JP7686548B2 - Behavior Recognition System - Google Patents

Description

The present invention is a technology related to a system that recognizes the behavior of people living in a residential environment by combining detection information from sensors such as home appliances and human presence sensors installed in the residential environment with sensors that detect and identify people.

For example, Patent Document 1 is an example of a conventional technology that uses operation information and sensor information from home appliances to control devices according to the status of a user residing in a living environment and to grasp the user's activity status. Patent Document 1 discloses a remote care system for apartment buildings that uses white goods used in general households to provide and operate remote care, in which home appliances installed in living spaces that make up an apartment building are equipped with detection units that detect the behavior of residents in the living spaces, and home appliances installed in multiple living spaces are remotely controlled and remotely monitored based on the detection information provided by the detection units on the home appliances.

JP 2017-011698 A

In a system for remotely checking the safety of residents, in a typical household, multiple residents share the living space, so the behavioral information needed to determine the safety of residents requires the detection and identification of each resident's individual behavior. However, individualizing the behavioral information of residents requires the installation of multiple image, audio, and other sensors in various locations within the living space, which creates issues with ensuring the privacy of residents, as well as installation and operating costs.

For example, in a two-family home, if a safety confirmation system that identifies behavioral information based solely on motion sensors and home appliance operation information is introduced, when multiple residents share their behavior in a shared space such as the dining room or living room, and one person remains in the shared space while the rest move to another living space, it will be impossible to distinguish or identify which resident's behavior the motion sensor detection information comes from. For this reason, if a safety confirmation system that identifies behavioral information based solely on motion sensors and home appliance operation information is introduced in a living space where a living environment is shared by two or more people, it will be impossible to detect or identify the individual behavior of the residents. On the other hand, if multiple image or audio sensors are installed in the living environment to solve the above problems, issues will arise in terms of ensuring privacy and the costs of installation and operation.

The present invention has been made to address the above-mentioned problems, and aims to provide a behavior recognition system that can accurately detect and classify the behavioral information of individuals in a living space occupied by multiple people, even when the number of human recognition sensors installed is limited to ensure privacy.

In order to solve the above problems, the behavior recognition system of the present invention has a sensor group including multiple human presence sensors that detect people and multiple human recognition sensors that detect people and identify the people, and a behavior estimation device that recognizes the behavior of the person identified by the human recognition sensors based on detection information received from the sensor group. The behavior estimation device has a memory unit that stores at least movement information related to the path a person moves from an area where at least one of the sensor group is installed to another area where at least one other of the sensor group is installed, and a calculation unit that performs a time series analysis of the information detected by the sensor group, creates candidate movement path information along which the person to be recognized may move based on the analysis result and the movement information, and recognizes the behavior of the person identified by the multiple human recognition sensors based on the candidate movement path information and information detected by the multiple human recognition sensors.

According to the present invention, by connecting a plurality of human presence sensors and human recognition sensors installed in a residential environment to a behavior estimation device installed in the residential environment, it becomes possible to determine the behavior of each resident living in the target residential environment by an estimation unit that estimates the behavior of each resident. Furthermore, since human behavior information is created by utilizing detection information from human presence sensors and home appliances, the number of sensors that recognize people can be reduced, and therefore a human behavior recognition system that takes user privacy into consideration can be provided.
Further features related to the present invention will become apparent from the description of the present specification and the accompanying drawings. Furthermore, the objects, configurations and effects other than those described above will become apparent from the following description of the embodiments.

1 is a schematic diagram showing a configuration of a behavior recognition system according to the present invention. FIG. 1 is a diagram showing an example of installing a plurality of sensors in a living environment. 3A to 3C are diagrams for explaining various types of information used in the behavior recognition system of the present invention. FIG. 4 is a flow diagram showing a procedure for processing information detected by a sensor in the behavior recognition system of the present invention. 3A to 3C are diagrams for explaining behavior detection information and individual behavior detection information used in the behavior recognition system of the present invention. FIG. 4 is a flow diagram showing a process of identifying movement information of an individual based on information detected by a sensor in the behavior recognition system of the present invention. FIG. 4 is a diagram for explaining travel route candidate information.

