Behavior Attribute Analysis System and Information Processing Apparatus

Abstract

To provide an action attribute analysis system and an information processing device which can analyze a human flow for each pattern of an action of a person.SOLUTION: An action attribute analysis system comprises: a detection device which is installed in a plurality of spots, and acquires user identification information from a communication apparatus carried by a person in a predetermined range; a collection device which collects the user identification information acquired by the detection device, detection identification information for identifying the detection device, and information on date and time when the user identification information was acquired, as collection information; and an information processing device which acquires the collection information. The information processing device comprises: a storage unit which stores correspondence information between the detection identification information and the type of the spot; and an action attribute giving unit which gives an action attribute which is combination of the type of the action of the person at the spot, and the type of the spot, for each user identification information, based on the collection information and the correspondence information.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a behavior attribute analysis system and an information processing apparatus for analyzing the flow of people.

Background Art

[0002] Conventionally, techniques for analyzing the movement path of people have been proposed. For example, Patent Document 1 discloses a data analysis apparatus that analyzes sensor data from sensors installed at a plurality of points. This data analysis apparatus includes a data reception unit that receives sensor data via a wireless network, and a data analysis unit that aggregates the movement amounts between two points in detail based on the sensor data received by the data reception unit and analyzes a series of movement paths by following the path with a large movement amount between two points.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] According to the data analysis apparatus described in Patent Document 1, it is possible to display analysis results such as the movement amount between two points, the total detection count, and the detection count for each spot. However, the technique described in Patent Document 1 has a problem that it is not possible to analyze the flow of people according to different patterns of human behavior.

[0005] The present invention has been made to solve the above problems, and an object thereof is to provide a behavior attribute analysis system and an information processing apparatus capable of analyzing the flow of people according to different patterns of human behavior.

Means for Solving the Problems

[0006] The behavior attribute analysis system according to the present invention includes a detection device installed at a plurality of locations for acquiring user identification information from communication devices carried by people within a predetermined range, a collection device for collecting, as collection information, the user identification information acquired by the detection device, detection identification information for identifying the detection device, and information on the date and time when the user identification information was acquired, and an information processing device for acquiring the collection information. The information processing device includes a storage unit for storing correspondence information between the detection identification information and the type of the location, and based on the collection information and the correspondence information, A plurality of the type of behavior of the person at the location, A plurality of a behavior attribute assignment unit that assigns, for each user identification information, a behavior attribute that is a combination of the type of behavior of the person at the location and the type of the location, and the type of the location includes a type corresponding to the demographic attribute of the person .

Effects of the Invention

[0007] According to the present invention, based on the collection information and the correspondence information, a behavior attribute that is a combination of the type of behavior of a person at a location and the type of the location is assigned for each user identification information, so that the flow of people can be analyzed according to different patterns of people's behavior.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Embodiments for Carrying Out the Invention

[0009] Embodiment 1. The behavior attribute analysis system 1 of the present Embodiment 1 analyzes the flow of people in a specific area such as a commercial facility or an airport. Note that the area where the behavior attribute analysis system 1 analyzes the flow of people can be applied to any area regardless of whether it is indoors or outdoors. For example, it may be applied to an outdoor area such as a shopping street or a theme park.

[0010] (System Configuration) FIG. 1 is a schematic configuration diagram of the behavior attribute analysis system 1 according to Embodiment 1. As shown in FIG. 1, the behavior attribute analysis system 1 includes a plurality of detection devices 10, a collection device 20, and an information processing device 100.

[0011] The detection device 10 is installed at a plurality of points within the area to be analyzed for the flow of people, and communicates with the communication device 2 carried by a person P located within a predetermined range. The communication device 2 is an information processing terminal such as a mobile phone including a smartphone, a tablet terminal, a watch-type information terminal, a notebook PC (Personal Computer), etc.

[0012] The detection device 10 acquires user identification information included in the communication signal with the communication device 2. The communication between the detection device 10 and the communication device 2 is, for example, a beacon signal, and the user identification information is the information included in the beacon signal. Specifically, the detection device 10 is composed of, for example, a Wi-Fi packet sensor, a Wi-Fi access point, or a PC capable of Wi-Fi communication. The detection device 10 receives the Probe Request packet transmitted from the communication device 2, and acquires, as user identification information, unique information for identifying the communication device 2, such as the MAC address (Media Access Control address) or SSID (Service Set Identifier) included in the Probe Request packet. Note that Wi-Fi is a registered trademark.

[0013] The collection device 20 is composed of a PC or a server, and is communicably connected to a plurality of detection devices 10 via a network, either wired or wirelessly. The collection device 20 collects, as collection information, the user identification information acquired by the detection device 10, the detection identification information for identifying the detection device 10, and the information on the date and time when the user identification information was acquired. Further, the collection device 20 is communicably connected to the information processing device 100 via a network, either wired or wirelessly, and transmits the collection information to the information processing device 100 periodically or each time.

[0014] The information processing device 100 is composed of a PC or a server, and analyzes the flow of people. Details will be described later.

