Monitoring system, monitoring method, and program

Abstract

A monitoring system (100) according to the present disclosure is provided with: a plurality of image-capturing means (101) installed at a plurality of places within an area to be monitored; a specifying means (102) for specifying objects to be monitored from video images captured by the plurality of image-capturing means; a determination means (103) for determining, among the specified objects to be monitored, whether or not the same object to be monitored is displayed in the video images of the plurality of image-capturing means that have been captured in the same time segment; and a control means (104) for selecting at least one image-capturing means on the basis of a predetermined condition from among the image-capturing means which have captured the video images displaying the same object to be monitored, and controlling the image-capturing by the selected image-capturing means or controlling supply of the video images.

Description

[Technical Field]

[0001] The present disclosure relates to a monitoring system, a monitoring method, and a non-transitory computer-readable medium having a program stored thereon. [Background technology]

[0002] Various surveillance systems have been proposed that monitor the behavior of a target using surveillance images. Such surveillance systems generally include multiple surveillance cameras or other imaging devices installed at multiple locations, and continuously capture images of the target area.

[0003] For example, Patent Document 1 describes a system in which an image sensor installed around a user's bed in a hospital, nursing home, etc. captures images within a capture range, and issues a specific notification if an abnormality in the user is detected from the user's behavior. Also, Patent Document 2 describes a system in which the behavior of a monitored object in a monitoring image is recognized and it is determined whether or not a response is required for the monitored object's behavior. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-045573 [Patent Document 2] Japanese Patent Application Publication No. 2020-194493 Summary of the Invention [Problem to be solved by the invention]

[0005] In such a surveillance system, when various processes such as a process for recognizing the behavior of a monitored object are performed by a computing device located away from the image capturing devices, the surveillance images captured by the multiple image capturing devices are transmitted to the computing device via a network. If all of the surveillance images are transmitted, there are concerns about a shortage of communication resources and a shortage of human resources due to a heavy workload on the monitors who monitor the monitored object based on the surveillance images.

[0006] An object of the present disclosure is to provide a monitoring system, a monitoring method, and a non-transitory computer-readable medium storing a program, which can eliminate resource shortages. [Means for solving the problem]

[0007] A surveillance system according to one embodiment comprises a plurality of imaging means installed at a plurality of locations within a surveillance area, an identification means for identifying surveillance targets from images captured by the plurality of imaging means, a determination means for determining whether the same surveillance target appears in images captured by the plurality of imaging means during the same time period among the identified surveillance targets, and a control means for selecting at least one imaging means from the imaging means that captured images capturing the same surveillance target based on predetermined conditions, and controlling the imaging by the selected imaging means or the supply of images.

[0008] An information processing method according to one embodiment identifies a monitoring target from images captured by multiple imaging means installed at multiple locations within a monitoring area, determines whether the same monitoring target appears in images captured by multiple imaging means during the same time period among the identified monitoring targets, selects at least one imaging means from the imaging means that captured images of the same monitoring target based on predetermined conditions, and controls the imaging by the selected imaging means or controls the supply of images.

[0009] In one embodiment, a non-transitory computer-readable medium stores a program that causes a computer to execute the following process: identify a monitoring target from images captured by multiple imaging means installed at multiple locations within a monitored area; determine whether the same monitoring target appears in images captured by multiple imaging means during the same time period among the identified monitoring targets; select at least one imaging means from the imaging means that captured images capturing the same monitoring target based on predetermined conditions; and control the imaging by the selected imaging means or the supply of images. [Effects of the Invention]

[0010] According to the above-mentioned aspects, a monitoring system, a monitoring method, and a non-transitory computer-readable medium storing a program are provided, which can resolve resource shortages. do It is possible. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a configuration of a monitoring system according to an embodiment. [Figure 2] 1 is a diagram illustrating an example of the configuration of a monitoring system including an event detection device according to an embodiment. [Figure 3] FIG. 1 is a diagram showing an example of the arrangement of a plurality of imaging devices in a hospital. [Figure 4] FIG. 10 is a diagram illustrating an example of a risk calculation condition DB. [Figure 5] FIG. 1 is a flow diagram illustrating an example of the flow of a monitoring method according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. Note that the following description and drawings have been omitted and simplified as appropriate for clarity of explanation. In addition, in each of the following drawings, the same elements are given the same reference numerals, and duplicate explanations are omitted as necessary. Furthermore, specific numerical values ​​shown below are merely examples to facilitate understanding of the present disclosure, and are not intended to be limiting.

[0013] The embodiments relate to various monitoring systems that monitor the behavior of a monitoring target using monitoring video. Fig. 1 is a block diagram showing the configuration of a monitoring system 100 according to the embodiment. As shown in Fig. 1, the monitoring system 100 according to the embodiment includes an imaging unit 101, an identification unit 102, and a determination unit 103.

