JP2018026006A - Watching system, watching method, and watching program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique capable of improving determination accuracy whether a detected state of an object person is proper.SOLUTION: A watching system 1 includes: a state detection part 21 detects a state an object person; an influence level acquisition section 22 that acquires an influence level of an influence factor which gives an influence on the state of the object person; and a determination part 23 that determines whether the state of the object person detected by the state detection part 21 is proper using the influence level of the influence factor acquired by the influence level acquisition section 22.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for watching a subject's daily life.

  Conventionally, there is a system that watches the daily life of the subject and determines that the subject's state is not appropriate (if it is judged that something has changed in the subject), the system outputs that fact. This system often targets elderly people living alone. This system includes a configuration described in Patent Documents 1 to 4 and the like.

  Patent Document 1 discloses an operation manual that serves as a standard for a normal life pattern of a target person based on detection outputs of a human sensor that detects the movement of a resident (target person) and a device sensor that detects the operation status of various facilities in a house. Build or modify. This system generates a life pattern of a subject in a certain period based on detection outputs of the human sensor and device sensor for a certain period. The system compares the operation manual with the life pattern of the subject over a certain period, and determines the subject's situation.

  Further, Patent Document 2 extracts a feature amount of a detection output of a sensor, and performs a multivariate analysis on the daily life of the subject using a period flag obtained by converting date / time information. This system compares the daily life level obtained by multivariate analysis with the daily life level reference value, and determines the situation of the subject. The daily life of the subject is a power sensor that measures the power consumption of electrical products, a temperature sensor that measures the temperature of a heat source such as a cooking appliance or a heating appliance, an illuminance sensor that measures ambient brightness, and an infrared sensor that detects the human body , A pressure sensor that detects the passage of the human body, a proximity sensor that detects the opening and closing of doors, an infrared sensor that detects transmission signals of infrared remote controls for various electrical products, a flow sensor that measures the amount of water used, and a gas usage It is detected by a gas flow sensor to be measured.

  Further, Patent Document 3 estimates the location of the subject at a certain time based on the detection output of the human sensor, and estimates the motion of the resident whose occurrence time is specified based on the detection output of the acceleration sensor. This system determines a situation such as a resident falling in the bathroom or a person other than the resident entering the residence based on the detection output of the human sensor and the detection output of the acceleration sensor.

  Furthermore, patent document 4 determines whether a subject exists based on the detection output of a temperature sensor or a human sensor. This system determines that the subject is normal if the subject's vital signs (signs that appear when maintaining life) are good, and otherwise determines that the subject is abnormal.

JP 2006-65886 A JP 2008-102884 A JP 2014-89494 A Japanese Patent Laid-Open No. 2006-103178

  However, the above-described conventional system compares the detection output of the sensor with a predetermined criterion (reference value, reference change pattern, etc.) for the sensor, and determines whether or not the subject's state is appropriate. ing. This criterion is determined based on the behavior pattern of the daily life of the subject.

  On the other hand, the behavior pattern of the subject varies depending on various factors (influencing factors). For example, when the subject has a special event, the behavior pattern before and after the event changes. Special events are behaviors that are not routinely performed by the target, such as going out to local festivals, entertainment facilities (movie theaters, museums, aquariums, etc.). Further, when the surrounding environment changes, the subject changes in the behavior pattern. Furthermore, the behavior pattern of the subject changes depending on changes in physiological desire.

  Therefore, in the configuration in which the detection output of the sensor is compared with a determination criterion determined based on the behavior pattern of the subject's daily life and the like, the behavior pattern of the subject is determined. When is changed for some reason, the subject's condition is misjudged. Specifically, even if the subject's state is appropriate, if the subject's behavior pattern changes due to some factor, the subject's behavior pattern at this time deviates from the daily behavior pattern, and the subject's state is Misjudging that it is not appropriate.

  An object of the present invention is to provide a technique for improving the determination accuracy of whether or not the detected state of the subject person is appropriate.

  The watching device of the present invention is configured as follows in order to achieve the above object.

  The state detection unit detects the state of the target person. A state detection part detects a subject's state, for example using the detection output of the sensor connected to the sensor connection part. The influence degree acquisition unit acquires the influence degree of an influence factor that affects the state of the subject.

  For example, influential factors that affect the sleep state of the subject include the behavior of the subject before sleep, the behavior schedule after waking up, and the like. In addition, influential factors that influence the bathing of the subject are the behavior of the subject, the climate of the day, and the like. In addition, the influence factors that influence the cooking behavior of the subject are the behavior of the subject that day, the behavior schedule of the day, and the like. In addition, the influencing factors that affect the use of the toilet of the subject are the behavior of the subject that day. In addition, the influencing factors that influence the removal of the postings from the subject's mailbox are the behavior of the subject that day. In addition, the influencing factors that affect the use of the target's air conditioner are the climate of the day.

  The determination unit determines whether or not the state of the subject detected by the state detection unit is appropriate using the influence level of the influence factor calculated by the influence level acquisition unit. Thereby, the determination precision of whether the detected subject's state is appropriate can be improved.

