JPH11175597A - Merchandise selection behavior information calculating method and its execution system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To collect (customers') merchandise selection behavior information from customer movement information and merchandise movement information within the store by connecting the customer movement information and the merchandise movement information in a store and calculating the information of a product that is selected, examined or purchased by a customer. SOLUTION: This method reads one behavior number, measurement time and positional information which correspond to the number from a customer movement information table 1112 of a storage device 111. It calculates a behavior number from read customer movement information and from merchandise movement information that is read from a merchandise movement information table 1113 of the same device 111 and stores it in a merchandise selection behavior information table 1116 of the device 111. Next, it calculates a merchandise that is purchased with a behavior number hat corresponds to one behavior number which is read from the read customer movement information and merchandise purchase information table 114 of the device 111 and erects a purchase flag of the merchandise number in the table 1116 of the device 111.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a product selection behavior information calculation system for collecting behavior information when a customer selects a product, and more particularly to clothing, accessories (shoes, bags, hats, ties, scarves, accessories, etc.). ), In a retail store that sells highly-preferred products such as CDs, product selection behavior information when customers select products in the store is accurately collected without giving the customer discomfort, and store sales analysis The present invention relates to a technology effective when applied to a product selection behavior information calculation system suitable for use as information or product analysis information for product planning.

[0002]

2. Description of the Related Art Conventionally, in a store that sells merchandise, movement information of customers in the store is collected and used as data for increasing sales. The system is described in Japanese Patent Publication No. Hei 3-38528. The outline is that a small transmitter for transmitting different transmitter identification information is attached to a shopping basket provided to a customer at a store, and the small transmitter is only transmitted when the small transmitter approaches a predetermined distance. The sensor panels that receive the transmitter identification information transmitted by the customer are installed at a plurality of predetermined positions through which customers in the store pass, and the reception information and the reception time are collected.

In another system, an image input device is installed in a store, a change point is analyzed from a change in each frame of image information collected by the image input device, and the change point is tracked to move information. There is something to collect. In addition, moving object detection and detection
The extraction apparatus and method are described in JP-A-8-221577.

[0004] As a conventional system for collecting information on movement of commodities, a sensor panel is installed on furniture displaying the commodities, and when the commodities displayed on the furniture are taken out of the furniture, the sensor panel senses and moves. There is one that collects the product information and the travel time.

[0005]

In the above-mentioned conventional system, when a small transmitter approaches a predetermined distance, sensor panels installed at a plurality of predetermined positions in a passage through which a customer passes indicate the transmitter identification information. By receiving the information, the movement status of the customer is grasped.

However, in this system, although the customer's movement information can be collected, it is not possible to grasp which product the customer has selected and finally purchased which product in the course of the movement. Further, when a customer leaves a shopping basket at a certain point and moves in a store, accurate customer movement information cannot always be collected.

In the above-described system for performing tracking based on image information, the movement information of the customer can be collected, but the product selected by the customer cannot be specified unless a person views and records the image information.

Further, in a system for collecting movement information of a product by sensing a sensor panel of a fixture, it is possible to specify a moving product, but not a customer who has selected and moved the product.

As described above, in the conventional system, even if the customer's movement information and the product's movement information can be collected, it is not possible to link them and to specify which customer has selected which product. Therefore, it is not possible to analyze the selection behavior such as which product the customer has selected and considered before purchasing the product, and it is not possible to analyze whether the selected product actually purchased. There is a problem.

As described above, in the conventional system, although the movement information of each of the customer and the merchandise can be collected, the movement information is not linked to the selection action until the actual purchase, and is used as information for taking a specific measure. There is a problem that it cannot be used effectively.

An object of the present invention is to solve the above-mentioned problem and to provide a technique capable of collecting customer's product selection behavior information from customer movement information and product movement information in a store.

Another object of the present invention is to provide a technique capable of taking measures for further increasing sales in store sales and product planning as needed.

[0013]

According to the present invention, in a product selection behavior information calculation system for calculating customer product selection behavior information, a specific customer is associated with customer movement information, product movement information and product purchase information. Calculate information on the products you have selected, considered or purchased between entering and leaving the store,
It is collected as the product selection behavior information of the customer.

In the commodity selection behavior information calculation system of the present invention, customer position information in a store is calculated at specified time intervals from image information in the store, and customer movement information is collected.

[0015] The position information of the product is calculated at specific time intervals based on information from the transmission / reception device attached to the product, and the product movement information is collected.

The collected customer movement information and merchandise movement information are associated with each other on the basis of their position information and measurement time, and the merchandise information selected and examined during the period from when the specific customer enters the store until the customer leaves the store. Information is calculated and collected as product selection behavior information of the customer.

When the customer purchases a product selected and examined, the product purchase information is collected, and the collected customer movement information and product purchase information are corresponded based on their location information and measurement time. In addition, the information of the merchandise that has been selected, examined, and purchased between the time the specific customer enters the store and the time the merchandise exits is calculated, and collected as merchandise selection behavior information of the customer.

As described above, in the commodity selection behavior information calculation system of the present invention, the customer movement information is correlated with the commodity movement information and the commodity purchase information. The correspondence between actions and products can be made clearer and more specific.