[0014] A plurality of imaging means 101 are installed at a plurality of locations within the monitored area and capture images of the monitoring target. Identification means 102 identifies the monitoring target from the images captured by the plurality of imaging means. Determination means 103 determines whether or not the same monitoring target appears in images captured by the plurality of imaging means during the same time period among the identified monitoring targets. Control means 104 selects at least one imaging means from the imaging means that have captured images of the same monitoring target based on predetermined conditions, and controls the imaging by the selected imaging means or controls the supply of images.

[0015] In this way, by controlling the shooting by at least one selected imaging means among the imaging means that have captured images showing the same monitored object, or by controlling the supply of images, it is possible to resolve resource shortages.

[0016] Next, a specific example of a monitoring system 100 according to an embodiment will be described with reference to Figures 2 and 3. Figure 2 is a diagram showing an example of the configuration of a monitoring system 100 including an event detection device 10 according to an embodiment. As shown in Figure 2, the monitoring system 100 includes the event detection device 10, multiple image capture devices 20, and a notification device 30. Figure 3 is a diagram showing an example of the arrangement of the multiple image capture devices 20. In the following description, the multiple image capture devices will be collectively referred to as image capture devices 20, and different reference numerals will be used when referring to individual image capture devices.

[0017] 2, the monitoring system 100 monitors the behavior of a patient or a care recipient, which is a monitoring target, using surveillance video in a hospital, a care facility, etc. The monitoring system 100 grasps the behavior of the monitoring target by image analysis or the like, and issues a specific notification corresponding to the type of the grasped behavior.

[0018] The monitored object is not limited to a person, but also includes an object that moves within the monitored area, such as a cleaning robot or a transport robot. In the following, an example will be described in which the monitoring system 100 monitors a patient 1 in a hospital. The monitor who monitors the patient 1 is a medical staff member 2. The monitoring system 100 can notify the medical staff member 2 of a request for action regarding the patient 1 by outputting a notification corresponding to the detected movements of the patient 1 to the notification device 30.

[0019] In the monitoring system 100, the event detection device 10, the photographing device 20, and the notification device 30 can be placed at appropriate distances from each other. For example, the event detection device 10, the photographing device 20, and the notification device 30 are connected via a network N and communicate in real time. Here, the network N is a communication network such as the Internet, a dedicated line, 5G, or LTE. In the case of a public network such as the Internet, it is preferable that a closed network is formed using a Virtual Private Network (VPN) or the like. However, the monitoring system 100 is not limited to such a placement. For example, part of the configuration of the event detection device 10 may be mounted within the photographing device 20.

[0020] The multiple image capturing devices 20 are so-called surveillance cameras, and are installed at multiple locations within the hospital. The image capturing devices 20 continuously capture images of a predetermined capture range (image capturing angle of view) at a predetermined capture rate, and generate video data consisting of multiple image data in time series. Note that the image capturing device 20 may be configured to move the capture range at a predetermined time interval.

[0021] An example of the arrangement of multiple imaging devices 20 will now be described with reference to Fig. 3. In the example shown in Fig. 3, two private rooms 32A and 32B are provided on one side of a corridor 31 located in the center of the hospital 3, and a shared space 33 and a monitoring room 34 are provided on the other side. Also, a staircase 35 is provided at the end of the corridor 31 for moving from this floor to other floors. A free space 36 is provided between the staircase 35 and the private room 32B.

[0022] One imaging device 20 is provided in each of the private rooms 32A and 32B where patients stay, and two imaging devices 20 are provided in the shared space 33. For the sake of explanation, the imaging device installed in the private room 32A is referred to as 20A, the imaging device installed in the private room 32B as 20B, and the imaging devices installed in the shared space 33 as 20C and 20D. In addition, imaging devices 20E and 20F are provided at the ends of the corridor 31, imaging device 20G is provided in the free space 36, and imaging device 20H is provided on the staircase 35.

[0023] 3, the imaging ranges of the imaging devices 20F, 20G, and 20H are indicated by dashed lines. As can be seen from the figure, the imaging range of the imaging device 20F partially overlaps with the imaging range of the imaging device 20G, and another portion overlaps with the imaging range of the imaging device 20H. Therefore, the patient 1 imaged by the imaging devices 20F, 20G, and 20H is included in the images generated by each of the imaging devices 20F, 20G, and 20H.

[0024] A notification device 30 that receives notifications from the event detection device 10 is installed in a monitoring room 34 where the medical staff 2 are staying. The notification device 30 may be a personal computer, a tablet terminal capable of input and output via a touch panel, a smartphone, or the like. The notification device 30 may be a dedicated device, or may be realized by installing a dedicated application on a general-purpose information terminal. Furthermore, it is preferable that the monitoring system 100 be configured so that small information terminals carried by medical staff 2 in locations other than the monitoring room 34 can also be used as the notification device 30.

