JP2012198079A - Water level monitoring device, water level monitoring method and water level monitoring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a water level to be monitored without requiring preparations in advance.SOLUTION: A water level monitoring device 10 acquires an observation image picked up by a camera 30 so disposed as to be able to photograph a prescribed water area and covering an observation point that is to be observed out of the water area. Further, the water level monitoring device 10 detects, out of a plurality of lines that are obtained from the observation image by edge detection and whose mutually relative directions are similar, a first line whose shape does not vary among the frames of the observation image and is positioned uppermost on the observation image. Further, the water level monitoring device 10 detects, out of the plurality of lines that are obtained from the observation image by edge detection and whose mutually relative directions are similar, a second line whose shape varies among the frames of the observation image. Further, the water level monitoring device 10 monitors the water level at the observation point on the basis of the first line and the second line.

Description

  The present invention relates to a water level monitoring device, a water level monitoring method, and a water level monitoring program.

  As one aspect of technology for preventing natural disasters, technology for monitoring the water level in water areas such as rivers, seas, and lakes is known. For example, an image obtained by capturing an image of a water level meter provided in a river is used for monitoring. When monitoring the water level in this way, it is necessary to install a water level gauge at each location to be monitored. For this reason, when monitoring the river with high density by narrowing the interval between water level gauges, and when monitoring the rivers over a wide area, the installation cost of the water level gauge becomes enormous. Monitoring the water level is not practical.

  An example of a technique for monitoring the water level without using such a water level meter is a water level observation system using a CCTV (Closed-Circuit TeleVision) camera. This water level observation system detects the water surface from an image of the water surface and a measurement object such as a structure existing in the river by a CCTV camera, and detects the water surface position and the three-dimensional spatial information or reference image measured in advance. Use to measure the water level.

JP 2008-57994 A

  However, the above-described conventional technology has a problem that the water level cannot be monitored without advance preparation as described below.

  That is, the water level observation system measures the water level by comparing the image captured by the CCTV camera with the three-dimensional spatial information or the reference image measured in advance. For this reason, in the above water level observation system, it is necessary to prepare and register three-dimensional spatial information and a reference image in advance. Further, in the above water level observation system, when the environment of the location to be observed changes, the three-dimensional spatial information and the reference image must be registered again.

  The disclosed technology has been made in view of the above, and an object thereof is to provide a water level monitoring device, a water level monitoring method, and a water level monitoring program capable of monitoring the water level without preparation in advance.

  The water level monitoring device disclosed in the present application has an acquisition unit that acquires an observation image captured including an observation point to be observed in the water area from an imaging apparatus provided so as to be able to image a predetermined water area. Further, the water level monitoring device, among a plurality of lines similar in mutual direction obtained by edge detection from the observation image acquired by the acquisition unit, the shape does not change between the frames of the observation image, and the A first line located at the highest position on the observation image is detected. Furthermore, the detection unit detects a second line whose shape changes between frames of the observation image. Furthermore, the water level monitoring device has a monitoring unit that monitors the water level at the observation point based on the first line and the second line detected by the detection unit.

  According to one aspect of the water level monitoring device disclosed in the present application, there is an effect that the water level can be monitored without prior preparation.

FIG. 1 is a diagram illustrating a system configuration of the disaster prevention system according to the first embodiment. FIG. 2 is a functional block diagram illustrating the configuration of the water level monitoring apparatus according to the first embodiment. FIG. 3 is a diagram illustrating an example of observation image metadata. FIG. 4 is a diagram illustrating an example of camera metadata. FIG. 5 is a diagram illustrating an example of river line data. FIG. 6 is a diagram illustrating an example of dangerous point data. FIG. 7 is a diagram illustrating an example of a schematic electronic map. FIG. 8 is a diagram illustrating an example of a line image. FIG. 9 is a flowchart illustrating the procedure of the entire process of the water level monitoring apparatus according to the first embodiment. FIG. 10 is a flowchart illustrating the procedure of the camera control process according to the first embodiment. FIG. 11 is a flowchart illustrating the procedure of the line detection process according to the first embodiment. FIG. 12 is a flowchart illustrating a procedure of notification control processing according to the first embodiment. FIG. 13 is a diagram for explaining the effect of the water level monitoring apparatus according to the first embodiment. FIG. 14 is a diagram for explaining the effect of the water level monitoring apparatus according to the first embodiment. FIG. 15 is a diagram for explaining the effect of the water level monitoring apparatus according to the first embodiment. FIG. 16 is a schematic diagram illustrating an example of a computer that executes a water level monitoring program according to the first and second embodiments.

  Hereinafter, embodiments of a water level monitoring device, a water level monitoring method, and a water level monitoring program disclosed in the present application will be described in detail with reference to the drawings. Note that this embodiment does not limit the disclosed technology. Each embodiment can be appropriately combined within a range in which processing contents are not contradictory.

[System configuration]
First, the system configuration of the disaster prevention system according to the present embodiment will be described. FIG. 1 is a diagram illustrating a system configuration of the disaster prevention system according to the first embodiment. In the disaster prevention system 1 illustrated in FIG. 1, a water level monitoring device 10, cameras 30 </ b> A to 30 </ b> C, a river information management device 50, and a user terminal 70 are accommodated. In the example of FIG. 1, the water level of the river at the observation point is obtained using observation images obtained by imaging the observation point of the river with the cameras 30 </ b> A to 30 </ b> C provided with the water level monitoring devices 10 scattered in the same river basin. Assume that you want to monitor

  The water level monitoring device 10, the cameras 30A to 30C, the river information management device 50, and the user terminal 70 are communicably connected via the network 3. As an aspect of the network 3, a communication network such as the Internet, a LAN (Local Area Network), and a VPN (Virtual Private Network) is employed.

  In the example of FIG. 1, one water level monitoring device 10, three cameras, one river information management device, and one user terminal 70 are illustrated, but the disclosed system is not limited to the illustrated configuration. That is, the disaster prevention system 1 only needs to accommodate at least one set of cameras and water level monitoring devices, and can accommodate any number of water level monitoring devices, cameras, river information management devices, and user terminals. In the following, when the respective devices of the camera 30A to the camera 30C are described without distinction, they may be expressed as “camera 30”.

  The camera 30 is an imaging device that captures an image. An example of such a camera 30 is an imaging device using a charge coupled device (CCD), a complementary metal oxide semiconductor (CMOS), or the like.

  Furthermore, the camera 30 functions as a network camera, and transmits an observation image obtained by imaging a river observation point to the water level monitoring apparatus 10 via the network 3. Here, it is not always necessary to install a new camera when imaging the observation point of the river. For example, a CCTV (Closed-circuit Television) camera that is already installed for the purpose of river management can be used as the camera 30. By using CCTV cameras installed in a wide area and with high density in a large number of rivers in this way, the initial cost related to camera installation can be reduced. In the present embodiment, from the viewpoint of detecting increase in water due to local heavy rain, description will be made assuming that observation points are set on the center line of the river at intervals of 1 km.

  The river information management device 50 is a device that collects and manages comprehensive information about rivers, such as rainfall, flow rate, water intake, water quality, and the like. For example, the river information management device 50 acquires the rainfall measured by a radar or a telemeter (not shown). Then, the river information management device 50 calculates the average rainfall in the rainfall area included in the river basin from the rainfall measured by the radar and the telemeter provided in the river basin of the radar and telemeter, and then the water level monitoring device 10 is notified.

