JP2020009378A - Monitoring device - Google Patents

Abstract

To provide a monitoring device that detects an operation for each of monitored persons.SOLUTION: A monitoring device 1 that monitors a specific object person comprises: an operation learning unit 411 that extracts a feature value of an operation of an object person on the basis of image data for the operation of the object person, and generates a learning model for the operation of the object person; an operation model storage unit 221 that stores the learning model for the operation generated by the operation leaning unit 411; and an operation determination unit 221 that determines the operation of the object person on the basis of the leaning model for the operation stored by the operation determination unit 211.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a monitoring device that monitors a specific target person in a home, a facility, or the like.

  2. Description of the Related Art Conventionally, there has been known an apparatus that determines whether or not an operation of a target person is abnormal based on image data captured by a surveillance camera and reports the abnormality (for example, see Patent Document 1). In the device of Patent Literature 1, a series of standard operations when using the automatic transaction device is learned, and when the target person's operation deviates from the standard operation, it is determined to be abnormal.

JP 2018-10332 A

  In the device described in Patent Document 1, in order to determine whether or not the motion of the subject is abnormal by comparing the motion of the subject with a standard motion, for the characteristic motion corresponding to the physical condition of the subject, It cannot be determined accurately. However, particularly when monitoring the posture for 24 hours for the purpose of nursing care, it is desired to determine an abnormal operation (unusual) for each subject having a different physical condition or the like.

  One embodiment of the present invention is a monitoring device that monitors a specific target person, an operation learning unit that learns the operation of the target person, an operation storage unit that stores the operation learned by the operation learning unit, and an operation storage An operation determining unit that determines an operation of the subject person based on the operation stored by the unit.

  According to the present invention, an operation can be detected for each monitoring target person.

The figure which shows schematically an example of installation of the monitoring apparatus which concerns on embodiment of this invention. FIG. 1 is a perspective view of a monitoring device according to an embodiment of the present invention. FIG. 2 is a cross-sectional view of the monitoring device according to the embodiment of the present invention. FIG. 1 is a block diagram schematically showing a configuration of a monitoring device according to an embodiment of the present invention. 5 is a flowchart illustrating an example of an operation determination process based on image data, which is executed by the controller group in FIG. 4. 5 is a flowchart illustrating an example of a posture determination process based on image data, which is executed by the controller group in FIG. 4. 5 is a flowchart illustrating an example of an abnormality determination process based on posture data, which is executed by the controller group in FIG. 4. 5 is a flowchart illustrating an example of an abnormality determination process based on measured values, which is executed by the controller group in FIG. 4. 5 is a flowchart illustrating an example of a learning process of an operation performed by the controller group in FIG. 4. The figure for explaining operation of the monitoring device concerning the embodiment of the present invention.

  Hereinafter, an embodiment of the present invention will be described with reference to FIGS. The monitoring device according to the embodiment of the present invention is installed in a living room of a nursing care facility, for example, to constantly monitor a specific target person, and a normal operation is performed according to a physical condition such as a bent back. The determination is made by learning various operations including an abnormal operation to be reported for a different target person.

  FIG. 1 is a diagram schematically illustrating an example of installation of monitoring devices 1A and 1B (hereinafter, collectively referred to as 1) according to an embodiment of the present invention. The monitoring device 1 is configured by a monitoring camera or the like, and is installed in a place where the watching target person spends alone, such as a living room of a nursing care facility or a single elderly person's home, and monitors the watching of the target person. The image data and the like output by the monitoring device 1 are transmitted to a terminal installed at a nurse station of a nursing facility, a mobile terminal of a family member of the watching target, and the like. FIG. 1 is an example of a case where the monitoring device 1 is installed in a living room of a care facility. The living room is, for example, as shown from the left side in FIG. 1, the entrance of the living room, a partitioned area AR1 (such as a toilet), an area AR2 where the subject walks and moves (such as a wash area), and the subject is mainly in a prone position. Is constituted by an area AR3 (an area where a bed or the like is placed).

