11 was photographed by the flood depth information input device 2. The image data 111 includes a reference point 91, and includes an image in which the flood depth is at least 0cm. For example, the image data 111a at the top of Fig. 3 is an image taken at the time of flooding, and the image data 111n at the bottom of Fig. 3 is an image in which the flood depth is 0cm and there is no flooding. As in this example, it is preferable to learn images of the same reference point with a flood depth of 0cm and other images.

[0024] The flood depth information 112 constituting the learning data 11 includes at least object information, object part information and depth information. The object information is the object that is the object when comparing the height of the water surface, and may be, for example, a utility pole, a building, a fence, a tree, a bank, etc., as long as the height of the object is known. The object part information is the water surface or the ground. The water depth information is the height of the water surface. When the object part information is the ground, the water depth information is 0 cm. For example, the flood depth information 112a in the upper part of Figure 3 is an object that is a utility pole, an object part that is the water surface, and a flood depth of 50 cm. Moreover, the flood depth information 112n in the lower part of Figure 3 is an object that is a utility pole, an object part that is the water surface, and a flood depth of 0 cm.

[0025] The learning model 10 employs, for example, a neural network structure, and includes an input layer 101, an intermediate layer 102, and an output layer 103. Synapses (not shown) that connect each neuron are laid between each layer, and each synapse is associated with a weight. A group of weight parameters consisting of the weights of each synapse is adjusted by machine learning.

[0026] The input layer 101 has neurons whose number corresponds to the maximum number of pixels in image data 111 as input data, and the value of each pixel is input to each neuron. The output layer 103 has neurons whose number corresponds to inundation depth information 112 as output data, and the prediction result (inference result) of the inundation depth information 112 for the image data 111 is output as output data.

[0027] The number of learning models 10 stored in the trained model storage unit 33 is not limited to one, and multiple learning models 10 may be stored that have different conditions, such as different machine learning methods, different objects, different specifications of the flood depth information input device 2 (resolution, type of image sensor, etc.), and different types of data included in the flood depth information 112. In this case, the learning data storage unit 32 may store multiple types of learning data 11 having data configurations that respectively correspond to the multiple learning models 10 with different conditions.

[0028] (Machine learning methods) FIG. 4 is a flowchart showing an example of a machine learning method performed by the machine learning device 3 of this embodiment.

[0029] First, in step S110, the learning data acquisition unit 301 acquires a desired number of learning data 11 as a preliminary preparation for starting machine learning, and stores the acquired learning data 11 in the learning data storage unit 32. The number of learning data 11 prepared here may be set in consideration of the inference accuracy required for the learning model 10 finally obtained.

[0030] Next, in step S120, the machine learning unit 302 prepares the pre-learning learning model 10 in order to start machine learning. The pre-learning learning model 10 prepared here is configured with a neural network model, and the weights of each synapse are set to initial values.

[0031] Next, in step S130, the machine learning unit 302 acquires, for example, one set of learning data 11 randomly from the multiple sets of learning data 11 stored in the learning data storage unit 32.

[0032] Next, in step S140, the machine learning unit 302 inputs input data (image data 111) contained in one set of learning data 11 to the input layer 101 of the prepared learning model 10 before learning (or during learning). As a result, output data (flood depth information 112) is output as an inference result from the output layer 103 of the learning model 10, but this output data was generated by the learning model 10 before learning (or during learning). Therefore, in the pre-learning (or during learning) state, the output data output as an inference result indicates information different from the correct label (flood depth information 112) contained in the learning data 11.

[0033] Next, in step S150, the machine learning unit 302 performs machine learning by comparing the correct label included in the set of learning data 11 acquired in step S130 with the output data output from the output layer 103 as an inference result in step S140 and performing a process of adjusting the weight of each synapse (backpropagation). In this way, the machine learning unit 302 causes the learning model 10 to learn the correlation between the input data and the output data.

[0034] Next, in step S160, the machine learning unit 302 determines whether a predetermined learning termination condition has been satisfied, for example, based on an evaluation value of an error function based on the correct label included in the learning data 11 and the output data output as the inference result, or the remaining number of unlearned learning data 11 stored in the learning data memory unit 32.

[0035] In step S160, if the machine learning unit 302 determines that the learning end condition is not satisfied and that machine learning is to be continued (No in step S160), the process returns to step S130, and performs steps S130 to S150 on the learning model 10 under learning multiple times using unlearned learning data 11. On the other hand, in step S160, if the machine learning unit 302 determines that the learning end condition is satisfied and that machine learning is to be ended (Yes in step S160), the process proceeds to step S170.

