JP2016081498A - Association program and information processing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an association program for associating a person with a vehicle, and an information processing apparatus.SOLUTION: An information processing apparatus 1 includes: person extraction means 101 and vehicle extraction means for extracting a person and a vehicle from an image captured by a camera; identifying means 104 which identifies whether the person extracted by the person extraction means 101 has entered or exited the vehicle extracted by the vehicle extraction means; and association means 105 which associates the person with the vehicle where the person has entered or exited when the identifying means 104 identifies the motion.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an association program and an information processing apparatus.

  As a conventional technique, an information processing apparatus that groups a plurality of persons in a moving image has been proposed (see, for example, Patent Document 1).

  The information processing apparatus disclosed in Patent Document 1 detects a person and the position of a person from a moving image, totals the relative distances of the detected persons, and calculates a relative distance threshold from an appearance distribution of the relative distances. A person whose relative distance between persons continues to be lower than the threshold for a long time is determined as a group performing the same action.

Japanese Patent No. 4506181

  An object of the present invention is to provide an association program and an information processing apparatus for associating a person with a vehicle.

  In order to achieve the above object, one aspect of the present invention provides the following associated program and information processing apparatus.

[1]
Extracting means for extracting a person and a vehicle from an image taken by a camera;
Sixth specifying means for specifying the movement of the person into or out of the vehicle;
An association program for causing a person having a predetermined relationship with the vehicle in which the person entered or exited to function as an association means for associating with the vehicle when the sixth specifying unit specifies the movement.
[2] Further function as third specifying means for specifying movement of the container between the plurality of persons,
The association program according to [1], wherein the association unit associates the plurality of persons when the third identification unit identifies movement.
[3] Further function as a fourth specifying means for specifying the line of sight or the orientation of the face of the extracted person,
The association unit according to [1] or [2], wherein the association unit associates a person who has a predetermined relationship with the line of sight or the face direction identified by the fourth identification unit and the extracted person. program.
[4] Further function as voice acquisition means for acquiring the voice of the extracted person,
The association program according to any one of [1] to [3], wherein the association unit associates a person who has a predetermined relationship with the voice acquired by the voice acquisition unit and the extracted person.
[5] Further function as fifth specifying means for specifying the movement of the extracted person's mouth,
The association unit according to any one of [1] to [4], wherein the association unit associates the person who has a predetermined relationship with the mouth movement identified by the fifth identification unit and the extracted person. Program.
[6] Attribute acquisition means for acquiring the attribute of the extracted person;
The association according to any one of [1] to [5], which further functions as a group attribute specifying unit that specifies an attribute of a group composed of a plurality of persons related by the association unit based on the acquired attribute. program.
[7] Further function as an acquisition means for acquiring information related to the product purchased by the extracted person,
The association program according to any one of [1] to [6], wherein the association unit associates the information with the plurality of associated persons.
[8] Further function as acquisition means for acquiring member information of the extracted person,
The association means according to any one of [1] to [6], wherein the association unit associates the member information with the plurality of associated persons.
[9] The association program according to [6], wherein the attribute acquisition unit acquires an attribute of the extracted person from a license plate of the vehicle.
[10] The association program according to any one of [1] to [9], wherein the sixth specifying unit does not specify the movement of the person when the license plate of the vehicle is predetermined. .
[11] extraction means for extracting a person and a vehicle from an image photographed by the camera;
Sixth specifying means for specifying the movement of the person into or out of the vehicle;
An information processing apparatus comprising: association means for associating a person having a predetermined relationship with the vehicle in which the person has entered or exited in association with the vehicle when the sixth specifying means has specified the movement.