The influence level acquisition unit may be configured to acquire the influence level of the influence factor using the detection output of the sensor connected to the sensor connection unit, or the target person's action schedule may be obtained from the target person's action schedule. The monitoring device may be configured to acquire the degree of influence of the influencing factor when the state of the subject detected by the state detection unit is determined to be inappropriate by the determination unit. A configuration including an output unit for outputting is preferable.

  According to the present invention, it is possible to improve the determination accuracy of whether or not the detected state of the subject is appropriate.

It is a block diagram which shows the structure of the principal part of a monitoring apparatus. It is a figure which shows the information memorize | stored in the determination information storage part. It is a flowchart which shows operation | movement of the monitoring apparatus concerning this example. It is a figure showing a watching system provided with a watching device of another example.

  Hereinafter, a monitoring device according to an embodiment of the present invention will be described.

  FIG. 1 is a block diagram showing a configuration of a main part of a monitoring device according to an embodiment of the present invention. The monitoring device 1 according to this example includes a control unit 2, a sensor connection unit 3, a display / operation unit 4, a determination information storage unit 5, and a communication unit 6. This watching device 1 watches the daily life of the subject. Specifically, this system determines whether or not the state of the subject is appropriate, and if it is determined that the subject is not appropriate (something has changed in the subject), it outputs that fact. In this example, the monitoring device 1 will be described as being installed in the residence of the subject, which is a space for watching the subject.

  The control unit 2 controls the operation of each part of the monitoring device 1 main body. The control unit 2 includes a state detection unit 21, an influence degree acquisition unit 22, and a determination unit 23, which will be described later. Moreover, the control part 2 performs the watching method concerning this invention. The watching program according to the present invention is installed in the control unit 2.

  A human sensor 51, a biological sensor 52, an environmental sensor 53, a position sensor 54, a device state detection sensor 55, and a facility state detection sensor 56 shown in FIG. 1 are connected to the sensor connection unit 3. As shown below, the human sensor 51, the biosensor 52, the environment sensor 53, the position sensor 54, the device state detection sensor 55, and the equipment state detection sensor 56 shown in FIG. 1 are each configured by one or more sensors. The

  The human sensor 51 is a sensor used for detecting the presence of the target person. The human sensor 51 may be any sensor that can detect whether a subject exists, such as an image sensor, an infrared sensor, a sound wave sensor, a vibration sensor, or a pressure sensor. The human sensor 51 is installed in a living room, a kitchen, a bedroom, a bathroom, a toilet or the like of the subject's residence. The human sensor 51 is composed of one or more sensors. Moreover, the sensor which comprises the human sensitive sensor 51 is 1 or more types.

  The biological sensor 52 is a sensor used for detecting the biological information (heartbeat state / respiration state / body movement) of the subject. The biological sensor 52 is a body temperature sensor, blood pressure sensor, heart rate sensor, acceleration sensor, temperature / humidity sensor, illuminance sensor, absolute pressure sensor, acoustic wave sensor, radio wave sensor, image sensor, or the like that can detect the biological information of the subject. Good. Some biological sensors 52 are installed in the room of the subject's residence, while others are attached to the subject. The biosensor 52 is composed of one or more sensors. Moreover, the sensor which comprises the biosensor 52 is 1 or more types.

  The environmental sensor 53 is a sensor used for detecting the surrounding environment of the subject's residence. The environmental sensor 53 may be any sensor that can detect the surrounding environment, such as an image sensor, a temperature / humidity sensor, an illuminance sensor, an absolute pressure sensor, a noise sensor, and an acceleration sensor. Some environmental sensors 53 are installed in the room of the subject's residence, while others are installed outdoors such as a veranda. The environmental sensor 53 is composed of one or more sensors. In addition, the environment sensor 53 includes one or more types of sensors.

  The position sensor 54 is a sensor used for detecting the position of the subject. The position sensor 54 may be any sensor that can detect the position of the subject, such as a GPS sensor, an image sensor, or an infrared sensor. The position sensor 54 may be installed in the room of the subject's residence or may be attached to the subject. The position sensor 54 is composed of one or more sensors. Further, the position sensor 54 is composed of one or more types of sensors.

  The device state detection sensor 55 is a sensor that is used for detecting the use state of electrical appliances (air conditioners, televisions, rice cookers, pots, washing machines, etc.) installed in the residence of the subject. The device state detection sensor 55 may be any device that can detect the usage state of the electrical appliance, such as a temperature / humidity sensor, an image sensor, an infrared sensor, and a current sensor. The device state detection sensor 55 may be installed in the room of the subject's residence, or may be mounted on an electrical appliance. Further, the device state detection sensor 55 may be configured to detect not only appliances but also other types of devices such as bicycles and automobiles. The device state detection sensor 55 is composed of one or more sensors. Moreover, the sensor which comprises the apparatus state detection sensor 55 is 1 or more types.