[0015] FIG. 2 is a schematic layout diagram showing an installation example of the detection device 10 of the behavior attribute analysis system 1 according to Embodiment 1. In this Embodiment 1, a case where the behavior attribute analysis system 1 is applied to the analysis of the indoor pedestrian flow in a complex commercial facility having a plurality of retail stores, restaurants, entertainment facilities, etc. will be described as an example.

[0016] As shown in FIG. 2, the detection device 10 is installed at a plurality of locations within the commercial facility. The detection device 10 can communicate with the communication device 2 located within a predetermined range from the installed location. In FIG. 2, the predetermined range within which the detection device 10 can communicate with the communication device 2 is indicated by a dotted circle. This predetermined range includes, for example, the range of the store area of each retail store or restaurant. Also, the predetermined range includes, for example, the ranges of facilities such as the entrances and exits of the commercial facility, men's toilets, women's toilets, baby rooms, kids spaces, escalators, and elevators.

[0017] In this way, the detection device 10 is installed at a plurality of locations within the commercial facility that have different uses and purposes. Hereinafter, the type or classification of the use and purpose of the location where the detection device 10 is installed will be referred to as the "type of location". Here, the type of location where the detection device 10 is installed includes types corresponding to the demographic attributes of the person P. Demographic attributes are attributes based on demographic attributes such as, for example, a person's gender, age, occupation, family composition, etc. For example, the type of location corresponding to male gender is the men's toilet. Also, for example, the type of location corresponding to the age of children is a cram school. Also, for example, the type of location corresponding to the occupation of employees of the commercial facility is the employee entrance. Also, for example, the type of location corresponding to a family with children is the kids space.

[0018] FIG. 3 is a control block diagram of the information processing apparatus 100 of the behavior attribute analysis system 1 according to Embodiment 1. As shown in FIG. 3, the information processing apparatus 100 includes an acquisition unit 110, a behavior attribute assignment unit 120, a storage unit 130, an extraction processing unit 140, an input unit 150, and an output unit 160. The acquisition unit 110, the behavior attribute assignment unit 120, and the extraction processing unit 140 are functional units realized by the CPU executing a program. The input unit 150 is an interface with an input device and receives a selection operation of a behavior attribute described later. The output unit 160 is an interface with an output device and outputs the extraction result of user identification information described later to the output device.

[0019] The storage unit 130 is composed of a non-volatile memory such as a RAM, a ROM, a flash memory, and an HDD. The storage unit 130 stores the correspondence information between the detection identification information and the type of location. This correspondence information will be described with reference to FIG. 4.

[0020] FIG. 4 is a diagram showing the data structure of the correspondence information of the behavior attribute analysis system 1 according to Embodiment 1. As shown in FIG. 4, the correspondence information has detection identification information (Sensor ID) and the type of location. The detection identification information is information for identifying the detection device 10. The type of location is the name of the use and purpose type or classification of the location where the detection device 10 is installed. Note that the type of location may be a general name indicating the use and purpose of the location, or a unique store name or facility name.

[0021] (Operation) The operation of the information processing apparatus 100 will be described. The acquisition unit 110 periodically or at any time acquires the collection information from the collection device 20 and stores the acquired collection information in the storage unit 130. Note that the acquisition unit 110 may perform data cleansing such as deleting duplicate data on the information acquired from the collection device 20. In the storage unit 130, the collection information is accumulated each time the acquisition unit 110 acquires the collection information. This collection information will be described with reference to FIG. 5.

[0022] FIG. 5 is a diagram showing the data structure of the collected information of the behavior attribute analysis system 1 according to Embodiment 1. As shown in FIG. 5, the collected information has user identification information (User ID), detection identification information (Sensor ID), and date and time information (Date Time). The user identification information is information acquired by the detection device 10 from the communication device 2. The detection identification information is information for identifying the detection device 10 that acquired the user identification information. The date and time information is the date and time information when the detection device 10 acquired the user identification information. In the following description, the location of the detection device 10 with the detection identification information "S01" may be referred to as "Location 1", and the location of the detection device 10 with the detection identification information "S02" may be referred to as "Location 2", etc.

[0023] Based on the collected information, the behavior attribute assignment unit 120 assigns the type of a person's behavior at the location where the detection device 10 is installed for each user identification information. Here, the type of a person's behavior is a type or classification of a person's characteristic behavior pattern, and is, for example, as follows.

[0024] [Stay] When the type of behavior is "stay", it means a state in which person P stays at a certain location (place) for a predetermined time or longer. When the user identification information is continuously acquired at one location for a predetermined time or longer, the behavior attribute assignment unit 120 assigns "stay" as the type of behavior. For example, the predetermined time is 5 minutes. When the type of behavior is "stay", it can be distinguished that person P is a person who uses the store or facility at that location.