  As shown in FIG. 1, the monitoring device 1A is installed on a wall or the like opposite to the entrance. The monitoring device 1A monitors the subject's entry into and exit from the room, monitors the operation of the subject in the area AR2 such as walking, and monitors the state of the subject in bed in the area AR3 from above. . The monitoring device 1B is installed at a position where the target person does not care, such as the ceiling of the partitioned area AR1. Although FIG. 1 shows an example in which the monitoring devices 1A and 1B are installed, the number of monitoring devices 1 may be one or three or more depending on the size and layout of the living room.

  FIG. 2 is a perspective view of the monitoring device 1. The monitoring device 1 includes a housing 11 having a substantially light bulb shape. The housing 11 has, at one end, a conductive base 12 in which a screw portion 12a is formed. The screw portion 12 a is screwed into a socket at the tip of a stand that can be attached to a wall surface, for example, and power is supplied to the main body via the base 12. You may make it screw into the socket of the duct rail attached to the wall surface. A cover 13 made of a light-transmitting resin material is attached to the other end of the housing 11. By adopting such a light bulb shape, the monitoring device 1 can be easily installed like a commercially available light bulb, and can reduce the consciousness that a general surveillance camera gives to a target person that the watch is being watched. it can.

  FIG. 3 is a sectional view of the monitoring device 1. As shown in the drawing, a camera 14 for capturing an image of a watching target, a plurality of (for example, eight) LEDs 15 arranged at equal circumferential intervals around the camera 14, Three controllers 20, 30, 40 (controller group 50) arranged at equal intervals in the circumferential direction are arranged, and electric power is supplied to these from outside via the base 12 and a cable (not shown). The camera 14 is connected to each of the controllers 20, 30, and 40 via a cable (not shown) or the like. The housing 11 is provided with a slit 11s for discharging heat (FIG. 1). The controller group 50 includes a notification unit such as a nurse call base unit (nurse call board) installed in a nurse station, and a management terminal for creating and managing a care record (for example, a notebook computer for a care staff). Are electrically connected by wire or wirelessly. Although not shown, each of the controllers 20, 30, and 40 is provided with an external input port such as a USB (Universal Serial Bus).

  The camera 14 constantly captures an image of the watching target, and transmits the image data to the controller group 50. As the camera 14, for example, a normal RGB camera having an optical zoom function of about 5 to 10 times is used. As shown in FIG. 3, the camera 14 is disposed at the center in the housing 11 so that the lens unit is parallel to the surface of the cover 13. The camera 14 has a lens portion fixed to the cover 13 and a main body portion movable in the length direction of the housing 11 indicated by an arrow along a rail 11r provided on the housing 11, thereby appropriately adjusting the magnification. can do.

  The LED 15 is configured by, for example, a high-brightness white LED. The LED 15 is arranged so that the irradiation surface is parallel to the surface of the cover 13. In the case of monitoring while the living room is dark, such as when the subject is sleeping, an infrared camera is used as the camera 14 instead of a normal RGB camera, and an infrared LED is used as the LED 15 instead of a high-brightness white LED. May be. An RGB camera from which an infrared cut filter has been removed may be used as the camera 14, and a high-brightness white LED and an infrared LED may be alternately arranged around the camera 14 at equal intervals in the circumferential direction.

  Here, problems of a conventional watching system such as a nurse call or a surveillance camera will be described. In the conventional nurse call monitoring system, when the target person feels something unusual, the target person himself / herself operates the nurse call slave unit in the living room to notify the care staff and the like. For this reason, when the target person loses consciousness or cannot move, the notification cannot be performed. In a conventional monitoring system using a surveillance camera, a care staff or the like looks at an image of the surveillance camera and determines whether or not there is any abnormality in the target person. For this reason, when one care staff watches over multiple subjects, or when a single family member constantly monitors a single person, delay in noticing the target person's abnormality due to oversight or oversight is delayed. There is. For example, if the subject loses consciousness or becomes unable to move due to a fall or depression, the subject is left unattended until the care staff or the like notices.

  Therefore, in the embodiment of the present invention, various operations of the watching target person including an abnormal operation (abnormal operation) when an abnormality occurs are learned as an operation model, and the watching target captured by the monitoring camera (camera 14) is learned. By determining whether or not the user's operation is an abnormal operation, an abnormality of the target person is automatically detected.