  The facility state detection sensor 56 is a sensor used for detecting the use state of residential facilities such as water supply, gas, and indoor lighting. The facility state detection sensor 56 may be any device that can detect the use state of the housing facility, such as a water or gas flow sensor, an image sensor, an infrared sensor, or an illuminance sensor. The equipment state detection sensor 56 may be installed in the room of the subject's residence, or may be attached to the residence equipment. The equipment state detection sensor 56 is composed of one or more sensors. Moreover, the sensor which comprises the equipment state detection sensor 56 is 1 or more types.

  One sensor connected to the sensor connection unit 3 includes two or more of the human sensor 51, the biosensor 52, the environment sensor 53, the position sensor 54, the device state detection sensor 55, and the equipment state detection sensor 56. May belong. For example, the configuration may be such that the specific image sensor belongs to the human sensor 51, the environment sensor 53, the position sensor 54, the device state detection sensor 55, and the equipment state detection sensor 56. In this way, the number of sensors connected to the sensor connection unit 3 can be suppressed.

  Each sensor connected to the sensor connection unit 3 is given an identification number. The sensor connection unit 3 inputs the detection output input from the connected sensor to the control unit 2 together with the sensor identification number. Therefore, the control unit 2 can determine which sensor the detection output input from the sensor connection unit 3 is. The connection of various sensors in the sensor connection unit 3 may be wired or wireless. Moreover, the structure connected via networks, such as the internet, may be sufficient.

  The display / operation unit 4 includes a display device such as a display device and input devices such as a keyboard, a mouse, a microphone, and a camera. The input device may be another device such as a touch panel integrated with a display, a smartphone linked with wireless communication, or the like. A subject or the like can input desired data into the device 1 by operating the input device. For example, the target person can watch the contents of the action schedule (type of event, date and time, etc.) such as going out to festivals, entertainment facilities (movie theaters, museums, aquariums, etc.) in the area where the target person lives. 1 can be entered. The content of this subject's action schedule is referred to as event-related information here.

  As shown in FIG. 2, the determination information storage unit 5 stores detection information 31, event related information 32, detection item information 33, influence degree information 34, determination criterion information 35, and the like.

  The detection information 31 is information obtained by accumulating, in time series, the detection output input from the sensor or the feature amount acquired by processing the detection output for each sensor connected to the sensor connection unit 3.

  The event-related information 32 is information indicating the contents of an action schedule such as going out to festivals and entertainment facilities (movie theaters, museums, aquariums, etc.) in the area where the target person resides. The event-related information 32 is not limited to information input by the subject or the like operating the input device of the display / operation unit 4, and may include information acquired from the Internet or the like.

  The detection item information 33 is information in which a sensor used to detect the state of the subject related to the item is associated with each item for detecting the state of the subject. The detection item information 33 also includes information for detecting the state of the subject from the detection output of the associated sensor for each item for detecting the state of the subject.

  The influence degree information 34 is information in which an event or a sensor related to an influence factor affecting the item is associated with each item for determining the state of the subject. The influence level information 34 also includes information for estimating the magnitude of the influence level of the influence factor from the related information of the associated event and the detection output of the sensor for each item for detecting the state of the target person. It is out. The influence factor mentioned here is an event that affects an action relating to an item for determining the state of the subject. For example, the influencing factors that affect the sleep of the subject are the living behavior of the subject before sleep, the schedule after waking up, and the like.

  The determination criterion information 35 is information in which a criterion (determination criterion) used for determining whether or not the state is appropriate is associated with each item for determining the state of the subject.

  The communication unit 6 communicates with a counterpart device that notifies the state of the target person via a network such as a public line network or the Internet network. The watching device 1 is preset with a counterpart device that notifies the state of the subject. For example, this counterpart device is a terminal of the family of the subject who does not live together, a terminal of a collaborator (volunteer, police, firefighting, etc.) who lives in the neighborhood of the subject, and a provider of this monitoring service. It is a terminal. The communication unit 6 corresponds to the output unit referred to in the present invention.

  Next, the state detection part 21, the influence degree acquisition part 22, and the determination part 23 with which the control part 2 is provided are demonstrated. The state detection unit 21 refers to the detection item information 33 for each item for detecting the state of the subject, and determines a sensor corresponding to the item. The state detection unit 21 detects the state of the subject based on the detection output of the sensor determined here. The detection output of each sensor is stored as detection information 31 in the determination information storage unit 5.

  The influence degree acquisition part 22 estimates the magnitude of the influence degree of the influence factor of the item for determining the state of the subject. The impact level acquisition unit 22 refers to the impact level information 34 for each item for determining the state of the subject, and determines event related information and sensors corresponding to the item. The influence level acquisition unit 22 estimates and acquires the magnitude of the influence level of the influence factor based on the event-related information determined here and the detection output of the sensor. Information used to estimate the magnitude of the influence degree of the influence factor is stored in the determination information storage unit 5 as the influence degree information 34.

  The determination unit 23 determines whether or not the state of the subject person is appropriate. The determination unit 23 determines the state of the subject detected by the state detection unit 21 for the item for determining whether or not the state of the subject is appropriate, and the influence level of the influence factor estimated and acquired by the influence acquisition unit 22. This determination is performed using the determination criterion acquired with reference to the determination criterion information 35.