[0025] [Passing] When the type of behavior is "passing", it means a state in which person P does not stay at a certain location (place) for a predetermined time or longer. When the user identification information is not continuously acquired at one location for a predetermined time or longer, the behavior attribute assignment unit 120 assigns "passing" as the type of behavior. For example, the predetermined time is 5 minutes. When the type of behavior is "passing", it can be distinguished that person P is a person who passed through the entrance, exit, or passage at that location, or a person who used a means of movement such as an elevator or escalator.

[0026] [The place where the person passed through first] The place where the type of action first passed refers to the point (place) where person P first passed on a specific day. The action attribute assignment unit 120 designates, as the type of action, the place where user identification information was first obtained among a plurality of points on a day as the "place where it first passed". When the type of action is the "place where it first passed", on a specific day, it is possible to distinguish the entrance and exit where person P entered the facility.

[0027] [Last passed place] The place where the type of action last passed refers to the point (place) where person P last passed on a specific day. The action attribute assignment unit 120 designates, as the type of action, the place where user identification information was last obtained among a plurality of points on a day as the "place where it last passed". When the type of action is the "place where it last passed", on a specific day, it is possible to distinguish the entrance and exit where person P left the facility.

[0028] [Total stay time] The type of action being the total stay time refers to the time from the time when person P first entered the facility to the time when person P last left the facility on a specific day. The action attribute assignment unit 120 calculates the time from the time when user identification information was first obtained to the time when it was last obtained on a day, and designates that time as the "total stay time" as the type of action. Based on this total stay time, on a specific day, it is possible to distinguish whether person P is a long-term visitor or a short-term visitor.

[0029] [Number of visits] The type of action being the number of visits refers to the number of days on which person P entered the facility during a specific period or a predetermined period in the past from a specific day. The action attribute assignment unit 120 extracts the number of days on which user identification information was obtained within the predetermined period, and designates it as the "number of visits" as the action attribute. Based on this number of visits, it is possible to distinguish whether person P is a first-time visitor (a casual visitor) to the facility or a person with a large number of visits (a regular visitor).

[0030] The action attribute assignment unit 120 assigns at least one of the above-described plurality of "types of actions" for each user identification information. A specific example is shown in FIG. 6.

[0031] FIG. 6 is a diagram showing an example of assigning types of actions of the action attribute analysis system 1 according to Embodiment 1. Note that the example shown in FIG. 6 describes only a part of the types of actions. Also, the example shown in FIG. 6 does not match the information illustrated in other figures.

[0032] In the example shown in FIG. 6, the cases where "staying" and "passing" are assigned as the types of actions are shown. The action attribute assigning unit 120 calculates, based on the collected information, for each combination of user identification information and detection identification information, the time during which the user identification information is continuously acquired as the staying time.

[0033] The action attribute assigning unit 120 determines whether the calculated staying time is equal to or longer than a predetermined time. If it is equal to or longer than the predetermined time, "staying" is assigned as the type of action, and if it is less than the predetermined time, "passing" is assigned. In the example shown in FIG. 6, a person P carrying the communication device 2 with User ID "jSIK" moves in the order of Sensor ID "S01", "S02", "S01", "S02", ··· [S04], that is, in the order of location 1 → location 2 → location 1 → location 2 ··· location 4, and either "staying" or "passing" is assigned as the type of action at each location.

[0034] Next, the action attribute assigning unit 120 assigns, for each user identification information, an action attribute, which is a combination of the type of a person's action at the location where the detection device 10 is installed and the type of the location, based on the collected information and the corresponding information. Specifically, the action attribute assigning unit 120 assigns, for each user identification information included in the collected information, the "type of location" corresponding to the detection identification information and at least one of the above-described multiple "types of actions". A specific example is shown in FIG. 7.

[0035] FIG. 7 is a diagram showing an example of assigning action attributes of the action attribute analysis system 1 according to Embodiment 1. Note that the example shown in FIG. 7 describes only a part of the action attributes. Also, the example shown in FIG. 7 does not match the information illustrated in other figures.

[0036] In the example shown in FIG. 7, for each User ID, if the type of action at each location corresponds to "passing through" or "staying", "TRUE" is assigned. Also, for each User ID, the number (detection identification information) of the location corresponding to the "first location passed through" and the "last location passed through" in the type of action is assigned. Further, for each User ID, the calculated value of the "total stay time" and the calculated value of the "number of visits" in the type of action are assigned. As a specific example, referring to the record in the first row of FIG. 7, Person P who carries communication device 2 with User ID "HhpD" first passes through the north entrance and enters the facility, stays at the café, 100-yen shop, and kids space, and then passes through the south entrance last and exits the facility, and the total stay time on the detection date 2022-10-1 is 125 minutes, and the number of visits so far is 7 times. The action attributes are assigned to this person.

[0037] Next, when the extraction processing unit 140 receives a selection operation of action attributes from the input unit 150, it extracts the user identification information among the plurality of user identification information whose action attributes match the selection operation of the input unit 150. A specific example will be described with reference to FIGS. 8 to 10.

[0038] FIGS. 8 and 9 are diagrams showing the input screen 200 of the action attribute analysis system 1 according to Embodiment 1. FIG. 10 is a diagram showing the output screen 300 of the action attribute analysis system 1 according to Embodiment 1. Note that the examples shown in FIGS. 8 to 10 describe a part of various information. Also, they do not match the information illustrated in other figures.

