Monitoring system

Abstract

To improve the convenience of a monitoring system for predicting and distributing flooding conditions to local residents.SOLUTION: A monitoring system includes: a monitoring camera 15 that captures images of a flooding monitoring location 2 where flooding may occur; a sensor 16 that measures the water level at the flooding monitoring location 2; a flooding situation estimation unit 23 that estimates the current extent of flooding at the flooding monitoring location 2 based on data captured by the monitoring camera 15 and the measurement data by the sensor 16; a database 14 that accumulates extent of flooding estimated in the past and measurement data at the time of the flooding at the flooding monitoring location 2; a flooding situation prediction unit 24 that predicts the future extent of flooding and water level at the flooding monitoring location 2 using the accumulated data in the database 14; and a flooding image creation unit 25 that creates a flooding image map including the results of estimation by the flooding situation estimation unit 23 and / or the results of prediction by the flooding situation prediction unit 24.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a monitoring system for monitoring waterlogging monitoring locations where waterlogging may occur.

Background Art

[0002] Conventionally, various countermeasures have been taken against flood disasters such as waterlogging, river flooding, and inundation due to floods. For example, detecting dangerous waterways and roads based on measurement data from sensors such as waterlogging sensors, water level gauges, and rain gauges, and detecting dangerous waterways and roads by image processing of video data from surveillance cameras. Also, predicting the waterlogging range (or inundation range) based on the water volume measured by sensors and distributing alerts to administrators and local residents based on that information. Here, "waterlogging" where roads and fields are submerged in water and "inundation" where water enters buildings are originally words with different concepts, but in the following explanation, "waterlogging" shall include the concept of "inundation".

[0003] Examples of the prior art in the technical field related to the present invention are as follows. For example, Patent Document 1 discloses an imaging device that photographs a river and a monitoring device that determines the state of the river based on the photographed river video. The monitoring device calculates the flow velocity of the river from the optical flow on the river video at the time of determination based on the correspondence between the flow velocity of the river measured in advance using a flow velocity meter and the optical flow on the river video photographed at the time of that measurement, and determines whether the river is in a dangerous state by comparing it with a predetermined threshold value.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Conventional measures against water-related disasters had the following problems: (1) In the event of severe weather such as heavy rain, the equipment and cables on the surveillance camera side may be submerged in water, or a power outage may occur, preventing power from being supplied to the surveillance camera, which may result in the inability to acquire video data from the surveillance camera. (2) The background subtraction method is used as an image processing technique to estimate the extent of flooding. However, the flooded area may include not only the underpass portion where the road is lower than the surrounding area, but also the portion that extends beyond the underpass. However, it is difficult to estimate the extent of flooding in the portion that extends beyond the underpass, which is easily affected by the surrounding environment, using the background subtraction method. (3) When flooding occurs, water level information and information on flooded areas are distributed, but it is difficult to convey exactly how far the flooding has occurred. (4) While flooding is predicted based on current water levels, it is not possible to estimate future flooding conditions.

[0006] This invention has been made in view of the above-mentioned conventional circumstances, and aims to improve the convenience of a monitoring system for predicting flood conditions and distributing that information to local residents. [Means for solving the problem]

[0007] To achieve the above objectives, a monitoring system according to one aspect of the present invention is configured as follows: In a monitoring system for monitoring flood-prone locations where flooding may occur, the system includes a camera for photographing the flood-prone locations, a sensor for measuring the water level at the flood-prone locations, a flood situation estimation unit for estimating the current flooded area at the flood-prone locations based on the camera's image data and the sensor's measurement data, a database for accumulating previously estimated flooded areas and their measurement data at those times for the flood-prone locations, a flood situation prediction unit for predicting future flooded areas and water levels at the flood-prone locations using the accumulated data in the database, and the estimation results from the flood situation estimation unit and / or The system includes a flood image creation unit that creates a flood image diagram including the prediction results from the flood situation prediction unit, and the camera has the function of transmitting video data by wireless communication and is a camera that can remotely control a first state in which it takes pictures and transmits video data and a second state in which it does not perform these operations, and the flood image creation unit is characterized by creating a flood image diagram by superimposing a first line that shows the current water level represented by measurement data from sensors and a second line that shows the future water level predicted by the flood situation prediction unit in different forms onto a schematic diagram that shows a cross-sectional view of the area including the flood monitoring location.