[0036] Then, in step S170, the machine learning unit 302 stores the trained learning model 10 (adjusted weight parameter group) generated by adjusting the weights associated with each synapse in the trained model storage unit 33, and ends the series of machine learning methods shown in Fig. 15. In the machine learning method, step S110 corresponds to a learning data storage step, steps S120 to S160 correspond to a machine learning step, and step S170 corresponds to a trained model storage step.

[0037] As described above, the machine learning device 3 and machine learning method of this embodiment can provide a learning model 10 that can predict (infer) flood depth information 112 from image data 111 captured by the flood depth information input device 2.

[0038] (Flood depth estimation device 4) 5 is a block diagram showing an example of a flood depth estimation device 4 of this embodiment. The flood depth estimation device 4 estimates the flood depth of a reference point 91 by inputting the flood depth input information input by the flood depth information input device 2 into a learning model 10 provided by the machine learning device 3. The flood depth estimation device 4 includes a control unit 40 that acquires, generates or processes various information, a communication unit 41 that is connected to the flood depth information input device 2, the machine learning device 3, the flood situation estimation device 5, the flood damage estimation device 6, the route estimation device 7, the user terminal device 8, etc. and functions as a communication interface for transmitting and receiving various data, and a memory unit 42 that stores various information such as various programs (operating systems, user terminal programs, etc.) and data (learning model 10) used in the operation of the flood depth estimation device 4.

[0039] The control unit 40 functions as an image data acquisition unit 401 and a flood depth information generation unit 402. The input unit 44 accepts various input operations, and the output unit 45 functions as a user interface by outputting various information via a display screen or audio.

[0040] The image data acquisition unit 401 acquires image data 111 captured by the flood depth information input device 2. In this embodiment, the image data acquisition unit 401 acquires (receives) the image data 111 from the flood depth information input device 2 via the communication unit 41 and the network 9.

[0041] The inundation depth information generation unit 402 generates inundation depth information 112 based on the image data 111 acquired by the image data acquisition unit 401 and the learning model 10 stored in the memory unit 42. The generated inundation depth information 112 is stored as learning data 11 in the learning data memory unit 32 of the machine learning device 3.

[0042] The memory unit 42 pre-stores various types of information, such as various programs (operating systems, user terminal programs, etc.) and data (learning model 10) used in the operation of the flood depth estimation device 4. It also stores the flood depth information 112 generated by the flood depth information generation unit 402 as a record of the completion of work. The memory unit 42 may be substituted by a memory unit of an external computer (for example, a server-type computer or a cloud-type computer), in which case the control unit 40 only needs to access the external computer.

[0043] (Flooding Situation Estimation Device 5) 6 is a block diagram showing an example of a flood situation estimation device 5 of this embodiment. The flood situation estimation device 5 acquires a water surface elevation 93 within a target range 92 based on a reference point 91 input by a user, the flood depth of the reference point 91 estimated by the flood depth estimation device 4, a target range 92 input by the user, and various information such as map information 521, elevation information 522, and social information 623, and estimates a flood depth distribution 94 within the target range 92.

[0044] The flood situation estimation device 5 comprises a control unit 50 that acquires, generates or processes various information, a communication unit 51 that is connected to the flood depth information input device 2, the machine learning device 3, the flood depth estimation device 4, the flood damage estimation device 6, the route estimation device 7 and the user terminal device 8, etc. and functions as a communication interface for sending and receiving various data, and a memory unit 52 that stores various information such as various programs (operating systems, user terminal programs, etc.) and data (learning model 10) used in the operation of the flood situation estimation device 5.

[0045] The control unit 50 functions as a reference point acquisition unit 501, a flood depth acquisition unit 502, a target range acquisition unit 503, a water surface elevation acquisition unit 504, and a flood depth distribution generation unit 505. The input unit 54 accepts various input operations, and the output unit 55 functions as a user interface by outputting various information via a display screen or voice.

[0046] The reference point acquisition unit 501 identifies the position of the reference point 91 input by the user in the flood depth information input device 2 based on the map information 521 stored in the memory unit 52, and acquires the identified position information. The position information may be latitude and longitude, etc. The reference point acquisition unit 501 may also store in advance in the memory unit 52 a number of reference points 91 whose positions have been identified using the map information 521, and allow the user to select one reference point 91 from the multiple reference points 91.