[0025] 2 shows the hardware configuration of the event detection device 10 as well as logical functional blocks that execute the calculation processes required for monitoring and the calculation processes for resolving resource shortages. As shown in Fig. 2, the event detection device 10 includes an acquisition unit 11, an identification unit 12, a determination unit 13, a control unit 14, a detection unit 15, a calculation unit 16, a notification unit 17, and a storage unit 18, which are interconnected by a bus (not shown). The event detection device 10 is a calculation processing device such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit).

[0026] The identification unit 12, the determination unit 13, the control unit 14, the detection unit 15, the calculation unit 16, and the notification unit 17 correspond to the identification means, the determination means, the control means, the detection means, the calculation means, and the notification means set forth in the claims. The storage unit 18 includes a person DB (database) and a risk calculation condition DB.

[0027] The acquisition unit 11 acquires surveillance video captured by the multiple image capture devices 20 during surveillance operation. The identification unit 12 refers to the person DB in the storage unit 18, detects the surveillance target from the video captured by the multiple image capture devices 20, and identifies the patient 1. In the person DB, for example, feature amounts related to the facial image of the patient 1 who is the surveillance target are registered in association with a person ID for identifying each patient 1.

[0028] In this case, the identification unit 12 can extract the facial area of ​​the person included in the surveillance video and compare it with the feature amount stored in the person DB to identify the patient 1. However, the identification unit 12 is not limited to identifying the patient 1 based on the facial image of the patient 1, and may identify the patient 1 based on, for example, the body shape, gait, skeleton, etc. of the patient 1. Furthermore, the patient 1 may be identified by combining multiple features.

[0029] The identification of the monitoring target can be performed by various known image analysis methods. For example, the identification unit 12 can be configured using a machine learning image analysis device having a neural network. It is also preferable that the identification unit 12 simultaneously identifies multiple monitoring targets in parallel.

[0030] The determination unit 13 determines whether or not the same patient 1 appears in images of the identified patients 1 captured by the multiple imaging devices 20 during the same time period (at the same time). As described above, in the example shown in Fig. 3, the same patient 1 is captured by the imaging devices 20F, 20G, and 20H. The determination unit 13 determines that the imaging devices 20F, 20G, and 20H are the imaging means that captured the images capturing the same patient 1.

[0031] As described above, transmitting all of the surveillance images captured by the multiple image capturing devices 20 to the event detection device 10 increases the network load and may result in a shortage of communication resources. Furthermore, in the case of a pay-per-use network, communication costs increase. Therefore, in this embodiment, the control unit 14 selects at least one image capturing device 20 based on predetermined conditions from among the image capturing devices 20 that have captured images of the same patient 1, and controls the image capturing by the selected image capturing device 20 or the supply of images.

[0032] The specified condition may be, for example, the orientation or range of the same patient 1, such as whether the face of the patient 1 is shown in the surveillance footage, whether the entire body is shown, or whether the patient 1 is facing forward.

[0033] For example, in a surveillance video showing the whole body of the patient 1, it is possible to detect the staggering of the patient 1. Also, in a surveillance video showing the face of the patient 1, it is possible to measure the biological information of the patient.

[0034] Here, biological information refers to information about a living body that can be measured from an image of a person, such as heart rate, blood pressure, oxygen saturation, etc. From a surveillance video showing the face of patient 1, biological information can be obtained based on information indicating changes in the condition of the body surface.

[0035] An example of information indicating a change in the state of the body surface is a change in the color of the body surface based on the flow of blood in the patient. Each frame of video data capturing the patient includes a change in the color of the patient's body surface. For example, if the patient 1 is breathing heavily in the surveillance video, the percutaneous arterial oxygen saturation (SpO2, hereinafter referred to as oxygen saturation) of the patient 1 can be measured based on the surveillance video.

[0036] Oxygen saturation is the percentage of hemoglobin in the blood that is bound to oxygen. When light is shone on a body surface such as the face, the intensity of reflected light changes over time due to fluctuations in the amount of hemoglobin in the blood vessels. The event detection device 10 can estimate SpO2 by analyzing the video. In this way, the control unit 14 selects at least one imaging device 20 from the imaging means that have captured video of the same patient 1 according to the purpose.

[0037] The control unit 14 then controls the shooting of the selected image capturing device 20 or the supply of video images. "Controlling shooting" includes, for example, controlling the start and stop of shooting by the image capturing device 20, and changing the image quality by controlling the optical zoom or digital zoom of the image capturing device 20. "Controlling the supply of video images" includes controlling the start and stop of the supply of video images by the image capturing device 20, and changing the image quality by changing the compression rate of the surveillance video images sent from the image capturing device 20 to the event detection device 10.

[0038] The imaging device 20 typically compresses and encodes the monitoring image to reduce the data size when transmitting the monitoring image. By lowering the compression rate of the target region compared to the compression rate of the non-target region, the data size of the entire monitoring image can be reduced while the target region has higher image quality than the non-target region. For example, when measuring the oxygen saturation of a patient 1, changing the compression rate can make the facial region (target region) of the patient 1 in the monitoring video have higher image quality than the non-facial region.