[0008] Furthermore, a monitoring system according to another aspect of the present invention is configured as follows: In a monitoring system for monitoring flood-prone locations where flooding may occur, the system includes a camera for photographing the flood-prone locations, a sensor for measuring the water level at the flood-prone locations, a flood-prone location estimation unit that estimates the current flooded area at the flood-prone locations based on the camera's image data and the sensor's measurement data, a database for storing previously estimated flooded areas and their measurement data at those times for the flood-prone locations, a flood-prone location prediction unit that uses the stored data in the database to predict future flooded areas and water levels at the flood-prone locations, and the estimation results from the flood-prone location estimation unit and / or The system includes a flood image creation unit that creates a flood image diagram including the prediction results from the flood situation prediction unit, and the camera has the function of transmitting video data by wireless communication and is a camera that can remotely control a first state in which it takes pictures and transmits video data and a second state in which it does not perform these operations, and the flood image creation unit is characterized by creating a flood image diagram by superimposing a first frame showing the current flooded area estimated by the flood situation estimation unit and a second frame showing the future flooded area predicted by the flood situation prediction unit in different forms onto a schematic diagram of the area including the flood monitoring location in a plan view.

[0009] In the monitoring system according to the present invention, the flooding situation estimation unit may be configured to estimate the current flooded area at the flooding monitoring location using an AI model that has learned from video data captured by a camera when a heavy rain warning or heavy rain advisory was issued in the past.

[0010] Furthermore, the surveillance system according to the present invention may be configured to use an AI model that has been trained using the SSD (Single Shot Multi Detector) algorithm on video data captured under any conditions, including daytime and nighttime.

[0011] Furthermore, in the monitoring system according to the present invention, the camera may be configured to be remotely controlled to activate when a heavy rain warning or heavy rain advisory is issued for an area including a flood monitoring location, and to deactivate when the heavy rain warning or heavy rain advisory is lifted for the area including the flood monitoring location. [Effects of the Invention]

[0012] According to the present invention, it is possible to improve the convenience of a monitoring system for predicting flood conditions and distributing that information to local residents. [Brief explanation of the drawing]

[0013] [Figure 1] This figure shows an example configuration of a flood monitoring system according to one embodiment of the present invention. [Figure 2] This figure shows an example of the configuration of the flooding situation estimation unit in a flooding monitoring server. [Figure 3] This figure shows an example of the configuration of the flooding situation prediction unit in a flooding monitoring server. [Figure 4] This figure shows an example of a flood image diagram created by the flood monitoring server. [Figure 5] This figure shows another example of a flood image diagram created by the flood monitoring server. [Figure 6] This figure shows an example flowchart illustrating the operation of a flood monitoring server. [Modes for carrying out the invention]

[0014] One embodiment of the present invention will be described with reference to the drawings. Figure 1 shows an example configuration of a flood monitoring system according to one embodiment of the present invention. The flood monitoring system shown in Figure 1 is a system for estimating and predicting the occurrence of flooding, and has equipment deployed at a management center 1 attached to the office of each local government, and equipment deployed at any flood monitoring location 2 where flooding may occur. Examples of flood monitoring locations 2 include roads located at a lower elevation than the surrounding terrain, or underpasses that pass under other roads or railway lines.

[0015] The devices deployed in Management Center 1 include waterlogging monitoring server 11, operation terminal 12, distribution server 13, database 14, etc. The devices deployed at waterlogging monitoring location 2 include monitoring camera 15, sensor 16, etc. In FIG. 1, only one waterlogging monitoring location 2 is shown, but the number of waterlogging monitoring locations 2 is arbitrary, and monitoring cameras 15 and sensors 16 are deployed at each waterlogging monitoring location 2.

[0016] Monitoring camera 15 is a device that captures the situation in the vicinity of the installation location and has the function of transmitting the video data obtained by the capture to waterlogging monitoring server 11. Here, an LTE (Long Term Evolution) battery camera will be used as monitoring camera 15. Since the LTE battery camera has the function of transmitting video data by wireless communication using the LTE line, it is not necessary to prepare other wireless devices on the waterlogging monitoring location 2 side for cable connection. Also, since the LTE battery camera operates with a built-in battery (for example, dry battery), it is not necessary to connect to an external power supply device to receive power supply. Thus, by using the LTE battery camera, it is possible to simplify the configuration on the waterlogging monitoring location 2 side. Also, the LTE battery camera can be remotely controlled from waterlogging monitoring server 11 to switch between the startup state of performing capture and transmitting video data and the stop state (or standby state) where these operations are not performed. Therefore, the consumption of the battery of the LTE battery camera can be suppressed, and it is possible to operate for a long period without battery replacement.