[0047] The flood depth acquisition unit 502 acquires the flood depth information 112 of the reference point 91 estimated by the flood depth estimation device 4. In another embodiment, when the machine learning device 3 and the flood depth estimation device 4 are not used, the user may input the flood depth directly from the flood depth information input device 2, and the flood depth input by the user may be used as the flood depth information 112.

[0048] The target range acquisition unit 503 acquires the target range 92 input by the user. The target range 92 can be input by the user selecting it on the map using a mouse or the like, for example, by dragging a circle around the map displayed on the screen, or by using polygon data created in advance, such as a hazard map.

[0049] The water surface elevation acquisition unit 504 acquires the water surface elevation 93 of the flooded area within the target range 92 based on the flood depth information 112 and the elevation information 522 stored in the memory unit 52. It is assumed that the water surface elevation 93 spreads uniformly within the target range 92.

[0050] The flood depth distribution generating unit 505 generates a flood depth distribution 94 within the target range 92. The flood depth distribution 94 is displayed on a map displayed on the screen.

[0051] The memory unit 52 pre-stores various types of information, such as various programs (such as an operating system or user terminal program) and data (learning model 10) used in the operation of the flood situation estimation device 5. It also stores the flood depth distribution 94 generated by the flood situation estimation device 5 as a record of the completion of work. The memory unit 52 may be substituted by the memory unit 52 of an external computer (for example, a server-type computer or a cloud-type computer), in which case the control unit 50 simply needs to access the external computer.

[0052] The map information 521 is used as the background of the screen, and may be any information that allows the reference point 91 and the target range 92 to be set and confirmed, such as a Geospatial Information Authority of Japan map. The elevation information 522 may be any information that allows the elevation of the reference point 91 and the target range 92 to be confirmed, such as a basic map information digital elevation 5m mesh model.

[0053] (Flood damage estimation device 6) 7 is a block diagram showing an example of a flood damage estimation device 6 of this embodiment. The flood damage estimation device 6 estimates flood damage such as flooded roads and flooded buildings based on various information such as the flood depth distribution 94 generated by the flood depth distribution generation unit 505 of the flood situation estimation device 5, map information 621, altitude information 622, and social information 623.

[0054] The flood damage estimation device 6 comprises a control unit 60 that acquires, generates or processes various information, a communication unit 51 that is connected to the flood depth information input device 2, the machine learning device 3, the flood depth estimation device 4, the flood situation estimation device 5, the route estimation device 7 and the user terminal device 8, etc. and functions as a communication interface for sending and receiving various data, and a memory unit 62 that stores various information such as various programs (operating systems, user terminal programs, etc.) and data (learning model 10) used in the operation of the flood damage estimation device 6.

[0055] The control unit 60 functions as a flood damage estimation unit 601. The input unit 64 accepts various input operations, and the output unit 65 functions as a user interface by outputting various information via a display screen or audio.

[0056] The flood damage estimation unit 601 acquires water surface elevations 93 of roads and the like within the target range 92 based on the flood depth information 112 and social information 623 stored in the storage unit 62, and estimates the damage situation within the target range 92. The social information 623 may use infrastructure facility information including industrial base facilities such as roads, railways, waterworks, power networks, ports, dams, and life base facilities such as schools and hospitals, as well as statistical information such as population. For example, road centerline data from a digital map (basic land information) provided by the Geospatial Information Authority of Japan or building perimeter line data from the Geospatial Information Authority of Japan basic map information may be used.

[0057] (Route Estimation Device 7) 8 is a block diagram showing an example of a route estimation device 7 of this embodiment. The route estimation device 7 estimates an access route from a starting point to a destination point specified by a user, based on various information such as a flood damage distribution 95 estimated by a flood damage estimation unit 601 of a flood damage estimation device 6, map information 621, altitude information 622, and social information 623.

[0058] The route estimation device 7 comprises a control unit 70 that acquires, generates or processes various information, a communication unit 71 that is connected to the flood depth information input device 2, the machine learning device 3, the flood depth estimation device 4, the flood situation estimation device 5, the flood damage estimation device 6 and the user terminal device 8, etc., and functions as a communication interface for sending and receiving various data, and a memory unit 72 that stores various information such as various programs (operating systems, user terminal programs, etc.) and data (learning model 10) used in the operation of the route estimation device 7.