According to the invention which concerns on Claim 1 or 11, a person and a vehicle can be linked | related.
According to the invention which concerns on Claim 2, persons can be linked | related based on the movement of the container between several persons.
According to the third aspect of the present invention, it is possible to associate a person's line of sight or face orientation with a person who has a predetermined relationship.
According to the invention which concerns on Claim 4, the person who has a predetermined relationship based on an audio | voice can be linked | related.
According to the fifth aspect of the present invention, it is possible to associate a movement of a person's mouth with a person who has a predetermined relationship.
According to the invention which concerns on Claim 6, the attribute of a group can be specified based on the attribute of a person.
According to the invention which concerns on Claim 7, the related person and the information regarding the purchased goods can be linked | related.
According to the invention which concerns on Claim 8, the related person and member information can be linked | related.
According to the invention which concerns on Claim 9, attribute information can be acquired from the license plate of a vehicle.
According to the invention which concerns on Claim 10, in the case of the vehicle of a predetermined number plate, it can avoid specifying a person's motion.

FIG. 1 is a schematic diagram illustrating a configuration example of an actual store. FIG. 2 is a block diagram illustrating a configuration example of the information processing apparatus according to the embodiment. FIG. 3 is a schematic diagram illustrating an example of a configuration of association information. 4A and 4B are schematic diagrams illustrating an example of a motion pattern specified by the specifying unit. FIGS. 5A and 5B are schematic views showing other examples of movement patterns specified by the specifying means. FIGS. 6A to 6C are schematic diagrams illustrating other examples of movement patterns specified by the specifying unit.

[Embodiment]
(Composition of actual store)
FIG. 1 is a schematic diagram illustrating a configuration example of a store.

  The actual store 2 is a retail store such as a supermarket, for example, and a plurality of products 200a, 200b (hereinafter collectively referred to as the product 200) are installed on the shelves 20a-20e. The customer 3a-3k enters from the entrance 21 of the actual store 2, travels around the actual store 2 together with the carts 5a-5d, transports the product 200 by the cart 5a-5d, and accounts at the cash register by the store clerk 4a-4c. I'll do it. When payment is made at the cash register, sales data such as a POS system (Point of sale system) is stored in a database (not shown).

  Cameras 13a-13d are installed in the actual store 2, and the customer 3a-3k is recognized as a person from the video imaged by the camera 13 by the information processing apparatus 1 to be described later. It is recognized whether it is a group. Here, the “group” refers to a set of customers associated with each other.

  The camera 13 can be an omnidirectional camera in addition to a normal camera. The camera 13 may capture a moving image, or may capture a still image at a predetermined interval. Further, it is assumed that the installation position of the camera 13 is registered in advance, and the coordinates in the image of the captured moving image are associated with the position coordinates in the actual store 2.

  Here, the “moving image” is a set of a plurality of frames (still images) captured along a time series, and the plurality of frames are reproduced along the time series.

(Configuration of information processing device)
FIG. 2 is a block diagram illustrating a configuration example of the information processing apparatus according to the embodiment.

  The information processing apparatus 1 is configured by a CPU (Central Processing Unit) and the like, and controls each unit and executes various programs, and a storage unit 11 configured by a storage medium such as a flash memory and stores information. A communication unit 12 that communicates with the outside via a network, and a camera 13 that can capture a moving image or a still image.

  The control unit 10 executes an association program 110, which will be described later, so that the moving image reception means 100, person extraction means 101, container extraction means 102, product extraction means 103, identification means 104, association means 105, sales data acquisition means 106, functions as a person attribute acquisition unit 107, a member information acquisition unit 108, and the like.

  The moving image receiving unit 100 receives moving image information generated by being captured by the camera 13.

  The person extraction unit 101 extracts a person when the image of the person is included in all or a part of the frames of the moving image information received by the moving image reception unit 100. Note that the person extracting unit 101 may identify each person when there are a plurality of persons, and may identify a person registered in advance (such as a store clerk) based on a previously registered image. An image registered in advance may be converted into a feature amount. As an example, the feature amount is obtained by extracting a feature point from an image by a Difference of Gaussian operation and extracting a SIFT feature amount from the feature point. As another example of the feature amount, a FIT (Fast Invariant Transform) feature amount generated from gradient information between the extracted feature point and a higher scale point may be used.