[0017] Sensor 16 is a device that detects and measures various environmental data such as water level and flow rate in the vicinity of the installation location and has the function of transmitting the detection data and measurement data to waterlogging monitoring server 11. As sensor 16, for example, various detectors such as waterlogging sensors, water level gauges, flow meters, thermometers, hygrometers, and microphones can be used. Sensor 16 may operate constantly, may operate only when a predetermined event occurs, or may operate according to the control from waterlogging monitoring server 11. [[ID=ll]]

[0018] The flood monitoring server 11 is a server that estimates and predicts the flood situation at the flood monitoring location 2 based on the data obtained from the monitoring cameras 15 and sensors 16. The flood monitoring server 11 includes a weather information receiving unit 21, a data acquisition unit 22, a flood situation estimation unit 23, a flood situation prediction unit 24, and a flood image creation unit 25. The flood monitoring server 11 is, for example, a computer equipped with hardware resources such as a processor and a memory, and is configured to read a program related to the operations of the above functional units 21 to 25 from the memory and execute it by the processor. In FIG. 1, the flood monitoring server 11 is constituted by one computer, but the flood monitoring server 11 may be constituted by a plurality of computers that can operate in cooperation with each other.

[0019] The weather information receiving unit 21 receives an email containing information such as heavy rain warnings and heavy rain advisories from the Japan Meteorological Agency 3 (or other weather forecasting agencies), analyzes the email content to identify the target areas of heavy rain warnings and heavy rain advisories, and determines whether the flood monitoring location 2 is included in the target areas. If the flood monitoring location 2 is included in the target area, the information is sent to the data acquisition unit 22.

[0020] Based on the information received from the weather information receiving unit 21, the data acquisition unit 22 sends a start command to the monitoring camera 15 at the flood monitoring location 2 included in the target area of the heavy rain warning or heavy rain advisory to start shooting at the flood monitoring location 2. Also, when the sensor 16 operates in a form that waits for a start command, the data acquisition unit 22 also sends a start command to the sensor 16. After that, the data acquisition unit 22 acquires the video data transmitted from the monitoring camera 15 and the environmental data transmitted from the sensor 16. Then, when the heavy rain warning or heavy rain advisory is cancelled, the data acquisition unit 22 sends a stop command to the monitoring camera 15 (and the sensor 16) to stop the operation.

[0021] The flooding situation estimation unit 23 estimates the current flooded area of ​​the flooding monitoring location 2 based on the current video data and environmental data of the flooding monitoring location 2 acquired by the data acquisition unit 22. As shown in Figure 2, the flooding situation estimation unit 23 includes an AI model 31 and a flooding information management unit 32.

[0022] AI Model 31 was trained using a large amount of video data of flooding captured under various conditions, including daytime and nighttime, during past heavy rain warnings or advisories, as well as environmental data (e.g., water level) obtained at that time. AI Model 31 is created using an algorithm such as SSD (Single Shot Multi Detector). By inputting video data and environmental data into AI Model 31, it estimates the extent of flooding and outputs the result.

[0023] In this way, by using AI to estimate the extent of flooding, it becomes possible to estimate the extent of flooding in areas extending beyond the underpass with higher accuracy than conventional methods (for example, estimation using background subtraction). In this example, the SSD algorithm is used for AI training, but this is just one example, and other algorithms such as Faster-RCNN may also be used for AI training.

[0024] The flood information management unit 32 converts the estimated current flood area output from the AI ​​model 31 into a format suitable for subsequent processing and transmits it to the flood image creation unit 25 as estimated current flood area data along with the current water level measurement data.

[0025] Furthermore, the flooding situation estimation unit 23 transmits estimated data of the current flooded area, as well as video and environmental data used for estimation, to the database 14, where it stores this data along with time information such as the date and time. Consequently, the database 14 will accumulate estimated data of the flooded area, as well as video and environmental data, from past heavy rain warnings and advisories.

[0026] The flooding situation prediction unit 24 predicts the future flooding area and water level at the flooding monitoring location 2 based on estimated flooding area data, video data, and environmental data stored in the database 14 at the time of past heavy rain warnings and advisories, as well as current video data and environmental data acquired by the data acquisition unit 22, and transmits this predicted data to the flooding image creation unit 25. As shown in Figure 3, the flooding situation prediction unit 24 includes a water level information prediction unit 35 and a flooding area prediction unit 36.