[0059] The control unit 70 functions as an evacuation route estimation unit 701. The input unit 74 accepts various input operations, and the output unit 75 functions as a user interface by outputting various information via a display screen or audio.

[0060] The evacuation route estimation unit 701 estimates an appropriate access route connecting a departure point and a destination point specified by the user within the target range 92, based on the flood damage distribution 95 and the social information 723 stored in the memory unit 72. The user may specify predetermined points on a map as the departure point and the destination point, or may select a point such as a shelter that is stored in advance. The evacuation route estimation unit 701 may set an appropriate access route based on the circumstances such as time, safety level, and the number of people at the shelter.

[0061] (User terminal device 8) 9 is a block diagram showing an example of a user terminal device 8. The user terminal device 8 is used when estimating flood damage, and operates the flood situation estimation device 5, the flood damage estimation device 6, the route estimation device 7, etc. The user terminal device 8 of this embodiment is a device such as a stationary personal computer, or a portable smartphone or tablet. The user terminal device 8 includes a control unit 80, a communication unit 81, a storage unit 82, an image capture unit 83, an input unit 84, and an output unit 85.

[0062] The control unit 80 functions as an image data transmission processing unit 801, a flood depth information processing unit 802, and a flood damage information processing unit 803. The communication unit 81 is connected to external devices (e.g., the flood depth information input device 2, the machine learning device 3, the flood depth estimation device 4, the flood situation estimation device 5, the flood damage estimation device 6, the route estimation device 7, etc.) via the network 9, and functions as a communication interface for transmitting and receiving various data. The storage unit 82 stores various programs (such as an operating system and a user terminal program) and data used in the operation of the user terminal device 8. The user terminal device 8 of this embodiment can operate the flood damage estimation system 1 using a browser for browsing websites on the network 9.

[0063] The image capturing unit 83 is composed of a camera (image sensor) such as a CMOS sensor or a CCD sensor having a predetermined resolution (number of pixels). By capturing an image of the reference point 91 with this camera, the user terminal device 8 can be used as the flood depth information input device 2.

[0064] The input unit 84 may be a keyboard, a touch panel, a microphone for voice input, etc., and accepts various input operations. In this embodiment, the input unit 84 inputs numerical values ​​for flood depth, the boundary lines of the target range 92, etc. The output unit 85 functions as a user interface by outputting various information via a display screen or voice, and may be a liquid crystal or organic electroluminescence display, a speaker, etc. The output unit 85 displays the operation screens and estimation results of the flood situation estimation device 5, the flood damage estimation device 6, and the route estimation device 7.

[0065] The image data transmission processing unit 801 transmits the image data 111 captured by the image capturing unit 83 to the flood depth information input device 2 via the communication unit 81 and the network 9.

[0066] When a user inputs flood depth information 112 from the input unit 84, the flood depth information processing unit 802 transmits the flood depth information 112 to the flood depth estimation device 4 via the communication unit 81 and the network 9.

[0067] The flood damage information processing unit 803 processes the forward damage information received from the flood situation estimation device 5, the flood damage estimation device 6, and the route estimation device 7, and notifies the user via voice or a display screen from the output unit 85 of the user terminal device 8.

[0068] (Hardware configuration of each device) 10 is a hardware configuration diagram showing an example of a computer 900. The flood depth information input device 2, machine learning device 3, flood depth estimation device 4, flood situation estimation device 5, flood damage estimation device 6, route estimation device 7, and user terminal device 8 are configured by a general-purpose or dedicated computer 900. The computer 900 functions as a flood damage estimation system 1 by a flood damage estimation program.

[0069] 8, the computer 900 includes, as its main components, a bus 910, a processor 912, a memory 914, an input device 916, an output device 917, a display device 918, a storage device 920, a communication I / F (interface) unit 922, an external device I / F unit 924, an I / O (input / output) device I / F unit 926, and a media input / output unit 928. Note that the above components may be omitted as appropriate depending on the application of the computer 900.

[0070] The processor 912 is composed of one or more arithmetic processing devices (such as a central processing unit (CPU), a micro-processing unit (MPU), a digital signal processor (DSP), a graphics processing unit (GPU), or a neural processing unit (NPU)) and operates as a control unit that controls the entire computer 900. The memory 914 stores various data and programs 930, and is composed of, for example, a volatile memory (DRAM, SRAM, etc.) that functions as a main memory, a non-volatile memory (ROM), a flash memory, etc.