  A learning model may be generated from a plurality of pre-registered images, and a person may be identified using the learning model. Note that the person extracting unit 101 may exclude the salesclerk from the extracted person. For example, when there is a store clerk uniform, a feature amount is generated in advance from an image of the uniform. Then, using this feature amount, it is possible to exclude the clerk from the extracted person image. Note that the identification of the store clerk and the customer is not limited to the uniform, and the feature amount may be generated using another image such as a name plate.

  In addition, for the container extraction unit 102 and the product extraction unit 103 described below, containers and products are extracted using the same technique.

  When the image of the cart is included as an example of the container in all or some of the frames of the moving image received by the container extracting unit 102 and the moving image receiving unit 100, the cart is extracted. Note that letters, numbers, marks, etc. may be attached to the cart in order to identify the cart.

  The product extraction unit 103 extracts the product when the image of the product is included in all or some of the frames of the video received by the video receiving unit 100. Note that letters, numbers, marks, etc. may be attached to the product in order to identify the product.

  The specifying unit 104 determines whether the person movement extracted by the person extraction unit 101, the cart movement extracted by the container extraction unit 102, or the movement of the product extracted by the product extraction unit 103 corresponds to a predetermined movement pattern. Identify. Note that the specifying unit 104 may specify only a single pattern, or may identify and specify a plurality of patterns. Further, as will be described later, the specifying means 104 is a first specifying means that specifies the movement of a product in or out of a container, and the second specifying means is a movement of a product between a plurality of persons. Is used as the third specifying means, and the fourth specifying means is used to specify the line of sight of the person or the face direction. As the specifying means, what specifies the movement of the person's mouth is realized, and as the sixth specifying means, what specifies the movement of the person entering or leaving the vehicle is realized.

  The association unit 105 performs association based on the pattern specified by the specifying unit 104. For example, as the first association pattern, when the person extracted by the person extracting unit 101 and the container extracted by the container specifying unit 102 are associated, a plurality of patterns extracted by the person extracting unit 101 are used as the second association pattern. When associating persons with each other, the third pattern may include associating a plurality of persons extracted by the person extracting means with the container specifying means 102.

  Note that the association unit 105 may implement only a single association pattern, or may implement a plurality of association patterns. Further, the number of associations performed by the association unit 105 is not limited to one. That is, the association can be performed repeatedly. For example, first, the association unit 105 associates the cart 5a with the customer 3a. Thereafter, the association unit 105 can associate the cart 5a and the customer 3b. In this case, the cart 5a, the customer 3a, and the customer 3b are associated with each other. In other words, customers 3a and 3b are recognized as a group. The association means 105 stores information associating a person in the storage unit 11 as association information 111.

  The sales data acquisition means 106 acquires sales data from a database (not shown) in which information such as a POS system is stored when the checkout is made at the cash register, and stores the sales data as sales data 112 in the storage unit 11. The sales data mentioned here is data that can be acquired from the POS system, such as purchased product type, number of purchases, payment amount, and payment time.

  The person attribute acquisition unit 107 acquires attribute information such as age, gender, and region information (details will be described later) from the person image extracted by the person extraction unit 101 and stores the attribute information in the storage unit 11 as the person attribute information 113.

  The member information acquisition unit 108 acquires member information in which a customer's face photo, name, gender, age, address, telephone number, purchase history, and the like are recorded from a database (not shown) and stores the information as member information 114 in the storage unit 11. .

  The storage unit 11 stores an association program 110, association information 111, sales data 112, person attribute information 113, member information 114, and the like that cause the control unit 10 to operate as the above-described units 100-108.

(Operation of information processing device)
Next, the operation of the present embodiment will be described.

  First, the camera 13 photographs customers 3a-3k, products 200a, 200b, carts 5a-5d, and the like in the actual store 2 to generate moving image information.

  The moving image receiving means 100 of the information processing apparatus 1 receives moving image information that is captured and generated by the camera 13.

