Monitoring device and monitoring method

Abstract

To issue appropriate warning so as to enhance safety of a monitored person.SOLUTION: A monitoring device 1 comprises: an image data acquisition section 131 for acquiring captured image data including a monitoring target area; an identification section 132 for identifying a monitoring target person and a monitoring person included in the captured image data; a visual field estimation section 133 for estimating a visual field of the monitoring person; and a warning section 135 for allowing a warning-generating device to issue a warning when the monitoring target area is not included in the visual field of the monitoring person estimated by the visual field estimation section 133 and also the monitoring target person exists within a predetermined range from the monitoring target area.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a monitoring device and a monitoring method for monitoring a target area. [Background technology]

[0002] 2. Description of the Related Art Conventionally, there is known a monitoring device that issues a warning when the distance between a child using play equipment and a supervisor is greater than a predetermined distance (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-27320 A Summary of the Invention [Problem to be solved by the invention]

[0004] A supervisor who is close to a child may not be able to monitor the child to prevent the child from falling into a dangerous situation. If a supervisor who is close to a child does not see an area that should be monitored (e.g., a dangerous area or an area where a child is present) and no warning is issued, the child may fall into a dangerous situation. On the other hand, if a warning is issued simply because a child is in a dangerous area, meaningless warnings will be issued more frequently.

[0005] The present invention has been made in consideration of these points, and has an object to make it possible to issue an appropriate warning so as to increase the safety of the person being monitored. [Means for solving the problem]

[0006] A surveillance device of a first aspect of the present invention comprises an image data acquisition unit that acquires captured image data including a monitored area, an identification unit that identifies the monitored subject and the surveillance person included in the captured image data, a field of view estimation unit that estimates the field of view of the surveillance person, and a warning unit that generates a warning in a device that generates a warning when the monitored area is not included in the field of view of the surveillance person estimated by the field of view estimation unit and the monitored subject is within a predetermined range of the monitored area.

[0007] The monitoring device may further have a state identification unit that identifies the orientation of the person being monitored, and the warning unit may issue the warning on the condition that the area to be monitored is in the orientation of the person being monitored identified by the state identification unit.

[0008] The monitoring device may further have a state identification unit that identifies the movement of the monitored person, and the warning unit may issue the warning on the condition that it determines that the monitored person is approaching the monitored area based on the movement identified by the state identification unit.

[0009] The condition identification unit may identify a required time for the monitored person to reach the monitored area from the monitored person's current location, and the warning unit may change the content of the warning as the required time identified by the condition identification unit becomes shorter.

[0010] The monitoring device may further have a memory unit that stores a plurality of the devices in association with a plurality of the monitored areas, and the warning unit may generate the warning for one or more of the devices among the plurality of devices that are associated with one or more of the monitored areas in which one or more of the monitored persons are included within the specified range.

[0011] A monitoring method of a second aspect of the present invention includes the steps of: acquiring captured image data including a monitored area; identifying the monitored person and the monitor included in the captured image data; estimating the monitor's field of view; and, if the monitored area is not included in the estimated field of view of the monitor and the monitored person is within a predetermined range of the monitored area, generating an alert in an alert-generating device, which is executed by a computer. Effect of the Invention

[0012] According to the present invention, it is possible to provide an appropriate warning to enhance the safety of the person being monitored. [Brief description of the drawings]

[0013] [Figure 1] 1 is a diagram for explaining an overview of a monitoring device 1. FIG. [Diagram 2] 1 is a diagram showing a configuration of a monitoring device 1. [Diagram 3] FIG. 11 is a diagram showing an example of monitoring target area data indicating the position of a monitoring target area. [Figure 4] FIG. 11 is a diagram for explaining supervisor data indicating the status of a supervisor. [Diagram 5] This is monitoring status data indicating the monitoring status of the area to be monitored. [Figure 6] This is monitoring target data indicating the status of the monitoring target. [Figure 7] 4 is a flowchart showing a process flow in the monitoring device 1. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] [Overview of monitoring device 1] Fig. 1 is a diagram for explaining an overview of a monitoring device 1. The monitoring device 1 is a device for generating a warning when a dangerous situation occurs in an area R that includes one or more areas to be monitored (hereinafter referred to as "monitored areas"). In the example shown in Fig. 1, monitored areas A to D are included, but the number of monitored areas is arbitrary.