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows an outline of the configuration of the behavior recognition system of the present invention. The behavior recognition system 1000 of the present invention has a sensor group including multiple human presence sensors 100 (low-cost, simple human presence sensors or home appliances) that detect people, multiple human recognition sensors 200 (image, audio, ToF, acceleration sensors, etc. that can detect and identify people), and a behavior estimation device 300 that identifies and recognizes people and their behavior based on the detection information of the sensor group. In this embodiment, the behavior recognition system 1000 further has a remote management unit 400 that can receive information from the behavior estimation device 300 and remotely confirm the safety and condition of the resident being detected.

The multiple human presence sensors 100 each have a power supply unit 110 for activating the human presence sensor 100, a communication unit 120 for transmitting detection information of the human presence sensor 100 and the unique ID of the human presence sensor 100 to the behavior estimation device 300, a memory unit 130 for storing the unique information of the human presence sensor 100, a detection unit 140 for detecting people, and a control unit 150 for controlling the operation of the human presence sensor 100.

Similarly, each of the multiple human recognition sensors 200 has a power supply unit 210 for activating the human recognition sensor 200, a communication unit 220 for transmitting the detection information of the human recognition sensor 200 and the sensor's unique ID to the behavior estimation device 300, a memory unit 230 for storing the unique information of the human recognition sensor 200, a detection unit 240 for detecting and identifying people, and a control unit 250 for controlling the operation of the human recognition sensor 200.

The behavior inference device 300 further includes a communication unit 310, an external communication unit 320, a control unit 330, a calculation unit 340, a memory unit 350, a timer 360, and an external power supply 370. The behavior inference device 300 is configured as a server, a PC, or the like.

The communication unit 310 receives detection information and sensor-specific IDs transmitted from the multiple human presence sensors 100 and multiple human recognition sensors 200. The external communication unit 320 receives and transmits detection information and analysis information to an external network other than the sensor network constructed by the multiple human presence sensors 100 and multiple human recognition sensors 200 and the behavior estimation device 300. The control unit 330 analyzes information from the sensor network, external network, etc., and controls the equipment based on the analysis results. The calculation unit 340 classifies the behavior of each resident based on input information from the external communication unit 320, the detection information of the control unit 330, the multiple human presence sensors 100, and the multiple human recognition sensors 200, spatial movement information, residence information, etc. described below, and determines the safety of the resident. The memory unit 350 stores movement route information that determines the behavior of each resident determined by the calculation unit 340. The timer 360 evaluates the time when the detection information of the multiple human presence sensors 100 and multiple human recognition sensors 200 is received. In addition, the remote management unit 400 can send and receive information regarding the safety of residents analyzed by the behavior inference device 300 via the external communication unit 320.

By constructing a behavior recognition system with the above configuration, it is possible to transmit detection information of residents detected by multiple human presence sensors 100 and multiple human recognition sensors 200 installed in the living environment and the unique ID of the sensor to the behavior estimation device 300. The calculation unit 340 of the behavior estimation device 300, which receives the detection information and sensor ID, manages the reception time using the timer 360, and if the sensor that transmitted the detection information was the human presence sensor 100, the calculation unit 340 can determine the room in which the sensor reacted (the place where the behavior occurred) and the time from the sensor ID and sensor installation information, and store this in the memory unit 350 as behavior detection information.

Similarly, the calculation unit 340 recognizes the type of sensor from the sensor's unique ID, and if the sensor that transmitted the detection information was the human recognition sensor 200, the calculation unit 340 can identify the resident being detected by comparing it with the resident information and store it as personal detection information in the memory unit 350.