  The user terminal 70 is an information processing device used by various users. As one aspect of the user terminal 70, there are a monitoring monitor provided as a monitoring facility of the disaster prevention system 1, an information processing apparatus used by a person concerned with the disaster prevention system 1, and the like. As another aspect, there is a server device provided as a media broadcasting facility. As a further aspect, there is an information processing apparatus used by an end user who accesses the network 3. The information processing apparatus is not limited to a fixed terminal such as a personal computer (PC), but may be a mobile terminal such as a mobile phone, PHS (Personal Handyphone System) or PDA (Personal Digital Assistant). Can be adopted.

  The water level monitoring device 10 is a device that monitors the water level of a river. As an example, when the water level monitoring apparatus 10 receives the average rainfall in the rainy area included in the river basin from the river information management apparatus 50, the water level monitoring apparatus 10 identifies an observation point existing in the rainy area. And the water level monitoring apparatus 10 selects the camera 30 which can image the observation point which exists in a rainfall area, and makes the camera 30 image the observation point of a river. In addition, the water level monitoring device 10 determines whether or not the water level at the observation point detected from the observation image captured by the camera 30 satisfies a predetermined water increase standard. At this time, if the water level monitoring device 10 determines that the water level at the observation point satisfies the water increase standard, the water level monitoring device 10 notifies the user terminal 70 of the observation point where the water level has increased.

  Here, the water level monitoring apparatus 10 according to the present embodiment acquires an observation image captured including observation points to be observed in the river basin from the camera 30 provided so as to be able to image the river basin. . And the water level monitoring apparatus 10 which concerns on a present Example does not change a shape between the frames of an observation image among the several lines with which the mutual direction obtained by edge detection from an observation image is similar, and on an observation image The first line located at the top is detected. Further, the water level monitoring apparatus 10 according to the present embodiment detects a second line whose shape changes between frames of the observation image among a plurality of lines similar in mutual direction obtained by edge detection from the observation image. . Then, the water level monitoring apparatus 10 according to the present embodiment monitors the water level at the observation point based on the first line and the second line detected previously.

  As described above, the water level monitoring device 10 according to the present embodiment is observed by detecting the first line that is located at the top of the substantially parallel line group detected from the observation image of the river and whose shape does not change with time. The embankment line that appears in the image can be estimated. Furthermore, the water level monitoring apparatus 10 according to the present embodiment detects a water surface line reflected in the observation image by detecting a second line whose shape changes over time from a group of substantially parallel lines detected from the observation image of the river. Can be estimated. Thereby, in the water level monitoring apparatus 10 which concerns on a present Example, the water level of an observation point can be monitored using the bank line and water surface line detected from the observation image.

  Therefore, the water level monitoring apparatus 10 according to the present embodiment can monitor the water level at the observation point only by using the observation image without comparing the three-dimensional spatial information or the reference image measured in advance with the observation image. . Therefore, according to the water level monitoring apparatus 10 according to the present embodiment, the water level can be monitored without prior preparation. Furthermore, since the water level monitoring apparatus 10 according to the present embodiment does not use the pre-measured three-dimensional spatial information or the reference image, even if the river environment at the observation point changes, the three-dimensional spatial information or the reference image is displayed. There is no need to re-register.

[Configuration of water level monitoring device]
Then, the structure of the water level monitoring apparatus which concerns on a present Example is demonstrated. FIG. 2 is a functional block diagram illustrating the configuration of the water level monitoring apparatus according to the first embodiment. As shown in FIG. 2, the water level monitoring apparatus 10 includes a communication I / F (Interface) unit 11, a storage unit 12, and a control unit 13. In addition, the water level monitoring apparatus 10 has functions such as various functional units included in a known computer other than the functional units illustrated in FIG. 2, such as various input devices, audio output devices, and display devices.

  Among these, the communication I / F unit 11 is an interface that performs communication control with other devices such as the camera 30, the river information management device 50, and the user terminal 70. As an example, the communication I / F unit 11 transmits a control command of the camera 30 to the camera 30 and receives an observation image from the camera 30. As another example, the communication I / F unit 11 receives, from the river information management device 50, the average rainfall in a rainfall area included in the river basin. As a further example, the communication I / F unit 11 transmits an electronic map, an observation image, an e-mail, and the like regarding the observation point where the water is increasing. As an aspect of the communication I / F unit 11, a network interface card (NIC) such as a LAN card or a modem can be employed.

  The storage unit 12 is a storage device such as a semiconductor memory device such as a flash memory, a hard disk, or an optical disk. The storage unit 12 is not limited to the above-mentioned types of storage devices, and may be a RAM (Random Access Memory) or a ROM (Read Only Memory).

  The storage unit 12 stores various programs such as an OS (Operating System) executed by the control unit 13 and a water level monitoring program for monitoring the water level at the observation point from the observation image. Furthermore, the storage unit 12 stores data necessary for executing the program executed by the control unit 13. As an example, the storage unit 12 stores observation image metadata 12a, camera metadata 12b, electronic map data 12c, river line data 12d, and danger point data 12e.

  Among these, the observation image metadata 12a is metadata for managing observation images. As an aspect of the observation image metadata 12a, data in which a spot number, water system, river, kp (kilo post), camera ID (identification), capture date and time, file name, and the like are associated can be employed. Note that the observation image metadata 12 a is registered by the acquisition unit 14 described later when an observation image in which a river observation point is captured by the camera 30 is stored in the storage unit 12.

  The “point number” here refers to a number for identifying an observation point. In addition, “water system” refers to flowing water including a tributary that is centered on the flow of one river. In addition, “river” refers to running water such as tributaries and mainstreams. “Kp” refers to a distance sign indicating a distance from an estuary or a confluence. “Camera ID” refers to an identifier for identifying the camera 30. The “capture date and time” refers to the date and time when the observation image was captured from the camera 30 and is represented by “mm / dd / hh / mm / ss”, for example. The “file name” indicates the file name of the observation image, and is generated from the point number and the capture date and time by the acquisition unit 14 described later, for example.

  FIG. 3 is a diagram illustrating an example of the observed image metadata 12a. FIG. 3 shows metadata of an observation image in which the observation point with the point number “000001” is captured. In the example of FIG. 3, since the water system “xx” and the river “XX” are different, the observation point of the point number “000001” is a tributary. Furthermore, in the example of FIG. 3, since it is 0 kilometer post from the merging point from the main stream, it shows that it is a merging point of the tributary and the main stream. Further, in the example of FIG. 3, the observation image acquired from the camera 30 is captured every other second from “September 21, 10:00:00”. In this example, the file name of the observed image is stored as metadata, but the observed image itself may be stored in binary format.

  The camera metadata 12b is metadata for managing the camera 30. As one aspect of the camera metadata 12b, data in which a camera ID, latitude, longitude, multicast address, port number, water system, river, bank, usable control command, reference imaging position, and the like are associated can be employed. . The “control command” here refers to a command for controlling the camera 30, and examples thereof include camera attitude control such as panning and tilting, magnification control using a zoom function, and battery control. Further, the “reference imaging position” refers to the center position or the center of gravity position in the imaging range captured by the camera 30 in a neutral state where the posture control of the camera 30 is not performed. The camera metadata 12b is added or deleted when the camera 30 is newly installed or removed.