In the commodity selection behavior information calculation system according to the present invention, the customer movement information and the commodity movement information are associated and collected as the commodity selection behavior information, so that the customer can purchase the commodity regardless of whether or not he / she has purchased the commodity. The product selection behavior information can be collected from the stage before considering.

The collected product selection behavior information can be used for more specific product analysis, sales analysis, and measures for sales promotion as compared with the conventional case of only customer movement information and product movement information.

As described above, according to the commodity selection behavior information calculation system of the present invention, the information of the commodity selected, examined or purchased by the customer is calculated by linking the customer movement information and the commodity movement information in the store. It is possible to collect customer product selection behavior information from customer movement information and product movement information within a store.

[0022]

BEST MODE FOR CARRYING OUT THE INVENTION A product selection behavior information calculation system according to an embodiment for calculating customer's product selection behavior information when a customer moves in a store, selects a product, and considers purchasing a specific product, is described below. Will be described.

First, an outline of processing in the commodity selection behavior information calculation system of the present embodiment will be described. The customer's product selection behavior is defined as a behavior in which the customer selects a specific product, removes it from the display fixtures, and draws it to his / her own hand.

In the merchandise item selection behavior information calculation system of the present embodiment, first, during the period from when the customer enters the store until the customer exits the store,
The customer is tracked at each specific time interval, and position information of the customer in the store at each specific time interval is calculated and stored as customer movement information.

Next, transmission / reception of information at each specific time interval between the transmission / reception device attached to the product and the reception / transmission distance calculation processing device installed in the store, and the reception / transmission distance calculation processing device And information exchange between the processing device installed in the store, the position information of the product in the store for each time interval is calculated, the information of the product taken out from the display fixture by the customer and moved, that is, Store as product movement information.

Next, a search is made for information in which the position information and the measurement time of the customer movement information and the product movement information match within a specific range, and these information are specified and stored as the customer's product selection behavior information. As a result, it is possible to specify the product selected at the stage of examination regardless of whether the customer has purchased the product before the customer enters the store and exits.

Next, a search is made for a purchased product in which the measurement time of the customer movement information and the product purchase time match within a specific range, and the customer position information of the customer movement information and the position information of the checkout machine match within a specific range. I do. This means that the customer moves to a certain point of the checkout machine and purchases a product. As a result of the customer's selection, the purchase flag is set on the product as a purchased product. This makes it possible to link and specify the behavior of the customer in the store and the actually purchased product.

Next, the classification of the product selection behavior, that is, the process of classifying the plurality of products selected by the customer from what viewpoints, for example, whether they have been selected for comparison or selected for combination. I do.

In this process, the products are classified according to the selection behavior by searching for products having a product classification that matches the classification condition of the selection behavior registered in advance, and the information is stored in the storage device. This makes it possible to classify what action was taken when selecting a product. The above is the outline of the processing.

FIG. 1 is a diagram showing a schematic configuration of the commodity selection behavior information calculation system of the present embodiment. As shown in FIG. 1, the product selection behavior information calculation system of the present embodiment includes a furniture sensor 101, a checkout machine 102, and an image input device 103.
, Transmission / reception device 104, reception / transmission distance calculation processing device 10
5, a communication device 106, an input device 107, a processing device 108, a screen display device 109, an output device 110,
And a storage device 111.

The furniture sensor 101 is a sensor for detecting whether or not a product has been taken out of the furniture 11, and setting the receiving / transmitting device 104 to a state where transmission is possible when the product is taken out of the furniture 11. It is desirable to install the furniture 11 as close to the front as possible.

The payment machine 102 collects merchandise purchase information indicating the merchandise purchased by the customer, and transmits the POS (Point of Sale) to the processing device 108 via the communication device 106.
s) A device such as a register system. Image input device 1
Reference numeral 03 denotes a device such as a video camera for inputting customer image information by tracking the customer who has entered the store.

The receiving / transmitting device 104 is attached to a product,
It is a device such as an IC card that transmits information for specifying the position of a product. The transmission / reception distance calculation processing device 105 instructs the transmission / reception device 104 attached to the product to transmit information for specifying the position of the product, receives the information returned from the reception / transmission device 104, These are a plurality of devices that calculate the distance between the receiving and transmitting device 104.

The communication device 106 is a device for transmitting information transmitted from the settlement machine 102, the image input device 103, and the transmission / reception distance calculation processing device 105 to the processing device 108. The input device 107 is a device that mainly performs input when registering product attribute information in advance or analyzing accumulated various data.

The processing device 108 includes a customer movement information calculation processing unit 1081, a product movement information calculation processing unit 1082, and a product selection behavior calculation processing unit 1083, and controls the operation of the entire product selection behavior information calculation system. The screen display device 109 is a device that displays information output from the processing device 108 as several types of screens. First, a screen for receiving an information display request from a user, and second, a screen that matches the information display request. Screen for displaying various information, 3rd
Is a device for displaying a screen for receiving a request (instruction) for the information displayed on the screen, and for displaying other screens as necessary.

The output device 110 is a device such as a line printer or a laser printer for printing the result calculated by the processing device 108 or the result displayed on the screen display device 109 on output paper or the like. The storage device 111 stores the product attribute information table 1
111, a customer movement information table 1112, a commodity movement information table 1113, a commodity purchase information table 1114, a selection behavior classification condition table 1115, and a commodity selection behavior information table 1116.