[0015] One or more surveillance cameras 2 are installed in area R. In the example shown in Fig. 1, surveillance cameras 2-1 to 2-4 are installed, and surveillance cameras 2-1 to 2-4 can capture images of all areas of monitored areas A to D. The surveillance cameras 2 may be monocular cameras that generate captured image data of a two-dimensional plane, or may be compound-eye cameras that generate captured image data indicating distance in the depth direction (i.e., captured image data including three-dimensional spatial coordinates).

[0016] The coordinates of each pixel in each captured image data can be converted into absolute coordinates within the region R based on the installation position and orientation of the surveillance camera 2 in the region R. As an example, absolute coordinates within the region R are assigned to each pixel in each captured image data, and the surveillance device 1 stores the absolute coordinates corresponding to each pixel in each captured image data.

[0017] The surveillance camera 2 transmits captured image data generated by capturing an image of the captureable range to the monitoring device 1. The surveillance camera 2 transmits the captured image data to the monitoring device 1, for example, by wireless communication, but may also transmit the captured image data to the monitoring device 1 by wired communication. The surveillance camera 2 may transmit the captured image data to the monitoring device 1 in association with a camera ID for identifying the surveillance camera 2.

[0018] Warning devices 3 capable of issuing warnings are installed near the areas to be monitored. In the example shown in FIG. 1, warning devices 3-1 to 3-4 are installed near areas A to D to be monitored. The warning devices 3 include, for example, speakers that emit warning sounds. The warning devices 3 may include a display that displays warning characters or images. The warning devices 3 may be information terminals such as smartphones used by people in the vicinity of the areas to be monitored.

[0019] The monitoring device 1 is, for example, a computer. The monitoring device 1 analyzes captured image data received from the monitoring camera 2, and when it detects that a person to be monitored is approaching a monitored area, it transmits warning generation instruction data to the warning device 3 to generate a warning. The person to be monitored is, for example, a child, but may also be a previously set object other than a child (for example, an animal or a vehicle). In the following description, a case where the person to be monitored is a child will be exemplified.

[0020] 1, when a surveillance camera 2 is installed corresponding to each of a plurality of areas to be monitored, a surveillance device 1 analyzes a plurality of captured image data received from the plurality of surveillance cameras 2. When the surveillance device 1 detects that a person to be monitored is approaching the area to be monitored based on the captured image data, the surveillance device 1 generates an alarm in an alarm device 3 installed in the vicinity of the area to be monitored corresponding to the captured image data.

[0021] The monitoring device 1 may be configured integrally with any of the monitoring cameras 2, or may be a computer separate from the monitoring camera 2. The monitoring device 1 may be installed within the area R, or may be installed outside the area R (for example, on the cloud).

[0022] Incidentally, even if a child approaches a monitored area, an adult (e.g., the child's parent) may be watching the child. If the warning device 3 issues a warning even though the child is safe because an adult is watching the child, the child will not be able to approach the monitored area at all, and the child's behavior will be restricted. In addition, if the warning device 3 issues a warning even though the child is safe, even if the warning device 3 issues a warning when a dangerous situation occurs, the person who notices the warning will not recognize that a dangerous situation is occurring.

[0023] Therefore, the monitoring device 1 is characterized in that it estimates the visual field of an adult in the area R, and if the visual field of the adult includes the monitored area, it excludes the monitored area from areas for which a warning is to be issued (i.e., monitored areas). In the following description, an object that is monitored to prevent it from getting too close to the monitored area, such as a child or a small animal, is referred to as a monitored person, and a person who can monitor the monitored person, such as a parent or a monitor, may be referred to as a monitor.