  FIG. 4 is a diagram illustrating an example of the camera metadata 12b. In the example of FIG. 4, it is shown that the camera with the camera ID “00001” is installed on the right bank of the river “OO”, that is, at a point of latitude “35.49.20” and longitude “139.54.29”. The camera with the camera ID “00001” is given a multicast address “239.196.73.150” and a network address with a port number “6037”. Further, the camera with the camera ID “00001” can use commands A to F among the control commands, and the latitude “35.50.00”, the longitude “139.55.00”, the high level when the camera posture is not controlled. Now take a picture of the spot at “15.05 (m)”.

  In the example of FIG. 4, it is shown that the camera with the camera ID “00002” is installed on the right bank of the river “OO”, that is, at a point of latitude “35.49.17” and longitude “139.54.30”. The camera with the camera ID “00002” is given a multicast address “239.196.73.151” and a network address with a port number “6038”. Further, the camera with the camera ID “00002” can use commands A to G among the control commands, and the latitude “35.20.00”, the longitude “139.55.00”, the high level when the camera posture is not controlled. Take a picture of the spot at “5.30 (m)”.

  In the example of FIG. 4, it is shown that the camera with the camera ID “XXXXX” is installed on the left bank of the river “XX”, that is, at a point of latitude “35.31.39” and longitude “139.14.08”. The camera with the camera ID “XXXXX” is assigned a multicast address “239.196.152.41” and a network address with a port number “8001”. Furthermore, the camera with the camera ID “XXXXX” can use commands A to Z among the control commands, and the latitude “35.35.00”, the longitude “139.16.00”, the high level when the camera posture is not controlled. Take a picture of the spot at “11.20 (m)”.

  The electronic map data 12c is map data including a river basin to be monitored by the water level monitoring device 10. As one aspect of the electronic map data 12c, data in which river center lines and observation point position information are associated with river map images expressed in a raster format or a vector format can be employed. The electronic map data 12c can also download new electronic map data from an external device (not shown) when the river basin map is updated.

  The river line data 12d is data relating to various river lines detected from the observation image. As an example, in the river line data 12d, in order to monitor the water level of the river, information related to the embankment line and the water surface line detected from the observation image is registered by the detection unit 16 described later. The “embankment line” mentioned here is a line that is detected by being estimated as an embankment from an observation image by a first detection unit 16b described later. This dike line is a line presumed as a dike and may actually be a riverbank instead of a dike. Further, the “water surface line” is a line detected by estimating the water surface from the observation image by a second detection unit 16c described later. As another example, the river line data 12d is referred to by the monitoring unit 17 described later in order to determine whether or not the observation point is flooded.

  As one aspect of the river line data 12d, data in which a spot number, a water system, a river, kp, a camera ID, a detection date and time, a water surface height, and a bank height are associated can be employed. The “detection date and time” referred to here indicates the date and time when the levee line and the water surface line were detected by the detection unit 16 described later. The “water level” refers to the distance from the river center line to the water surface line in the observed image. “Elevation height” refers to the distance from the river center line to the embankment line in the observed image.

  FIG. 5 is a diagram illustrating an example of the river line data 12d. In the example of FIG. 5, the observation point to be imaged by the camera with the camera ID “00001” is the location number “000001” of the 0 km post from the junction of the river, the location number “000002” of the 1 km post, and the location number “000003” of the 2 km post. It shows that it changed in the order of 10 minutes. Among these, at the observation point with the point number “000001” shown in FIG. 5, the water level is “5.0 (m)”, “5.1 (m)” and “5.2 (m)” compared to the dike height “15.0 (m)”. It represents a change of 0.1 m per minute. In addition, at the observation point with the point number “000002” shown in FIG. 5, the water level is “5.1 (m)”, “5.2 (m)”, “5.3 (m)”, and “1” (15.2 (m)). It shows that it is changing by 0.1m per minute. Furthermore, at the observation point with the spot number “000003” shown in FIG. 5, the water surface height is “5.3 (m)”, “5.4 (m)”, “5.5 (m)”, and 1 for the bank height “15.3 (m)”. It shows that it is changing by 0.1m per minute.

  In the example of FIG. 5, the observation points to be imaged by the camera with the camera ID “00002” are 10 in the order of the spot number “000004” of the 3 km post and the spot number “000005” of the 4 km post from the confluence of the river “XX”. Indicates that it has been changed in minutes. Among these, at the observation point with the point number “000004” shown in FIG. 5, the water level is “5.2 (m)”, “5.3 (m)” and “5.4 (m)” compared to the dike height “10.3 (m)”. It represents a change of 0.1 m per minute.

  The dangerous point data 12e is data relating to a dangerous point where there is a risk of water increase among the observation points of the river. As an example, in order to extract the observation point to be notified, the dangerous point data 12e is an observation point, which will be described later, as an observation point determined that the water level detected from the observation image satisfies a predetermined water increase standard. 17 is registered. As another example, the dangerous point data 12e is referred to by a notification control unit 18 described later in order to notify the dangerous point.

  As one aspect of the dangerous point data 12e, data in which a point number, water system, river, kp, camera ID, multicast address, port number, determination date, latitude, and longitude are associated can be employed. The “determination date and time” here refers to the date and time when the observation point is determined as a dangerous point by the monitoring unit 17 described later. Note that the metadata of the camera 30 that captured the dangerous spot is registered in the above-described columns of camera ID, multicast address, port number, determination date, latitude, and longitude.

  FIG. 6 is a diagram illustrating an example of the dangerous point data 12e. In the example of FIG. 6, the observation points with the location numbers “000001” to “000005” from 0 km post, 1 km post, 2 km post, 3 km post and 4 km post from the point where the river “○○” joins the main stream are listed as dangerous points. Indicates that it is up. In addition, the observation points having the point numbers “000001” to “000003” among the observation points having the point numbers “000001” to “000005” are observation points captured by the camera having the camera ID “00001”. Furthermore, the observation points with the point numbers “000004” to “000005” are observation points captured by the camera with the camera ID “00002”.

  The control unit 13 has an internal memory for storing programs defining various processing procedures and control data, and executes various processes using these. As shown in FIG. 2, the control unit 13 includes an acquisition unit 14, a camera control unit 15, a detection unit 16, a monitoring unit 17, and a notification control unit 18.

  The acquisition unit 14 is a processing unit that acquires an observation image captured from the camera 30 including a river observation point. Explaining this, the acquisition unit 14 acquires an observation image captured by the camera control unit 15 (described later) with horizontal rotation and / or adjustment of the elevation angle and the magnification of the camera 30. At this time, when the observation image transmitted by the camera 30 is compressed by a moving image compression method such as MPEG, the acquiring unit 14 moves the moving image received from the camera 30 at a predetermined interval, for example, every frame / second. Is transcoded to a still image compression method such as JPEG. Then, the acquisition unit 14 stores the transcoded still image observation image in the storage unit 12 and generates a file name from the acquisition date and time when the observation image was acquired from the camera 30 and the camera ID. In addition, the acquisition unit 14 registers the information included in the header of the observation image, for example, the point number, the water system, the river, the kp, and the camera ID together with the capture date and file name as the observation image metadata 12a in the storage unit 12. . Note that “MPEG” is an abbreviation of “Moving Picture Experts Group”, and “JPEG” is an abbreviation of “Joint Photographic Experts Group”.