[0039] This allows the area required for measuring biometric information to be acquired with higher image quality than other areas while saving communication resources compared to when all video data is transmitted without compression or when all video data is compressed and transmitted at a fixed compression rate, making it possible to measure biometric information with high reliability.

[0040] Another example of the predetermined condition is the environment around the patient 1 within the imaging range captured by each of the multiple imaging devices 20. For example, the control unit 14 can select the imaging device 20H that is capturing an image of a staircase 35 (a caution area) where there is a risk of tripping or falling, from among the imaging devices 20F, 20G, and 20H that have captured images of the same patient 1.

[0041] The control unit 14 can control the camera 20 so that it transmits only the surveillance video of the camera 20H, which is installed so that its imaging range includes such a caution area, to the event detection device 10, and does not transmit the surveillance video of the other camera devices (camera devices 20F, 20G). This can prevent an increase in network traffic.

[0042] Note that, when one monitoring image shows the entire body of the patient 1 and another monitoring image shows the facial region of the same patient 1, the control unit 14 can also select both of the image capturing devices 20 that captured these monitoring images. By selecting multiple image capturing devices 20 in this way, it is possible to perform stagger detection using the monitoring image from one image capturing device 20 and measure oxygen saturation using the monitoring image from the other image capturing device 20.

[0043] In this case, when multiple image capturing devices 20 are selected, the control unit 14 can control each selected image capturing device 20 to perform different processing. That is, one image capturing device 20 can perform processing to cut out only the peripheral area of ​​the patient 1, which is necessary for detecting staggering, from the monitoring video. In addition, the other image capturing device 20 can provide a higher image quality for the facial area of ​​the patient 1 than for the non-facial area in order to measure oxygen saturation. This allows the medical staff 2 to more reliably check the condition of the patient 1.

[0044] Furthermore, if the patient 1 is not included in the surveillance video and / or if there is no movement in the surveillance video, the control unit 14 controls the camera 20 that captured the video to stop supplying the surveillance video. That is, only the camera 20 capturing the video that includes the patient 1 performing a movement transmits the surveillance video to the event detection device 10. This reduces the amount of data of the surveillance video transmitted from the camera 20 to the event detection device 10, thereby saving communication resources. Furthermore, if a person other than the patient 1 is captured in the surveillance video of the selected camera 20, the image quality of the other person in addition to the patient 1 can be made higher than that of the other areas.

[0045] Meanwhile, there is a need to automatically detect when a patient 1 is in a situation requiring attention within a hospital 3, so that medical staff 2 can quickly respond. Examples of situations requiring attention include when a patient becomes unsteady due to poor health, when the patient goes to a cautionary area, or when a patient 1 who needs to be accompanied acts alone.

[0046] However, because symptoms, personality, and other characteristics differ from patient 1 to patient 1, the degree of danger in a given situation differs for each patient 1. Even when the degree of danger to patients 1 is low, if medical staff 2 were to respond to each and every notification issued in a uniform manner, the workload on medical staff 2 would increase, raising concerns about a shortage of human resources, and also placing a heavy psychological burden on patients.

[0047] Therefore, a configuration for resolving the shortage of human resources will be described below. The detector 15 acquires video after the control unit 14 has controlled the imaging or the supply of video, and detects the condition of the patient 1 in the video. For example, the detector 15 can acquire video with a controlled compression rate so that a predetermined target area has higher image quality than areas other than the target area.

[0048] The situation of the patient 1 detected by the detection unit 15 includes, for example, the patient's staggering, entering or leaving a room, destination, and other movements of the patient 1. The movements can be detected by various conventionally known image analysis methods. For example, the detection unit 15 can be configured using an image recognition device using deep learning technology.

[0049] The destination can be estimated based on the installation position of the photographing device 20 that photographed the patient 1 and the direction of the feet of the patient 1. Furthermore, the condition of the patient 1 also includes whether there is a medical staff member 2 or a family member accompanying the patient 1, and the above-mentioned biological information such as the heart rate and oxygen saturation level.

[0050] The calculation unit 16 calculates the degree of risk by referring to the risk calculation condition DB in the storage unit 18 based on the conditions preset for the patient 1 and the situation of the patient 1. The risk calculation condition DB stores, for example, the degree of risk for each situation condition linked to a person ID. Fig. 4 shows an example of the risk calculation condition DB.

[0051] 4, the risk calculation condition DB stores a combination of symptoms, situation conditions, and risk of each patient 1 for each person ID, which is the identification information of each patient 1. The risk is set individually for each patient 1. That is, in the embodiment, the risk of the patient 1's movement is set to a different value depending on the patient's symptoms.