[0024] 1, the visual field of monitor M1 is an area surrounded by a dashed line including monitored area A, and the visual field of monitor M2 is an area surrounded by a dashed line including monitored area D. In this case, the monitoring device 1 excludes monitored area A and monitored area D from the areas subject to warning generation, and sets monitored area B and monitored area C as the areas subject to warning generation.

[0025] As a result, the warning device 3 issues a warning when a child approaches a monitored area that an adult is not watching, but does not issue a warning even if a child approaches a monitored area that an adult is watching, allowing the child to approach the monitored area as long as the adult is watching. Also, since the warning device 3 issues a warning only when there is a truly dangerous situation, when the warning device 3 issues a warning, a person who notices the warning can recognize that a dangerous situation is occurring and take prompt action.

[0026] [Configuration of monitoring device 1] 2 is a diagram showing the configuration of the monitoring device 1. The monitoring device 1 has a communication unit 11, a storage unit 12, and a control unit 13. The control unit 13 has an image data acquisition unit 131, an identification unit 132, a visual field estimation unit 133, a state identification unit 134, and a warning unit 135.

[0027] The communication unit 11 includes a communication interface for transmitting and receiving data to and from other devices. The communication unit 11 receives, for example, captured image data transmitted from the surveillance camera 2. The communication unit 11 inputs the received captured image data to the image data acquisition unit 131. In addition, the communication unit 11 transmits warning generation instruction data to the warning device 3 in response to an instruction from the warning unit 135.

[0028] The storage unit 12 includes a storage medium such as a ROM (Read Only Memory), a RAM (Random Access Memory), or an SSD (Solid State Drive). The storage unit 12 stores a program executed by the control unit 13. The storage unit 12 also temporarily stores captured image data received by the communication unit 11.

[0029] The storage unit 12 stores data related to the surveillance camera 2 and the warning device 3. The storage unit 12 stores a camera ID for identifying the surveillance camera 2, which is used when receiving captured image data from the surveillance camera 2. The storage unit 12 also stores absolute coordinates (coordinates indicating a position in the region R) of each pixel in the captured image data generated by each surveillance camera 2.

[0030] The storage unit 12 also stores a plurality of warning devices 3 in association with a plurality of areas to be monitored. For example, the storage unit 12 stores a device ID for identifying the warning device 3 installed closest to the area to be monitored in association with the position of the area to be monitored.

[0031] Furthermore, the storage unit 12 stores various data necessary for the control unit 13 to cause the warning device 3 to issue a warning when a child approaches a monitored area that is not being watched by an adult. Figures 3 to 6 are diagrams showing examples of data stored in the storage unit 12.

[0032] FIG. 3 is a diagram showing an example of monitored area data indicating the position of a monitored area. The monitored area data includes coordinates on the contour line of the monitored area. The monitored area data shown in FIG. 3 includes coordinates of the four corners of each of the monitored areas A to D shown in FIG. 1. The coordinates shown in FIG. 3 are three-dimensional coordinates, but may be two-dimensional coordinates. In this case, the two-dimensional coordinates are, for example, coordinates corresponding to the positions of the four corners of the monitored area in captured image data generated by capturing an image of the monitored area represented by the three-dimensional coordinates shown in FIG. 3 by the monitoring camera 2, which is a monocular camera.

[0033] Fig. 4 is a diagram for explaining monitor data indicating the status of a monitor. In the monitor data shown in Fig. 4(a), a monitor, a position of the monitor, a viewing direction of the monitor, and a monitored area are associated with each other. The position of the monitor is expressed by three-dimensional coordinates. The viewing direction of the monitor is expressed as polar coordinates (θ, φ) by an angle θ from the X-axis in an XY plane in a polar coordinate space shown in Fig. 4(b) and an angle φ from the XY plane.