  Here, the case where the moving image of the observation image transmitted by the camera 30 is converted into a still image is illustrated, but the observation image may be taken into the storage unit 12 as a moving image. In addition, here, the case where the moving image received from the camera 30 is converted into a still image every frame / second and taken into the storage unit 12 is illustrated, but the rate taken into the storage unit 12 can be arbitrarily changed.

  The camera control unit 15 is a processing unit that controls the camera 30. Explaining this, the camera control unit 15 refers to the electronic map data 12c stored in the storage unit 12 when the river information management device 50 is notified of the average rainfall in the rainfall region included in the river basin. Extract observation points in the rainy area. At this time, the camera control unit 15 can also extract the observation points included in the rain region where the average rainfall is equal to or greater than a predetermined value among the observation points existing in the rain region. Subsequently, the camera control unit 15 uses the latitude and longitude of the camera 30 defined in the camera metadata 12b and the latitude and longitude of the observation point extracted earlier to determine the distance from the camera 30 to the observation point. Calculate for each. And the camera control part 15 selects the camera 30 with the shortest distance to an observation point among the cameras 30 as a camera used for imaging of an observation point.

  Subsequently, the camera control unit 15 extracts the latitude and longitude of the imaging point to be imaged by the camera 30 selected as the camera used for imaging the observation point. As an example, the camera control unit 15 refers to the electronic map data 12c, and the extension line obtained by extending from the installation position of the camera 30 to the position of the observation point is that the camera 30 is installed on both banks of the river. The latitude and longitude of the point that intersects the opposite bank and the opposite bank are extracted as imaging points.

  Thereafter, the camera control unit 15 determines the imaging parameter based on the reference imaging position of the camera 30 defined in the camera metadata 12b, the position of the observation point, and the relative value of the imaging point extracted earlier. The “imaging parameter” here refers to a parameter relating to imaging such as the horizontal rotation and elevation angle correction amount of the camera 30 and the magnification of the camera 30. As an example, the camera control unit 15 calculates the correction amount of the horizontal rotation and the elevation angle of the camera 30 from the amount of deviation between the reference imaging position of the camera 30 and the position of the imaging point. As another example, the camera control unit 15 determines the magnification of the camera 30 so that the observation point is included in the imaging range of the camera 30.

  Then, the camera control unit 15 generates a control command for the camera 30 in accordance with the previously determined imaging parameter and transmits the control command to the camera 30 to control pan, tilt, and / or zoom of the camera 30. To do. In this way, by controlling the pan and tilt of the camera 30, the horizontal rotation and elevation angle of the camera 30 are corrected. Furthermore, the magnification of the camera is adjusted by controlling the zoom of the camera 30. Thereafter, the acquisition unit 14 acquires an observation image in which the observation point is captured by the camera 30.

  FIG. 7 is a diagram illustrating an example of a schematic electronic map. In the example of FIG. 7, the center line of the river is illustrated by a one-dot chain line. In the example of FIG. 7, the observation points 20A to 20C are illustrated by black circles, and the imaging points 22 are illustrated by white circles. Furthermore, in the example of FIG. 7, the imaging range 23 of the camera 30B is illustrated by a dotted line. In the example of FIG. 7, it is assumed that the observation point 20B is extracted as an observation target among the observation points 20A to 20C.

  In the example shown in FIG. 7, the camera control unit 15 calculates the distance between the camera 30A and the observation point 20B, the distance between the camera 30B and the observation point 20B, and the distance between the camera 30C and the observation point 20B. Of the calculated distances between the cameras 30 and the observation points 20B, the distances between the cameras 30B and the observation points 20B are the shortest. Therefore, the camera control unit 15 selects the camera 30B as a camera to be used for imaging the observation point 20B. Subsequently, the latitude and longitude of the point where the extension line obtained by extending from the installation position of the camera 30B to the position of the observation point 20B intersects with the opposite bank of the river bank where the camera 30B is installed on both banks of the river is the camera. The image is extracted as the imaging point 22 by the control unit 15. Thereafter, the correction amount of the horizontal rotation and elevation angle of the camera 30 is calculated by the camera control unit 15 from the amount of deviation between the reference imaging position of the camera 30B and the position of the imaging point 22. Further, the magnification of the camera 30 is determined by the camera control unit 15 so that the observation point 22 is included in the imaging range 23 of the camera 30B. In this way, the imaging parameters such as the horizontal rotation and elevation angle correction amount of the camera 30 and the magnification of the camera 30 are determined.

  The detection unit 16 is a processing unit that detects an embankment line and a water surface line from the observation image stored in the storage unit 12. As shown in FIG. 2, the detection unit 16 further includes an edge processing unit 16a, a first detection unit 16b, and a second detection unit 16c.

  Among these, the edge processing unit 16a is a processing unit that generates a line image including a line connecting edges in similar directions within a predetermined range among edges detected from the observed image stored in the storage unit 12. is there. As an example, the edge processing unit 16a refers to the observation image metadata 12a, reads the observation images captured in the storage unit 12 by the acquisition unit 14 in the order of acquisition date, and detects the edge of each of the read observation images. In the above, the following processing is executed.

  Here, in the following, an observation image captured every second is read out in units of one minute, and a first detection unit 16b and a second detection unit 16c, which will be described later, detect a dike line and a water surface line to detect the line. Is assumed to be repeatedly detected for 10 minutes, that is, 10 times. At this time, the administrator of the water level monitoring device 10 can arbitrarily set the frame of the observation image for reading the observation image in one minute and the number of frames. Further, the administrator of the water level monitoring apparatus 10 can arbitrarily set the time unit for reading the observation image and the number of repetitions of line detection.

  Explaining this, the edge processing unit 16a uses a GIS (Geographical Information System) function to superimpose the river centerline stored as the electronic map data 12c in the storage unit 12 on the observation image in which the edge is detected. In this way, the river center line is superimposed on the observation image so that the candidate line of the dike line and the water surface line is based on the river center line estimated to have a direction similar to the dike line and the water surface line. It is for extracting.

  After that, the edge processing unit 16a extracts an edge having a direction similar to the river center line and having a length equal to or greater than a predetermined number of pixels from the edges detected from the observed image. When the edges are extracted in this way, if edges that are too far from the river center line are included, many edges unrelated to the embankment line and the water surface line are extracted. For this reason, it is preferable to extract only edges that are substantially parallel to the river center line, for example, edges whose deviation from the river center line is within ± 5 °. Furthermore, even if the edge has a direction similar to the center line of the river, it may include an edge of a building or a vehicle that is unrelated to the embankment line and the water surface line. Therefore, it is preferable to extract only a length estimated to be an edge of a building or a vehicle, for example, an edge having 10 pixels or more.

  Further, the edge processing unit 16a extracts an edge that exists within a predetermined number of pixels on the extension line of the previously extracted edge and has a direction similar to the extracted edge. In addition, the edge processing unit 16a connects the previously extracted edge and the extracted edge based on the edge to form a line that can be a candidate for a dike line or a water surface line. Thereafter, the edge processing unit 16a outputs an observation image in which a line is formed by connecting the edges to each other to a first detection unit 16b and a second detection unit 16c described later. Hereinafter, an image in which a line is formed by connecting detected edges to an observed image may be referred to as a “line image”.

  In this way, the edge processing unit 16a performs edge detection of each observation image read out in the order of acquisition date and then connects edges that satisfy a certain condition to thereby connect the line image for each frame. Generate.

  The first detection unit 16b is similar to the direction of the center line of the river set with reference to the electronic map data 12c corresponding to the observation point, the shape does not change between frames of the line image, and on the line image. It is a processing unit that detects a line located at the top as a bank line.