The calculation unit 340 identifies the movement information of an individual by analyzing the movement route of the individual using the personal detection information and behavior detection information stored in the memory unit 350, as well as the residence information and spatial movement information. The memory unit 350 can then store the movement route of the identified individual. By monitoring this stored movement route, if detection information that deviates from the daily movement information of an individual is detected, it can be determined that some abnormality has occurred in the individual being detected, and the abnormality information can be transmitted to the remote management unit 400 via the external communication unit 320.

Figure 2 is a diagram showing an overview of the behavior recognition system 1000 configured as shown in Figure 1 when it is introduced into a living environment including a living space 2000. The human presence sensor 100 is composed of a human presence sensor that detects the presence or absence of a person, a door opening/closing sensor that detects the opening/closing of a door, an illuminance sensor that detects the brightness of the room, and the like. These sensors simply detect the presence or absence of a person, and do not identify personal information. Therefore, there are no restrictions on where they can be installed within the living space 2000 so as not to violate the privacy of the resident living in the detection target. In Figure 2, the human presence sensor 100 is installed in the living room L and dining room D.

On the other hand, the human recognition sensor 200 includes sensors that acquire personal information such as images, voice, and ToF to identify individuals, as well as sensors that measure the acceleration of door opening and closing. Since these sensors detect information that includes the residents' privacy, privacy must be taken into consideration when installing them. For example, it is preferable to install them in places that are relatively common, such as entrances and hallways. In FIG. 2, the human recognition sensor 200 is installed at the entrance E, etc.

Note that while FIG. 2 illustrates a robot vacuum cleaner 500 as a representative home appliance, any home appliance that handles image information, such as a robot vacuum cleaner, can be used as the human recognition sensor 200. Furthermore, when sensors or operation information of a refrigerator, washing machine, air conditioner, etc. are used as sensor detection information, these home appliances can be used as the human presence sensor 100. The behavior estimation device 300 and remote management unit 400 are not shown in the figure.

Note that FIG. 2 is an example showing an outline of a method for installing the human presence sensor 100 and the human recognition sensor 200 in the living space 2000 in the present invention, and the method for installing the human presence sensor 100 and the human recognition sensor 200 in the present invention is not limited to that shown in FIG. 2.

The process steps for detecting and recognizing the behavioral information of each target resident in the behavior recognition system 1000 configured as described above will be explained with reference to Figures 3 to 7.

First, various types of information used in the behavior recognition system 1000 according to the present invention will be described. A plurality of human presence sensors 100 and a plurality of human recognition sensors 200 are installed in the residential space 2000 to be detected, and basic information on the residents, the house, and the sensors is registered in advance in the memory unit 350 of the behavior estimation device 300. Information to be registered includes, for example, resident information, room information, sensor installation information, residence information, spatial movement information, etc. An example of this information will be described with reference to FIG. 3.

As shown in FIG. 3, resident information is constructed from the resident's name, a personal ID for distinguishing one resident from another, and feature data related to the resident's features for identifying an individual using the human recognition sensor 200. Features for identifying an individual include information that can be detected by various sensors, such as height, physique, posture, walking style, dominant hand, images, and sounds. Room information is constructed from the room name and room ID. Sensor installation information is information that manages the installation information of the human presence sensor 100 and human recognition sensor 200, and is constructed from the sensor ID, the room ID where it is installed, and the type of sensor (0: human recognition sensor, 1: non-human recognition sensor).

The residence information is a database of the spaces where the resident of the detection target is usually present/absent, and is composed of the personal ID, characteristic date and time information (for example, day of the week or weekend, etc., in this embodiment, the cases are divided into WD (weekday)/WE (weekend), but the case division is not limited to this embodiment), time period (for example, in this embodiment, the cases are divided into 0: morning (4:00-10:00), 1: afternoon (10:00-16:00), 2: night (16:00-22:00), 3: late night (22:00-4:00), but the division method is not limited to this embodiment), and space utilization status (cases are divided into 0: present/1: absent/-: unknown, but the case division is not limited to this embodiment).