  FIG. 3 is a flowchart showing the operation of the watching apparatus according to this example. The monitoring device 1 determines whether or not there is an item (hereinafter referred to as a determination target item) that determines whether or not the state of the subject is appropriate at that time (s1). The determination target items are sleep depth, sleep time, behavior during sleep, presence / absence of bathing, bathing time, cooking behavior, use of toilet, use of air conditioner, and the like.

  The monitoring device 1 refers to the detection information stored in the determination information storage unit 5 if there is a determination target item for determining whether or not the state of the subject is appropriate at that time, and sets the determination target item as the determination target item. A corresponding sensor is determined (s2). Moreover, the watching apparatus 1 acquires the detection output of the sensor judged by s2 with reference to the detection information 31 (s3). The watching device 1 detects the state of the subject using the detection output of the sensor acquired in s3 (s4). The processing relating to s2 to s4 is performed by the state detection unit 21.

  If the watching apparatus 1 determines that there is no determination target item for determining whether or not the state of the subject person is appropriate in s1, the process related to s1 is repeated.

  Further, the watching device 1 refers to the influence information 34 stored in the determination information storage unit 5 to determine an event or a sensor related to an influence factor that affects the current determination target item. Information for estimating and acquiring the degree of influence is acquired (s5). The watching device 1 refers to the detection information and the event related information 32 stored in the determination information storage unit 5, and estimates and acquires the magnitude of the influence factor determined in s5 (s6). The processes related to s5 and s6 are performed by the influence level acquisition unit 22.

  The monitoring device 1 may be configured to execute the processes related to s2 to s4 after executing the processes related to s5 and s6. Moreover, the monitoring apparatus 1 may perform the process concerning s2-s4 after performing the process of s5, and may perform the process concerning s6 after that. Furthermore, the watching device 1 may execute the processes related to s5 and s6 after executing the process of s2, and then execute the processes related to s3 and s4. That is, the order of the processes concerning s2 to s6 may be any order as long as it is an order in which the state of the target person can be detected and the influence degree of the influence factor can be acquired (estimated).

  The watching device determines whether or not the state of the subject detected in s4 is appropriate using the magnitude of the influence estimated in s6 (s7). In s7, with reference to the determination criterion information 35 memorize | stored in the determination information storage part 5, the determination criterion used for determination of whether it is appropriate about a determination object item is acquired. In s7, instead of determining whether the determination target item is appropriate by comparing the state of the subject detected in s4 with the determination criterion, the influence degree of the influencing factors affecting the determination target item is determined. It is determined whether it is appropriate in consideration of the size of. The determination unit 23 performs the process related to s7.

  If the watching device 1 determines that it is appropriate in s7, it returns to s1. On the other hand, if the watching device 1 determines that it is not appropriate in s7, it performs a notification process (s8) and returns to s1. The notification process of s8 is a process in which the communication unit 6 notifies a predetermined partner terminal that the state of the target person is not appropriate. The other party's terminal is the terminal of the target person's family who does not live together, the terminal of a collaborator (volunteer, police, firefighter, etc.) who lives in the vicinity of the target person, the terminal of the company providing this watch-over service, etc. is there. This notification may be executed for a plurality of counterpart terminals.

  Thus, the monitoring device 1 according to this example does not determine whether the state of the subject is appropriate by comparing the state of the subject detected from the detection output of the sensor and the determination criterion, Judgment is made using the degree of influence of the influence factors that influence the condition of the subject. Therefore, even if the subject's behavior pattern deviates from the daily behavior pattern due to some factor, it is possible to prevent erroneous judgment that the subject's state is not appropriate. That is, it is possible to improve the determination accuracy of the monitoring device 1 as to whether or not the state of the subject is appropriate.

  It is also possible to adopt a configuration in which the processing shown in FIG. 3 (especially processing related to s2 to s7) is executed by a computer (so-called artificial intelligence (AI)) that learns by utilizing deep learning. . If comprised in this way, the determination precision of whether a subject's state is appropriate will improve as learning of a computer progresses.

  Next, a specific example of the determination target item will be shown, and determination on whether or not the state of the target person in the watching device 1 is appropriate will be described.

  (1) The determination target items are sleep depth and sleep time.

  The depth of sleep and the sleep time of the subject can be determined by the detection outputs of the human sensor 51 and the biological sensor 52. For example, the watching apparatus 1 determines that the state in which the subject is substantially stationary on the bed is the start of the subject's sleep. The state in which the subject is substantially stationary on the bed can be determined by the detection output of the human sensor 51 such as an image sensor, a sound wave sensor, a vibration sensor, or a pressure sensor. In addition, the breathing of the subject during sleep and the heartbeat of the subject can be acquired by the detection output of the biological sensor 52 such as a sound wave sensor or a pressure sensor. The state detection unit 21 uses the detection outputs of the human sensor 51 and the biological sensor 52 to start and end the sleep of the subject, the depth of sleep during the sleep of the subject, and the awakening of the subject during sleep. By acquiring behavior during sleep such as turning over in time series, it is possible to detect the sleep depth and sleep time of the subject.