  The operation of the watching target who needs care, such as the elderly, may be different from the standard operation of a standard adult depending on the state of the body such as bending. In such a case, it is difficult to learn the standard operation in advance, and to detect an abnormality as compared with the operation of the watching target imaged by the monitoring camera. For example, there is a possibility that an operation that has no problem for the subject, such as an operation of moving while the hip is bent, may be determined as an abnormal operation.

  In order to prevent such erroneous detection, the monitoring device 1 learns an operation model for each watching target person, determines an operation for each target person, and detects an abnormality. Specifically, based on the nursing record of the watching target and the like, various operations including an abnormal operation to be notified assumed for each target person are performed by the care staff in advance and learned as an operation model. For example, in the case of the monitoring device 1 that monitors a target person whose waist is bent, an operation of moving while the waist is bent or an operation of falling over are learned as an operation model. In addition, by further learning the actual motion of the target person as a motion model during the monitoring of the target person, the motion model is updated to a more appropriate one according to the actual condition of the target person.

  The configuration of the monitoring device 1 capable of realizing the above-described learning of the operation model and the determination of the operation and the abnormality will be described.

  FIG. 4 is a block diagram schematically illustrating a configuration of the monitoring device 1. The monitoring device 1 includes a camera 14, a controller group 50 electrically connected to the camera 14, a notification unit 16 electrically connected to the controller group 50 by wire or wirelessly, and a management terminal 17. The controller group 50 includes the controllers 20, 30, and 40 that are configured to be able to communicate with each other.

  The notification unit 16 is, for example, a nurse call master device installed at a nurse station of a nursing facility, and notifies a nursing staff or the like according to a command from the controller group 50. The notification unit 16 is not limited to the one installed in the same premises as the monitoring device 1 and may be a mobile terminal such as a smartphone owned by the family of the watching target. When the monitoring device 1 is installed in the residence of the subject of solitary living, the notification unit 16 may be a terminal of a medical institution or the like. The notification unit 16 may be formed by combining the above terminals.

  The management terminal 17 is, for example, a notebook computer used by a care staff at a nursing facility for creating and managing a care record. The management terminal 17 includes text data such as an action schedule of the subject included in the care record and a determination result of the action of the subject. Is transmitted to and received from the controller group 50. The management terminal 17 is not limited to the one installed in the same premises as the monitoring device 1 and may be a mobile terminal such as a smartphone owned by the family of the watching target. When the monitoring device 1 is applied for watching a target person who is hospitalized in a medical facility, the management terminal 17 may be used as a terminal of a medical institution to create a care record. Further, the management terminal 17 may be a combination of a plurality of terminals.

  The controller 20 includes a computer having a GPU (Graphics Processing Unit: image processing device) 21, a memory 22 such as a ROM and a RAM, and other peripheral circuits.

  The GPU 21 has, as a functional configuration, an operation determination unit 211 that determines a current operation based on the movement model of the watching target person, a posture determination unit 212 that determines the current posture based on the posture model of the target person, An output unit 213 that outputs the posture determined by the determination unit 212 to the controller 30. The motion determining unit 211 and the posture determining unit 212 may be configured by different GPUs.

  The memory 22 includes, as functional components, an operation model storage unit 221 that stores a model of a target person's operation including normal operations such as movement, eating, and sleep, and abnormal operations such as falling and aspiration, And a posture model storage unit 222 for storing a model of the posture of the subject such as a sitting position or a lying position.

  The controller 30 includes a computer having a CPU (Central Processing Unit) 31, a memory 32 such as a ROM and a RAM, and other peripheral circuits.

  The CPU 31 has, as a functional configuration, an abnormal posture that determines whether or not the change in the posture of the subject output by the output unit 213 is abnormal based on the determination condition of the abnormal posture change and specifies the type of the abnormal change. A change determination unit 311, a vital measurement unit 312 that measures vital signs such as a heart rate, a blood pressure, a respiratory rate, a blood oxygen saturation (SPO2), and a body temperature of the subject based on image data from the camera 14, and a vital An abnormality determination unit 313 that determines whether an abnormality has occurred in the subject based on a current measurement value or the like by the measurement unit 312 and specifies a type of the abnormality, and determinations by the abnormal posture change determination unit 311 and the abnormality determination unit 313. An abnormal information output unit 314 that outputs abnormal information to the notification unit 16 and the management terminal 17 according to the result, an abnormal posture change determination unit 311 and an abnormality determination unit 3 A learning time determining unit 315 for determining times tps to tpe and tvs to tve to be learned as abnormal motions of the subject person in the image data from the camera 14 in accordance with the type of the abnormality specified by 3; The image data at the times tps to tpe and tvs to tve determined by 315 is extracted, and the learning data of the subject's motion (abnormal motion) is displayed together with the type of change specified by the abnormal posture change determination unit 311 and the change determination unit 313. And a learning data output unit 316 for outputting to the controller 40.