[0071] The input device 916 is, for example, a keyboard, a mouse, a numeric keypad, an electronic pen, etc., and functions as an input unit. The output device 917 is, for example, a sound (audio) output device, a vibration device, etc., and functions as an output unit. The display device 918 is, for example, a liquid crystal display, an organic EL display, electronic paper, a projector, etc., and functions as an output unit. The input device 916 and the display device 918 may be integrally configured, such as a touch panel display. The storage device 920 is, for example, a HDD, an SSD, etc., and functions as a storage unit. The storage device 920 stores various data necessary for the execution of the operating system and the program 930.

[0072] The communication I / F unit 922 is connected to a network 940 (which may be the same as the network 9 in FIG. 1) such as the Internet or an intranet by wire or wirelessly, and functions as a communication unit that transmits and receives data to and from other computers according to a predetermined communication standard. The external device I / F unit 924 is connected to an external device 950 such as a camera, a printer, a scanner, a reader / writer by wire or wirelessly, and functions as a communication unit that transmits and receives data to and from the external device 950 according to a predetermined communication standard. The I / O device I / F unit 926 is connected to an I / O device 960 such as various sensors and actuators, and functions as a communication unit that transmits and receives various signals and data, such as detection signals from sensors and control signals to actuators, between the I / O device 960. The media input / output unit 928 is composed of a drive device such as a DVD drive or a CD drive, a memory card slot, and a USB connector, and reads and writes data from and to a medium (non-temporary storage medium) 970 such as a DVD, a CD, a memory card, or a USB memory.

[0073] In the computer 900 having the above configuration, the processor 912 calls up the program 930 stored in the storage device 920 into the memory 914, executes the program, and controls each unit of the computer 900 via the bus 910. The program 930 may be stored in the memory 914 instead of the storage device 920. The program 930 may be recorded in the medium 970 in an installable file format or an executable file format, and provided to the computer 900 via the media input / output unit 928. The program 930 may be provided to the computer 900 by downloading it via the network 940 via the communication I / F unit 922. In addition, the computer 900 may realize various functions realized by the processor 912 executing the program 930, for example, with hardware such as a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC).

[0074] The computer 900 is, for example, a stationary computer or a portable computer, and is an electronic device of any form. The computer 900 may be a client-type computer, a server-type computer, or a cloud-type computer, or may be, for example, an embedded computer called a control panel, a controller (including a microcomputer, a programmable logic controller, and a sequencer). The computer 900 may also be applied to devices other than the flood depth information input device 2, the machine learning device 3, the flood depth estimation device 4, the flood situation estimation device 5, the flood damage estimation device 6, the route estimation device 7, and the user terminal device 8, etc.

[0075] In this way, according to the flood damage estimation system 1 of this embodiment, the flooded area and flood depth distribution 94 can be instantly estimated with simple operations on a terminal.

[0076] (Flood damage estimation method) 11 is a flowchart showing an example of a flood damage estimation method according to this embodiment. In the following, an example will be described in which the flood damage estimation system 1 estimates flood damage. A user can use the flood damage estimation system 1 by opening a browser on a user terminal device 8.

[0077] First, in step ST1, the flood situation estimation device 5 acquires a reference point 91 (ST1). The reference point 91 is a location for which the user wants to know or knows the flood depth information 112. The location of the reference point 91 is specified by the user inputting it onto a map displayed on the display of the user terminal device 8. The user may input an address or place name, etc. in advance to narrow the display range on the screen.

[0078] FIG. 12 shows an example of a reference point 91 used in the flood damage estimation method of this embodiment. The reference point 91 in this embodiment is indicated by an x ​​on the map, but any distinguishable symbol may be used. The position of the reference point 91 may be specified by the user touching the map displayed on the display of the user terminal device 8. Alternatively, multiple candidates for the reference point 91 may be set to be displayed on the map in advance, and the user may select one of them. For example, the candidate for the reference point 91 may be the position where the camera is installed. The position of the reference point 91 may be specified by the touched position, or may be specified by the position of the reference point 91 that is closest to the touched position among the multiple candidates for the reference point 91.

[0079] Furthermore, the point where the user captured an image using the user terminal device 8 may be set as the reference point 91. In this case, the user terminal device 8 may be used as the flood depth information input device 2, and the position of the reference point 91 may be identified using the position information of the user terminal device 8.