Spatial movement information (movement information) is information that shows a list of ON and OFF sensor ID strings for multiple sensors that react when a person moves from one room to another, and the movement path can be estimated by comparing the spatial movement information with the sensor detection information, as described below.

The above basic information is registered in the storage unit 350 of the behavior estimation device 300, and the behavior recognition system 1000 is operated. FIG. 4 is a flow diagram showing the process in which the behavior recognition system 1000 receives a detection signal from the human presence sensor 100 or the human recognition sensor 200, and then stores the signal in the storage unit 350 as behavior detection information or individual detection information.

When the behavior recognition system 1000 is operational, it first proceeds to operation step S100. After proceeding to operation step S100, the behavior recognition system 1000 determines whether either the human presence sensor 100 or the human recognition sensor 200 has detected the behavior of a resident. If either the human presence sensor 100 or the human recognition sensor 200 has detected the behavior of a resident and the sensor has responded, it proceeds to operation step S110, where it proceeds to a process of selecting/identifying sensor information.

If the sensor does not respond, the operation step proceeds to S200, and it is determined whether the operating time (T) from the start of detection of the sensor response is equal to or greater than a predetermined elapsed time (Tan). The predetermined elapsed time (Tan) can be set to, for example, 6 hours. If the operating time (T) is equal to or greater than the elapsed time (Tan) in operation step S200, the operation step proceeds to S210 in FIG. 6, and the process proceeds to estimation of a moving person. This process will be described later. If the operating time (T) < the elapsed time (Tan), the elapsed time is added to T + ΔT, and the operation step returns to S100.

The behavior recognition system 1000 moves to operation step S110, selects the sensor information, and moves to the operation step of accumulating the detection data according to the sensor information. After reading the sensor installation information in operation step S110, the behavior recognition system 1000 moves to operation step S120.

In operation step S120, the behavior recognition system 1000 compares the detection sensor ID obtained from the detection information received from the human presence sensor 100 or the human recognition sensor 200 with the sensor installation information, and determines whether the sensor that transmitted the detection information is the human recognition sensor 200 or not (the human presence sensor 100). In operation step S120, if the sensor transmitting the detection information is not the human recognition sensor 200, that is, if it is determined that it is the human presence sensor 100, the behavior recognition system 1000 proceeds to operation step S121, creates behavior detection information, proceeds to operation step S160, and stores the behavior detection information in the memory unit 350 of the behavior estimation device 300.

Here, the behavior detection information is a data string consisting of a room ID and a detection time, as shown in FIG. 5. In other words, the behavior detection information does not contain information that can identify a person, but does contain information regarding the place and time at which the sensor reacted. If it is determined in operation step S120 that the sensor emitting the detection information is the human recognition sensor 200, the behavior recognition system 1000 transitions to operation step S130.

After the behavior recognition system 1000 shifts the process to operation step S130, it reads the resident information from the memory unit 350 of the behavior estimation device 300 and then shifts the process to operation step S140.

The behavior recognition system 1000, which has moved to operation step S140, compares the detection information received from the human recognition sensor 200 with the individual characteristic data registered in the resident information. If there is no individual characteristic data that matches the detection information, the behavior recognition system moves to operation step S121 and stores the information in the memory unit 350 as behavior detection information. If there is individual characteristic data that matches the detection information, the behavior recognition system 1000 moves to operation step S150.

When the behavior recognition system 1000 transitions to the behavior step at S150, it determines that the detection information is personal behavior information, and creates personal detection information and behavior detection information. As shown in FIG. 5, the personal detection information is configured as a data string that is made up of a room ID, a personal ID, and a detection time. In other words, the personal detection information includes information about the identified individual in addition to the behavior detection information.