[0034] Fig. 5 is monitoring status data showing the monitoring status of the areas to be monitored. The monitoring status data shown in Fig. 5 indicates that the area to be monitored A is being monitored by a monitor M1, and the area to be monitored D is being monitored by a monitor M2. The monitoring status data shown in Fig. 5 also indicates that the areas to be monitored B and C are not being monitored by a monitor, and are being monitored by the monitoring device 1.

[0035] Fig. 6 shows monitoring subject data indicating the state of the monitoring subject. In the monitoring subject data shown in Fig. 6, the monitoring subject, the current location of the monitoring subject, the previous location of the monitoring subject, the movement speed, the area of ​​the movement target, and whether or not a warning is required are associated with each other. The previous location is the location of the monitoring subject detected at the previous detection timing by the monitoring device 1, which detects the location of the monitoring subject at a predetermined time interval. The movement speed is calculated by dividing the distance between the current location and the previous location by the time interval at which the monitoring device 1 detects the location of the monitoring subject.

[0036] 2, the control unit 13 will be described in detail. The control unit 13 has, for example, a CPU (Central Processing Unit). The control unit 13 executes a program stored in the storage unit 12 to function as an image data acquisition unit 131, a specification unit 132, a visual field estimation unit 133, a state specification unit 134, and a warning unit 135.

[0037] The image data acquisition unit 131 acquires captured image data including a monitoring target area via the communication unit 11. The image data acquisition unit 131 inputs the acquired captured image data to the identification unit 132. The image data acquisition unit 131 may store the acquired captured image data in the storage unit 12 and input the captured image data to the identification unit 132 via the storage unit 12.

[0038] The identification unit 132 identifies the surveillance target and the surveillance person included in the captured image data. First, the identification unit 132 identifies a plurality of feature points in the captured image data, and compares the shape of the detected object connecting the plurality of feature points with a reference shape stored in the storage unit 12. The reference shape is the shape of a silhouette of a person. The identification unit 132 identifies the detected object as a person when the similarity between the shape of the detected object identified in the captured image data and the reference shape is equal to or greater than a shape threshold value.

[0039] When the detected object is a person, the identification unit 132 identifies whether the detected object is an adult who is the monitor or a child who is the subject of monitoring based on the size of the detected object. As an example, the identification unit 132 determines that the detected object is an adult when the vertical length of the detected object is equal to or greater than the identification threshold, and determines that the detected object is a child when the vertical length of the detected object is less than the identification threshold.

[0040] Since the detected object becomes smaller at a more distant position in the captured image data, the identification unit 132 may use a discrimination threshold corresponding to a position in the captured image data. For example, the identification unit 132 sets a discrimination threshold corresponding to an upper position in the captured image data smaller than a discrimination threshold corresponding to a lower position.

[0041] In addition, the identification unit 132 may identify adults or children in the captured image data by inputting the captured image data into a machine learning model that has undergone deep learning using image data including images of adults and image data including images of children as teacher data.

[0042] When the detected object is an adult and the adult is not registered in the monitor data as a monitor, the identification unit 132 registers the adult as a new monitor in the monitor data. For example, when a monitor associated with a position within a predetermined distance from the position of the detected adult is not registered in the monitored person data, the identification unit 132 determines that the detected person is unregistered. In this case, the identification unit 132 assigns identification information (ID) to the detected adult, associates the assigned ID with the position of the detected adult, and registers it in the monitored person data. For example, the identification unit 132 determines the position of the adult to be the center position of the adult's head in the captured image data. The predetermined distance is, for example, a distance that is predetermined as a distance that a person can move between the previous detection timing and the latest detection timing.

[0043] When the detected object is a watcher, the identification unit 132 may identify the face direction of the watcher by analyzing the captured image data, and identify the viewing direction based on the identified face direction. The identification unit 132 updates the viewing direction associated with the watcher in the watcher data stored in the storage unit 12.