  Next, the person extracting unit 101 extracts a person when all or some of the frames of the moving image received by the moving image receiving unit 100 include a person image. The person extracting unit 101 identifies each person when there are a plurality of persons, and identifies a person registered in advance (such as a store clerk or a customer who frequently visits the store) based on a pre-registered image. .

  Next, when the image of the cart is included as an example of the container in all or a part of the frames of the moving image received by the container extracting unit 102 and the moving image receiving unit 100, the cart is extracted. If there are multiple carts in the image, each is identified. In the case of identification, characters or images for identifying each cart may be attached, or the cart once identified may be traced thereafter.

  Next, the product extraction unit 103 extracts the product when the image of the product is included in all or some of the frames of the moving image received by the moving image receiving unit 100. When there are a plurality of products in the image, each may or may not be identified. In the case of identification, characters, numbers, marks or the like for identifying each product may be added, or an image of the product may be registered in advance and identified by image recognition.

  Next, the specifying unit 104 uses any one of the predetermined movement patterns in which the movement of the person extracted by the person extraction unit 101, the movement of the cart extracted by the container extraction unit 102, and the movement of the product extracted by the product extraction unit 103 are set in advance. Identify if applicable.

  Hereinafter, the movement pattern specified by the specifying unit 104 will be described.

(Pattern 1)
4A and 4B are schematic diagrams illustrating an example of a motion pattern specified by the specifying unit 104. FIG.

  As shown in FIG. 4A, the customer 3a pushing the cart 5a or in the vicinity of the cart 5a picks up the product 200c, and puts the product 200c into the cart 5a as shown in FIG. 4B. In this case, the specifying unit 104 specifies the movement of the customer 3a, the cart 5a, and the product 200c as “pattern 1” (first specifying unit).

  Next, the associating means 105 associates the person extracted by the person extracting means 101 with the container extracted by the container specifying means 102 based on the movement pattern specified by the specifying means 104.

  When the “pattern 1” shown in FIGS. 4A and 4B is specified, the associating unit 105 associates the customer 3a with the cart 5a.

  In the above “Pattern 1”, the case where the product 200c is put in the cart 5a is shown. However, the specifying unit 104 specifies the movement of putting out the product 200c in the cart 5a as “Pattern 1 ′”, and the “pattern” When specifying “1 ′”, the associating means 105 may associate the customer 3a with the cart 5a.

  In addition, in the “pattern 1”, when the customer 3a pushes the cart 5a for a predetermined time or more or stays within a certain range from the cart 5a regardless of the presence or absence of the product 200c, the association unit 105 The customer 3a may be associated with the cart 5a.

  FIG. 3 is a schematic diagram illustrating an example of the configuration of the association information 111.

  The association information 111 has a cart ID of the cart extracted by the container extraction unit 102 and a person associated with the cart ID, and has one or more person IDs extracted by the person extraction unit 101.

  For example, when two people with the person IDs “1113” and “1112” are associated with the cart with the cart ID “001”, “1113” is written in the person 1 column and “1112” is written in the person 2 column. Is done. Person 1, person 2, person 3... May be described in the order of extraction, or may be described in the order of the distance to the cart or the time within a certain range.

  Further, each time the association unit 105 performs association, the information of the association information 111 is updated. If a predetermined condition is met, the person ID associated with the cart ID described in the association information 111 may be reset. Predetermined conditions include a case where a person described in the association information 111 completes payment at a cash register or a case where the cart described in the association information 111 is returned to the cart storage area. Whether or not the person described in the association information 111 has completed the payment at the cash register can be confirmed using the sales data acquisition means 106. Further, whether or not the cart described in the association information 111 has been returned to the cart storage area can be confirmed by whether or not there is a cart in a predetermined place (for example, the cart storage area). Note that the conditions for resetting and the confirmation method are not limited to those described above. The resetting of the association information 111 can be similarly applied to other patterns described later.