After the behavior recognition system 1000 creates individual detection information and behavior detection information in operation step S150, or creates behavior detection information in operation step S121, it transitions to operation step S160. Then, after storing the behavior detection information and individual detection information in the storage unit 350 of the behavior estimation device 300, it transitions to operation step S200 again, and when the operating time (T) is equal to or greater than the elapsed time (Tan), it transitions to operation step S210 and transitions to the moving person estimation process, which is the process of estimating movement information and identifying the individual held by the movement information, as shown in FIG. 6.

In the operation step S210, the behavior recognition system 1000 reads the spatial movement information from the storage unit 350, and then compares the behavior detection information divided by the analysis time width w with the spatial movement information to create room-to-room movement information. Here, the room-to-room movement information is information about the room (the room in which the reacting sensor is installed) to which a person moved during a certain time period included in the behavior detection information. Specifically, for example, in the behavior detection information of FIG. 5, a person moves from room R1 to room R2 between time t1 and time t2. With reference to this and the spatial movement information of FIG. 3, it can be seen that there are two routes for movement from room R1 to room R2 as the reaction sensor sequence: S1 → S2 and S1 → S6 → S5 → S2. In other words, the room-to-room movement information is information about these two routes. This room-to-room movement information is calculated according to the detection time included in the behavior detection information.

The created inter-room movement information is combined to create candidate movement route information, and the operation step moves to S220. As shown in FIG. 7, the candidate movement route information is information that links together inter-room movement information that occurred within an elapsed time (Tan), and each piece of candidate movement route information is arranged in chronological order based on the pair of the room ID to which the movement was made and the time when the movement was completed.

After moving to operation step S220, the behavior recognition system reads the residence information from the memory unit 350, and then uses the presence/absence/unknown information for each room ID for each personal ID included in the personal detection information and the residence information to determine the room ID and weighting coefficient (w) for the time indicated by each candidate travel route information in operation steps S235 to S265.

Here, the weighting coefficient (w) is an index that indicates the likelihood that the candidate route is a route taken by a person identified by a personal ID, and is calculated using the following flow.

The behavior recognition system 1000, which has moved to operation step S230, determines whether there is personal detection information that includes a room ID and detection time that match the room ID and inter-room movement time indicated in the movement route candidate information. Here, the inter-room movement time refers to, for example, the time between t1 and t2 for the movement from R5 to R2 in #1 of the movement route candidate information in FIG. 7. This is compared with the personal detection information to determine whether there is detection information that indicates a presence in room R2 between times t1 and t2. This process is performed for all route candidates.

If it is determined in S230 that there is individual detection information that matches the candidate travel route information, the behavior recognition system 1000 proceeds to operation step S235, sets the weighting coefficient to +1, and proceeds to operation step S270. If there is no matching individual detection information in S230, the behavior recognition system proceeds to operation step S240.

The behavior recognition system 1000, which has moved to operation step S240, determines whether there is a room ID that matches the room ID and room movement time indicated in the movement route candidate information and matches the "presence" in the residence information. If there is a room ID that matches the presence in the residence information DB, the behavior recognition system 1000 moves to operation step S245, sets the weighting coefficient to +1, and moves to operation step S270. For example, in #1 of the movement route candidate information in FIG. 7, room R2 at time t2 is listed as a candidate. With reference to the residence information in FIG. 3, for example, if this time t2 is TimeZone: 2 on Day: WD, the person specified by personal ID: P1 corresponds, so a weighting of +1 is applied as a candidate for this person's behavior. If there is no room ID that matches the presence in the residence information in operation step S240, the behavior recognition system moves to operation step S250.