  Influencing factors that affect sleep depth and sleep time include daytime behavior before sleep, stress state before sleep, daytime physical burden before sleep, and the like. The daytime behavior before sleeping can be acquired by using the detection output of the biological sensor 52 such as an acceleration sensor attached to the subject. Moreover, the stress state before sleep can be acquired by using the detection output of the biosensor 52 such as a body temperature sensor attached to the subject. The mental burden during the day before sleeping can be estimated and acquired from the action schedule of the subject on the day. The influence level acquisition unit 22 estimates and acquires the magnitude of the influence level based on the detection output of the biological sensor 52 and the action schedule of the subject on the day for the influence factors that affect the sleep depth and sleep time. To do.

  The determination unit 23 compares the sleep depth and sleep time of the subject detected by the state detection unit 21 with the determination criteria, and affects the sleep depth and sleep time acquired by the influence level acquisition unit 22. This is done in consideration of the degree of influence of the influencing factors. Specifically, the range determined to be appropriate with respect to the determination criterion is determined according to the magnitude of the influence factor that affects the sleep depth and sleep time acquired by the influence acquisition unit 22. Adjust.

  (2) The determination target items are a bedtime and a wake-up time.

  The bedtime and the wake-up time can be determined based on the detection outputs of the human sensor 51 and the biosensor 52 as described above.

  Influencing factors that affect the bedtime and the wake-up time are the action schedule of the subject after wake-up, the weather (temperature and humidity, snow and snow, wind), noise, and the like. The action schedule of the subject after getting up can be acquired by referring to the event related information 32. The weather (temperature and humidity, rain and snow, wind) and noise can be acquired by using the detection output of the environment sensor 53. Specifically, the discomfort index can be acquired from the detection output of the temperature and humidity sensor used as the environment sensor 53. Further, the amount of rain and snow due to low atmospheric pressure can be acquired from the detection output of the absolute pressure sensor used as the environmental sensor 53. Further, the noise level due to wind and rain can be acquired from the detection output of the noise sensor used as the environment sensor 53. The influence level acquisition unit 22 estimates and acquires the magnitude of the influence level, which affects the bedtime and the wake-up time, based on the detection output of the environmental sensor 53.

  The determination unit 23 compares the bedtime and the wake-up time of the subject detected by the state detection unit 21 with the determination criteria, and influence factors that affect the sleep time and the wake-up time acquired by the influence level acquisition unit 22. This is done in consideration of the degree of influence. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the influence level of the influence factor that affects the sleeping time and the wake-up time acquired by the influence level acquisition unit 22. .

  (3) The determination target item is an action during sleep.

  As for the behavior during sleep, for example, the behavior frequently performed in the toilet during sleep can be detected by the detection output of an image sensor, a sound wave sensor, or the like used as the human sensor 51.

  The influencing factors of the behavior of the subject going to the toilet during sleep are the activities before sleep and the contents of eating and drinking. The subject's activities before sleep and the contents of food and drink are stimulated by the detection output of the image sensor used as the human sensor 51 and the odor sensor used as the environment sensor, etc. It can be acquired that the meal was strong. The influence level acquisition unit 22 estimates and acquires the magnitude of the influence level based on the detection output of the human sensor 51 for an influence factor that affects the behavior during sleep.

  The determination unit 23 compares the behavior of the subject in sleep detected by the state detection unit 21 with the determination criterion, and the influence degree of the influence factor that influences the sleep behavior acquired by the influence degree acquisition unit 22. Considering the size of. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the influence degree of the influence factor that affects the sleeping behavior acquired by the influence degree acquisition unit 22.

  (4) The determination target item is the presence or absence of bathing.

  The presence or absence of bathing can be determined by the detection output of the human sensor 51 installed in the bathroom or the equipment state detection sensor 56 (for example, a sensor that detects use of a water heater).

  The influencing factors related to the presence or absence of the subject's bathing are the amount of activity of the day, the temperature and humidity of the day, etc. The amount of activity of the subject for the day can be acquired from the detection output of the biosensor 52 (for example, an acceleration sensor) attached to the subject. The temperature and humidity of the day can be acquired from the detection output of the environmental sensor 53 (for example, the temperature and humidity sensor). Further, by using a temperature sensor attached to the subject as the biosensor 52, the amount of sweating of the subject can be acquired from the detection output. Further, by using a GPS sensor or a sunshine sensor attached to the subject as the biosensor 52, the subject's outdoor activity and indoor activity can be distinguished from the detected output. The influence level acquisition unit 22 estimates and acquires the magnitude of the influence level based on the detection output of the biological sensor 52 and the environment sensor 53 for the influence factors related to the presence or absence of bathing.

  The determination unit 23 compares the presence / absence of the subject's bathing detected by the state detection unit 21 and the determination criterion with respect to the degree of influence of the influencing factor on the presence / absence of the subject's bathing acquired by the influence level acquisition unit 22. Take into account the size. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the magnitude of the influence factor of the influence factor related to the presence or absence of bathing acquired by the influence degree acquisition unit 22.

  (5) The determination target item is a bathing time.

  The bathing time can be determined by the detection output of the human sensor 51 installed in the bathroom or the equipment state detection sensor 56 (for example, a sensor that detects use of a water heater).