  As an example, the first detection unit 16b extracts a line similar to the direction of the center line of the river from the line image generated by the edge processing unit 16a. For example, the first detection unit 16b extracts a line whose deviation from the river center line is within ± 5 ° among the lines included in the line image. Then, the first detection unit 16b determines whether or not the shape of the line located at the top of the line image among the previously extracted lines changes between the frames of the line image. Then, the first detection unit 16b detects the line positioned at the top of the line image as the bank line when the shape of the line positioned at the top of the line image does not change between the frames of the line image.

  In this way, a line similar to the direction of the river center line is targeted for detection because the embankment is constructed along the river and the outline line formed by the bank is similar to the river center line. It is because it can be estimated that it has. Furthermore, the reason why the line whose shape changes between frames of the line image is set as the detection target is to detect a dike line by excluding moving bodies such as the water surface. Furthermore, the reason why the topmost line on the line image is the detection target is that it can be estimated that the embankment line is located on the top of the line image unless the water surface line exceeds the embankment line due to flooding or the like.

  The second detection unit 16c detects, as a water surface line, a line that is similar to the direction of the river center line set with reference to the electronic map data 12c corresponding to the observation point and whose shape changes between frames of the line image. Is a processing unit.

  As an example, the second detection unit 16c extracts a line similar to the direction of the center line of the river from the line image generated by the edge processing unit 16a. For example, the second detection unit 16c extracts a line that is within ± 5 ° from the center line of the river and is not a straight line from the lines included in the line image. Then, the first detection unit 16b determines whether there is a line whose shape changes between frames of the line image among the previously extracted lines. At this time, when there is a line whose shape changes between frames of the line image, the first detection unit 16b detects the line as a water surface line.

  Here, the reason why the detection target is a non-straight line among the lines similar to the direction of the center line of the river is that when the water increases due to heavy rain, the fluctuation of the water surface becomes large and the water surface does not become a straight line. . At this time, if a line that is not a straight line is a detection target, it is assumed that the water surface line is not detected during drought with mild fluctuations in the water surface. However, since there is little danger even if people approach the river basin during drought, there is no problem even if the water surface line at the time of drought cannot be detected if the water surface line at the time of water increase can be detected. Further, the reason why the line whose shape changes between frames of the line image is set as the detection target is to detect the water surface line by excluding a dike or a building that is not a moving body unlike the water surface.

In this way, when the levee line and the water surface line are detected by the first detection unit 16b and the second detection unit 16c, the detection unit 16 determines the height of the river center line from the height of the levee line on the line image. by subtracting is to calculate the levee height W _a. Furthermore, the detection unit 16 calculates the water surface height W _b by subtracting the height of the river center line from the height of the water surface line on the line image. After that, the detection unit 16 includes the levee height W _a and the water surface height calculated in advance for the spot number, water system, river, kp, and camera ID of the first frame among the observation images used for the detection of the levee line and the water surface line. River line data in which W _b is associated with the detection date is registered in the storage unit 12. Thus, the river line data of the bank height W _a and the water surface height W _b detected in a predetermined time unit, every minute in this example, is registered in the storage unit 12 ten times.

In calculating the dike height W _a and the water surface height W _b , an arbitrary coordinate system can be used. As an example, the detection unit 16 sets the height of the lowest horizontal line of the line image to zero, and the height from the zero point of the highest pixel among the pixels forming the embankment line, the water surface line, and the river center line. Can be calculated as the height of the embankment line, the height of the water surface line, and the height of the river center line. At this time, in order to monitor the water level more safely, the height from the zero point of the lowest pixel among the pixels forming each line is set as the height of the dike line and the center line of the river. It may be the height of the river center line.

FIG. 8 is a diagram illustrating an example of a line image. In the example of FIG. 8, a case where six lines 41A to 41F are detected by the edge processing unit 16a in the line image 40 is illustrated. Among these, the line 41F illustrated by the alternate long and short dash line is a river centerline that is overlaid using the electronic map data 12c. A line 41E illustrated by a two-dot chain line is a line in which a change in shape is detected with another line image. In this case, of the lines 41A to 41E having a direction similar to the river center line 41F, the line 41A is located at the top of the line image 40 and the shape change between the frames of the line image is not detected. Is detected by the first detector 16b as a dike line. In addition, the line 41E which is not a straight line and whose shape changes between frames of the line image among the lines 41A to 41E is detected as a water surface line by the second detection unit 16c. Thereafter, the levee height W _a and the water surface height W _b are calculated by the detection unit 16 from the height of the levee line, the height of the water surface line, and the height of the river center line.

  The monitoring unit 17 is a processing unit that monitors the water level at the observation point based on the levee line and the water surface line detected by the detection unit 16. As an example, the monitoring unit 17 determines whether or not the relative distance between the dike line and the water surface line is less than a predetermined threshold value, or whether the time change of the relative distance between the dike line and the water surface line is equal to or greater than a predetermined threshold value. The water level at the observation point is monitored by judging whether or not.

Explaining this, the monitoring unit 17 uses the levee height W _a and the water surface height W _b stored as the river line data 12d in the storage unit 12, and the water level D, which is the distance from the water surface to the levee, An increment ΔL is calculated. As an example, the monitoring unit 17 calculates the level of each time unit by subtracting the water surface height W _b from the embankment height W _a which is registered as a river line data 12d. Of the water levels for each time unit calculated in this manner, the latest water level is used as the water level D to be determined. Furthermore, the monitoring unit 17 calculates the water level increment ΔL by subtracting the oldest water level from the latest water level.

Thereafter, the monitoring unit 17, the water level D is equal to or less than a predetermined threshold D _TH. In this case, the monitoring unit 17 further determines if the water level D is equal to or smaller than the threshold D _TH, increment [Delta] L of the water level to or greater than the predetermined threshold value [Delta] L _TH. Here, when the water level D is equal to or less than the threshold value D _TH and the water level increment ΔL is equal to or greater than the threshold value ΔL _TH , the water surface line is set to the embankment line at the observation point, and the water surface rapidly rises. Therefore, it can be estimated that there is a risk of water increase at the observation point. In this case, the monitoring unit 17 re-executes water level monitoring for the observation point adjacent to the observation point that is the target of water level monitoring this time. Thereafter, when the monitoring unit 17 determines that there is a risk of water increase even at adjacent observation points, the monitoring unit 17 registers information regarding each observation point determined to have a risk of water increase in the storage unit 12 as dangerous point data 12e. As an example of the dangerous point data 12e, in addition to the point number of the dangerous point, water system, river, kp, camera ID of the camera 30 that captured the dangerous point, multicast address, port number, latitude, longitude, it is determined as a dangerous point. The determined determination time is registered.

Here, the case where it is determined that there is a risk of water increase when the water level D is equal to or less than the threshold value D _TH and the water level increment ΔL is equal to or greater than the threshold value ΔL _TH is described. It can also be determined that there is a risk of water increase.