The behavior recognition system 1000 that has moved to the operation step S250 determines whether there is a room ID that matches the room ID and room movement time shown in the movement route candidate information and matches the "absent" in the residence information DB. If there is a room ID that matches the absence in the residence information DB, the behavior recognition system 1000 moves the operation step to S255, sets the weighting coefficient to -1, and moves the operation step to S270. For example, in the residence information in FIG. 3, if there is sensor detection information for room ID: R1 on Day: WD, Time Zone: 1, this is a time period in which the person identified by personal ID: P1 is "absent," and therefore the possibility of this person being a route candidate is low, so a weighting of -1 is applied. If there is no room ID that matches the absence in the residence information DB, the behavior recognition system 1000 moves the operation step to S265, sets the weighting coefficient to 0, and moves the operation step to S270.

The behavior recognition system 1000, which has moved to the operation step S270, calculates an occupancy score using the occupancy time and weighting factor in each room ID of each candidate movement route, and moves the operation step to S280. The occupancy score S _i is obtained by multiplying the weighting factor w (RID _k ) for each room by the occupancy time (T _k+1 -T _k ) in the room, and adding the value obtained from the measurement start time to the measurement end time.

After the behavior recognition system 1000 advances to the operation step S280, it determines whether or not calculations have been performed for all of the candidate movement routes. If calculations for all of the occupancy scores have been completed, it advances to the operation step S290. If calculations for all of the occupancy scores have not been completed, it advances to the operation step S220.

The behavior recognition system 1000, which has shifted the operation step to S290, determines the candidate movement route with the highest presence score among the results of the weighting calculation for each candidate movement route for the individual ID to be calculated as the movement route of that individual, and shifts the operation step to S300. The behavior recognition system 1000, which has shifted the operation step to S300, determines whether the movement routes of all residents of the living environment to be detected have been determined, and if the movement routes of all residents have been determined, shifts the operation step to S310.

If the movement routes of all residents have not been determined, the operation step proceeds to S220. After proceeding to S310, the behavior recognition system 1000 stores the movement information of the individual subject to detection in the memory unit 350 of the behavior estimation device 300, resets the operating time (T) to 0, and then proceeds to S100.

According to the embodiment of the present invention described above, the following advantageous effects are obtained.
(1) A behavior recognition system according to an embodiment of the present invention includes a sensor group including a plurality of human presence sensors that detect a person and a plurality of human recognition sensors that detect the person and identify the person; and a behavior estimation device that recognizes the behavior of the person identified by the human recognition sensors based on detection information received from the sensor group, wherein the behavior estimation device includes a memory unit that stores at least movement information related to a route along which a person moves from an area in which at least one of the sensor group is installed to another area in which at least one other of the sensor group is installed, and a calculation unit that performs a time series analysis of the information detected by the sensor group, creates candidate movement route information along which the person to be recognized may move based on a result of the analysis and the movement information, and recognizes the behavior of the person identified by the multiple human recognition sensors based on the candidate movement route information and information detected by the multiple human recognition sensors.

With the above configuration, by connecting multiple sensors and a human presence sensor installed in the living environment to a behavior estimation device installed in the living environment, it becomes possible to determine the behavior of residents by an estimation unit that estimates the behavior of each resident living in the target living environment. Furthermore, since human behavior information is created using detection information from human presence sensors and home appliances, the number of sensors that recognize people can be reduced, making it possible to provide a human behavior recognition system that takes user privacy into consideration.

(2) The behavior estimation device further has an external communication unit that transmits information about human behavior to outside the area where the sensor group is installed via a network. This makes it possible to notify an external monitoring device of an abnormality in a resident when it is determined that there is an abnormality in the resident's movement information.

(3) The memory unit further stores personal information related to features for identifying the person being detected, and the human recognition sensor identifies the person by comparing the detection content with the personal information. This makes it possible to use information that can be detected by various sensors, such as height, physique, posture, walking style, dominant hand, images, and audio, and therefore to identify individuals from various angles, improving the accuracy of identification.

(4) The memory unit further stores a record of the person's behavior within the area where the sensor group is installed, and the calculation unit recognizes the person's behavior based on the behavior record in addition to the information detected by the multiple human recognition sensors and the candidate movement route information. This makes it possible to recognize the person's behavior based on the pre-stored behavior record (e.g., which room belongs to whom, etc.) even if the person cannot be identified based on the features related to the person.