[0044] The visual field estimation unit 133 estimates the visual field of the observer. For example, the visual field estimation unit 133 estimates that the visual field of the observer is a predetermined angle range centered on the viewing direction of the observer identified by the identification unit 132 in the captured image data. As an example, the visual field estimation unit 133 estimates that the visual field is a cone-shaped space of ±30 degrees in the vertical and horizontal directions with respect to the viewing direction. The visual field estimation unit 133 may estimate that the visual field is a sector-shaped range of ±30 degrees in the horizontal directions centered on the viewing direction, starting from the center position of the observer's head in the captured image data.

[0045] Next, the visual field estimation unit 133 judges whether or not the estimated visual field includes a monitored area. In the example shown in Fig. 1, the visual field estimation unit 133 judges that the visual field of the monitor M1 includes the monitored area A, and judges that the visual field of the monitor M2 includes the monitored area D. The visual field estimation unit 133 notifies the warning unit 135 of information (e.g., an area ID) for identifying the monitored area included in the visual field of either monitor.

[0046] The state identification unit 134 identifies the state of the monitored person. For example, the state identification unit 134 identifies at least one of the position or orientation of the monitored person. The state identification unit 134 may identify the movement of the monitored person. The state identification unit 134 may further identify the moving speed of the monitored person. The state identification unit 134 notifies the warning unit 135 of the monitored person's state by storing the identified state of the monitored person in the memory unit 12. For example, when the state identification unit 134 detects a change in the monitored person's state, the state identification unit 134 notifies the warning unit 135 of the monitored person's state by updating the monitored person data.

[0047] The state identification unit 134 may identify the time required for the person to be monitored to reach the area to be monitored from the current location of the person to be monitored. The state identification unit 134 refers to the coordinates of the area to be monitored stored in the storage unit 12, and calculates the distance between the current location of the person to be monitored in the area to be monitored and a location closest to the current location of the person to be monitored. The state identification unit 134 identifies the time required for the person to be monitored to reach the area to be monitored from the current location by dividing the calculated distance by the moving speed of the person to be monitored. The state identification unit 134 notifies the warning unit 135 of the identified time required.

[0048] The warning unit 135 issues a warning to the warning device 3 when the monitored area is not included in the field of view of the monitor estimated by the field of view estimation unit 133 and the monitored person is within a predetermined range from the monitored area. The predetermined range is stored in the storage unit 12 in association with the monitored area. The predetermined range is represented, for example, by the coordinates of multiple positions on the outline of the predetermined range. The warning unit 135 issues a warning to the warning device 3 corresponding to the monitored area in which the monitored person has been detected within the predetermined range by transmitting warning generation instruction data via the communication unit 11.

[0049] When the region R monitored by the monitoring device 1 includes multiple monitored areas, the warning unit 135 issues a warning to one or more warning devices 3 associated with one or more monitored areas in which one or more monitored persons are included within a predetermined range, among the multiple warning devices 3. When a monitored person is within a predetermined range from each of multiple monitored areas that are not included in the field of view of the monitor, the warning unit 135 issues a warning to the warning devices 3 corresponding to each of the multiple monitored areas.

[0050] The warning unit 135 may generate warning sounds with different sound quality or sound patterns for each of the multiple warning devices 3. By each of the multiple warning devices 3 emitting a different warning sound, even if multiple warning devices 3 emit warning sounds at the same time, a person who hears the warning sound can easily identify in which monitored area the warning sound has been emitted.

[0051] The warning unit 135 may generate a warning on the condition that there is a monitoring target area in the direction of the monitoring target person identified by the state identification unit 134. Specifically, the warning unit 135 determines whether or not there is a monitoring target area on a straight line extending from the current position of the monitoring target person as a starting point in the direction of the monitoring target person, and if there is a monitoring target area on the straight line, transmits warning generation instruction data to the warning device 3 corresponding to the monitoring target area. If there is no monitoring target area in the direction of the monitoring target person, the warning unit 135 may not generate a warning even if the monitoring target person is within a predetermined range from the monitoring target area that is not monitored by the monitor. By operating in this manner, the warning unit 135 can suppress the issuance of a warning even if there is a low probability that the monitoring target person will be in a dangerous state.