[0080] Next, in step ST2, the inundation situation estimation device 5 acquires the inundation depth of the reference point 91 obtained in step 1 (ST2). The inundation situation estimation device 5 of this embodiment acquires the inundation depth estimated by the inundation depth estimation device 4. The inundation depth estimation device 4 of this embodiment estimates the inundation depth for the image of the reference point 91 obtained in step 1 using the learning model 10 learned by the machine learning device 3.

[0081] Figure 13 shows an example of a flood depth acquisition method used in the flood damage estimation method of this embodiment. The flood depth estimation device 4 of this embodiment estimates the flood depth using a learning model 10 for the image of the reference point 91 shown in Figure 11(a) and the image without flooding shown in Figure 11(b) input from the flood depth information input device 2. The flood depth estimated by the flood depth estimation device 4 is transmitted to the flood situation estimation device 5.

[0082] Next, in step ST3, the flood situation estimation device 5 acquires a target range 92 (ST3). The target range 92 is a range for which the user wants to know flood depth information 112. The target range 92 is set to surround a reference point 91. The flood depth within the reference point can be estimated based on the flood depth information 112 of the reference point 91.

[0083] Fig. 14 shows an example of a target range 92 used in the flood damage estimation method of this embodiment. The position and range of the target range 92 of this embodiment is specified by the user dragging the periphery of a predetermined range on the map displayed on the display of the user terminal device 8. In Fig. 12, a line formed to surround the x mark indicating the reference point 91 indicates the periphery of the target range 92.

[0084] Next, in step ST4, the flood situation estimation device 5 acquires the water surface elevation 93 (ST4). The flood situation estimation device 5 acquires the water surface elevation 93 of the target range 92 based on the flood depth information 112, map information 521, and elevation information 522 of the target range 92. The water surface elevation 93 is acquired based on the flood depth information 112 and elevation information 522 of the reference point 91.

[0085] Next, in step ST5, the inundation depth distribution generation unit 505 generates a inundation depth distribution 94 based on the water surface elevation 93 and the elevation information 522 of the target range 92 (ST5).

[0086] Fig. 15 shows an example of a flood depth distribution 94 used in the flood damage estimation method of this embodiment. The flood depth distribution 94 of this embodiment displays the flood depth within a target area in different colors or shades according to the depth. By creating the flood depth distribution 94 in this way, dangerous locations can be understood at a glance.

[0087] Next, in step ST5, the flood damage estimation device 6 acquires the flood damage distribution 95 (ST5). The flood damage estimation device 6 acquires the flood damage distribution 95 based on the water surface elevation 93 of the target range 92 and the social information 623. The flood damage distribution 95 estimates the current situation, for example, whether roads, etc. are flooded.

[0088] FIG. 16 shows an example of a flood damage distribution 95 used in the flood damage estimation method of this embodiment. The flood damage distribution 95 of this embodiment displays the flood depth on roads within the target range 92 estimated by the flood damage estimation device 6 in different colors or shades according to the depth. By creating the flood damage distribution 95 in this way, dangerous locations can be understood at a glance.

[0089] In this way, according to the flood damage estimation method of this embodiment, the flooded area and flood depth distribution 94 can be instantly estimated with simple operations on a terminal. Also, the flood damage distribution 95 can be estimated. The flood damage distribution 95 may display the flooded state of roads, the number of flooded buildings, etc. Furthermore, the damage may be ranked according to the results of estimated economic losses.

[0090] FIG. 17 shows an example of an evacuation route distribution 96 used in the flood damage estimation method of this embodiment. The evacuation route distribution 96 of this embodiment displays an access route from a departure point to a destination point designated by the user within the target range 92 estimated by the route estimation device 7. In this way, by estimating the evacuation route distribution 96, the user can understand the evacuation route at a glance.

[0091] In this way, according to the flood damage estimation method of this embodiment, the evacuation route distribution 96 can be instantly estimated by simply operating a browser on a terminal. The user may specify predetermined points on a map as the starting point and destination point, or may select a pre-stored point such as an evacuation shelter. The evacuation route estimation unit 701 may set an appropriate access route based on the circumstances such as time, safety level, and number of people at the evacuation shelter.