The processing device 108 has a customer movement information calculation processing unit 1081, a product movement information calculation processing unit 1082, and a product selection behavior calculation processing unit 1083.

The customer movement information calculation processing unit 1081 tracks a specific customer based on the image information in the store sent from the image input device 103 via the communication device 106, and determines the position information of the customer in the store at a specific time interval. This is a processing unit that calculates and collects as customer movement information.

The product movement information calculation processing unit 1082 determines the transmission / reception device based on the distance to the transmission / reception device 104 calculated by the plurality of transmission / reception distance calculation devices 105 and the position information of each reception / transmission distance calculation processing device 105. Reference numeral 104 denotes a processing unit that calculates position information of a product to which the product is attached at a specific time interval and collects it as product movement information.

The merchandise selection behavior calculation processing unit 1083 calculates information on merchandise selected, studied, or purchased from the customer movement information, the merchandise movement information, and the merchandise purchase information until a specific customer enters and exits the store. The processing unit collects the information as the product selection behavior information of the customer, classifies the information of the product according to a preset classification condition, and collects information indicating what selection behavior the product has been selected. is there.

A program for causing the processing unit 108 to function as the customer movement information calculation processing unit 1081, the product movement information calculation processing unit 1082, and the product selection behavior calculation processing unit 1083 is recorded on a recording medium such as a CD-ROM and stored on a magnetic disk. Etc., and then loaded into a memory and executed. The medium for recording the program is C
A medium other than the D-ROM may be used.

The storage device 111 stores a product attribute information table 1111, a customer movement information table 1112, a product movement information table 1113, and a product purchase information table 11.
14, a selection behavior classification condition table 1115, and a product selection behavior information table 1116.

The product attribute information table 1111 is a table for storing the attributes of the products registered by the input device 107 in advance. The customer movement information table 1112 is a table for storing customer movement information collected by the customer movement information calculation processing unit 1081.

The product movement information table 1113 is a table for storing the product movement information collected by the product movement information calculation processing unit 1082. Product purchase information table 1114
Is a table for storing merchandise purchase information transmitted from the checkout machine 102 via the communication device 106.

The selection action classification condition table 1115 is a table for storing the classification pattern of the selection action of the product registered in advance from the input device 107 or the like. The product selection behavior information table 1116 is a table for storing customer's product selection behavior information when a customer moves in a store, selects a product, and considers purchasing a specific product.

Store layout example 10 is an example showing the layout of a store of clothing. Furniture 11 is a device such as a display shelf for displaying products.

FIG. 1 shows an example of installation of various devices in a layout in a store. A fixture sensor 101, a settlement machine 102, an image input device 103, and a transmission / reception distance calculation processing device 105 are installed in the store and a communication device. 106 and a communication line.

The image input device 103 is used to continuously track the movement of the customer in the store and to calculate the accurate position of the customer. Must be installed in a place where the entire store can be photographed. When the blind spot is generated by the furniture 11 or the like, it is necessary to install a plurality of image input devices 103 so that the blind spot does not occur.

FIG. 2 is a diagram showing a configuration example of the transmission / reception device 104 attached to a product of the present embodiment. As shown in FIG. 2, the transmission / reception device 104 of the present embodiment has a transmission / reception device identification ID.
201, transmission state information 202, and a control device 203.

The transmission / reception device identification ID 201 is an ID number for identifying each transmission / reception device 104. The transmission state information 202 is information set by the operation of the furniture sensor 101, and is information indicating whether or not the transmission / reception device 104 can transmit information for specifying the position of a product.

The control device 203 always receives the signal transmitted from the transmission / reception distance calculation processing device 105, and, when the transmission status information 202 is a value capable of transmitting, the transmission / reception device identification ID 2
01 is a device that sends back information to the transmission / reception distance calculation processing device 105 after adding the information 01.

The transmission / reception device identification ID 201 and the transmission status information 202 are stored in a memory in the transmission / reception device 104, and the product ID of the reception / transmission device ID 20
It is assumed that whether the receiving / transmitting device 104 is attached is registered in advance.

FIG. 3 is a diagram showing a configuration example of the transmission / reception distance calculation processing device 105 of the present embodiment. As shown in FIG. 3, the transmission / reception distance calculation processing device 105 of the present embodiment includes a clock unit 30.
1, the identification ID 302, and the transmission / reception device identification ID recognition unit 3
03, the transmission / reception distance calculation processing unit 304, and the communication control unit 3
05 and a communication control unit 306.

The clock section 301 is a transmission / reception distance calculation processing device 1
A circuit 05 measures the time so that processing can be performed at each specific measurement time, and notifies the transmission time and the reception time. Identification I
D302 is a different ID number for identifying each transmission / reception distance calculation processing device 105.

The transmission / reception device identification ID recognition unit 303 receives the transmission / reception device identification ID 20 transmitted from the transmission / reception device 104.
1 is a processing unit for recognizing No. 1. Transmission / reception distance calculation processing unit 30
4 is a receiving / transmitting device 104 and a receiving / transmitting distance calculation processing device 105
This is a processing unit that calculates the distance between.

The communication control unit 305 is a processing unit that controls transmission and reception of information between the transmission / reception device 104 and the transmission / reception distance calculation processing device 105. The communication control unit 306 is a processing unit that controls transmission of information from the transmission / reception distance calculation processing device 105 to the communication device 106.