  The vital measurement unit 312 measures various vital signs of the subject based on image data from the camera 14. For example, the heart rate can be measured based on a change in the intensity of the light received by the camera 14. Of the three component lights received by the RGB camera, the blue (B) component light having a short wavelength is reflected on the skin surface of the subject, and the green (G) component light having a longer wavelength is the presence of capillaries. Penetrates into the dermis layer, and is absorbed by hemoglobin contained in red blood cells in the blood. That is, the received light intensity of the blue (B) component varies according to the brightness (environmental light) of the living room, and the received light intensity of the green (G) component varies according to the pulsation of the subject. Therefore, the pulsation cycle (heart rate) can be measured based on the fluctuation of the received light intensity without being affected by the ambient light.

  The memory 32 has, as a functional configuration, an action schedule storage unit 321 that stores an action schedule of the target person such as a meal or entrance / exit that is input in advance based on the nursing record of the target person, and when the target person has an abnormality. An abnormal posture change determination condition storage unit 322 that stores the determination condition of an abnormal posture change (posture change) for each type of abnormality, and an abnormality determination condition based on a measured value of a subject's vital sign and the like. It has an abnormal change determination condition storage unit 323 that stores the measured values of vital signs of the subject measured by the vital measurement unit 312, and a vital measurement value storage unit 324 that stores the vital sign measurement values measured by the vital measurement unit 312. As an example of a determination condition of an abnormal posture change, for example, when determining a fall, a condition that the posture of the subject changes from a standing position to a lying position within a predetermined time (for example, within 3 seconds) is set as a condition. I do. In addition, when aspiration is determined, it is assumed that it is estimated that the user is eating based on the action schedule, and that the heart rate exceeds a predetermined value (for example, 100 times / minute).

  The controller 40 includes a GPU (Graphics Processing Unit: image processing device) 41, a memory 42 such as a ROM and a RAM, and a computer having other peripheral circuits.

  The GPU 41 has, as a functional configuration, an operation learning unit 411 that extracts and learns a feature amount corresponding to a feature of an operation based on learning data of the operation of the object person and generates a learning model of the operation of the object person. A posture learning unit 412 that extracts and learns a feature amount of the posture based on the learning data of the posture of the subject and generates a learning model of the posture of the subject, An output unit 413 for outputting to the model storage unit 221 and the posture model storage unit 222;

  The memory 42 has, as a functional configuration, an operation learning data storage unit 421 that stores learning data of the motion of the subject output by the learning data output unit 316 for each type of motion, and a subject that has been captured by the camera 14 in advance. And a posture learning data storage unit 422 for storing the posture learning data (image data) for each posture type. The motion learning data storage unit 421 is configured to store, in addition to the learning data output from the learning data output unit 316, learning data of the motion of the subject captured in advance by the camera 14 for each type of motion. Is also good.

  The motion learning unit 411 and the posture learning unit 412 perform supervised learning of image data by a known machine learning program suitable for learning image data, for example, Deep Learning, which is a machine learning method using a multilayer neural network. . That is, the operation learning unit 411 learns image data (image data at times tps to tpe, tvs to tve, etc.) stored and accumulated for each type of operation by the operation learning data storage unit 421, and Generate an operation model. The posture learning unit 412 also learns image data stored and accumulated for each posture type by the posture learning data storage unit 422, and generates a posture model for each posture type. The motion model and the posture model after learning are output to the controller 20 by the output unit 413, and are stored and updated in the motion model storage unit 221 and the posture model storage unit 222, respectively.