[0092] FIG. 18 shows a graph comparing the time it takes to grasp the flooding situation. It takes about three hours at the earliest to grasp the flooding situation by aerial photographs taken by a drone, taking into account the weather and the like. In contrast, in the flood damage estimation system 1 of this embodiment, when a SNS capable of identifying a location is used, it takes two and a half hours to grasp the flooding situation, and when a SNS with GPS information is used, it takes one and a half hours to grasp the flooding situation. Furthermore, when a report is made by an IoT camera or the like with clear location information, it takes less than one hour, and it is possible to grasp the flooding situation within a few minutes to a few tens of minutes. Furthermore, the flood damage estimation system 1 of this embodiment can utilize reports from cameras, SNS, etc. from the initial response to flooding to the damage investigation stage.

[0093] As described above, the flood damage estimation system 1 of this embodiment includes a flood depth information input device 2 for acquiring flood depth input information of a reference point 91 on map information 521 selected by a user, a flood depth estimation device 4 for estimating flood depth information 112 of the reference point 91 based on the flood depth input information input by the flood depth information input device 2, a flood situation estimation device 5 for acquiring water surface elevation 93 indicating the elevation of the water surface within the target range 92 based on the flood depth information 112 of the reference point 91 estimated by the flood depth estimation device 4, a target range 92 including the reference point 91, and elevation information 522 indicating the elevation of the land, and estimating a flood depth distribution 94 within the target range 92, and a user terminal device 8 for displaying the flood depth distribution 94 estimated by the flood situation estimation device 5. Thus, according to the flood damage estimation system 1 of this embodiment, the flooded range and flood depth distribution 94 can be instantly estimated with simple operations on a browser.

[0094] Moreover, in the flood damage estimation system 1 of this embodiment, the flood situation estimation device 5 has a reference point acquisition unit 501 that identifies the position of the reference point 91 acquired by the flood depth information input device 2 based on map information 521 and acquires the identified position information, a flood depth acquisition unit 502 that acquires flood depth information 112 of the reference point 91 estimated by the flood depth estimation device 4, a target range acquisition unit 503 that acquires the input target range 92, a water surface elevation acquisition unit 504 that acquires the water surface elevation 93 within the target range based on the flood depth information 112 and elevation information 522, and a flood depth distribution generation unit 505 that estimates the flood depth distribution 94 within the target range. Therefore, according to the flood damage estimation system 1 of this embodiment, the flood depth distribution 94 can be generated quickly and appropriately.

[0095] Furthermore, the flood damage estimation system 1 of this embodiment is equipped with a flood damage estimation device 6 that estimates a flood damage distribution 95 including at least one of road flooding information and the number of flooded buildings within a target area based on the flood depth distribution 94 estimated by the flood situation estimation device 5 and social information 623, and a user terminal device 8 displays at least one of the road flooding information and the number of flooded buildings estimated by the flood damage estimation device 6. Therefore, the flood damage estimation system 1 of this embodiment can quickly and appropriately estimate the damage situation of roads and buildings.

[0096] Furthermore, the flood damage estimation system 1 of this embodiment includes a route estimation device 7 that estimates at least one of an evacuation route and a rescue route from a starting point on the map information 721 selected by the user to a destination on the map information 721 based on the flood damage distribution 95 estimated by the flood damage estimation device 6, road flooding information, and social information 723, and the user terminal device 8 displays at least one of the evacuation route and rescue route estimated by the route estimation device 7. Therefore, the flood damage estimation system 1 of this embodiment can quickly and appropriately estimate an evacuation route or rescue route.

[0097] Furthermore, the flood damage estimation system 1 of this embodiment is equipped with a machine learning device 3 that operates as the main actor in the machine learning learning phase for the flood depth input information acquired by the flood depth information input device 2, and the flood depth estimation device 4 estimates flood depth information 112 for the reference point 91 by inputting the flood depth input information acquired by the flood depth information input device 2 into the learning model 10 provided by the machine learning device 3. Therefore, the flood damage estimation system 1 of this embodiment can quickly and appropriately estimate flood depth through machine learning.

[0098] Furthermore, in the flood damage estimation system 1 of this embodiment, the flood depth input information is image data 111, and the machine learning device 3 has a learning data acquisition unit 301 that acquires multiple learning data 11 consisting of the image data 111 as input data and flood depth information 112 as output data, and a machine learning unit 302 that trains the learning model 10 to learn the correlation between the image data 111 and the flood depth information 112 contained in the multiple learning data 11. Therefore, according to the flood damage estimation system 1 of this embodiment, the flood depth information 112 can be quickly and appropriately estimated from the image data 111 through machine learning.