Transmission / reception device identification ID 201 and identification ID 30
Data format different from 2 or data without duplication,
It is assumed that erroneous recognition between the transmission / reception device identification ID 201 and the identification ID 302 is avoided.

The transmission / reception distance calculation processing device 105 is a computer such as a personal computer.
The identification ID 302, the transmission / reception device identification ID recognition unit 303, the transmission / reception distance calculation processing unit 304, the communication control unit 305, and the communication control unit 306 are stored in a memory in the transmission / reception distance calculation processing unit 105, and are used for the transmission / reception distance calculation processing. C in device 105
It shall be processed by PU.

FIG. 4 is a diagram showing an example of the data configuration of the product attribute information table 1111 of this embodiment. As shown in FIG. 4, the product attribute information table 1111 of the present embodiment includes a product number 401, a product classification 402, a product name 403,
It has a color 404, a size 405, and a unit price 406.

The product number 401 is a number for identifying a product such as absolute single item information that can also identify the color 404 and the size 405, for example. The product category 402 is information indicating a product type such as a shirt, a blouse, and a jacket.

The product name 403 is information indicating the name of the product. The color 404 is information indicating the color of the product. Size 4
05 is information indicating the size of the product. The unit price 406 is information indicating the price of a single product.

FIG. 5 is a diagram showing an example of the data structure of the customer movement information table 1112 of the present embodiment. As shown in FIG. 5, the customer movement information table 1112 of the present embodiment has an action number 501, a measurement time 502, and position information 503.

The action number 501 is a number for distinguishing a specific customer's continued action from entering the store to leaving the store. The measurement time 502 is the time when the position of the specific customer in the store is measured. The position information 503 is information indicating the position of the customer measured at the measurement time 502, and is, for example, two-dimensional coordinate information represented by an x-axis and a y-axis.

One corresponding to one measurement time 502
One piece of position information 503 exists, and a plurality of measurement times 502 and position information 503 exist for one action number 501. An example of the position information 503 will be described later.

FIG. 6 is a diagram showing an example of the data structure of the product movement information table 1113 of this embodiment. As shown in FIG. 6, the product movement information table 1113 according to the present embodiment includes a product number 601, a measurement time 602, and product position information 603.
And

The product number 601 is a number for identifying a product. The measurement time 602 is the time when the position of the product in the store was measured. Product location information 603 is measurement time 6
02 is information indicating the position of the product measured, for example, x
This is two-dimensional coordinate information represented by an axis and a y-axis.

The one corresponding to one measurement time 602
There is one product position information 603 and one product number 60
A plurality of measurement times 602 and product position information 60 for one
There are three. An example of the product position information 603 will be described later.

In the commodity selection behavior information calculation system of this embodiment, the commodity selected by the customer by searching for data whose time and position match within a specific range in the customer movement information table 1112 and the commodity movement information table 1113. , The time information for setting the measurement time 502 and the measurement time 602 needs to match within a specific error range.

It is necessary that the position information 503 of the customer movement information table 1112 and the data format of the coordinate system of the commodity position information 603 of the commodity movement information table 1113 also match.

FIG. 7 is a diagram showing an example of the data structure of the product purchase information table 1114 of this embodiment. As shown in FIG. 7, the product purchase information table 1114 of the present embodiment includes a purchase number 701, position information 702 of a checkout machine, a purchase time 7
03, product number 704, unit price 705, and quantity 706
And the total amount 707.

The purchase number 701 is a number for identifying the contents of a purchase made by the customer at one time. The payment machine position information 702 is information indicating the position of the payment machine 102 in the store.
For example, it is two-dimensional coordinate information represented by an x-axis and a y-axis.

The purchase time 703 is the time when the product was purchased. The product number 704 is a number for identifying a purchased product. The unit price 705 is information indicating a single item price of a purchased product. The quantity 706 is information indicating the number of purchased commodities. The total price 707 is the total price of the products purchased at one time.

The information on the merchandise purchased by the customer is
2 is transmitted via the communication device 106 and stored in the product purchase information table 1114 of the storage device 111.
Product number 704, unit price 7 for one purchase number 701
05 and the product quantity 706 may be plural. An example of the position information 702 of the checkout machine will be described later.

Payment machine position information 702 and purchase time 70
3 and the measurement time 50 of the customer movement information table 1112
2 and the position information 503 are used to identify a customer who was at the position of the checkout machine 102 at that time by searching for data whose time and position matched within a specific range, and to specify a product purchased by the customer. Therefore, the time information for setting the measurement time 502 and the purchase time 703 needs to match within a specific error range.

It is necessary that the data format such as the coordinate system unit of the position information 702 of the checkout machine and the position information 503 of the customer movement information table 1112 also match.

FIG. 8 is a diagram showing an example of the data configuration of the selection action classification condition table 1115 according to the present embodiment. As shown in FIG. 8, the selection behavior classification condition table 1115 of this embodiment
Has a selection action 801 and a classification condition 802.

The selection action 801 includes, for example, comparison, combination,
This is information for classifying an action when a specific product is selected from a plurality of products such as another selection. The classification condition 802 is a condition for classifying the behavior and determining the selection behavior 801.

For example, when purchasing clothing, the customer examines and selects products by “comparing” or “combining” the products. Then, in the case of “comparing” and selecting products, the same product such as a shirt and a shirt, and similar products such as a shirt and a blouse are selected.