[0052] For example, even if the situation condition is the same, "unsteadiness," a patient (P1) with severe symptoms will have a higher risk value than a patient (P2) with mild symptoms. Note that although the risk value is set to differ only depending on the symptoms of patient 1, the risk value may be set according to two or more characteristics, including not only symptoms but also other factors such as personality.

[0053] For example, if a patient (P1) with serious symptoms leaves his room and does not return within a predetermined time, the risk level can be increased, or if the serious patient (P1) is heading to a caution area such as a staircase. Also, different risk levels may be set for when the patient 1 is in a private room and when he is in the shared space 33. Note that if the patient 1's symptoms improve, the risk level calculation condition DB can be rewritten so that the risk level value becomes smaller.

[0054] The notification unit 17 outputs notification information to predetermined notification destinations based on the risk level calculated by the calculation unit 16. The predetermined notification destinations include the notification device 30 installed in the monitoring room 34, small information terminals carried by medical staff 2 in locations other than the monitoring room 34, as well as a nurse call system, a speaker at a nurse station, etc. Furthermore, the notification unit 17 may output notification information only to limited notification destinations, such as an information terminal carried by the medical staff 2 in charge of the identified patient 1 or an information terminal carried by the medical staff 2 closest to the patient 1.

[0055] The notification information includes the identified patient 1, the patient 1's location, and the level of danger. The notification device 30 can output the notification information from the notification unit 17 in the form of video, text, audio, etc. that can be recognized by the medical staff 2. When the level of danger is equal to or higher than a predetermined threshold, the notification device 30 can vary the content of the notification depending on the level of danger. Examples of notification content include the volume of the warning sound, the size of the display, and the font. For example, the higher the level of danger, the louder the warning sound the notification device 30 emits so that the medical staff 2 recognizes the urgency.

[0056] In this way, by changing the content of the notification depending on the degree of risk that takes into account the characteristics of the patient 1, the burden on the medical staff 2 can be reduced compared to when notifications are issued uniformly.

[0057] The storage unit 18 is a non-volatile storage device such as a hard disk drive (HDD), a solid state drive (SSD), or a flash memory. The storage unit 18 stores a program that realizes each function of the components included in the event detection device 10. The event detection device 10 executes this program to realize each function of the components included in the event detection device 10.

[0058] The storage unit 18 may include a memory such as a RAM (Random Access Memory) or a ROM (Read Only Memory). When executing the above programs, the event detection device 10 may read these programs into memory before executing them, or may execute them without reading them into memory. As described above, the storage unit 18 also stores a person DB and a risk calculation condition DB.

[0059] Furthermore, the identification unit 12 can identify the type of person included in the video captured by the multiple image capture devices 20 by referring to the person DB in the storage unit 18. In addition to feature amounts related to the facial image of the patient 1 who is the monitoring target, the person DB can also store feature amounts related to the facial images of medical staff 2 such as doctors and nurses, and of people accompanying the patient 1. The identification unit 12 can classify the people in the recognized video into types such as "patient," "medical staff," and "attendant." Furthermore, among "patients," people with severe symptoms who particularly require attention from the medical staff 2 can be classified into the type of "person requiring caution."

[0060] The identification unit 12 can identify the type of person using various known methods, and for example, it is possible to identify a person in the video based on their clothing, posture, gait, etc. The control unit 14 can change the image quality of the video captured by the selected camera device 20 depending on the type of person in the video captured by that camera device 20.

[0061] When transmitting video from the selected imaging device 20, the image quality of persons of a category other than the patient 1 can be lowered than that of the patient 1 before transmission. By transmitting the image quality of persons of a category other than the patient 1 at a lower quality than that of the patient 1, the amount of data can be reduced and tracking of the patient 1 can be made easier. Furthermore, even if the person is a patient, if it can be confirmed from the facial image, posture, walking style, etc. that they are not a suspicious person, the image quality can be lowered and transmitted in the same manner as above.

[0062] When delivering video from a selected imaging device 20 capturing a caution area such as a staircase where falls are common, all patients 1 can be displayed with higher image quality than the surrounding area, even if they are not persons of concern. Similarly, when a patient 1 is in a different location than usual, all patients 1 can be displayed with higher image quality than the surrounding area, even if they are not persons of concern.

[0063] Furthermore, in order to collect information such as the body shape and gait of the monitoring target, which is used to identify the monitoring target, the image quality of the monitoring target area in the monitoring image captured by the image capturing device 20 installed in the hallway can be made higher than that of other areas. This allows the detailed characteristics of multiple patients to be more accurately reflected, improving the accuracy of person recognition.

[0064] Furthermore, the detection unit 15 can detect whether the patient 1 is together with another person. For example, when the patient 1 and another person are within a predetermined distance for a predetermined time, the detection unit 15 can determine that the patient 1 is together with another person. When the patient 1 is together with another person, the calculation unit 16 can reduce the risk level. Furthermore, the degree of reduction in the risk level calculated by the calculation unit 16 can be changed depending on the type of other person who is together with the patient 1.