  The notification control unit 18 is a processing unit that controls to notify the user terminal 70 of the dangerous point. Explaining this, when the monitoring point 17 registers the dangerous point data 12d in the storage unit 12, the notification control unit 18 reads the dangerous point data 12d from the storage unit 12 and sends the following notification to the user terminal 70. Execute. As an example, the notification control unit 18 uses the electronic map data 12c stored in the storage unit 12 to publish or distribute content data in which dangerous points are mapped on the electronic map on the web. The content data is provided to the user terminal 70. As another example, the notification control unit 18 publishes or distributes a video obtained by connecting observation images related to a plurality of dangerous spots on the web, and distributes a video that can be browsed by a scroll operation using a browser or the like. This is provided to the user terminal 70. As a further example, the notification control unit 18 transmits an email describing the latitude, longitude, and address of each danger point to the user terminal 70.

[Process flow]
Next, the process flow of the water level monitoring apparatus according to the present embodiment will be described. Here, (1) overall processing, (2) camera control processing, (3) line detection processing, and (4) notification control processing executed by the water level monitoring device 10 will be described in this order.

(1) Overall Process FIG. 9 is a flowchart showing the overall process procedure of the water level monitoring apparatus according to the first embodiment. This overall process is started when the river information management device 50 notifies the average rainfall in the rainfall area included in the river basin.

  As illustrated in FIG. 9, the camera control unit 15 refers to the electronic map data 12c stored in the storage unit 12 and extracts observation points that exist in the rainfall region (step S101). Then, the camera control unit 15 performs “camera control processing” in which the camera 30 is pointed at the imaging point to image the observation point, and the acquisition unit 14 acquires the observation image (step S102).

Subsequently, the detection unit 16 performs a “line detection process” for detecting a dike line and a water surface line using the observation image acquired by the acquisition unit 14 (step S103). Thereafter, the monitoring unit 17 uses the dike height W _a and the water surface height W _b calculated from the dike line and the water surface line detected by the detection unit 16, and the water level D that is the distance from the water surface to the dike and the water level Is calculated (step S104).

Thereafter, the monitoring unit 17, the water level D is equal to or less than a predetermined threshold _{D TH} (step S105). At this time, when the water level D is equal to or smaller than the threshold _{D TH} (step S105: Yes), the monitoring unit 17 further determines increment [Delta] L of the water level to or greater than a predetermined threshold value [Delta] L _TH (step S106) .

If the water level increment ΔL is greater than or equal to the threshold value ΔL _TH (Yes at Step S106), the monitoring unit 17 determines whether or not there is a possibility of water increase at a plurality of observation points (Step S107). At this time, when it is determined for the first time that there is a risk of water increase at the observation point (No at Step S107), the monitoring unit 17 extracts an observation point adjacent to the observation point that is the target of water level monitoring at this time (Step S107). S108). Then, the process regarding the monitoring of the water level from step S101 to S106 is performed anew.

  Here, when it is determined that there is a risk of water increase at a plurality of observation points (Yes in step S107), the monitoring unit 17 stores information regarding each observation point determined to be a risk of water increase as dangerous point data 12e. 12 is registered (step S109). Thereafter, the notification control unit 18 executes a “notification control process” for notifying the user terminal 70 of the dangerous point (step S110), and ends the process.

If the water level D is greater than the threshold value D _TH or the water level increment ΔL is less than the threshold value ΔL _TH (No in step S105 or S106 negative), there is no risk of water increase, and the process is terminated.

(2) Camera Control Processing FIG. 10 is a flowchart illustrating the procedure of camera control processing according to the first embodiment. This camera control process is a process for step S102 shown in FIG. 9, and starts when an observation point to be monitored for water level is extracted.

  As shown in FIG. 10, the camera control unit 15 uses the latitude and longitude of the camera 30 defined in the camera metadata 12b and the latitude and longitude of the observation point extracted in advance, from the camera 30 to the observation point. The distance is calculated for each camera 30 (step S301). And the camera control part 15 selects the camera 30 with the shortest distance to an observation point among the cameras 30 as a camera used for imaging of an observation point (step S302).

  Subsequently, the camera control unit 15 extracts the latitude and longitude of the imaging point to be imaged by the camera 30 selected as the camera used for imaging the observation point (step S303). Thereafter, the camera control unit 15 determines the imaging parameter based on the reference imaging position of the camera 30 defined in the camera metadata 12b, the position of the observation point, and the relative value of the previously extracted imaging point (step S304). .

  Then, the camera control unit 15 generates a control command for the camera 30 in accordance with the previously determined imaging parameter and transmits the control command to the camera 30 to control pan, tilt, and / or zoom of the camera 30. (Step S305).

  Thereafter, the acquisition unit 14 acquires an observation image captured by the camera control unit 15 that has been subjected to horizontal rotation and / or adjustment of the elevation angle of the camera 30 and the magnification of the camera 30 (step S306), and acquires the acquired observation image. The data is stored in the storage unit 12 and the process ends.

(3) Line Detection Processing FIG. 11 is a flowchart illustrating a procedure of line detection processing according to the first embodiment. This line detection process is a process corresponding to step S <b> 103 shown in FIG. 9, and is activated when an observation image is taken into the storage unit 12 by the acquisition unit 14.

  As shown in FIG. 11, the edge processing unit 16a refers to the observation image metadata 12a, reads the observation image captured by the acquisition unit 14 into the storage unit 12, and detects the edge of the read observation image (step). S501).

  Then, using the GIS function, the edge processing unit 16a superimposes the river center line stored as the electronic map data 12c in the storage unit 12 on the observation image in which the edge is detected (step S502). Subsequently, the edge processing unit 16a connects the edges in the observation image to generate a line image (step S503).

  Thereafter, the first detection unit 16b is similar to the direction of the river center line set with reference to the electronic map data 12c corresponding to the observation point, the shape does not change between frames of the line image, and the line image The embankment line positioned at the top is detected (step S504).

  Further, the second detection unit 16c detects a water surface line that is similar to the direction of the center line of the river set with reference to the electronic map data 12c corresponding to the observation point, and whose shape changes between frames of the line image. (Step S505).

Thereafter, the detection unit 16 calculates the embankment height W _a by subtracting the height of the center line of the river from a height of embankment line on line image, high center line of the river from a height of water surface line the calculating the water surface height _{W b} by subtracting of (step S506).

The detection unit 16 includes the spot number, water system, river, kp, and camera ID of the first frame among the observation images used for the detection of the levee line and the water surface line, the levee height W _a , the water surface height W _b, and the detection date and time. The associated river line data is registered in the storage unit 12 (step S507).

  Note that the processing of steps S504 and S505 in FIG. 11 is not necessarily performed in the order shown. That is, the order of each process can be changed and executed, or each process can be executed in parallel.

(4) Notification Control Processing FIG. 12 is a flowchart illustrating a procedure of notification control processing according to the first embodiment. This notification control process is started when it is determined that there is a risk of water increase at a plurality of observation points and the dangerous point data 12e is registered in the storage unit 12.

  As illustrated in FIG. 12, the notification control unit 18 reads the dangerous spot data 12d from the storage unit 12 (step S701). Then, the notification control unit 18 uses the electronic map data 12c to publish or distribute the content data in which the dangerous point is mapped on the electronic map to the user terminal 70, thereby distributing the content data to the user terminal 70. Provide (step S702).

  Subsequently, the notification control unit 18 publishes or distributes a video obtained by connecting observation images related to a plurality of dangerous spots on the web, and distributes a video that can be browsed by a scroll operation using a browser or the like. 70 (step S703).

  Thereafter, the notification control unit 18 transmits an e-mail in which the latitude, longitude, and address of each danger point are described to the user terminal 70 (step S704), and ends the process.