  The operation determination unit 211 of the controller 20 determines the current operation of the target person based on the latest operation model of the target person stored in the operation model storage unit 221 and determines whether the current operation is an abnormal operation. judge. Further, the posture determination unit 212 determines the current posture of the subject based on the latest posture model of the subject stored in the posture model storage unit 222.

  More specifically, the motion determining unit 211 extracts the feature amount of the current motion of the subject in the image data from the camera 14 and compares the feature amount with the motion model of the subject to determine the current motion of the subject. And determine whether or not the operation is abnormal. In addition, the posture determination unit 212 extracts the feature amount of the current posture of the target person in the image data, compares the feature amount of the target posture model stored in the posture model storage unit 222 with the characteristic amount of the target person, and Is determined. The motion or posture having the highest degree of correlation with the learning model (abnormal motion model or posture model) is determined as the current motion or posture. The motion determining unit 211 can also calculate the absolute amount of the motion (the amount of activity) based on the current motion of the subject in the image data. By recording the change in the activity amount together with the vital signs, it is possible to determine not only accidents and diseases but also initial symptoms of dementia.

  FIG. 5A is a flowchart illustrating an example of an operation determination process based on image data, which is executed by the GPUs 21 and 41 and the CPU 31 of the controller group 50 in accordance with programs stored in the memories 22, 32 and 42 of the controller group 50 in advance. The processing shown in this flowchart is started, for example, when the power of the monitoring apparatus 1 is turned on, and is repeated at a predetermined cycle. First, in step S1, the current image data captured by the camera 14 is read by the processing of the operation determining unit 211. Next, in step S2, the motion of the subject in the current image data is determined based on the motion model stored in the motion model storage unit 221, and in step S3, it is determined whether the motion is an abnormal motion. If affirmative in step S3, the process proceeds to step S4, and if negative, the process ends. In step S <b> 4, by the processing in the incident information output unit 314, the incident information is output to the notification unit 16 and the management terminal 17.

  FIG. 5B is a flowchart illustrating an example of a posture determination process performed by the controller group 50 based on image data. First, in step S10, the current image data captured by the camera 14 is read by the processing of the posture determination unit 212. Next, in step S11, the posture of the subject in the current image data is determined based on the posture model stored in the posture model storage unit 222. Next, in step S <b> 12, data of the current posture of the subject determined by the posture determination unit 212 is output to the abnormal posture change determination unit 311 by the processing of the output unit 213.

  FIG. 6A is a flowchart illustrating an example of an anomaly determination process performed by the controller group 50 based on the posture data. First, in step S20, the data of the current posture of the subject output by the output unit 213 is read by the processing of the abnormal posture change determination unit 311. Next, in step S21, the posture change between the posture read before the previous time and the posture read this time is abnormal based on the abnormal posture change determination condition stored in the abnormal posture change determination condition storage unit 322. Is determined, and the type of the incident is specified. If affirmative in step S21, the process proceeds to step S22, and if negative, the process ends. In step S22, the abnormal information output unit 314 outputs the abnormal information to the notification unit 16 and the management terminal 17 by the processing.

  Next, in step S23, the learning time determination unit 315 determines the times tps to tpe to be learned as the abnormal motion of the subject based on the type of the abnormalities specified by the abnormal posture change determination unit 311. Next, in step S24, the learning data output unit 316 performs processing to extract the image data from the time tps to tpe determined by the learning time determination unit 315 from the image data from the camera 14 and learn the operation of the subject. Output as data.

  FIG. 6B is a flowchart illustrating an example of the abnormal change determination process based on the measurement value, which is executed by the controller group 50. First, in step S30, the current image data captured by the camera 14 is read by the processing in the vital measurement unit 312. Next, in step S31, the current vital sign of the subject is measured based on the image data. Next, in step S32, the values measured by the vital measurement unit 312, the target person's action schedule stored in the action schedule storage unit 321 and the change determination condition storage unit 323 are stored by the processing in the anomaly determination unit 313. Based on the determination condition of the abnormality, whether or not the target person has an abnormality is determined, and the type of the abnormality is specified. If affirmative in step S32, the process proceeds to step S33, and if negative, the process ends. In step S33, the abnormal information output unit 314 outputs abnormal information to the notification unit 16 and the management terminal 17 by the processing.