[0065] For example, when patient 1 is with medical staff 2, even if patient 1 is in a dangerous situation, medical staff 2 can immediately confirm the situation and respond to patient 1, so the risk level can be reduced to a level that does not output notification information to the notification device 30. On the other hand, when patient 1 is with an attendant, the risk level can be reduced by a smaller amount than when patient 1 is with medical staff 2.

[0066] Here, a monitoring method according to the embodiment will be described with reference to Fig. 5. Fig. 5 is a flow diagram illustrating an example of the flow of the monitoring method according to the embodiment. First, the identification unit 12 identifies patients 1 from monitoring footage captured by multiple image capture devices 20 installed at multiple locations within the monitoring target area (S11). Then, it is determined whether or not the same patient 1 appears in the monitoring footage captured by the multiple image capture devices 20 during the same time period among the identified patients 1 (S12).

[0067] If the same patient 1 is not captured (S12, NO), the process ends. If the same patient 1 is captured (S12, YES), at least one imaging device 20 is selected based on predetermined conditions from the imaging devices 20 that captured monitoring footage showing the same monitoring target, and the imaging by the selected imaging device 20 or the supply of video is controlled (S13).

[0068] For example, the imaging device 20 transmits to the event detection device 10 a monitoring video in which a specific area (for example, a face area) of the same patient 1 has higher image quality than other areas. This makes it possible to acquire the area necessary for measuring biometric information with higher image quality than other areas while saving communication resources, compared to when all video data is transmitted without compression or when all video data is transmitted compressed at a fixed compression rate, and makes it possible to measure biometric information with high reliability.

[0069] Furthermore, when a patient 1, who is a person, is identified as the monitoring target, the detection unit 15 acquires video after the control unit 14 has controlled the shooting or the supply of video, and detects the status of the monitoring target person in the video (S14). Then, a risk level is calculated based on conditions preset for the patient 1 and the status of the monitoring target person, the patient 1 (S15). Then, notification information is output to a predetermined notification destination based on the calculated risk level (S16). As described above, since highly reliable biological information can be measured, it is possible to provide appropriate treatment for the person and reduce the workload of the medical staff 2.

[0070] As described above, according to the embodiment, among a plurality of camera devices that capture surveillance images, at least one camera device 20 is selected based on predetermined conditions from the camera devices 20 that have captured surveillance footage showing the same surveillance target, and the camera device 20 is controlled to capture images and transmit images, thereby making it possible to resolve resource shortages.

[0071] Each functional block shown in the drawings that performs various processes can be configured in hardware with a processor, memory, and other circuits. The processes described above can also be implemented by having a processor execute a program. Therefore, these functional blocks can be implemented in various forms, such as hardware only, software only, or a combination of both, and are not limited to any one of these.

[0072] The above-described program can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic storage media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical storage media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may also be supplied to a computer by various types of transitory computer-readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer-readable media can supply the program to a computer via a wired communication path such as an electric wire or optical fiber, or via a wireless communication path.