[0099] In addition, in the flood damage estimation system 1 of this embodiment, the flood depth estimation device 4 has an image data acquisition unit 401 that acquires image data 111 from the flood depth information input device 2, and a flood depth information generation unit 402 that generates flood depth information 112 based on the image data 111 acquired by the image data acquisition unit 401 and the learning model 10.

[0100] It should be noted that the present invention is not limited to the embodiments, and that although many specific details are included for illustrative purposes in the description of the embodiments, those skilled in the art may make various variations and modifications to these details.

[0101] For example, the flood damage estimation system 1 of this embodiment can also be used as a simulation. When used as a simulation, the machine learning device 3 and the flood depth estimation device 4 are not used, and the user inputs an assumed flood depth into the flood depth information input device 2, and the flood depth input by the user is used as the flood depth information 112. [Explanation of symbols]

[0102] 1...Flood damage estimation system, 2...Flood depth information input device, 3...Machine learning device, 4...Flood depth estimation device, 5...Flood situation estimation device, 6...Flood damage estimation device, 7...Evacuation route estimation device, 8...User terminal device, 9...Network, 10…Learning model, 11…Learning data

Claims

1. a flood depth information input device for acquiring flood depth input information of a reference point on map information selected by a user; A flood depth estimation device that estimates flood depth information of the reference point based on the flood depth input information acquired by the flood depth information input device; a flood situation estimation device that acquires water surface elevation indicating the elevation of the water surface within the target range based on the flood depth information of the reference point estimated by the flood depth estimation device, a target range including the reference point, and elevation information indicating the elevation of the land, and estimates the flood depth distribution within the target range; A user terminal device that displays the flood depth distribution estimated by the flood situation estimation device; and Equipped A flood damage estimation system characterized by the above.

2. The flooding situation estimation device includes: a reference point acquisition unit that identifies the positions of the reference points acquired by the flood depth information input device based on the map information and acquires the identified position information; a flood depth acquisition unit that acquires the flood depth information of the reference point estimated by the flood depth estimation device; A target range acquisition unit that acquires the input target range; a water surface elevation acquisition unit that acquires the water surface elevation within the target range based on the flood depth information and the elevation information and estimates the flood depth distribution within the target range; have 2. The flood damage estimation system according to claim 1 .

3. a flood damage estimation device that estimates a flood damage distribution including at least one of road flooding information and the number of flooded buildings within the target area based on the flood depth distribution estimated by the flood situation estimation device and social information, The user terminal device displays at least one of the road flooding information and the number of flooded buildings estimated by the flood damage estimation device.

2. The flood damage estimation system according to claim 1 .

4. a route estimation device that estimates at least one of an evacuation route and a rescue route from a starting point on the map information selected by the user to a destination on the map information based on the road flooding information and the social information included in the flood damage distribution estimated by the flood damage estimation device, The user terminal device displays at least one of the evacuation route and the rescue route estimated by the route estimation device.

4. The flood damage estimation system according to claim 3.

5. A machine learning device that operates as a subject of a machine learning learning phase for the flood depth input information acquired by the flood depth information input device, The flood depth estimation device estimates the flood depth information of the reference point by inputting the flood depth input information input by the flood depth information input device into a learning model provided by the machine learning device.

2. The flood damage estimation system according to claim 1 .

6. The flood depth input information is image data, The machine learning device includes: a learning data acquisition unit that acquires a plurality of learning data pieces that are composed of the image data as input data and the flood depth information as output data; a machine learning unit that causes the learning model to learn a correlation between the image data included in the plurality of learning data and the flood depth information; have 6. The flood damage estimation system according to claim 5.

7. The flood depth estimation device comprises: An image data acquisition unit that acquires the image data from the flood depth information input device; a flood depth information generation unit that generates the flood depth information based on the image data acquired by the image data acquisition unit and the learning model; have 7. The flood damage estimation system according to claim 6.

8. A flood damage estimation program that causes a computer to function as the flood damage estimation system according to any one of claims 1 to 7.

9. Obtaining flood depth input information for a reference point on map information selected by a user; performing a learning phase of machine learning on the flood depth input information; A step of estimating flood depth information of the reference point by inputting the flood depth input information into a learning model provided from the learning phase of the machine learning; A step of acquiring water surface elevation indicating the elevation of the water surface within the target range based on the flood depth information of the reference point, a target range including the reference point, and elevation information indicating the elevation of the land, and estimating the flood depth distribution within the target range; Displaying the estimated flood depth distribution; have A flood damage estimation method comprising:

Images