  The influencing factors concerning the bathing time of the subject are the amount of activity of the day, the temperature and humidity of the day, the time of going out, the presence or absence of drinking, and the like. As described above, the daily activity amount of the subject can be acquired from the detection output of the biosensor 52 (for example, an acceleration sensor) attached to the subject. The temperature and humidity of the day can be acquired from the detection output of the environmental sensor 53 (for example, the temperature and humidity sensor). Further, by using a temperature sensor attached to the subject as the biosensor 52, the amount of sweating of the subject can be acquired from the detection output. Furthermore, as the biosensor 52, the location of the subject's outing (distance from the subject's residence) or the like can be acquired from the detection output of the GPS sensor attached to the subject. Further, by using an image sensor as the human sensor 51, it is possible to acquire the presence or absence of drinking of the subject from the detection output. The influence level acquisition unit 22 estimates and acquires the magnitude of the influence level of the influence factor related to the bathing time based on the detection output of the biological sensor 52 and the environment sensor 53.

  The determination unit 23 compares the target person's bathing time detected by the state detection unit 21 with the determination criteria, and the degree of influence of the influencing factor on the subject's bathing time acquired by the influence level acquisition unit 22. Take this into consideration. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the magnitude of the influence factor of the influence factor related to the bathing time acquired by the influence acquisition section 22.

  (6) The determination target items are cooking time and work amount.

  The time when the subject person is cooking and the amount of work can be acquired from the detection output of the human sensor 51 such as an infrared sensor and a sound wave sensor installed in the kitchen or cooking equipment and the equipment state detection sensor 56.

  Factors that affect the time of cooking and the amount of work are the amount of activity of the day. If the subject has a large amount of activity on the day and is intense, the subject may delay the start of cooking or simplify cooking to shorten cooking time. The amount of activity of the subject for the day can be acquired from the detection output of the biosensor 52 (for example, an acceleration sensor) attached to the subject. The influence level acquisition unit 22 estimates and acquires the magnitude of the influence level based on the detection output of the biological sensor 52 for the influence factors related to the cooking time and the work amount.

  The determination unit 23 compares the subject's cooking time and work amount detected by the state detection unit 21 with the determination standard, and the subject's cooking time and work amount obtained by the influence degree obtaining unit 22. This is done in consideration of the magnitude of the influence factor. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the cooking time acquired by the influence acquisition unit 22 and the influence of the influence factor on the work amount.

  In addition, from the detection output of the sonic sensor, the image sensor, etc., the subject's detailed cooking behavior is acquired, the type of food or cooking is acquired from the detection output of the odor sensor, or the detection output of the equipment state detection sensor 56 You may determine what kind of cooking is performed by acquiring consumption of electricity, gas, and water, and may estimate and acquire the magnitude of an influencing factor using these.

  (7) The determination target item is the presence or absence of cooking action.

  The time when the subject person is cooking and the amount of work can be acquired from the detection output of the human sensor 51 such as an infrared sensor and a sound wave sensor installed in the kitchen or cooking equipment and the equipment state detection sensor 56. That is, the presence / absence of a cooking action can be determined based on the detection output of the human sensor 51 such as an infrared sensor and a sound wave sensor installed in a kitchen or cooking equipment, or the equipment state detection sensor 56.

  Factors affecting the presence or absence of cooking behavior include eating out, lunch boxes, delivery meals, and the use of ingredients that do not require cooking. Such information can be acquired by referring to the event related information 32. The influence level acquisition unit 22 estimates and acquires the magnitude of the influence level based on the information acquired with reference to the event related information 32 regarding the influence factors related to the presence or absence of cooking behavior.

  The determination unit 23 compares the presence / absence of the subject's cooking behavior detected by the state detection unit 21 with the determination criterion, and the influence of the influence factor on the presence / absence of the subject's cooking behavior acquired by the influence level acquisition unit 22. Take into account the magnitude of the degree. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the magnitude of the influence factor of the influence factor related to the presence or absence of the cooking action acquired by the influence degree acquisition unit 22.

  (8) The determination target item is use of the toilet.

  Whether the subject has used the toilet can be acquired from the detection output of the human sensor 51 installed in the toilet.

  An influencing factor for the use of toilets is the use of toilets in other places due to going out and outdoor activities. The influence degree acquisition unit 22 determines the action of the subject person and the proximity to the toilet from the detection output of the acceleration sensor, the atmospheric pressure sensor, the GPS sensor, and the like, which are the biosensors 52 attached to the subject person, and the magnitude of the influence degree. Estimate and get

  The determination unit 23 compares the use of the toilet detected by the state detection unit 21 with the determination criteria, and determines the degree of influence of the influence factor on the use of the toilet of the subject acquired by the influence acquisition unit 22. Take it into consideration. Specifically, the range that is determined to be appropriate with respect to the determination criterion is adjusted according to the magnitude of the influence factor of the influencing factor that is acquired by the influence acquisition unit 22 and that is related to the use of the toilet.

  (9) The determination target item is the usage time of the toilet.

  Whether the subject is using the toilet can be acquired from the detection output of the human sensor 51 installed in the toilet.