Therefore, the classification condition 802 of a product that takes a selection action 801 of “comparison action” for a certain product is as follows.
For example "shirt and shirt" or "shirt and blouse"
And the like. Similarly, selection action 8 called “combination”
An example of the classification condition 802 of a product that takes 01 is “jacket and pants”.

The selection action classification condition table 1115 has
A plurality of classification conditions 802 are provided for one selection action 801 by a preset number, and are input from the input device 107 or the like and stored in the storage device 111.

FIG. 9 is a diagram showing an example of the data configuration of the product selection action information table 1116 of this embodiment. As shown in FIG. 9, the product selection action information table 1116 of the present embodiment
Is the action number 901, the product number 902, and the product category 9
03, a purchase flag 904, a selection action 905, and a product number 906.

The action number 901 is a number for distinguishing a specific customer's continued action from entering the store to leaving the store. The product number 902 is a number for identifying the product selected in the action. The product category 903 is information for identifying the type of the product.

The purchase flag 904 is a flag indicating whether or not the product has been purchased as a result of selecting and examining the product. The selection action 905 is information indicating an action when the product is selected and examined. The product number 906 is the number of another product targeted for the selection action 905.

When a plurality of products are selected from one action number 901, a plurality of product numbers 902 are stored for one action number 901.

FIG. 10 is a diagram showing an example of position information display in a store according to the present embodiment. The starting point 1001 is a point on the bottom of the store, and is the origin of the coordinate axes. x-axis 10
02 is a horizontal coordinate axis passing through the starting point 1001. The y-axis 1003 is a vertical coordinate axis passing through the starting point 1001.
Customer position information example 1004 is an example of information indicating the position of the customer. The product position information example 1005 is an example of information indicating the position of a product.

The position information display shown in this example includes the position information 503 of the customer movement information table 1112, the product position information 6
03, a display common to the position information 702 of the checkout machine.

In the merchandise item selection behavior information calculation system of this embodiment, the position information is two-dimensional coordinate information represented by two axes x-axis 1002 and y-axis 1003, and the floor when the store is viewed from directly above. It represents the coordinates of the surface. The coordinate axis defines a point on the floor as a starting point 1001, and the starting point 1001
A straight line passing through the x-axis 1002 and the starting point 1001
, And a straight line perpendicular to the x-axis 1002 is set as the y-axis 1003.

Both the x-axis 1002 and the y-axis 1003 are straight lines parallel to the floor surface, one of which is the + (plus) direction and the other is the-(minus) direction, starting from the starting point 1001. For example, in the example of FIG. 10, it is assumed that the store has a rectangular floor, and a starting point 1001 is set at the end of the store. One side of the store passing through the starting point 1001 is an x-axis 1002,
The other side is set as the y-axis 1003, and the x-axis 1003 is set.
Both the 2 and y axes 1003 are in the + (plus) direction.

The position information is represented by the intersection of the distance from the starting point 1001 in the x-axis 1002 direction and the distance in the y-axis 1003 direction. For example, the customer location information example 1004 is the x-axis 1
600cm in 002 direction, 240c in y-axis 1003 direction
m and can be expressed as (600, 240). The product position information example 1005 is 150 cm in the x-axis 1002 direction and 280 cm in the y-axis 1003 direction, and can be expressed as (150, 280).

FIG. 11 is a flowchart showing a processing procedure of the customer movement information calculation processing unit 1081 of the present embodiment.
According to this flowchart, the calculation processing of the customer movement information will be described. In the product selection behavior information calculation system according to the present embodiment, the movement of a specific customer entering the store is tracked using the image information transmitted from the image input device 103, and the position information 503 is calculated for each measurement time 502. I do.

As a conventional method of tracking the movement of a specific customer, there is a method of tracking a moving object using a moving image. In this method, the same fixed place is first photographed as a moving image such as a video. Next, from information stored by cutting a series of motions of the moving image into frames, a certain frame and the next frame are observed, and a change point (an image that has moved with respect to an image that does not move such as a background) is calculated by a calculation process. And memorize. Next, the change point is registered as a tracking point, and the change point of the same image in each frame is continuously stored to track a specific moving object.

Also in the commodity selection behavior information calculation system of the present embodiment, an object (customer) moving into a store is specified and tracked based on the image information obtained from the image input device 103.

In step 1101, the image information acquired from the image input device 103 at each preset measurement time 502 is processed by the processing device 108 via the communication device 106.
To the customer movement information calculation processing unit 1081.

In step 1102, a change point newly entered in the image, that is, a customer entering the store is calculated to specify a tracking point. In step 1103, an action number 501 is assigned to the specified tracking point.

In step 1104, the position of the specified tracking point in the store is calculated. First, the image information of the background of the store in a state where there is no moving object is acquired in advance, the position information 503 of the floor surface shown in the image is analyzed, and the information is stored in the storage device 111.

The acquired image information specifying the tracking point is superimposed on the coordinate information of the background image analyzed in advance, and the position information 503 of the tracking point is analyzed and calculated from the positional relationship with the background image.

Next, in step 1105, it is determined whether or not the position information 503 of the calculated tracking point is within the measurement range, that is, the coordinates in the store. If the calculated position information 503 of the tracking point is not the coordinates in the store, it is assumed that the customer has already left the store, and the process ends without storing the information.