[0039] As shown in FIG. 8, the input screen 200 has a date input unit 201 for inputting a specific day or period, or day of the week to be analyzed for the flow of people, and an action type input unit 202 for inputting the type of action. The selection operation input from the date input unit 201 corresponds to a specific day or a predetermined period in the type of action described above. Also, the selection operation input from the action type input unit 202 corresponds to the type of action described above.

[0040] Figure 9 shows the state where "Places Stayed" and "Number of Visits" are selected in the action type input section 202 of the input screen 200. For example, when "Places Stayed" in the action type input section 202 is selected, the location input section 203 for selecting the type of the corresponding information point is displayed as a selection candidate. Also, for example, when "Number of Visits" in the action type input section 202 is selected, the selection candidates by number of times of the number of visits are displayed.

[0041] The user of the action attribute analysis system 1 performs an operation of selecting the conditions of the action attributes desired to be extracted from the selection items displayed on the input screen 200. For example, when it is desired to extract frequent male customers who used a ramen shop during a certain period, the corresponding period is input to the date input section 201, and as shown in Figure 9, "Ramen" and "Men's Toilet" in the "location input section 203" and the number of visits "10 times or more" in the "action type input section 202" are selected.

[0042] The extraction processing unit 140 extracts the user identification information among the plurality of user identification information whose action attributes match the selection operation of the input section 150, and outputs the extraction result to the output device via the output unit 160. The output device is arbitrary, such as display on the output screen or data output to other information processing devices.

[0043] As shown in Figure 10, for example, the extraction result is displayed on the output screen 300 of the output device such as a monitor. The output screen 300 has an action attribute display 301 for displaying the result of the selection operation of the action attributes and an extraction result display 302. The extraction result display 302, for example, displays the number of user identification information extracted by the extraction processing unit 140 for each date of the predetermined period as a bar graph. Note that the display method is not limited to this, and can be appropriately set according to the display items. As the display method, in addition to the bar graph, for example, there are a pie chart, a heat map, an OD table (Origin-Destination Table), and an information list. Also, it is not limited to the absolute number of the number of user identification information extracted by the extraction processing unit 140, and it may be the ratio to all users or the increase / decrease rate compared with a specific day.

[0044] (Effect) As described above, in the first embodiment, the action attribute assignment unit 120 assigns action attributes for each user identification information based on the collected information and the correspondence information. Therefore, it is possible to analyze the flow of people based on the action attributes caused by the behavior patterns of people.

[0045] In addition, the type of the location of the correspondence information includes a type corresponding to the demographic attributes of people. Therefore, even when there is no information regarding the demographic attributes of person P in the information acquired from the communication device 2, the demographic attributes of person P can be estimated. That is, even when the communication between the detection device 10 and the communication device 2 is a beacon signal, the demographic attributes of person P can be estimated.

[0046] In addition, the action attribute assignment unit 120 assigns, for each user identification information, the type of the behavior of people at the location where the detection device 10 is installed based on the collected information. Therefore, it is possible to distinguish the personal preferences and characteristic behavior patterns of person P.

[0047] In addition, the extraction processing unit 140 extracts the user identification information among the plurality of user identification information whose action attributes match the selection operation of the input unit 150. Therefore, it is possible to analyze the flow of people for the action attributes desired by the user of the action attribute analysis system 1.

[0048] (Modification example) In the above description, the configuration in which the detection device 10 receives the beacon signal transmitted by the communication device 2 and acquires the user identification information has been described, but the first embodiment is not limited to this. Each of the plurality of detection devices 10 may be configured to transmit a beacon signal including detection identification information. In this case, the communication device 2 receives the beacon signals from the detection devices 10 within a predetermined range and acquires the detection identification information. Then, the communication device 2 transmits the detection identification information, the user identification information of the communication device 2, and the information of the date and time when the beacon signal was received to the collection device 20.

[0049] In addition, any communication method can be applied to the communication method between the communication device 2 and the detection device 10. For example, BLE (Bluetooth Low Energy) communication may be utilized to transmit and receive beacon signals between the communication device 2 and the detection device 10. Note that Bluetooth is a registered trademark.

[0050] Embodiment 2. In this Embodiment 2, a form in which the "type of behavior" is further added will be described. Hereinafter, Embodiment 2 will be described centering on the differences from Embodiment 1.

[0051] The behavior attribute assignment unit 120 of this Embodiment 2 assigns at least one of the "type of behavior" of Embodiment 1 and the following "type of behavior".

[0052] [Visiting time zone] The type of behavior being the visiting time zone refers to the time zone to which the time when person P first entered the facility belongs on a specific day. The behavior attribute assignment unit 120 sets the time zone to which the time when the user identification information was first acquired on a day belongs as the "visiting time zone" as the type of behavior. By this visiting time zone, the behavior pattern of person P by time zone can be distinguished. Note that the time zone may be divided, for example, into morning, daytime, afternoon, etc., or may be divided by specific times such as 9:00 to 12:00, 12:00 to 16:00, etc.