[0052] The warning unit 135 may generate a warning on the condition that it is determined that the person to be monitored is approaching the area to be monitored based on the movement identified by the state identification unit 134. The warning unit 135 determines that the person to be monitored is approaching the area to be monitored when the area to be monitored is on a straight line connecting the previous position of the person to be monitored identified by the state identification unit 134 as a starting point and the current position.

[0053] The warning unit 135 may determine the contents of the warning based on the time required from the current location of the person to be monitored to the area to be monitored, which is identified by the state identification unit 134. As an example, the warning unit 135 changes the contents of the warning as the required time specified by the state identification unit 134 becomes shorter. Specifically, the warning unit 135 instructs the warning device 3 to increase the volume of the warning sound or increase the size of the characters displayed on the warning display as the required time becomes shorter. By the warning unit 135 operating in this manner, the closer the person to be monitored gets to the area to be monitored, the more people are likely to notice the warning.

[0054] [Processing flow in monitoring device 1] Fig. 7 is a flowchart showing the flow of processing in the monitoring device 1. The flowchart shown in Fig. 7 starts at the point when the monitoring start time preset in the monitoring device 1 arrives.

[0055] The image data acquisition unit 131 acquires captured image data from the surveillance camera 2 (S11). The identification unit 132 analyzes the captured image data acquired by the image data acquisition unit 131 and detects an image of a person in the captured image data (S12). The identification unit 132 determines whether the detected person is an adult or a child (S13).

[0056] When the identification unit 132 determines that the detected person is an adult (YES in S13), it determines that the detected person is a supervisor, and determines whether the detected person is registered in the supervisor data (S14). For example, when a supervisor associated with a position within a predetermined distance from the position of the detected adult is not registered in the supervisor data, the identification unit 132 determines that the detected adult is unregistered (YES in S14), and updates the supervisor data by registering the detected adult in the supervisor data (S15).

[0057] If a monitor associated with a position within a predetermined distance from the position of the detected person is registered in the monitor data, the identification unit 132 determines that the detected person is a registered person (NO in S14), and proceeds to S16. In S16, the identification unit 132 updates the monitor position by registering the position of the detected person in the monitor data (S16).

[0058] Further, the visual field estimation unit 133 estimates the visual field of the observer (S17). The visual field estimation unit 133 identifies the monitoring target area that the observer can monitor (i.e., the area being monitored) based on the estimated visual field (S18). The visual field estimation unit 133 updates the monitoring status data stored in the storage unit 12 based on the identification result.

[0059] In S13, when the identification unit 132 determines that the detected person is not an adult (NO in S13), it determines that the detected person is a child. When the identification unit 132 determines that the detected person is a child, it determines that the detected person is a person to be monitored, and determines whether the detected child is registered in the monitored person data (S19).

[0060] For example, if the monitored person is not registered in the monitored person data in association with a position within a predetermined distance from the detected child's position, the identification unit 132 determines that the detected person is unregistered (YES in S19), and updates the monitored person data by registering the detected child in the monitored person data (S20).Then, the status identification unit 134 identifies the status of the detected child (S21).

[0061] Next, the warning unit 135 judges whether the child's location is close to a non-monitored monitored area by referring to the monitoring status data and the monitored person data (S22). If the warning unit 135 judges that the child's location is within a predetermined range from the non-monitored monitored area (YES in S22), it generates a warning by transmitting warning generation instruction data to the warning device 3 installed in the monitored area (S23). If the warning unit 135 judges that the child's location is not within a predetermined range from the non-monitored monitored area (NO in S22), it advances the process to S24 without generating a warning.