[0027] The water level information prediction unit 35 predicts the water level after a predetermined time or number of days has elapsed, based on the water level fluctuations up to the present in relation to the current heavy rain warning / advisory and the water level fluctuations in the same region and under similar weather conditions in the past. As an example, three days after the lifting of the latest heavy rain warning / advisory, it receives the latest water level information from the database 14 and calculates the average value (in centimeters) of the water level prediction results for each weather condition up to that point and the latest water level information. In other words, by adding the actual water level results during the current heavy rain warning / advisory to the previously predicted water level, the accuracy of water level prediction is improved. As another example, under the assumption that future water levels will fluctuate in a similar trend to past cases similar to the current water level fluctuations, a function representing past water level fluctuations similar to the current water level fluctuations is derived from the accumulated data in the database 14, and future water levels are predicted in light of this function.

[0028] The flooding area prediction unit 36 ​​predicts the flooding area after a predetermined time or number of days has elapsed, based on the fluctuations in the flooding area up to the present in relation to the heavy rain warning / advisory and the fluctuations in the flooding area in the same region and under similar weather conditions in the past. As an example, three days after the lifting of the latest heavy rain warning / advisory, the latest flooding area calculation result is received from the database 14, and the average value (pixels) of the flooding area prediction results by weather information and region up to that point and the latest flooding area calculation result is calculated. In other words, the accuracy of flooding area prediction is improved by adding the actual flooding area calculation result for the current heavy rain warning / advisory to the previous flooding area prediction results. As another example, under the assumption that future flooding areas will fluctuate in a similar trend to past cases similar to the fluctuations in the flooding area of ​​the current situation, a function representing past flooding area fluctuations similar to the fluctuations in the flooding area up to the present is derived from the accumulated data of the database 14, and future flooding areas are predicted in light of this function.

[0029] The flood image creation unit 25 generates a flood image diagram showing the current estimated flood area and the future predicted flood area based on the estimated data of the current flood area and the measurement data of the current water level received from the flood situation estimation unit 23, and the predicted data of the future flood area and the predicted water level received from the flood situation prediction unit 24, and transmits it to the distribution server 13.

[0030] The flood image creation unit 25 creates two types of flood image diagrams: one showing a cross-sectional view of the flood situation as shown in Figure 4, and another showing an overhead view of the flood situation as shown in Figure 5. The flood image creation unit 25 is assumed to have pre-existing schematic diagram data that serves as the basis for the flood image, or to be able to obtain it from another server.

[0031] In the example in Figure 4, a schematic diagram showing a cross-sectional view of the area including flood monitoring location 2 displays a first line 41 representing the current water level (measured value), drawn based on current water level measurement data, and a second line 42 representing the future water level (predicted value), drawn based on future water level prediction data, superimposed on the diagram. In Figure 4, the first line 41 is drawn as a solid line and the second line 42 as a dashed line to distinguish them, but the form in which each line is drawn is not particularly limited, as long as each line is drawn in a different form so that it can be visually distinguished.

[0032] In the example in Figure 5, a schematic diagram of the area including flood monitoring location 2 is shown in plan view, with a first frame 51 representing the current flooded area (estimated value), drawn based on estimated data of the current flooded area, and a second frame 52 representing the future flooded area (predicted value), drawn based on predicted data of the future flooded area, superimposed on it. In Figure 4, the first frame 51 is drawn with a solid line and the second frame 52 is drawn with a dashed line to distinguish them, but the form in which each frame is drawn is not particularly limited, as long as they are drawn in a different form so that they can be visually distinguished. Also, in this example, a plan view is used to make it easier to understand the location and extent of flooding, but an overhead view looking down from diagonally above the area may also be used to make it easier to understand the topography (three-dimensional shape) of the area where flooding occurred.

[0033] The operation terminal 12 is a terminal used by local government officials to input text data, voice data, and other information necessary to inform local residents about flooding. The data entered into the operation terminal 12 is transmitted to the distribution server 13.

[0034] The distribution server 13 distributes flood image diagrams created by the flood image creation unit 25 of the flood monitoring server 11, as well as text data and voice data entered using the operation terminal 12, as flood situation information to user terminals (not shown) held by local residents. The distribution of flood situation information may be a push-type distribution where the distribution server 13 automatically provides it to pre-registered user terminals, or a pull-type distribution where the distribution server 13 responds to requests from user terminals.