[0053] [Moving route] The type of behavior being the moving route refers to the order in which person P moved through a plurality of points (places). The behavior attribute assignment unit 120 sets the order of a plurality of points at which the same user identification information was acquired as the "moving route" as the type of behavior. From this moving route, it is possible to extract in what order person P moved through each point in the facility. Also, for example, it is possible to distinguish between those who used a dedicated passage for a specific user provided in the facility and those who did not.

[0054] [Visiting interval] The type of action, the visit interval, refers to the interval between the days when person P visits the facility. The action attribute assignment unit 120 sets the interval between the days when the user identification information is acquired as the "visit interval" as the type of action. For example, by determining whether the visits are made at a specific cycle within a predetermined period, it is possible to distinguish whether the person is visiting at a specific cycle for a specific purpose. For example, it is possible to distinguish users of events or lessons regularly held within the facility.

[0055] [Event sensitivity] The type of action, the event sensitivity, refers to the ratio of the visits of person P on a specific day when an event is held at the facility or a discount measure is implemented. The action attribute assignment unit 120 sets the ratio of the day when the user identification information is acquired to the predetermined day as the "event sensitivity" as the type of action. For example, it is possible to aggregate the visit rates limited to the event implementation day or the discount implementation day.

[0056] [Movement amount] The type of action, the movement amount, refers to the quantity of person P moving between a plurality of areas by dividing a plurality of points (locations) into two or more areas on a specific day. The action attribute assignment unit 120 divides the plurality of points into two or more areas, aggregates the quantity of the areas to which the plurality of points where the same user identification information is acquired belong, and sets this quantity as the "movement amount" as the type of action. For example, it is possible to distinguish whether person P is a person with a large movement amount.

[0057] [Movement speed] The type of action, the movement speed, refers to the speed at which person P moves between a plurality of points (locations). The action attribute assignment unit 120 obtains the movement speed from the distance between the plurality of points where the user identification information is acquired and the time difference of the times when the same user identification information is acquired at the plurality of points, and sets this as the "movement speed" as the type of action. For example, depending on whether the movement speed is equal to or higher than a predetermined speed, it is possible to distinguish between a person who passed through the points quickly and a person who passed through them slowly.

[0058] [Congestion tolerant person] The type of behavior "crowding tolerant" refers to the classification of whether person P is a person who uses the facilities in the crowded area (location) even in a crowded area. The behavior attribute assignment unit 120 divides a plurality of points into two or more areas. When the user identification information is continuously acquired for a predetermined time or more in an area in a crowded state where the number of acquired different user identification information exceeds the threshold, the "crowding tolerant" as the type of behavior is assigned to the user identification information.

[0059] [Total stay time within a period] The type of behavior "total stay time within a period" refers to the total time of the stay time that person P stayed in the facility within a predetermined period. The behavior attribute assignment unit 120 sets the time obtained by summing up the above-mentioned overall stay time within a predetermined period as the "total stay time within a period" as the type of behavior.

[0060] (Effect) As described above, in the second embodiment, in addition to the type of behavior in the first embodiment, the above-mentioned type of behavior is further added. Therefore, the personal preferences and characteristic behavior patterns of person P can be further distinguished.

[0061] Embodiment 3. In this third embodiment, an operation of clustering a plurality of behavior attributes into a plurality of clusters based on similarity will be described. Hereinafter, the third embodiment will be described centering on the differences from the first and second embodiments.

[0062] FIG. 11 is a control block diagram of the information processing apparatus 100 of the behavior attribute analysis system 1 according to the third embodiment. As shown in FIG. 11, the information processing apparatus 100 includes a clustering unit 170. The clustering unit 170 is a functional unit realized by the CPU executing a program. Note that the information processing apparatus 100 may omit the input unit 150.

[0063] (Operation) The operation of the information processing apparatus 100 will be described. By operating in the same manner as in the first or second embodiment, the action attribute assigning unit 120 assigns, for each user identification information, an action attribute, which is a combination of the type of action and the type of location, based on the collected information and the corresponding information.

[0064] Next, the clustering unit 170 clusters a plurality of action attributes assigned by the action attribute assigning unit 120 into a plurality of clusters based on the degree of similarity. Specifically, the clustering unit 170 maps the action attributes to a multi-dimensional space using, as parameters, the type of action and the type of location in each user identification information, and clusters the mapped action attributes according to the proximity of distance using, for example, the k-means method. Note that the clustering method is not limited to non-hierarchical clustering such as the k-means method, and any method such as hierarchical clustering such as the shortest distance method can be used.

[0065] The extraction processing unit 140 extracts, for each cluster clustered by the clustering unit 170, the user identification information whose action attribute is included in the cluster. Further, the extraction processing unit 140 outputs the extraction result to an output device via the output unit 160. A specific example will be described with reference to FIG. 12.