If the calculated tracking point position information 503 is a coordinate in the store, the storage device 11 in step 1106
The numbered action number 501, the measurement time 502, and the position information 503 are stored in one customer movement information table 1112, and the coordinates of the next measurement time 502 are analyzed.

Thus, the process of calculating the position of the tracking point is continued until a customer comes out of the store. The above is the processing for calculating the customer movement information.

It should be noted here that the image input device 103 is fixed and the same place is photographed. Otherwise, an image that does not originally move, such as the background, also moves, and a change point cannot be captured.

Therefore, in the merchandise item selection behavior information calculation system of this embodiment, the image input device 103 is installed at the upper part of the store, and the whole store can be photographed. In order to install the image input device 103 so that the entire store can be photographed, the image input device 103 is necessarily installed at the upper part of the store. Must be installed at

In the commodity selection behavior information calculation system of the present embodiment, the position of the customer is grasped by the coordinates of the floor. However, in order to be installed in the upper part of the store and to track without being interrupted by the blind spot, the customer's head must be tracked. Normally, it is more accurate to capture the position of the foot that touches the floor surface, but the foot position often enters the blind spot due to the furniture 11, and the head position is less likely to enter the blind spot, so the tracking is continued. Because it is easy to do.

Then, since the height of each customer is different, if the image is viewed obliquely, an error occurs between the position of the head and the position of the floor. Therefore, in the product selection behavior information calculation system of the present embodiment, it is assumed that the customer can be photographed almost vertically from the top of the store to the floor, and the error between the position of the head to be tracked and the position on the floor is ignored. .

If the error between the position of the head to be tracked and the actual position of the floor surface becomes large due to the height, when the tracking point is specified at the entrance, the image of the customer from the side or a sensor installed at the entrance is used. When the height is measured and the position of the customer is analyzed in step 1104, the position may be corrected by adding height information.

The customer movement information table 1112 is information obtained by the processing unit 108 as described above. However, information on customers, children, clerks, suppliers, and the like that are not sold in the store can be stored by a person. By judging and deleting, it is possible to further improve the accuracy of the information.

FIG. 12 is a flowchart showing the processing procedure of the product movement information calculation processing section 1082 of the present embodiment.
The calculation processing of the product movement information will be described according to this flowchart. In the product selection behavior information calculation system of the present embodiment, a signal is transmitted and received between the transmission / reception device 104 and the reception / transmission distance calculation processing device 105 attached to the product at each measurement time 602, and the device is determined based on the transmission time of the signal. The product distance information 603 is calculated by calculating the distance between the products.

The receiving / transmitting device 104 has transmission status information 202, and the furniture sensor 101 indicates that the transmission status information 202 indicates that transmission is possible from a value indicating that transmission is possible or that transmission is possible from a value indicating that transmission is not possible. Value.

At the time when the product is displayed on the furniture 11, the transmission status information 202 has a value indicating that transmission is not possible. When a product is taken out of the furniture 11 by someone's hand, the receiving / transmitting device 104 detects a signal from the furniture sensor 101 attached to the furniture 11 and transmits the transmission status information 20.
2 is set to a value indicating that transmission is possible.

Only while the transmission status information 202 has a value indicating that transmission is possible, the processing of the flowchart described below is continuously performed. When the product is returned to the furniture 11, the receiving / transmitting device 104 senses the signal of the furniture sensor 101, and the transmission status information 202 is set from a value indicating that transmission is possible to a value indicating that transmission is not possible.

As a result, it is possible to acquire the product movement information only while the product is being taken out of the furniture 11 and is moving, to reduce the data amount of the product movement information, and at the time of associating with the customer movement information. It is possible to reduce the error and increase the accuracy of the information.

The receiving / transmitting device 104 of the product selection action information calculation system of the present embodiment can transmit information for specifying the position of a product only while the transmission status information 202 has a value indicating that transmission is possible. However, a switch is provided instead of the transmission state information 202, and the switch is turned on.
Only during the period, information for specifying the position of the product may be transmitted.

Hereinafter, the process of calculating the product movement information when the transmission status information 202 has a value indicating that transmission is possible will be described. First, in step 1201, the identification ID 302 and the measurement time 602 are transmitted from the transmission / reception distance calculation processing device 105 at every preset measurement time 602.

In step 1202, the information transmitted from the transmission / reception distance calculation processing device 105 is received by the transmission / reception device 104. In step 1203, the transmission / reception device 104
A reply is made by adding the information of the transmission / reception device identification ID 201 to the received information.

In step 1204, the transmission / reception distance calculation processing device 105 selects the step 120 from the returned information.
Only the information having the same ID as its own ID 302 transmitted in 1 is received, and the reception time is measured.

The round trip response time from transmitting and receiving information from the transmission / reception distance calculation processing device 105 to returning from the transmission / reception device 104 is very short, and the processing from step 1201 to step 1204 is performed at the next measurement time 602. When the process is completed sufficiently, the processing can be simplified such that only the identification ID 302 is transmitted from the transmission / reception distance calculation processing device 105 and the transmission / reception device 104 returns with the transmission / reception device identification ID 201 added. . This is the next measurement time 602
By this time, the process is completed, so that it is possible to determine which measurement time 602 the returned information is without transmitting the measurement time 602.