  Note that the processes in steps S702 and S704 in FIG. 12 are not necessarily executed in the order shown. That is, the order of each process can be changed and executed, or each process can be executed in parallel.

[Effect of Example 1]
As described above, the water level monitoring apparatus 10 according to the present embodiment detects the first line that is positioned at the top of the substantially parallel line group detected from the observation image of the river and whose shape does not change with time. To estimate the embankment line in the observed image. Furthermore, the water level monitoring apparatus 10 according to the present embodiment detects a water surface line reflected in the observation image by detecting a second line whose shape changes over time from a group of substantially parallel lines detected from the observation image of the river. presume.

  Therefore, the water level monitoring apparatus 10 according to the present embodiment can monitor the water level at the observation point only by using the observation image without comparing the three-dimensional spatial information or the reference image measured in advance with the observation image. . Therefore, according to the water level monitoring apparatus 10 according to the present embodiment, the water level can be monitored without prior preparation. Furthermore, since the water level monitoring apparatus 10 according to the present embodiment does not use the pre-measured three-dimensional spatial information or the reference image, even if the river environment at the observation point changes, the three-dimensional spatial information or the reference image is displayed. There is no need to re-register.

  Here, the water level monitoring apparatus 10 according to the present embodiment can monitor the water level using an observation image in which the observation point is imaged from an arbitrary direction by the camera 30. FIGS. 13-15 is a figure for demonstrating the effect of the water level monitoring apparatus 10 which concerns on Example 1. FIGS. The example of FIG. 13 shows an example of line detection when the direction in which the river flows and the direction in which the camera 30 captures images are substantially perpendicular. 14 and 15 show an example of line detection when the direction in which the camera 30 captures images is close to the direction in which the river flows.

  In the example shown in FIG. 13, an edge image 61A is generated by the edge processing unit 16a by detecting an edge from the observation image 60A. At this time, the river center line 62A is obtained by superimposing the river center line in the electronic map on the edge image 61A. Then, the edge processing unit 16a generates the line image 64A by connecting the edges with the river center line 62A as a reference. Here, of the four lines 63Aa to 63Ad included in the line image 64A, the line 63Ad is a line in which a change in shape is detected with respect to another line image. In this case, of the lines 63Aa to 63Ad having a direction similar to the river center line 62A, the line 63Aa that is located at the top of the line image 64A and in which no change in shape is detected between the frames of the line image is the levee line. As detected by the first detector 16b. Of the lines 63Aa to 63Ad, the line 63Ad whose shape changes between frames of the line image is detected as a water surface line by the second detection unit 16c.

  In the example shown in FIG. 14, an edge image 61B is generated by the edge processing unit 16a by detecting an edge from the observed image 60B. At this time, the river center line 62B is obtained by superimposing the river center line on the electronic map on the edge image 61B. Then, the edge processing unit 16a generates the line image 64B by connecting the edges with respect to the river center line 62B. Here, it is assumed that the line 63Bc among the three lines 63Ba to 63Bc included in the line image 64B is a line in which a change in shape is detected with respect to another line image. In this case, of the lines 63Ba to 63Bc having a direction similar to the river center line 62B, the line 63Ba located at the top of the line image 64B and in which no change in shape is detected between the frames of the line image is the dike line. As detected by the first detector 16b. Of the lines 63Ba to 63Bc, the line 63Bc whose shape changes between frames of the line image is detected as a water surface line by the second detection unit 16c.

  In the example shown in FIG. 15, an edge image 61C is generated by the edge processing unit 16a by detecting an edge from the observed image 60C. At this time, the river center line 62C is obtained by superimposing the river center line on the electronic map on the edge image 61C. The edge processing unit 16a generates the line image 64C by connecting the edges with respect to the river center line 62C. Here, among the six lines 63Ca to 63Cf included in the line image 64C, the line 63Cf is a line in which a change in shape has been detected with respect to another line image. In this case, of the lines 63Ca to 63Cf having a direction similar to the river center line 62C, the line 63Ca that is positioned at the top of the line image 64C and in which no change in shape is detected between the frames of the line image is the levee line. As detected by the first detector 16b. Of the lines 63Ca to 63Cf, the line 63Cf whose shape changes between frames of the line image is detected as a water surface line by the second detection unit 16c.

  As described above, in the water level monitoring apparatus 10 according to the present embodiment, even when the direction in which the river flows and the direction in which the camera 30 captures images are substantially perpendicular, the direction in which the camera 30 captures images is close to the direction in which the river flows. However, the dike line and the water surface line can be detected appropriately.

  In addition, the water level monitoring apparatus 10 according to the present embodiment generates a line image including a line connecting edges in similar directions within a predetermined range among edges detected from the observed image. In addition, the water level monitoring apparatus 10 according to the present embodiment detects a dike line and a water surface line from the line image. For this reason, in the water level monitoring apparatus 10 according to the present embodiment, it is possible to detect the levee line and the water surface line even if the edge forming the levee or the water surface is partially missing among the edges detected from the observation image. become.

  Furthermore, the water level monitoring apparatus 10 according to the present embodiment is similar to the direction of the river center line set based on the map information corresponding to the observation point, the shape does not change between the frames of the line image, and the line The top line on the image is detected as a bank line. Moreover, the water level monitoring apparatus 10 according to the present embodiment detects a line that is similar to the direction of the center line of the river and whose shape changes between frames of the line image as a water surface line. For this reason, in the water level monitoring apparatus 10 according to the present embodiment, the embankment line and the water surface line can be formed based on the center line of the river that is likely to have a direction similar to the embankment line and the water surface line. Therefore, the water level monitoring apparatus 10 according to the present embodiment can appropriately detect the levee line and the water surface line among a plurality of lines whose directions are similar in the line image.

  Further, the water level monitoring apparatus 10 according to the present embodiment is configured to determine whether or not the relative distance between the dike line and the water surface line is less than a predetermined threshold value, or whether the relative distance between the dike line and the water surface line changes with time. The water level at the observation point is monitored by determining whether or not the above is true. For this reason, in the water level monitoring apparatus 10 which concerns on a present Example, it can be monitored whether the water surface line has become the embankment line at the observation point, or whether the water surface is rising rapidly. Therefore, the water level monitoring apparatus 10 according to the present embodiment can accurately monitor the increase in water at the observation point.

  Furthermore, the water level monitoring apparatus 10 according to the present embodiment, when it is determined that the relative distance between the embankment line and the water surface line is less than the threshold value, or the time change of the relative distance is greater than or equal to the threshold value, Control to notify. For this reason, in the water level monitoring apparatus 10 according to the present embodiment, it is possible to make the observation point where there is a risk of water increase known as a dangerous point.

  In addition, the water level monitoring device 10 according to the present embodiment is determined for a plurality of observation points when the relative distance between the dike line and the water surface line is less than the threshold value, or when the change in the relative distance over time is equal to or greater than the threshold value. In some cases, notification is executed. For this reason, in the water level monitoring apparatus 10 according to the present embodiment, if the water increase occurs at one observation point, the observation point is limited to the case where the increase in water also occurs at the adjacent observation point when the trend is in accordance with the river increase tendency. Can be informed of danger points. Therefore, in the water level monitoring apparatus 10 according to the present embodiment, it is possible to prevent notification when one observation point determines that there is a risk of water increase due to an erroneous determination, and it is possible to prevent issuing a false alarm.