[0066] FIG. 12 is a diagram showing an output screen 400 of the action attribute analysis system 1 according to the third embodiment. As shown in FIG. 12, each cluster is represented by a circle on the output screen 400. The size of the circle is displayed according to the absolute number of user identification information whose action attribute is included in the cluster or the ratio to the total number of users. Further, inside the circle, the type of action and the type of location of the clustered action attributes are described. For example, if the action attribute of a certain cluster is such that the staying place is "kids space", the staying time is "1 to 2 hours", the visiting day is "weekday", the visiting time zone is "daytime", and the number of visits is "10 times or more", these pieces of information are displayed inside the circle, and the ratio of the number of user identification information belonging to this cluster to the total number of users is displayed below the circle.

[0067] For example, as shown in FIG. 12, the display of the type of action may be converted to other terms or names, such as displaying "1 to 2 hours" of stay time as "long time" and "10 times or more" of the number of visits as "regular visitor". Also, for example, a plurality of types of actions may be grouped and displayed as one term or name. For example, "30 minutes to 1 hour" and "1 to 2 hours" of stay time may be grouped and displayed as "long time".

[0068] (Effect) As described above, in the third embodiment, the clustering unit 170 clusters a plurality of action attributes into a plurality of clusters based on the similarity, and the extraction processing unit 140 extracts the user identification information in which the action attributes are included in each cluster. Therefore, the user can know the characteristic action attributes that the user of the action attribute analysis system 1 is not aware of.

[0069] In addition, the action attributes classified into each cluster can be grasped as a persona showing the portrait of a typical action pattern. That is, the user of the action attribute analysis system 1 can clearly grasp what kind of actions the visitors to the facility to be analyzed perform.

[0070] Embodiment 4. In the fourth embodiment, the operation of grasping the location where the person P exists by obtaining the position coordinates of the communication device 2 will be described. Hereinafter, the fourth embodiment will be described centering on the differences from the first to third embodiments.

[0071] The collection device 20 of the fourth embodiment obtains the position coordinates of the communication device 2 based on the radio waves transmitted or received between the communication device 2 and the detection device 10. Specifically, the collection device 20 performs position measurement using UWB (Ultra Wide Band).

[0072] In positioning using UWB, the communication device 2 serves as a transmitter, the detection device 10 serves as a receiver, and the transmitter transmits a positioning signal of a predetermined pattern. In positioning, the position of the transmitter is calculated using the distance from the transmitter to the receiver or the angle (direction) of the transmitter with respect to the receiver. As methods for measuring the distance, there are a method using the reception intensity of the positioning signal received by the receiver, a method using the time it takes for the positioning signal from the transmitter to reach the receiver, or a method using the time difference when the positioning signal transmitted from one transmitter is received by a plurality of receivers. As a method for measuring the angle, there is a method of calculating the angle at which the transmitter is located using the directivity of each antenna constituting the antenna array of the receiver, etc.

[0073] FIG. 13 is an explanatory diagram of the positioning of the behavior attribute analysis system 1 according to Embodiment 4. In FIG. 13, the positions of three detection devices 10-1 to 10-3 functioning as receivers and a communication device 2 functioning as a transmitter are schematically shown.

[0074] The communication device 2 to be positioned transmits a positioning signal. The detection devices 10-1 to 10-3 each receive the positioning signal and acquire its radio wave intensity. The collection device 20 collects the radio wave intensities acquired by each detection device 10. Information on the position coordinates of each of the plurality of detection devices 10 is stored in the collection device 20 in advance. Then, the collection device 20 calculates the distances (R1 to R3) between each detection device 10 and the communication device 2 from the radio wave intensities acquired by each detection device 10. Thereby, the collection device 20 can calculate the position coordinates of the communication device 2.

[0075] Next, the collection device 20 designates as the detection device 10 that has acquired the user identification information of the communication device 2, the detection device 10 whose position coordinates of the communication device 2 are included in the position coordinates within a predetermined range. For example, information on the position coordinates indicating a predetermined range of each of the plurality of detection devices 10 is stored in the collection device 20 in advance, and by comparing this position coordinate information with the position coordinates of the communication device 2, it is possible to determine at which point (location) the communication device 2 exists. The subsequent operations are the same as any of the above-described Embodiments 1 to 3.

[0076] Note that the method for positioning to obtain the position coordinates of the communication device 2 is not limited to UWB positioning. For example, any positioning method such as positioning using BLE-AOA (Bluetooth Low Energy Angle of Arrival) may be used.

[0077] (Effect) As described above, in the fourth embodiment, the position coordinates of the communication device 2 are obtained based on the radio waves transmitted or received between the communication device 2 and the detection device 10. Therefore, the location where the communication device 2 exists can be accurately detected.

[0078] Embodiment 5. In this fifth embodiment, an operation of acquiring the feature amount of the face of the person P from the image of the person P and assigning user identification information will be described. Hereinafter, the fifth embodiment will be described centering on the differences from the above-described first to fourth embodiments.