  Factors affecting the toilet usage time include smoking, reading, and cleaning. These influencing factors can be acquired from detection outputs of a smoke sensor, a sound sensor, a sound wave sensor, etc. installed in the toilet. The influence level acquisition unit 22 determines the behavior of the target person based on detection outputs from a smoke sensor, a sound sensor, a sound wave sensor, and the like installed in the toilet, and estimates and acquires the magnitude of the influence level.

  The determination unit 23 compares the use time of the toilet detected by the state detection unit 21 with the determination criteria, and the degree of influence of the influence factor on the use time of the target person's toilet acquired by the influence level acquisition unit 22 is large. Take this into consideration. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the magnitude of the influence factor of the influencing factor acquired by the influence degree acquisition unit 22 on the toilet usage time.

  (10) The determination target item is an action of taking out a post from the mailbox.

  Whether or not a postal matter such as a mail or a newspaper is contained in the mailbox can be acquired by a detection output of a facility state detection sensor 56 such as a weight sensor or an optical sensor installed in the mailbox. In other words, the state detection unit 21 can detect the action of taking out the post from the mailbox by the detection output of the equipment state detection sensor 56 such as a weight sensor or an optical sensor installed in the mailbox.

  The influential factors concerning the behavior of picking up a post from the mailbox are the actions of the target person such as long phone calls, cooking behavior, and outdoor work. These behaviors of the subject can be acquired from detection outputs of an image sensor, a sound wave sensor, and an acceleration sensor attached to the subject, which is the biosensor 52, which are human sensors 51 installed in the residence. The influence acquisition unit 22 determines the behavior of the target person from detection outputs of an image sensor, a sound wave sensor, and an acceleration sensor attached to the target person as the biosensor 52, which are the human sensor 51 installed in the residence. Then, the degree of influence is estimated and acquired.

  The determination unit 23 compares the behavior of taking out the post from the subject's mailbox detected by the state detection unit 21 and the determination criteria of the post from the subject's post received by the influence acquisition unit 22. This is done in consideration of the degree of influence of the influencing factors on the extraction behavior. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the degree of influence of the influence factor related to the action of taking out the post from the mailbox acquired by the influence acquisition unit 22. To do.

  (11) The determination target item is the use of an air conditioner.

  The use status of the air conditioner can be determined from the detection output of the device state detection sensor 55 or the temperature / humidity sensor which is the environmental sensor 53 installed in the residence.

  Factors affecting the use of air conditioners include temperature and humidity in the house, outdoor environment, the number of visitors, and the physical condition of the target person. The temperature and humidity in the residence can be obtained from the detection output of a temperature and humidity sensor or the like that is the environmental sensor 53 installed in the residence. In addition, the level of noise and dust caused by construction or crowding can be acquired in the vicinity by the detection output of the environmental sensor 53 such as an acoustic sensor or a dust (particle) sensor installed outdoors. In addition, the number of visitors can be acquired by the detection output of the human sensor 51 such as an image sensor installed indoors. Further, the physical condition of the subject can be acquired from the detection output of the biosensor 52 attached to the subject. The influence degree acquisition unit 22 estimates and acquires the influence degree of an influence factor related to the use of the air conditioner from the detection output of the sensor described above.

  The determination unit 23 compares the use condition of the air conditioner detected by the state detection unit 21 and the determination criteria in consideration of the magnitude of the influence factor of the influence factor related to the use of the air conditioner acquired by the influence degree acquisition unit 22. Do it. Specifically, the range determined to be appropriate with respect to the determination criterion is adjusted according to the magnitude of the influence factor of the influence factor related to the use of the air conditioner acquired by the influence degree acquisition unit 22.

(12) Others The daily behavior of the subject may be determined using changes in life rhythm due to events as an influencing factor. Moreover, you may determine use of electrical appliances, such as a target person's television, using the change of the operating condition of a use apparatus, or a utilization purpose as an influence factor. In addition, the response status of the target person to the visitor may be determined by using a change in sleep state as an influencing factor.

  Further, with respect to the above (1) to (12), the movement of the pet may be one of the influencing factors, and it may be determined whether or not the state of the subject is appropriate. Thereby, it can prevent misjudging whether an object person's state is appropriate by the influence by a pet's movement.

  In addition, even if the monitoring device 1 analyzes a conversation with a device equipped with a communication means such as a microphone or a robot equipped with a communication means, it is determined whether or not the state of the target person is appropriate. Good. For example, in a conversation with a device having a communication device such as a microphone in the watching device 1 or a robot having a communication device, the conversation time of the day, the tone of the conversation, whether or not there is a response to the robot call Home detection and action at home can be acquired based on the presence / absence of response / interval / voice tone depending on the interval / contents and type of call. Further, the voice volume and conversation speed in the conversation with the robot may be acquired to determine the physical condition and the degree of stress of the target person.

  Further, when the subject is sleeping or absent, the intrusion of the suspicious person may be determined by the detection output of the human sensor 51 such as an absolute pressure sensor, an illuminance sensor, or a noise sensor.