  Next, in step S34, the time tvs to tve to be learned as the abnormal motion of the subject is determined by the process of the learning time determination unit 315 based on the type of the abnormality identified by the abnormality determination unit 313. Next, in step S35, the learning data output unit 316 extracts the image data from the time tvs to tve determined by the learning time determination unit 315 from the image data from the camera 14 to learn the operation of the subject. Output as data.

  FIG. 7 is a flowchart illustrating an example of an operation learning process performed by the controller group 50. The processing shown in the flowchart is different from the processing in the flowcharts of FIGS. 5A to 6B and is started at any time in response to an instruction input by a user managing the monitoring device 1 through an appropriate switch or the like. First, in step S40, the learning data of the subject's motion output from the learning data output unit 316 and stored in the motion learning data storage unit 421 is read by the processing of the motion learning unit 411. Next, in step S41, the motion of the subject is learned, and in step S42, a motion model is generated. Next, in step S43, the operation model generated by the operation learning unit 411 is output to the operation model storage unit 221 of the controller 20 by the processing in the output unit 413.

  The posture learning process is executed in the same manner as the operation learning process. That is, the learning data of the posture of the subject stored in the posture learning data storage unit 422 by the processing of the posture learning unit 412 is read (S40), the posture of the subject is learned (S41), and a posture model is generated (S41). (S42), the posture model is output to the posture model storage unit 222 of the controller 20 by the processing in the output unit 413 (S43).

  The main operation of the monitoring device 1 according to the present embodiment will be described more specifically. FIG. 8 is a diagram for explaining the operation of the monitoring device 1. In FIG. 8, from the top of the figure, the posture of the subject determined by the posture determination unit 212, the behavior plan of the subject stored in the behavior plan storage unit 321, and the abnormal posture change by the abnormal posture change determination unit 311. The determination result, the heart rate as an example of the vital signs measured by the vital measurement unit 312, and the determination result of the abnormal change by the abnormal change determination unit 313 are shown in time series.

  For example, it is assumed that the monitoring apparatus 1 monitors an elderly subject with a bent waist entering a care facility. When the subject alone in the living room of the nursing facility falls from time tp to tpe, the monitoring device 1 installed in the living room determines that the posture change is abnormal and specifies that the fall has occurred ( S10 to S12, S20 to S21). The change in the subject is notified to the care staff via a nurse call board (notification unit 16) installed at the nurse station (S22). For this reason, even when the target person cannot use the nurse call by himself because he / she cannot move due to the fall, the care staff can be notified of the abnormal situation. In addition, even when the care staff is looking away from the image of the monitoring camera 14, the monitoring device 1 can notify the target person of anomaly by the notification. Further, the information on the change of the subject is also transmitted to the notebook computer (management terminal 17) of the care staff (S22). As a result, information such as the type and time of occurrence of the abnormality that has occurred in the subject is automatically recorded in the format of the nursing care record and the like, and the administrative burden on the nursing care staff can be reduced.

  Image data at times tps to tpe at which it is estimated that the subject has fallen is extracted and accumulated as learning data of the fall operation (S23, S24). The user of the monitoring device 1 connects the monitoring device 1 with the cover 13 removed from the socket to a user terminal such as a personal computer via a USB cable or the like, and learns the operation stored and accumulated in the memory 42 of the controller 40. It is possible to check the data storage state and the like. When the user confirms the accumulation of the amount of learning data necessary for learning the falling motion and instructs to start learning the falling motion, learning by the controller 40 is started, a learning model is generated, and the learning of the controller 20 is performed. The model is updated (S40 to S43). The learning model used to determine the motion is updated from the motion model generated based on the demonstration of the care staff to the learning model of the motion of the actual subject, and the motion of the subject is determined. The accuracy is improved.

  If the subject's aspiration occurs during a meal and the subject's heart rate rises above a predetermined value (for example, 100 times / minute) at time ta due to choking, the monitoring device 1 causes aspiration (abnormality). It is determined that this has occurred (S31 to S33). Image data at times tvs to tve, which is estimated to have caused aspiration and aspiration, are extracted and accumulated as learning data of the aspiration operation, and are learned in accordance with the user's instruction (S34, S35, S40). To S43). The time extracted as the learning data of the operation is determined according to the type of the incident, so that it is possible to learn including the operation before the occurrence of the incident. For example, learning can be performed including an inappropriate posture at the time of eating and drinking, which is an operation before aspiration occurs.