[0079] FIG. 14 is a schematic configuration diagram of the behavior attribute analysis system 1 according to the fifth embodiment. As shown in FIG. 14, the behavior attribute analysis system 1 according to the fifth embodiment includes a plurality of detection devices 11.

[0080] The detection devices 11 are installed at a plurality of points within the area to be analyzed for the flow of people. The detection device 11 is composed of a camera, and acquires the feature amount of the face of the person P from the image of the person within a predetermined range. Specifically, the detection device 11 extracts a multi-dimensional vector (for example, a vector of about 128 dimensions) from the face image, and uses this as the feature amount of the face of the person P. Then, the detection device 11 transmits the information on the feature amount of the face, the detection identification information of the detection device 11, and the information on the date and time when the feature amount of the face was acquired to the collection device 20.

[0081] The collection device 20 collects the feature amount of the face, the detection identification information for identifying the detection device 10, and the information on the date and time when the detection device 10 acquired the feature amount from the plurality of detection devices 11, and transmits the collected information to the information processing device 100.

[0082] The acquisition unit 110 of the information processing apparatus 100 performs face authentication based on the facial feature amounts, and assigns the same user identification information to the facial feature amounts determined to be of the same person. A specific example will be described with reference to FIGS. 15 and 16.

[0083] FIG. 15 is an explanatory diagram of face authentication of the behavior attribute analysis system 1 according to Embodiment 5. FIG. 15 schematically shows a case where the detection devices 11-A to 11-C have captured the faces of persons X, Y, and Z. In FIG. 15, the elliptical dotted line indicates a predetermined range captured by each detection device 11.

[0084] As shown in FIG. 15, person X moves within the facility and is captured by the detection devices 11-A and 11-B, and the facial feature amounts (Fx-A, Fx-B) are calculated by the respective detection devices 11. Similarly, persons Y and Z also move within the facility, and the facial feature amounts are calculated by a plurality of detection devices 11. Note that since the facial images captured by each detection device 11 differ depending on the facial expression, angle, etc., the facial feature amounts do not completely match even for the same person.

[0085] FIG. 16 is an explanatory diagram of user identification information assignment of the behavior attribute analysis system 1 according to Embodiment 5. The acquisition unit 110 maps the collected facial feature amounts into a multidimensional space, clusters them using a method such as the k-means method, for example, and determines that the feature amounts of each cluster are the facial feature amounts of the same person. Then, the acquisition unit 110 assigns the same user identification information (for example, U-ID01 to U-ID04, etc.) to the facial feature amounts determined to be of the same person.

[0086] Next, the acquisition unit 110 stores, in the storage unit 130 as collection information, a plurality of pieces of user identification information, detection identification information for identifying the detection device 10, and information on the date and time when the detection device 10 acquired the feature amounts. The subsequent operations are the same as those of any of the above-described Embodiments 1 to 4.

[0087] Note that the face authentication method is not limited to the above, and any method can be used. For example, when the difference between each feature amount is equal to or less than a predetermined value, it may be determined that they are the same person.

[0088] (Effect) As described above, in the fifth embodiment, face authentication is performed based on the facial feature amount of a person, and the same user identification information is assigned to the facial feature amount of the face determined to be the same person. Therefore, even when the person P is not carrying the communication device 2, the action attribute of the person P can be assigned.

[0089] (Modification) In the above description, the operation of performing face authentication based on the facial feature amount of the person P has been described. However, the fifth embodiment is not limited to this. The feature amount of the person P may be the feature amount of the whole body or an arbitrary part of the person P, such as the whole body posture or clothing of the person P. In this case, the detection device 11 acquires the feature amount of the whole body or an arbitrary part of the person P from the image of the person taken. Then, the detection device 11 transmits the information on the feature amount of the person, the detection identification information of the detection device 11, and the information on the date and time when the feature amount of the person was acquired to the collection device 20. The collection device 20 collects the feature amount of the person, the detection identification information for identifying the detection device 10, and the information on the date and time when the detection device 10 acquired the feature amount from a plurality of detection devices 11, and transmits the collected information to the information processing device 100. The acquisition unit 110 of the information processing device 100 performs authentication based on the feature amount of the person, and assigns the same user identification information to the feature amount of the person determined to be the same person.

Description of Reference Numerals

[0090] 1 Action attribute analysis system, 2 Communication device, 10 Detection device, 11 Detection device, 20 Collection device, 100 Information processing device, 110 Acquisition unit, 120 Action attribute assignment unit, 130 Storage unit, 140 Extraction processing unit, 150 Input unit, 160 Output unit, 170 Clustering unit, 200 Input screen, 201 Date input unit, 202 Action type input unit, 203 Location input unit, 300 Output screen, 301 Action attribute display, 302 Extraction result display, 400 Output screen.

Claims

1. A detection device installed at a plurality of locations for obtaining user identification information from communication devices carried by people within a predetermined range, a collection device for collecting, as collection information, the user identification information obtained by the detection device, detection identification information for identifying the detection device, and information on the date and time when the user identification information was obtained, an information processing device for obtaining the collection information, comprising: the information processing device has a storage unit for storing correspondence information between the detection identification information and the type of the location, and an action attribute assignment unit for assigning, for each user identification information, an action attribute that is a combination of the type of the person's action at a plurality of the locations and the type of the plurality of locations based on the collection information and the correspondence information, having wherein the type of the location includes a type corresponding to the demographic attribute of the person an action attribute analysis system.