[0073] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

[0074] Furthermore, some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0075] (Appendix 1) a plurality of imaging means installed at a plurality of locations within a monitoring target area; an identification means for identifying a surveillance target from the images captured by the plurality of imaging means; a determination means for determining whether or not the same surveillance target is captured in images captured by a plurality of image capture means during the same time period among the identified surveillance targets; a control means for selecting at least one of the image capturing means that have captured images of the same surveillance target based on a predetermined condition, and for controlling the image capturing by the selected image capturing means or for controlling the supply of the images; A monitoring system comprising: (Appendix 2) the control means controls the selected image capture means to change the image quality of the image when the image includes a surveillance target; 1. A monitoring system as described in Appendix 1. (Appendix 3) The identification means identifies a type of person included in the video; The surveillance system according to claim 2, wherein the control means changes the image quality depending on the type of person in the image captured by the selected imaging means. (Appendix 4) When transmitting the video captured by the imaging means, the control means controls the compression rate of a target area in the video, which includes the surveillance target, to be lower than that of an area other than the target area, thereby making the target area have a higher image quality than the area other than the target area. 4. The monitoring system of claim 2 or 3. (Appendix 5) the control means selects at least one of the plurality of image capturing means in accordance with the surrounding environment of the monitoring target within the image capturing range captured by each of the plurality of image capturing means; Control is performed so that only the images captured by the selected imaging means are transmitted. A monitoring system according to any one of appendices 1 to 4. (Appendix 6) When the image does not include a surveillance target and / or there is no movement in the image, the control means controls to stop supplying the image from the image capturing means that captured the image. A monitoring system according to any one of appendices 1 to 5. (Appendix 7) When a plurality of image capturing means are selected, the control means controls the selected image capturing means to perform different processes. A monitoring system according to any one of appendices 1 to 6. (Appendix 8) When the identification means identifies a person to be monitored as a person to be monitored, a detection means for acquiring an image after the control means has controlled the photographing or the supply of the image, and detecting the state of the person to be monitored in the image; a calculation means for calculating a risk level based on a condition preset according to the person to be monitored and the situation of the person to be monitored; a notification means for outputting notification information to a predetermined notification destination based on the risk level; Further provided with A monitoring system according to any one of appendices 1 to 7. (Appendix 9) The surveillance system according to claim 8, wherein the control means changes the image quality depending on the situation of the person being monitored in the image captured by the selected imaging means. (Appendix 10) the detection means detects whether the monitored person is with another person; the calculation means reduces the risk level when the person to be monitored is with another person; 10. The monitoring system of claim 8 or 9. (Appendix 11) the calculation means changes the extent of reduction of the risk level depending on the type of the other person; 11. The monitoring system of claim 10. (Appendix 12) Identifying a monitoring target from images captured by a plurality of imaging means installed at a plurality of locations within a monitoring target area; Among the identified surveillance targets, it is determined whether the same surveillance target appears in images captured by a plurality of image capturing means during the same time period; selecting at least one of the image capturing means that have captured images of the same surveillance target based on a predetermined condition, and controlling the image capturing by the selected image capturing means or controlling the supply of the image; Monitoring method. (Appendix 13) If the image includes a monitored object, controlling the selected image capture means to change the image quality of the image. 12. The monitoring method described in Appendix 12. (Appendix 14) Identifying the type of person included in the video; 13. A surveillance method as described in Appendix 12, wherein the image quality is changed depending on the type of person in the image of the selected imaging means. (Appendix 15) When transmitting the video captured by the imaging means, a compression rate of a target area in the video including the surveillance target is controlled to be lower than that of an area other than the target area, thereby making the image quality of the target area higher than that of the area other than the target area. 14. The monitoring method of claim 12 or 13. (Appendix 16) selecting at least one of the plurality of photographing means in accordance with the surrounding environment of the monitoring target within a photographing range photographed by each of the plurality of photographing means; Control is performed so that only the images captured by the selected imaging means are transmitted. A monitoring method according to any one of appendices 12 to 15. (Appendix 17) If the image does not include the monitored object and / or there is no movement in the image, control is performed to stop supplying the image from the image capturing means that captured the image. A monitoring method according to any one of appendices 12 to 16. (Appendix 18) When a plurality of image capturing means are selected, control is performed so that different processes are performed on the selected plurality of image capturing means. A monitoring method according to any one of appendices 12 to 17. (Appendix 19) If a person is identified as a person to be monitored, Acquire the video after the shooting or supply of the video has been controlled, and detect the status of the person being monitored in the video; Calculating a risk level based on a condition preset according to the person to be monitored and the situation of the person to be monitored; outputting notification information to a predetermined notification destination based on the risk level; A monitoring method according to any one of appendices 12 to 18. (Appendix 20) A monitoring method as described in Appendix 19, in which the image quality is changed depending on the situation of the person being monitored in the image of the selected imaging means. (Appendix 21) Detecting whether the monitored person is with another person; reducing the risk level if the monitored person is with other people; 21. A monitoring method according to claim 19 or 20. (Appendix 22) changing the extent of reduction of the risk level depending on the type of the other person; 2. The monitoring method described in Appendix 21. (Appendix 23) Identifying a monitoring target from images captured by a plurality of imaging means installed at a plurality of locations within a monitoring target area; Among the identified surveillance targets, it is determined whether the same surveillance target appears in images captured by a plurality of image capturing means during the same time period; selecting at least one of the image capturing means that have captured images of the same surveillance target based on a predetermined condition, and controlling the image capturing by the selected image capturing means or controlling the supply of the image; A non-transitory computer-readable medium that stores a program that causes a computer to execute a process. [Explanation of symbols]

[0076] 1 patient 2. Medical staff 3. Hospital 10 Event detection device 11 Acquisition Department 12 Specific part 13 Judgment section 14 Control Unit 15 Detector 16 Calculation section 17 Notification Department 18 Memory section 20 Imaging equipment 30 Notification device 31 Hallway 32A, 32B Private Rooms 33 Shared Space 34 Surveillance room 35 stairs 36 Free Space 100 Surveillance System 101 Photography Method 102 Specific means 103 Judgment means 104 Control Means N Network