  If the subject at the time of eating or drinking takes an inappropriate posture that leads to aspiration after the aspiration operation model is updated, the monitoring device 1 determines that the aspiration is an aspiration operation (operation that leads to aspiration), An abnormality of the subject is notified (S1 to S3). The nursing staff who receives the notification can rush to the living room of the target person, or pay attention to the posture of the target person at the time of eating and drinking by nurse call or the like. Even when the target person eats and drinks alone, aspiration can be prevented from occurring. In addition, since a posture leading to aspiration is learned for each subject, the present invention can be applied to a subject whose proper posture at the time of eating and drinking is different from a standard adult posture due to bending of the spine or hemiplegia. In addition, when the care staff assists the target person at the time of eating and drinking, according to the state of the target person, while watching the situation, determine the posture leading to aspiration, and the posture at that time as the operation of aspiration You can also learn. That is, an abnormal operation can be detected for each watching target person.

  In the monitoring device 1, the action schedule of the target person and the measured values of the vital signs are stored in the memory 32 of the controller 30 in time series. When the monitoring target visits a medical institution, these data can be presented to a doctor. By confirming the action schedule such as a meal or waking up and the change in vital signs, it can be used for medical treatment and diagnosis of the target person.

According to the present embodiment, the following effects can be obtained.
(1) The monitoring device 1 that monitors a specific watching target person extracts a feature amount of the watching target person's motion based on the image data of the watching target person's motion, and generates a learning model of the watching target person's motion. Based on the action learning unit 411, the action model storage unit 221 that stores the action learning model generated by the action learning unit 411, and the action learning model stored by the action model storage unit 221, the motion of the watching target person is performed. (See FIG. 4). By learning the motion for each subject, it is possible to determine a motion including an abnormal motion (unusual) for each subject having a different physical condition or the like.

(2) The monitoring device 1 determines whether or not an abnormality has occurred in the watching target based on the vital measurement unit 312 that measures the vital sign of the watching target and the vital sign measured by the vital measuring unit 312. It further includes an abnormality determination unit 313 (FIG. 4). The motion learning unit 411 extracts a feature amount of the motion of the watching target person based on the image data of the motion of the watching target person when the abnormality determination unit 313 determines that the watching target has an abnormality, and extracts the characteristic amount of the watching target person's motion. Generate a learning model for the abnormal behavior of. Thereby, based on the change in vital signs, it is possible to automatically extract the learning data of the operation when the abnormality occurs and the operation before and after the occurrence of the abnormality.

(3) The monitoring device 1 further includes a management terminal 17 that acquires the action schedule of the watching target (FIG. 4). The abnormality determination unit 313 further determines whether an abnormality has occurred in the watching target person based on the action schedule acquired by the management terminal 17. By confirming the behavior schedule of the target person and the change of the vital sign in comparison, it can be used for the medical treatment and diagnosis of the target person.

  The above embodiment can be modified into various forms. Hereinafter, modified examples will be described. In the above embodiment, the monitoring device 1 learns the operation of a specific target person, and determines the operation of the target person based on the learned operation. It is not limited to such. For example, a monitoring device is applied to a subject having a specific physical condition, a motion model corresponding to the physical condition is learned, and the motion of another subject having a similar physical condition is determined using the learned motion model. May be.

  In the above-described embodiment, the vital measurement unit 312 measures the vital sign of the subject based on the image data from the camera 14. However, the configuration of the vital information acquiring unit that acquires the vital information of the subject is such a configuration. Not limited to things. The measurement value of the vital sensor of the type worn by the subject may be acquired in a wired or wireless manner.

  In the above embodiment, the text data of the subject's behavior included in the nursing care record created by the management terminal 17 is transmitted to the controller group 50. However, the behavior plan acquisition unit that acquires the subject's behavior plan Is not limited to this. For example, as the determination of the operation performed on the controller group 50 side, the behavior of the target person such as eating and sleeping is automatically determined based on the image data from the camera 14, and the determined behavior is described as text data for nursing record. May be output to the management terminal 17.