  Moreover, you may judge a subject's risk degree by acquiring the vibration of a residence with the detection output of the equipment state detection sensors 56, such as a vibration sensor. Further, after the vibration has subsided, the action of the subject is acquired again by the detection output of the human sensor 51, and if the movement of the subject can be confirmed, it is determined that there is little possibility of injury, and the subject is placed in a specific place. If it stays and the movement is weak, it may be determined that the possibility of injury is high.

  In addition, the monitoring apparatus 1 is good also as a structure by which the detection output of various sensors is input via a repeater. Also, by providing a plurality of repeaters, the area over which the subject is watched can be widened.

  Further, the watching device 1A may be a server device on a network as shown in FIG. The communication unit 40 shown in FIG. 4 is installed in a subject's residence, and the detection output of various sensors is input, It transmits to the monitoring apparatus 1A via the network 41 of the input detection output of various sensors. The monitoring device 1A shown in FIG. 4 has substantially the same configuration as the monitoring device 1 shown in FIG. 1, but differs in that the sensor connection unit 3 is a communication unit (not shown) via a network. This communication unit controls communication with the communication unit 40.

  Moreover, the communication unit 40 may be provided with the structure concerning the monitoring apparatus 1 shown in FIG. In this case, the monitoring device 1 (this monitoring device 1 is provided in the communication unit 40) and the monitoring device 1A may share the items for determining the state of the subject. Moreover, it is good also as a structure which makes the method of judging the state of a subject differ in the monitoring apparatus 1 and the monitoring apparatus 1A, and the monitoring apparatus 1 and the monitoring apparatus 1A judge a subject's state separately. When the monitoring device 1 and the monitoring device 1A are configured to individually determine the state of the subject, when the state of the subject is determined to be inappropriate in at least one of the monitoring device 1 and the monitoring device 1A, the above-mentioned It is preferable that the notification process related to s8 is executed.

DESCRIPTION OF SYMBOLS 1, 1A ... Watching device 2 ... Control part 3 ... Sensor connection part 4 ... Display / operation part 5 ... Determination information storage part 6 ... Communication part 21 ... State detection part 22 ... Influence degree acquisition part 23 ... Determination part 31 ... Detection information 32 ... Event related information 33 ... Detection item information 34 ... Influence level information 35 ... Determination criteria information 51 ... Human sensor 52 ... Biosensor 53 ... Environmental sensor 54 ... Position sensor 55 ... Equipment state detection sensor 56 ... Equipment state detection sensor

Claims (13)

A state detection unit that detects the state of the target person;
An impact level acquisition unit that calculates the impact level of the influencing factors that affect the condition of the target person,
A monitoring device comprising: a determination unit that determines whether or not the state of the subject detected by the state detection unit is appropriate using the influence level of the influence factor acquired by the influence level acquisition unit. The sensor is connected, and a sensor connection part to which the detection output of the connected sensor is input is provided.
The monitoring apparatus according to claim 1, wherein the state detection unit detects the state of the subject using a detection output of a sensor connected to the sensor connection unit.   The monitoring apparatus according to claim 2, wherein the influence degree acquisition unit acquires the influence degree of the influence factor using a detection output of a sensor connected to the sensor connection part.   The monitoring device according to any one of claims 1 to 3, wherein the influence degree acquisition unit acquires the influence degree of the influence factor based on contents of an event related to the subject.   The monitoring device according to any one of claims 1 to 4, further comprising an output unit that outputs a message to the effect that the determination unit determines that the state of the subject detected by the state detection unit is not appropriate.   The monitoring device according to claim 1, wherein the influencing factor that affects the sleep state of the subject includes at least one of a behavior of the subject before sleep or a behavior schedule after waking up.   The monitoring device according to any one of claims 1 to 6, wherein the influencing factors affecting the bathing of the subject include at least one of the behavior of the subject and the climate of the day.   The monitoring device according to any one of claims 1 to 7, wherein the influence factor that affects the cooking behavior of the subject includes at least one of the behavior of the subject and the behavior schedule of the day.   The monitoring device according to claim 1, wherein the influential factor that affects the use of the toilet of the subject is the behavior of the subject that day.   The monitoring device according to any one of claims 1 to 9, wherein the influence factor that affects the removal of the posting from the mailbox of the subject is the behavior of the subject of the day.   The monitoring device according to claim 1, wherein the influential factor that affects the use of the air conditioner of the subject is the climate of the day. In the state detection unit, a state detection step for detecting the state of the target person,
In the influence level acquisition unit, an influence level acquisition step of acquiring the influence level of the influencing factors that affect the state of the target person,
The control unit determines whether the determination unit determines whether or not the state of the subject detected by the state detection unit is appropriate using the degree of influence of the influence factor acquired by the influence level acquisition unit. How to watch over. In the state detection unit, a state detection step for detecting the state of the target person,
In the influence level acquisition unit, an influence level acquisition step of acquiring the influence level of the influencing factors that affect the state of the target person,
A determination unit for determining whether or not the state of the subject detected by the state detection unit is appropriate by using the influence degree of the influence factor acquired by the influence degree acquisition unit; Watch program.

Images