(Another example of pattern 1)
FIGS. 5A and 5B are schematic diagrams illustrating other examples of movement patterns specified by the specifying unit 104.

  When the customer 3b in the vicinity of the cart 5a picks up the product 200c as shown in FIG. 5 (a) and puts the product 200c into the cart 5a as shown in FIG. The movement of the customer 3b, the cart 5a, and the product 200c is specified as “pattern 1”. Further, a customer 3a standing in the vicinity of the cart 5a who pushes the cart 5a, in which the product is put, stands. In such a case, the association means 105 may associate the customers 3a and 3b with the cart 5a, assuming that the cart 5a and the customer 3a have a predetermined relationship.

  In this case, on the condition that the customer 3a is already associated with the cart 5a (for example, when the customer 3a and the cart 5a correspond to the above-described pattern 1), the customers 3a and 3b are added to the cart 5a. You may associate. That is, if the customer 3a is not yet associated with the cart 5a, only the customer 3b and the cart 5b may be associated. Alternatively, first, the cart 5a and the customer 3b are associated with each other, and when the cart 5a and the customer 3a correspond to the above-described pattern 1, the customer 3a is additionally associated with the cart 5a associated with the customer 3b. Also good.

(Pattern 2)
FIGS. 6A to 6C are schematic diagrams illustrating other examples of movement patterns specified by the specifying unit 104.

  As shown to Fig.6 (a), the customer 3a who pushes the cart 5a or is in the vicinity of the cart 5a stands, and the customer 3b picks up the goods 200c. Next, as shown in FIG. 6B, when the product 200c is handed over from the customer 3b to the customer 3a, the specifying unit 104 specifies the movement of the customer 3a, the customer 3b, and the product 200c as “pattern 2”. (Second identification means). Next, as shown in FIG. 6C, when the customer 3a puts the product 200c into the cart 5a, the specifying unit 104 specifies the movement of the customer 3a, the cart 5a and the product 200c as “pattern 1”.

  First, when “pattern 2” is specified by the second specifying means of the specifying means 104, the customer 3a and the customer 3b are associated with each other. In this case, the association information 111 is associated with the person ID of the customer 3a and the person ID of the customer 3b. That is, the cart ID of the cart 5a does not necessarily have to be associated with the person ID of the customer 3a or the customer 3b. Next, when “pattern 1” is specified by the first specifying means, the customer 3a and the cart 5a are associated with each other. At this time, the person ID of the customer 3a and the person ID of the customer 3b are already associated with the association information 111. Accordingly, when “pattern 1” is specified, the person ID of the customer 3a, the person ID of the customer 3b, and the cart ID of the cart 5a are associated with each other.

  In addition, the movement that the cart 5a as a container is transferred or moved from the customer 3a to the customer 3b may be specified as “pattern 3” (third specifying means). In this case, the associating means 105 associates the customers 3a and 3b. The customers 3a and 3b may be associated with the cart 5a.

  Note that “Pattern 1” to “Pattern 3” described above describe the movement of goods at the cash register before accounting, but not only before accounting, but also during the accounting, the movement of goods, the customer after accounting. Patterns can also be set for product movements.

  For example, if a customer is late at the checkout and another customer brings a product late at the cashier and adds a product to the cart or puts the product in the cart, The customer may be associated with the cart when the merchandise is delivered to the store clerk.

  In addition, after the accounting, for example, when a product for which accounting has been completed is passed from one customer to another customer, they may be associated as belonging to the same group.

  Next, when any of the associated customers pays at the cash register, the sales data acquisition means 106 acquires the sales data 112 of the POS system or the like corresponding to the cash register from a database (not shown), and The data is stored in the storage unit 11 in association with the cart ID.

  Also, the person attribute acquisition unit 107 acquires attribute information such as age, gender, and region information (details will be described later) from the person image extracted by the person extraction unit 101 to obtain the person attribute information 113, and the association information 111. Are stored in the storage unit 11 in association with each user.