  In the above embodiment, the monitoring device 1 is configured to include the camera 14 in the main body, but the monitoring device that monitors a specific target person is not limited to this. For example, the image data of the subject may be acquired from a surveillance camera separate from the monitoring device by wire or wirelessly, and the learning and determination of the operation may be performed based on the acquired image data. In this case, for example, image data of surveillance cameras of different manufacturers can be used as learning data by complying with network camera standardization standards such as ONVIF (Open Network Video Interface Forum).

  The above description is merely an example, and the present invention is not limited by the above-described embodiments and modifications as long as the features of the present invention are not impaired. It is also possible to arbitrarily combine one or more of the above-described embodiment and the modifications, and it is also possible to combine the modifications.

1 (1A, 1B) monitoring device, 11 case, 11r rail, 11s slit, 12 base, 13 cover, 14 camera, 15 LED, 16 notification unit, 17 management terminal, 20, 30, 40 controller, 21, 41 GPU, 22, 32, 42 memory, 31 CPU, 50 controller group, 211 operation determination unit, 212 posture determination unit, 213 output unit, 221 operation model storage unit, 222 posture model storage unit, 311 abnormal posture change determination unit, 312 vital measurement section, 313 abnormal change determination section, 314 abnormal change information output section, 315 learning time determination section, 316 learning data output section, 321 action schedule storage section, 322 abnormal posture change determination condition storage section, 323 abnormal change determination condition storage section, 324 vital measurement storage unit, 411 operation learning unit, 412 posture learning unit, 413 output unit, 421 motion learning data storage unit, 422 posture learning data storage unit

One embodiment of the present invention is a monitoring device that monitors a specific subject, extracts a feature amount of a motion including a posture change of the subject based on image data of the motion of the subject, and A learning model generation unit that generates a learning model; a learning model storage unit that stores a learning model of the operation generated by the learning model generation unit; and a motion of the subject based on the learning model stored by the learning model storage unit. And an operation determination unit that determines

One embodiment of the present invention is a monitoring device that monitors a specific target person , and extracts a feature amount of an operation including a posture change unique to the target person based on image data of an operation unique to the target person. a learning model storage unit for storing a learning model of the specific behavior and learning model generating unit for generating a learning model of the specific behavior, the subject generated by the learning model generating unit to user, stored by the learning model storage unit And a motion determining unit that determines the motion of the target person based on the learning model of the motion specific to the target person and the image data of the motion specific to the target person.

One aspect of the present invention, generates a particular a monitoring device for monitoring a subject, extracting a feature amount of operation of the subject based on the image data of the operation subject, learning model of the operation of the subject a learning model generating unit to a learning model storage unit for storing a learning model of the operation generated by the learning model generating unit, based on the stored learned model by learning model storage unit, the operation of the subject An operation determining unit for determining , a biological information acquiring unit for acquiring biological information of the subject, and an abnormality determining unit for determining whether an abnormality has occurred in the subject based on the biological information acquired by the biological information acquiring unit. And . The learning model generation unit extracts a feature amount of the target person's movement based on the image data of the target person's movement when the change determination unit determines that the target person has an abnormality, and learns the abnormal movement of the target person. Generate a model.

Claims (3)

A monitoring device for monitoring a specific target person,
A learning model generation unit that extracts a feature amount of the motion of the target person based on the image data of the motion of the target person, and generates a learning model of the motion of the target person;
A learning model storage unit that stores a learning model of the operation generated by the learning model generation unit,
A monitoring device comprising: an operation determining unit that determines an operation of the subject based on a learning model stored in the learning model storage unit. The monitoring device according to claim 1,
A biological information acquisition unit that acquires the biological information of the subject;
An abnormality determination unit configured to determine whether an abnormality has occurred in the subject based on the biological information acquired by the biological information acquisition unit,
The learning model generating unit extracts a feature amount of the motion of the subject based on image data of the motion of the subject when it is determined that the abnormal has occurred in the subject by the abnormal determination unit, and A monitoring device for generating a learning model of abnormal behavior of a user. The monitoring device according to claim 2,
Further comprising an activity schedule acquisition unit for acquiring the activity schedule of the target person,
The monitoring device, wherein the abnormality determination unit further determines whether or not an abnormality has occurred in the subject based on the action schedule acquired by the action schedule acquisition unit.

Images