[0062] The control unit 13 repeats the processes from S11 to S24 at regular time intervals until the condition for ending monitoring is satisfied (NO in S24). The condition for ending monitoring is, for example, that the time period to be monitored has ended or that the power supply of the monitoring device 1 has been turned off. When the condition for ending monitoring is satisfied (YES in S24), the control unit 13 ends the monitoring process.

[0063] [Variations] In the above description, when the entire area to be monitored is included in the field of view of the monitor, the area to be monitored is excluded from targets for which a warning is issued, but when only a part of the area to be monitored is included in the field of view of the monitor, the part of the area to be monitored may be excluded from targets for which a warning is issued. In this case, the warning unit 135 issues a warning to the warning device 3 corresponding to the area to be monitored, on condition that the person to be monitored is within a predetermined range of the part of the area to be monitored included in the field of view. By operating the warning unit 135 in this manner, even if the entire area to be monitored is not included in the field of view of the monitor, the warning device 3 can issue an appropriate warning.

[0064] [Effects of monitoring device 1] As described above, the monitoring device 1 has a visual field estimation unit 133 that estimates the visual field of the monitor, and a warning unit 135 that issues a warning to the warning device 3 when the monitored area is not included in the visual field of the monitor estimated by the visual field estimation unit 133 and the monitored person is within a predetermined range from the monitored area. Since the monitoring device 1 has such a configuration, a warning is issued when a child is near a monitored area that is not being monitored, and a warning is not issued when a child is near a monitored area monitored by an adult, so that a warning is not issued when it is unnecessary and an appropriate warning can be issued to increase the safety of the monitored person.

[0065] Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist of the present invention. For example, all or part of the device can be configured by distributing or integrating functionally or physically in any unit. In addition, new embodiments resulting from any combination of multiple embodiments are also included in the embodiments of the present invention. The effect of the new embodiment resulting from the combination combines the effect of the original embodiment. [Explanation of symbols]

[0066] 1 Monitoring device 2. Surveillance cameras 3. Warning Devices 11 Communications Department 12 Storage section 13 Control section 131 Image data acquisition unit 132 Specific part 133 Field of view estimation unit 134 Status Identification Unit 135 Warning section

Claims

1. an image data acquisition unit that acquires captured image data including a monitoring target area; An identification unit that identifies a surveillance target and a surveillance person included in the captured image data; A visual field estimation unit that estimates the visual field of the observer; a warning unit that generates a warning in a device that generates a warning when the surveillance target area is not included in the surveillance target area estimated by the visual field estimation unit and the surveillance target person is within a predetermined range from the surveillance target area; A monitoring device having the above configuration.

2. A state identification unit that identifies a direction of the monitoring target person, the warning unit issues the warning on condition that the monitoring target area is in a direction in which the monitoring target person identified by the state identification unit is facing; The monitoring device according to claim 1.

3. A state identification unit that identifies a movement of the monitoring target person, the warning unit issues the warning on condition that it is determined that the person to be monitored is approaching the area to be monitored based on the movement identified by the state identification unit. The monitoring device according to claim 1.

4. the state identification unit identifies a required time for the person to be monitored to reach the monitoring target area from a current location of the person to be monitored, the warning unit changes the content of the warning as the required time identified by the state identifying unit becomes shorter. The monitoring device according to claim 3.

5. A storage unit that stores a plurality of the devices in association with a plurality of the monitoring target areas, The warning unit issues the warning to one or more of the devices associated with one or more of the monitoring target areas in which the one or more monitoring targets are included within the predetermined range, among the plurality of devices. The monitoring device according to claim 1.

6. The computer executes acquiring captured image data including a monitoring target area; Identifying a surveillance target and a surveillance person included in the captured image data; estimating the field of view of the observer; generating a warning in a device that generates a warning when the area to be monitored is not included in the estimated field of view of the monitor and the person to be monitored is within a predetermined range from the area to be monitored; The monitoring method includes:

Images