  Further, when the customer uses a membership card when the customer accounts at the cash register, the membership information acquisition unit 108 acquires the membership information associated with the membership card from a database (not shown) as the membership information 114, and associates information 111 And stored in the storage unit 11.

  The sales data 112, the person attribute information 113, and the member information 114 described above are used for group behavior analysis, sales analysis, and the like together with the association information 111 that defines the group. Moreover, you may give group ID to member information. In this case, the group ID may be updated each time the store visits or may be accumulated.

(Effect of embodiment)
According to the above-described embodiment, it is possible to identify each person such as a person in the vicinity of the cart as an example of a container and the movement of goods in and out of the cart, and associate the cart with the person by the movement. Or each movement, such as delivery of goods between persons and movement of a cart between persons, can be specified, and persons can be related by the movement.

[Other embodiments]
The present invention is not limited to the above embodiment, and various modifications can be made without departing from the spirit of the present invention.

  In the above-described embodiment, the customers who act in the actual store 2 are associated with each other. However, the present invention is not limited to the actual store 2, and the customers who act in the structure may be associated with each other. Good. For example, the structure includes a restaurant, a shopping mall, a building, an airport, a station, a hospital, a school, a leisure facility, and the like. Moreover, it may move like an airplane or a ship. The sales information related to the customer corresponds to the menu ordered by the customer or the amount paid at the cash register. In addition, in the case of a building or an airport, the action content of a store clerk working in the building or airport is specified.

  Further, the specifying unit 104 may specify the customer's line of sight or face orientation as a “fourth pattern” (fourth specifying unit). The associating means 105 may associate the customer with another customer whose line of sight or face is directed for a predetermined time or more. The specifying unit 104 may specify the movement of the mouth as the “fifth pattern” (fifth specifying unit). The associating means 105 may associate customers who have a conversation.

  The cart is an example of a container for storing products, and may be a basket or a shopping bag.

  Further, in the present embodiment, a person is associated with a container such as a cart. However, a person may be associated with a vehicle such as a car or a motorcycle, not limited to a container such as a cart. For example, a camera is installed in a parking lot and a car is photographed. Then, using the specifying unit 104, a movement in which a customer leaves or enters a certain car is specified. The specifying means 104 specifies the movement of a customer coming out of or entering a certain car as “pattern 6” (sixth specifying means).

  Next, the associating means 105 associates each customer with a car, and handles them as a group. For example, when a plurality of customers come out of a car, the plurality of people can be associated with each other. Furthermore, when a car license plate can be photographed, the area information may be acquired as attribute information from the photographed license plate information.

  The specifying unit 104 may be a vehicle such as a car or a motorcycle, and the movement may not be specified when the vehicle is a business vehicle such as a delivery vehicle that does not carry a passenger. In this way, when business vehicles such as delivery vehicles are excluded, it is only necessary to previously register images of vehicles to be excluded, license plates, etc., and use these to identify vehicles that are not eligible.

  The specifying unit 104 only needs to include a specifying unit that specifies at least one pattern. For example, a specifying unit having only “Pattern 1” may be used, or a specifying unit for specifying “Pattern 1” to “Pattern 6” may be provided.

  Further, as for the association means 105, for example, when the customer 3a has pushed the cart 5a for a predetermined time or more, or is within a certain range from the cart 5a, regardless of the presence or absence of the product 200c, the association means 105 May associate the customer 3a with the cart 5a. Thus, the associating means 105 determines whether or not the relationship between the customer and the cart is a predetermined relationship using image processing, and determines that the relationship is a predetermined relationship. You may have the structure which associates a customer and a cart. Note that the determination operation may be executed by different means, and is not necessarily executed by the association means 105.

  In addition, voice acquisition means for acquiring the voice of the conversation may be further provided. The acquired voice is analyzed to identify the pattern of the conversation content, and the associating means 105 identifies the customers based on the conversation content pattern. You may make it relate.