(5) Before referring to the behavioral records, the calculation unit refers to the information detected by multiple human recognition sensors to recognize human behavior. This allows the processing to be prioritized, which is expected to reduce the load on calculation processing and increase its speed.

(6) When the calculation unit is able to recognize human behavior by referring to information detected by multiple human recognition sensors, it assigns positive weighting and does not refer to the behavior record, and when the calculation unit is unable to recognize human behavior by referring to information detected by multiple human recognition sensors, it recognizes human behavior by referring to the behavior record without weighting. This makes it possible to quantitatively recognize human behavior based on detected information, improving recognition accuracy.

(7) The calculation unit performs positive weighting when it is able to recognize human behavior by referring to the behavior record, and performs negative weighting when it is unable to recognize human behavior. This makes it possible to quantitatively recognize behavior as in (6), and furthermore, by performing negative weighting, it becomes possible to eliminate information due to erroneous detection by the sensor.

The present invention is not limited to the above-mentioned examples, and various modifications are possible. For example, the above-mentioned examples have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to an embodiment that includes all of the configurations described. It is also possible to replace part of the configuration of one example with the configuration of another example. It is also possible to add the configuration of another example to the configuration of one example. It is also possible to delete part of the configuration of each example, or to add or replace other configurations.

100 Human presence sensor, 200 Human recognition sensor, 300 Behavior estimation device, 320 External communication unit, 340 Calculation unit, 350 Memory unit

Claims (7)

An activity recognition system for recognizing human activities,
A sensor group including a plurality of human presence sensors that detect people and a plurality of human recognition sensors that detect people and identify the people;
a behavior inference device that recognizes a behavior of a person identified by the human recognition sensor based on detection information received from the group of sensors,
The behavior estimation device includes:
a storage unit that stores at least movement information regarding a route along which a person moves from an area in which at least one of the sensor groups is installed to another area in which at least another of the sensor groups is installed;
a calculation unit that performs a time series analysis of information detected by the group of sensors, creates candidate movement route information along which the person to be recognized may move based on a result of the analysis and the movement information, and recognizes the behavior of the person identified by the plurality of human recognition sensors based on the candidate movement route information and information detected by the plurality of human recognition sensors;
The behavior recognition system is characterized by: The behavior recognition system according to claim 1 ,
the behavior inference device further includes an external communication unit that transmits information about the human behavior to an area outside an area where the group of sensors is installed via a network.
The behavior recognition system is characterized by: The behavior recognition system according to claim 1 ,
The storage unit further stores person information related to a feature amount for identifying a person to be detected,
The human recognition sensor identifies the person by comparing the detection content with the person information.
The behavior recognition system is characterized by: The behavior recognition system according to claim 1 ,
The storage unit further stores a behavior record of the person to be detected within the area in which the group of sensors is installed,
The calculation unit recognizes the behavior of the person based on the behavior record in addition to the information detected by the plurality of human recognition sensors and the movement route candidate information.
The behavior recognition system is characterized by: The behavior recognition system according to claim 4,
the calculation unit recognizes the behavior of the person by referring to information detected by the plurality of human recognition sensors before referring to the behavior record;
The behavior recognition system is characterized by: The behavior recognition system according to claim 5,
the calculation unit performs positive weighting and does not refer to the behavior record when the behavior of the person can be recognized by referring to the information detected by the plurality of human recognition sensors, and recognizes the behavior of the person by referring to the behavior record without performing weighting when the behavior of the person cannot be recognized by referring to the information detected by the plurality of human recognition sensors.
The behavior recognition system is characterized by: The behavior recognition system according to claim 6,
the calculation unit performs positive weighting when the action of the person can be recognized by referring to the action record, and performs negative weighting when the action of the person cannot be recognized.
The behavior recognition system is characterized by:

Images