[0035] Figure 6 shows an example flowchart illustrating the operation of the flood monitoring server 11. In the flood monitoring server 11, the weather information receiving unit 21 receives an email containing information such as heavy rain warnings and advisories provided by the Japan Meteorological Agency 3 (or other weather forecasting organizations) (step S11). The weather information receiving unit 21 analyzes the contents of the email to identify the areas covered by the heavy rain warnings and advisories (step S12), and determines whether the flood monitoring location 2 is included in the area covered by the heavy rain warnings and advisories (step S13).

[0036] If the target area does not include flood monitoring location 2 (Step S13; No), the process is terminated. On the other hand, if the target area does include flood monitoring location 2 (Step S13; Yes), the data acquisition unit 22 sends an activation command to the surveillance camera 15 at flood monitoring location 2, which is included in the target area for heavy rain warnings and advisories (Step S14), and acquires video data transmitted from the surveillance camera 15 and environmental data transmitted from the sensor 16 (Step S15).

[0037] Next, the flooding situation estimation unit 23 estimates the current flooded area of ​​the flood monitoring location 2 based on the current video data and environmental data acquired by the data acquisition unit 22 (step S16). Furthermore, if it is possible to predict future flooding conditions and water levels, the flooding situation prediction unit 24 uses the estimated flooding area data, video data, and environmental data stored in the database 14 at the time of past heavy rain warnings and advisories to predict the future flooded area and water level of the flood monitoring location 2.

[0038] Next, the flood image creation unit 25 creates a flood image diagram by superimposing the current estimated flood area, etc., based on the current flood area measured by the flood situation estimation unit 23 (step S17), and sends the flood image diagram to the distribution server 13 (step S18). After that, the weather information receiving unit 21 determines whether or not it has received a cancellation email from the Japan Meteorological Agency 3 (or another weather forecasting agency) containing information such as the cancellation of heavy rain warnings and advisories (step S19), and repeats the above process (steps S15 to S18) until a cancellation email is received. If a cancellation email is received (step S19; Yes), the data acquisition unit 22 sends a stop command to the surveillance camera 15 at the flood monitoring location 2 (step S20), stopping the operation of the surveillance camera 15.

[0039] As described above, the monitoring system in this example includes a monitoring camera 15 that photographs a flood monitoring location 2 where flooding may occur, a sensor 16 that measures the water level at the flood monitoring location 2, a flood situation estimation unit 23 that estimates the current flooded area at the flood monitoring location 2 based on the data captured by the monitoring camera 15 and the data measured by the sensor 16, a database 14 that stores the previously estimated flooded area and the measurement data at that time for the flood monitoring location 2, a flood situation prediction unit 24 that uses the data stored in the database 14 to predict the future flooded area and water level at the flood monitoring location 2, and a flood image creation unit 25 that creates a flood image diagram including the estimation results from the flood situation estimation unit 23 and / or the prediction results from the flood situation prediction unit 24. The monitoring camera 15 is an LTE battery camera that has the function of transmitting video data via wireless communication using an LTE line, operates on a built-in battery, and can be remotely controlled between an active state that takes pictures and transmits video data and a stopped state (or hibernation state) in which these operations are not performed. The flood image creation unit 25 creates a flood image diagram (cross-sectional view) as shown in Figure 4 by superimposing a first line 41, which represents the current water level represented by the measurement data from the sensor 16, and a second line 42, which represents the future water level predicted by the flood situation prediction unit 24, in different forms onto a schematic diagram of a cross-sectional view of the area including the flood monitoring location 2. The flood image creation unit 25 also creates a flood image diagram (plan view) as shown in Figure 5 by superimposing a first frame 51, which represents the current flooded area estimated by the flood situation estimation unit 23, and a second frame 52, which represents the future flooded area predicted by the flood situation prediction unit 24, in different forms onto a schematic diagram of a plan view of the area including the flood monitoring location 2.

[0040] Therefore, according to the monitoring system in this example, even in emergencies such as power outages near flood monitoring locations, it is possible to estimate the flooding situation in real time, create a flood image map, and distribute it. At this time, a flood image map (plan view) like Figure 4 is created and distributed, so local residents can grasp the extent of the flooding at a glance. Also, a flood image map (plan view) like Figure 5 is created and distributed, so local residents can grasp the location and extent of the flooding at a glance. In this way, the monitoring system in this example makes it possible to improve the convenience of a monitoring system for predicting and distributing flooding situations to local residents.