  In addition, according to the attribute and the number of persons belonging to the group acquired by the person attribute acquisition unit 107, a unit for specifying which category the family is a family, a couple, or an organization may be further provided. Further, the purchase pattern may be analyzed from the category of the specified group and the sales data.

  As another example of analysis, the difference between the product picked up by the customer and the actually purchased product shown in the sales data is taken to detect the information that was picked up but not bought. It may be used as

  In the above embodiment, people, products, carts, etc. are extracted from the image. However, the present invention is not limited to this, and the identification information and the current position may be acquired from an IC chip embedded in a mobile phone or a card. The current position may be acquired using GPS.

  In the above embodiment, the functions of the units 100 to 108 of the control unit 10 are realized by a program. However, all or part of the units may be realized by hardware such as an ASIC. The program used in the above embodiment can be provided by being stored in a recording medium such as a CD-ROM. In addition, replacement, deletion, addition, and the like of the above-described steps described in the above embodiment are possible within a range that does not change the gist of the present invention.

DESCRIPTION OF SYMBOLS 1 Information processing apparatus 2 Store 3a-3k Customer 4a-4c Store clerk 5a-5d Cart 10 Control part 11 Storage part 12 Communication part 13a-13d Camera 20a-20e Shelf 21 Entrance 100 Movie reception means 101 Person extraction means 102 Container extraction means 103 Product extraction means 104 Identification means 105 Association means 106 Sales data acquisition means 107 Person attribute acquisition means 108 Member information acquisition means 110 Association program 111 Association information 112 Sales data 113 Person attribute information 114 Member information 200a-200c

Claims (11)

Computer
Extracting means for extracting a person and a vehicle from an image taken by a camera;
Sixth specifying means for specifying the movement of the person into or out of the vehicle;
An association program for causing a person having a predetermined relationship with the vehicle in which the person entered or exited to function as an association means for associating with the vehicle when the sixth specifying unit specifies the movement. Further function as a third specifying means for specifying movement of containers between a plurality of persons,
The association program according to claim 1, wherein when the third specifying unit specifies movement, the association unit associates the plurality of persons. Further functioning as a fourth specifying means for specifying the line of sight of the extracted person or the orientation of the face,
The association program according to claim 1, wherein the association unit associates the person who has a predetermined relationship with the line of sight or the face direction identified by the fourth identification unit and the extracted person. Further functioning as voice acquisition means for acquiring the voice of the extracted person,
The association program according to any one of claims 1 to 3, wherein the association unit associates a person who has a predetermined relationship in the voice acquired by the voice acquisition unit with the extracted person. Further functioning as a fifth specifying means for specifying the movement of the extracted person's mouth,
5. The association according to claim 1, wherein the association unit associates the extracted person with a person who has a predetermined relationship with the mouth movement identified by the fifth identification unit. program. Attribute acquisition means for acquiring the attribute of the extracted person;
The association program according to any one of claims 1 to 5, further functioning as a group attribute identification unit that identifies an attribute of a group composed of a plurality of persons associated by the association unit based on the acquired attribute. . It further functions as an acquisition means for acquiring information related to the product purchased by the extracted person,
The association program according to any one of claims 1 to 6, wherein the association unit associates the information with the plurality of associated persons. Further function as an acquisition means for acquiring the extracted member information of the person,
The association program according to any one of claims 1 to 6, wherein the association unit associates the member information with the plurality of associated persons.   The association program according to claim 6, wherein the attribute acquisition unit acquires an attribute of the extracted person from a license plate of the vehicle.   The association program according to claim 1, wherein the sixth specifying means does not specify the movement of the person when the license plate of the vehicle is predetermined. Extracting means for extracting a person and a vehicle from an image taken by a camera;
Sixth specifying means for specifying the movement of the person into or out of the vehicle;
An information processing apparatus comprising: association means for associating a person having a predetermined relationship with the vehicle in which the person has entered or exited in association with the vehicle when the sixth specifying means has specified the movement.

Images