At step 1205, the difference between the reception time and the transmission time is calculated. This means that the transmission / reception distance calculation processing device 105
Is the information transfer time of the round trip between the transmission and reception device 104, and the following formula (information transfer time) * (information transfer speed) / 2 =
The distance is calculated by (the distance between the transmission / reception distance calculation processing device 105 and the transmission / reception device 104).

The information transfer time includes the processing time in the transmission / reception device 104, but is not so complicated in the commodity selection behavior information calculation system of the present embodiment, so that the processing time can be ignored here. It is assumed that

In step 1206, the measurement time 602
(I.e., the transmission time), the identification ID 302, the transmission / reception device identification ID 201 of the transmission / reception device 104 that has returned the information, and the distance between the transmission / reception distance calculation processing device 105 and the transmission / reception device 104 via the communication device 106. The information is transmitted to the product movement information calculation processing unit 1082 of the device 108.

In step 1207, the product movement information calculation processing unit 1082 sorts the transmitted information of each transmission / reception distance calculation processing device 105 for each transmission / reception device identification ID 201.

At step 1208, an intersection at which the distance from the previously registered position of the transmission / reception distance calculation processing device 105 to the distance between the transmission / reception device 104 is calculated. This is the position of the transmission / reception device 104, and the product position information 603
It is.

In step 1209, the product number 60 of the product to which the receiving / transmitting device 104 is attached is registered in advance.
1 is read, and the product number 601, the measurement time 602, and the calculated product position information 603 are stored in the product movement information table 1113 of the storage device 111.

In order to specify the point of intersection between the position of each transmission / reception distance calculation processing device 105 and the distance from the transmission / reception device 104 to one point on the two-dimensional coordinates, at least three reception / transmission distance calculation processing devices 105 Information on each location and information on the distance from it is required. Therefore, the transmission / reception distance calculation processing device 1
05 needs to be installed in at least three places in the store. Note that the transmission / reception distance calculation processing device 105 has at least four
It may be provided at three locations to specify the position of the transmitting / receiving device 104 on three-dimensional coordinates.

Next, the calculation processing of the customer's product selection action information will be described with reference to FIGS. FIG. 13 is a flowchart illustrating a processing procedure of the product selection behavior calculation processing unit 1083 of the present embodiment. In the product selection behavior information calculation system of the present embodiment, the product selected from the time the customer enters the store until the customer exits is specified using the calculated customer movement information and the product movement information.

At step 1301, one action number 5 is stored in the customer movement information table 1112 of the storage device 111.
01, the measurement time 502 corresponding to the number, and the position information 503 are read.

In step 1302, information on the product selected for each action number 501 (customer's product selection behavior) from the read customer movement information and the product movement information similarly read from the product movement information table 1113 of the storage device 111. Information) is calculated and stored in the product selection action information table 1116 of the storage device 111.

Next, in step 1303, step 13
From the customer movement information read out in step 01 and the product purchase information read out from the product purchase information table 1114 of the storage device 111, the product purchased in the action number 901 corresponding to one action number 501 read out in step 1301 is calculated. , The purchase flag 90 of the corresponding product number 902 in the product selection action information table 1116 of the storage device 111.
Make 4

In step 1304, all action numbers 50
It is checked whether the processing of steps 1302 and 1303 has been performed for No. 1, and when the above processing has been performed for all the action numbers 501, the processing ends. When there is an action number 501 that has not been performed, the processing returns to step 1301.

Next, the process of calculating the product selection action information in step 1302 in FIG. 13 will be described in detail with reference to the flowchart in FIG. FIG. 14 is a flowchart illustrating a procedure of a process of calculating product selection behavior information according to the present embodiment. In step 1401, one product number 601, measurement time 602 and product position information 60 corresponding to the product number 601 are stored in the product movement information table 1113 of the storage device 111.
Read 3

In step 1402, the measurement time 502 of the customer movement information read out in step 1301 of FIG.
And the measurement time 602 of the product movement information read in step 1401 matches within a specific range, and the position information 503 of the customer movement information and the product position information 603 of the product movement information
Searches for data that matches within a specific range.

In the merchandise item selection behavior information calculation system of the present embodiment, each of the position information 503 and the merchandise position information 6
A search is made for a value in which the numerical value of the x-axis 1002 and the numerical value of the y-axis 1003 of 03 match within a specific range. For example, if position information 503 of customer movement information = (a, b), product position information 603 of goods movement information = (c, d), and a specific range is δ, |
The position information 503 and the merchandise position information 603 satisfying ac | <δ and | bd | <δ are searched. This indicates that the customer has performed a product selection action, and thereby the product selected by the customer is specified.

At step 1403, it is checked whether or not there is data satisfying the condition. If data satisfying the condition is detected, the process proceeds to step 1404. If no data that satisfies the above condition is detected, it is determined that the product has not been selected, and steps 1406 to 1
Returning to 401, the product movement information of another product number 601 is read.

In step 1404, the process proceeds to step 1402 from the product attribute information table 1111 in the storage device 111.
The information of the product number 401 that matches the product number 601 of the matching data is searched, and the product classification 402 is read out.

In step 1405, step 1402
, The action number 501 read in step 1301 of FIG. 13, step 14
The product classification 402 read in 04 is stored in the product selection action information table 1116 of the storage device 111.