Claims

1. a plurality of imaging means installed at a plurality of locations within a monitoring target area; an identification means for identifying a surveillance target from the images captured by the plurality of imaging means; a determination means for determining whether or not the same surveillance target is captured in images captured by a plurality of image capture means during the same time period among the identified surveillance targets; a control means for selecting at least one of the image capturing means that have captured images of the same surveillance target based on a predetermined condition, and for controlling the image capturing by the selected image capturing means or for controlling the supply of the image; a detection means for acquiring an image after the control means has controlled the shooting or the supply of the image, when the identification means has identified the person to be monitored as the monitoring target, and detecting the status of the person to be monitored in the image; a calculation means for calculating a risk level based on a condition preset according to the person to be monitored and the situation of the person to be monitored; a notification means for outputting notification information to a predetermined notification destination based on the risk level; Equipped with the detection means detects whether the monitored person is with another person; When the person to be monitored is with another person, the calculation means changes the extent of reduction of the risk level according to the type of the other person; the monitored area includes a plurality of rooms and a corridor for accessing the rooms; the photographing means includes photographing means disposed in each of the plurality of rooms, and photographing means disposed in the corridor and taking photographs along the corridor; The control means controls the selected photographing means to change the image quality of the image when the image includes a monitored object, and when the selected photographing means is located in the corridor, controls the selected photographing means to make the image quality of the image higher than that of an image captured by a photographing means located in a place other than the corridor. Surveillance system.

2. The identification means identifies a type of person included in the video; the control means changes the image quality according to the type of person in the image captured by the selected imaging means; The monitoring system of claim 1 .

3. When transmitting the video captured by the imaging means, the control means controls the compression rate of a target area in the video, which includes the surveillance target, to be lower than that of an area other than the target area, thereby making the target area have a higher image quality than the area other than the target area. The monitoring system of claim 2 .

4. the control means changes the image quality according to the state of the person being monitored in the image captured by the selected image capture means; The monitoring system of claim 1 .

5. the control means selects at least one of the plurality of image capturing means in accordance with the surrounding environment of the monitoring target within the image capturing range captured by each of the plurality of image capturing means; Control is performed so that only the images captured by the selected imaging means are transmitted. The monitoring system according to any one of claims 1 to 4.

6. When the image does not include a surveillance target and / or there is no movement in the image, the control means controls to stop supplying the image from the image capturing means that captured the image. The monitoring system according to any one of claims 1 to 5.

7. When a plurality of image capturing means are selected, the control means controls the selected image capturing means to perform different processes. The monitoring system according to any one of claims 1 to 6.

8. Identifying a monitoring target from images captured by a plurality of imaging means installed at a plurality of locations within a monitoring target area; Among the identified surveillance targets, it is determined whether the same surveillance target appears in images captured by a plurality of image capturing means during the same time period; selecting at least one of the image capturing means that have captured images of the same surveillance target based on a predetermined condition, and controlling the image capturing by the selected image capturing means or controlling the supply of the image; When a person to be monitored is identified as the monitoring target, an image is acquired after the image capture is controlled or the image supply is controlled, and the situation of the person to be monitored in the image is detected; Calculating a risk level based on a condition preset according to the person to be monitored and the situation of the person to be monitored; outputting notification information to a predetermined notification destination based on the risk level; Detecting whether the monitored person is with another person; When the person to be monitored is with another person, the extent of reduction of the risk level is changed according to the type of the other person; the monitored area includes a plurality of rooms and a corridor for accessing the rooms; the photographing means includes photographing means disposed in each of the plurality of rooms, and photographing means disposed in the corridor and taking photographs along the corridor; When the image includes a monitored object, the selected image capturing means is controlled to change the image quality of the image, and when the selected image capturing means is located in the corridor, the selected image capturing means is controlled to change the image quality of the image to be higher than that of an image captured by an image capturing means located in a place other than the corridor. Monitoring method.

9. The surveillance is performed from images taken by multiple imaging devices installed at multiple locations within the surveillance area. Identify the visual target, Among the identified surveillance targets, the same image was captured in multiple images taken by multiple means during the same time period. Determine whether the surveillance target is visible or not, A small number of images are captured based on predetermined conditions from the means of capturing images of the same surveillance target. At least one photographing means is selected, and photographing by the selected photographing means is controlled or the image is Controlling supply, If a person to be monitored is identified as the subject of said monitoring, the recording will be controlled or the supply of the image will be restricted. Acquire an image after the control, and detect the situation of the person to be monitored in the image; Based on the conditions preset according to the person to be monitored and the situation of the person to be monitored, and calculate the risk. outputting notification information to a predetermined notification destination based on the risk level; Detecting whether the monitored person is with another person; If the person to be monitored is with other people, the risk is determined according to the type of the other people. Change the degree of decrease, the monitored area includes a plurality of rooms and a corridor for accessing the rooms; the photographing means includes photographing means arranged in each of the plurality of rooms, and photographing means arranged in the corridor and taking photographs along the corridor; When the image includes a monitored object, the selected image capturing means is controlled to change the image quality of the image, and when the selected image capturing means is located in the corridor, the selected image capturing means is controlled to change the image quality of the image to be higher than that of an image captured by an image capturing means located in a place other than the corridor. A program that causes a computer to perform a process.

Images