[0041] Although embodiments of the present invention have been described above, these embodiments are merely illustrative and do not limit the technical scope of the present invention. The present invention can take many other embodiments, and various modifications such as omissions and substitutions can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention as described herein, and are included in the scope of the invention and its equivalents as described in the claims.

[0042] Furthermore, the present invention can be provided not only as the devices described above or as systems composed of such devices, but also as methods executed by these devices, programs for a processor to realize the functions of these devices, and storage media for storing such programs in a computer-readable manner. [Industrial applicability]

[0043] This invention can be used in a monitoring system that monitors flood-prone areas where flooding may occur. [Explanation of symbols]

[0044] 1: Management Center, 2: Flood Monitoring Location, 3: Japan Meteorological Agency, 11: Flood Monitoring Server, 12: Operation Terminal, 13: Distribution Server, 14: Database, 15: Surveillance Camera, 16: Sensor, 21: Weather Information Receiving Unit, 22: Data Acquisition Unit, 23: Flood Situation Estimation Unit, 24: Flood Situation Prediction Unit, 25: Flood Image Creation Unit, 31: AI Model, 32: Flood Information Management Unit, 35: Water Level Information Prediction Unit, 36: Flood Area Prediction Unit

Claims

1. In a monitoring system that monitors flood-prone areas where flooding may occur, A camera for photographing the aforementioned flood monitoring location, A sensor for measuring the water level at the aforementioned flood monitoring location, A flooding situation estimation unit estimates the current flooded area at the flood monitoring location based on the captured data from the camera and the measured data from the sensor. A database that stores the estimated flood range and measurement data at the time of the aforementioned flood monitoring location, A flood situation prediction unit that uses the accumulated data in the aforementioned database to predict the future flood area and water level at the flood monitoring location, The system includes a flood image creation unit that creates a flood image diagram including the estimation results from the flood situation estimation unit and / or the prediction results from the flood situation prediction unit, The camera has the function of transmitting video data via wireless communication, and is a camera that can remotely control a first state in which it takes pictures and transmits video data, and a second state in which it does not perform these operations. The flood image creation unit is characterized by creating a flood image diagram by superimposing, in different forms, a first line indicating the current water level represented by the measurement data from the sensor and a second line indicating the future water level predicted by the flood situation prediction unit onto a schematic diagram showing a cross-sectional view of the area including the flood monitoring location.

2. In a monitoring system that monitors flood-prone areas where flooding may occur, A camera for photographing the aforementioned flood monitoring location, A sensor for measuring the water level at the aforementioned flood monitoring location, A flooding situation estimation unit estimates the current flooded area at the flood monitoring location based on the captured data from the camera and the measured data from the sensor. A database that stores the estimated flood range and measurement data at the time of the aforementioned flood monitoring location, A flood situation prediction unit that uses the accumulated data in the aforementioned database to predict the future flood area and water level at the flood monitoring location, The system includes a flood image creation unit that creates a flood image diagram including the estimation results from the flood situation estimation unit and / or the prediction results from the flood situation prediction unit, The camera has the function of transmitting video data via wireless communication, and is a camera that can remotely control a first state in which it takes pictures and transmits video data, and a second state in which it does not perform these operations. The flood image creation unit is characterized by creating a flood image diagram by superimposing, in different forms, a first frame showing the current flooded area estimated by the flooded area estimation unit and a second frame showing the future flooded area predicted by the flooded area prediction unit onto a schematic diagram of the area including the flooded area observation location in a plan view.

3. In the monitoring system according to claim 1 or claim 2, The flooding situation estimation unit is characterized by using an AI model that has learned from video data captured by the camera when past heavy rain warnings or advisories were issued to estimate the current flooded area at the flooding monitoring location.

4. In the monitoring system described in claim 3, The surveillance system is characterized by using an AI model that has been trained using the SSD (Single Shot Multi Detector) algorithm on video data captured under arbitrary conditions, including daytime and nighttime.

5. In the monitoring system according to claim 1 or claim 2, The monitoring system is characterized in that the camera is remotely controlled to activate when a heavy rain warning or heavy rain advisory is issued for the area including the flood monitoring location, and to deactivate when the heavy rain warning or heavy rain advisory is lifted for the area including the flood monitoring location.

Images