  Although the embodiments related to the disclosed apparatus have been described above, the present invention may be implemented in various different forms other than the above-described embodiments. Therefore, another embodiment included in the present invention will be described below.

[Application example]
In the above-described first embodiment, the case where the embankment line and the water surface line are detected based on the center line of the river is illustrated, but the disclosed apparatus is not limited thereto. For example, the disclosed apparatus first detects a water surface line whose shape changes between frames of a line image, and then selects a line located at the top of the lines existing above the water surface line on the line image as a dike line. Can also be detected.

[Scope of application]
In the first embodiment, the case where observation points are provided at regular intervals has been described. However, the setting of observation points in a river is not limited to the mode of the first embodiment. For example, the disclosed device may use a point where an image with high detection accuracy can be obtained by the camera 30 as an observation point. In addition, the disclosed apparatus can reduce the installation interval of observation points such as a river confluence and a point where the flow rate changes rapidly without setting observation points at regular intervals.

  In the first embodiment, the case where the water level of the river is monitored is illustrated, but the disclosed apparatus is not limited to this. That is, the disclosed device can be similarly applied to water areas such as seas and lakes other than rivers.

  In addition, each component of each illustrated apparatus does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, the acquisition unit 14, the camera control unit 15, the detection unit 16, the monitoring unit 17, or the notification control unit 18 may be connected as an external device of the water level monitoring device via a network. Moreover, another apparatus has the acquisition part 14, the camera control part 15, the detection part 16, the monitoring part 17, or the alerting | reporting control part 18, respectively, and the function of said water level monitoring apparatus is carried out by connecting with a network and cooperating. It may be realized.

[Water level monitoring program]
The various processes described in the above embodiments can be realized by executing a prepared program on a computer such as a personal computer or a workstation. Therefore, in the following, an example of a computer that executes a water level monitoring program having the same function as in the above embodiment will be described with reference to FIG.

  FIG. 16 is a schematic diagram illustrating an example of a computer that executes a water level monitoring program according to the first and second embodiments. As illustrated in FIG. 16, the computer 100 includes an operation unit 110a, a speaker 110b, a camera 110c, a display 120, and a communication unit 130. Further, the computer 100 includes a CPU 150, a ROM 160, an HDD 170, and a RAM 180. These units 110 to 180 are connected via a bus 140.

  As shown in FIG. 16, the HDD 170 has a water level monitoring program that exhibits the same functions as the acquisition unit 14, the camera control unit 15, the detection unit 16, the monitoring unit 17, and the notification control unit 18 described in the first embodiment. 170a is stored in advance. The water level monitoring program 170a may be integrated or separated as appropriate, similar to each component of the acquisition unit 14, camera control unit 15, detection unit 16, monitoring unit 17, and notification control unit 18 shown in FIG. good. In other words, all data stored in the HDD 170 need not always be stored in the HDD 170, and only data necessary for processing may be stored in the HDD 170.

  Then, the CPU 150 reads the water level monitoring program 170 a from the HDD 170 and develops it in the RAM 180. Accordingly, as shown in FIG. 16, the water level monitoring program 170a functions as a water level monitoring process 180a. The water level monitoring process 180a expands various data read from the HDD 170 in an area allocated to itself on the RAM 180 as appropriate, and executes various processes based on the expanded data. The water level monitoring process 180a is performed by the acquisition unit 14, the camera control unit 15, the detection unit 16, the monitoring unit 17, and the notification control unit 18 illustrated in FIG. 2, for example, the processes illustrated in FIGS. including. In addition, each processing unit virtually realized on the CPU 150 does not always require that all processing units operate on the CPU 150, and only a processing unit necessary for the processing needs to be virtually realized.

  Note that the water level monitoring program 170a is not necessarily stored in the HDD 170 or the ROM 160 from the beginning. For example, each program is stored in a “portable physical medium” such as a flexible disk inserted into the computer 100, so-called FD, CD-ROM, DVD disk, magneto-optical disk, or IC card. Then, the computer 100 may acquire and execute each program from these portable physical media. In addition, each program is stored in another computer or server device connected to the computer 100 via a public line, the Internet, a LAN, a WAN, etc., and the computer 100 acquires and executes each program from these. It may be.

DESCRIPTION OF SYMBOLS 1 Disaster prevention system 3 Network 10 Water level monitoring apparatus 11 Communication I / F part 12 Storage part 12a Observation image metadata 12b Camera metadata 12c Electronic map data 12d River line data 12e Hazardous point data 13 Control part 14 Acquisition part 15 Camera control part 16 Detection unit 16a Edge processing unit 16b First detection unit 16c Second detection unit 17 Monitoring unit 18 Notification control unit 30A, 30B, 30C Camera 50 River information management device 70 User terminal

Claims (8)

An acquisition unit that acquires an observation image captured including an observation point to be observed in the water area, from an imaging device provided to be able to image a predetermined water area;
Of the plurality of lines obtained by edge detection from the observation image acquired by the acquisition unit and having similar orientations, the shape does not change between frames of the observation image and is positioned on the topmost position on the observation image Detecting a first line and a second line whose shape changes between frames of the observation image;
And a monitoring unit that monitors the water level at the observation point based on the first line and the second line detected by the detection unit. The detector is
An edge processing unit that generates a line image including a line connecting edges in similar directions within a predetermined range of edges detected from the observation image acquired by the acquisition unit;
The water level monitoring apparatus according to claim 1, wherein the first line and the second line are detected from a line image generated by the edge processing unit. The detector is
The line that is similar to the direction of the reference line set based on the map information corresponding to the observation point, does not change in shape between frames of the line image, and is positioned at the highest position on the line image is the first line. A first detector for detecting as one line;
The apparatus according to claim 2, further comprising: a second detection unit that detects, as the second line, a line that is similar in direction to the reference line and whose shape changes between frames of the line image. Water level monitoring device.   The monitoring unit determines whether or not the relative distance between the first line and the second line is less than a predetermined threshold, or the time variation of the relative distance between the first line and the second line. The water level monitoring device according to claim 1, 2 or 3, wherein the water level at the observation point is monitored by determining whether or not is equal to or greater than a predetermined threshold value.   When the monitoring unit determines that the relative distance is less than the threshold value or the change in the relative distance with time is greater than or equal to the threshold value, the information control unit further controls to notify the observation point. The water level monitoring apparatus according to claim 4. The notification control unit
The notification is executed when the monitoring unit determines that the relative distance is less than a threshold value or a change in the relative distance with time is greater than or equal to a threshold value for a plurality of observation points. Item 6. The water level monitoring apparatus according to Item 5. Computer
From an imaging device provided to be able to image a predetermined water area, obtain an observation image captured including an observation point to be observed in the water area,
Of a plurality of lines obtained by edge detection from the observed image and having similar directions, the first line that is not changed in shape between the frames of the observed image and is positioned on the highest position on the observed image; Detecting a second line whose shape changes between frames of the observed image;
A water level monitoring method, comprising: executing a process of monitoring a water level at the observation point based on the first line and the second line. On the computer,
From an imaging device provided to be able to image a predetermined water area, obtain an observation image captured including an observation point to be observed in the water area,
Of a plurality of lines obtained by edge detection from the observed image and having similar directions, the first line that is not changed in shape between the frames of the observed image and is positioned on the highest position on the observed image; Detecting a second line whose shape changes between frames of the observed image;
A water level monitoring program that executes processing for monitoring the water level at the observation point based on the first line and the second line.

Images