2. The information processing device has an input unit for receiving a selection operation of the action attribute, and an extraction processing unit for extracting, from among a plurality of the user identification information, the user identification information for which the action attribute matches the selection operation of the input unit, The action attribute analysis system according to claim 1.

3. The information processing device has a clustering unit for clustering a plurality of the action attributes into a plurality of clusters based on similarity, and an extraction processing unit for extracting, for each cluster clustered by the clustering unit, the user identification information for which the action attribute is included in the cluster, The action attribute analysis system according to claim 1.

4. The type of the person's action is staying, which is a state in which the user identification information is continuously obtained at one of the locations for a predetermined time or more, passing, which is a state in which the user identification information is obtained at one of the locations without being continuous for a predetermined time or more, the first passed location, which is the location at which the user identification information was first obtained among a plurality of the locations in a day, the last passed location, which is the location at which the user identification information was last obtained among a plurality of the locations in a day, the total stay time, which is the time from the time when the user identification information was first obtained to the time when it was last obtained in a day, the total stay time within a period, which is the sum of the total stay times within a predetermined period, the number of visits, which is the number of days on which the user identification information was obtained within a predetermined period, the visit time zone, which is the time zone to which the time when the user identification information was first obtained belongs, the visiting interval, which is the interval between the days on which the user identification information was obtained The event sensitivity that the date when the user identification information was acquired is a predetermined ratio, The moving speed obtained from the distance between the plurality of locations where the user identification information was acquired and the time difference between the times when the user identification information was acquired at the plurality of locations, And, A congestion tolerance person who divides the plurality of locations into two or more areas and continuously acquires the user identification information for a predetermined time or more in the area where the number of different user identification information acquired exceeds a threshold value, Is at least one of The action attribute analysis system according to any one of claims 1 to 3.

5. The user identification information is, Information included in a beacon signal transmitted or received between the communication device and the detection device The action attribute analysis system according to any one of claims 1 to 3.

6. The collection device, The information of the position coordinates indicating the respective predetermined ranges of the plurality of detection devices is stored in advance, Obtain the position coordinates of the communication device based on the radio wave transmitted or received between the communication device and the detection device, The detection device whose position coordinates of the communication device are included in the position coordinates of the predetermined range is set as the detection device that has acquired the user identification information of the communication device The action attribute analysis system according to any one of claims 1 to 3.

7. A detection device installed at a plurality of locations and acquiring feature amounts of a person from an image obtained by photographing a person within a predetermined range, A collection device that collects the feature amounts of the person, detection identification information for identifying the detection device, and information on the date and time when the detection device acquired the feature amounts, An information processing device that acquires information from the collection device, Comprising, The information processing device, An acquisition unit that performs authentication based on the feature amounts of the person, assigns the same user identification information to the feature amounts of the person determined to be the same person, and acquires, as collection information, the plurality of user identification information, the detection identification information for identifying the detection device, and the information on the date and time when the detection device acquired the feature amounts, A storage unit that stores correspondence information between the detection identification information and the type of the location, An action attribute assignment unit that assigns, for each user identification information, an action attribute that is a combination of the type of the person's action at the plurality of locations and the type of the plurality of locations based on the collection information and the correspondence information, Having, The type of the location includes a type corresponding to the demographic attribute of the person Action attribute analysis system.

8. An information processing device that acquires, as collection information, user identification information of communication devices carried by people within a predetermined range, detection identification information for identifying the detection device, and information on the date and time when the detection device acquired the user identification information, which are acquired by detection devices installed at a plurality of locations. A storage unit that stores correspondence information between the detection identification information and the type of the location. An action attribute assignment unit that assigns, for each user identification information, an action attribute that is a combination of the type of action of the person at a plurality of the locations and the type of a plurality of the locations, based on the collection information and the correspondence information. It has: The type of the location includes a type corresponding to the demographic attribute of the person. Information processing device.

9. An information processing device that acquires the feature amount of a person within a predetermined range, detection identification information for identifying the detection device, and information on the date and time when the detection device acquired the feature amount, which are acquired by detection devices installed at a plurality of locations. An acquisition unit that performs authentication based on the feature amount of the person, assigns the same user identification information to the feature amount of the person determined to be the same person, and acquires, as collection information, a plurality of the user identification information, the detection identification information for identifying the detection device, and the information on the date and time when the detection device acquired the feature amount. A storage unit that stores correspondence information between the detection identification information and the type of the location. An action attribute assignment unit that assigns, for each user identification information, an action attribute that is a combination of the type of action of the person at a plurality of the locations and the type of a plurality of the locations, based on the collection information and the correspondence information. It has: The type of the location includes a type corresponding to the demographic attribute of the person. Information processing device.

Images