In this case, when there is a plurality of matching data in one product number 601, only one product number 601 is stored. This is because the measurement time 502 and the measurement time 602 are the same when the customer moves while holding the product.
This is because the product and the customer move in the same position, and the selected product is one.

Next, in step 1402, the matched data is cleared as processed, and it is checked in step 1406 whether or not all the product information has been processed.
If the process of 405 has been performed, the process ends. If there is a product number 601 that has not been processed, the process returns to step 1401.

It is to be noted that since the position information 503 and the product position information 603 are actually expected to have some errors, those within a specific range registered in advance are regarded as coincident.

It is expected that the error varies greatly depending on the characteristics of the product. For example, in the case of clothing, the user often applies the product to his / her body and looks in a mirror or tries on the product, so it can be seen that there is almost no error. Therefore, in the commodity selection behavior information calculation system of the present embodiment, it is assumed that the error of the commodity position information 603 of clothing can be ignored. Products worn like accessories for clothing (accessories, hats, bags, shoes, etc.) are considered to be similar to clothing.

[0138] On the other hand, for a product such as a CD, for example, it is expected that the error will be larger than that for clothing. Actually, the range of the error is considered to be the distance between the position where the customer stands and the product held in the hand. Since this distance is smaller than the distance between other customers when selecting a product in front of the furniture 11, it does not greatly affect the specification of the product that has taken the selection action 905 for each customer.

Other errors, such as errors in the position of the tracking point in the calculation of the customer movement information and errors in the processing time of the receiving / transmitting device 104 in the calculation of the product position information 603, are also detected before the furniture 11 If the distance is smaller than the distance between the customers when the item is selected, it is basically considered that there is no problem in obtaining the desired result of the product selection behavior information calculation system of the present embodiment.

Next, the process of calculating the product purchase information in step 1303 in FIG. 13 will be described in detail with reference to the flowchart in FIG. FIG. 15 is a flowchart illustrating the processing procedure of the commodity purchase information calculation processing according to the present embodiment.
In step 1501, the product purchase information table 1114 is read from the storage device 111. In step 1502, the measurement time 502 in the action number 501 of the customer movement information read out in step 1301 of FIG. 13 matches the purchase time 703 of the product purchase information within a specific range, and the position information 503 of the customer movement information. A search is made for a purchase number 701 that matches the position information 702 of the payment machine within a specific range.

In the merchandise item selection behavior information calculation system according to the present embodiment, the value of the x-axis 1002 and the value of the y-axis 1003 of the position information 503 and the position information 702 of the checkout machine can be regarded as the same within a specific range. Search for. For example, if position information 503 of customer movement information = (a, b), position information 702 of a checkout machine = (c, d), and a specific range is δ,
The position information 503 of | ac | <δ and | bd | <δ and the position information 702 of the checkout machine are searched. This means that the customer has moved to a location of the checkout machine 102 and purchased the product.

At step 1503, it is checked whether or not a purchase number 701 satisfying the above conditions exists.
When 1 is detected, the process proceeds to step 1504. If no matching purchase number 701 is detected, the process ends as a product is not purchased.

In step 1504, the product number 704 belonging to the detected purchase number 701 is read from the product purchase information table 1114, and the product number 902 stored in the product selection action information table 1116 of the storage device 111 in step 1405 in FIG. A number matching the product number 704 is searched.

In step 1505, step 1504
Product number 9 that matches the product number 704 detected in
A purchase flag 904 corresponding to 02 is set.

Next, the classification process of the customer's product selection behavior will be described with reference to the flowchart of FIG. FIG. 16 is a flowchart illustrating a processing procedure of a classification process of a customer's product selection behavior according to the present embodiment. In step 1601, the product selection behavior calculation processing unit 1083 reads one behavior number 901 and the product selection behavior information from the product selection behavior information table 1116 stored in the storage device 111.

In step 1602, one action number 90
Check if there are multiple product numbers 902 in one,
If there are a plurality of product numbers 902, the next step 160
Proceeding to step 3, if there is only one product number 902, the process returns to step 1601 to read another action number 901.

At step 1603, one product number 902 and its product category 9 are read from the read product selection action information.
Read 03. In step 1604, the storage device 11
Selection action classification condition table 1115 stored in No. 1
One of the selection actions 801 and the corresponding classification condition 802 are read out.

In step 1605, step 1603
Product classification 903 and classification condition 802 of step 1604
Then, another product number 902 having a product classification 903 that matches the classification condition 802 within the same action number 901 is detected. This classifies what kind of selection action 801 such as “comparison” or “combination” is taken in the product selection action to select a product.

In step 1606, step 1604
The selection action 801 in the product selection action information table 1116 of the storage device 111 and the selection action 801 read out in step 1605 and the other product number 902 detected in step 1605 as selection action information corresponding to the product number 902 read out in step 1603. This is stored in the product number 906, respectively.

In Step 1607, all the selection actions 80
It is checked whether the processing of steps 1604 to 1606 has been performed for No. 1 and the processing proceeds to step 1608 if the processing has been performed for all the selection actions 801, and returns to step 1604 if there is any selection action 801 that has not been performed.

At step 1608, the action number 901
It is checked whether or not the processing of Steps 1603 to 1606 has been performed for all the product numbers 902, and if the processing has been performed for all the product numbers 902, Step 1
Proceed to 609. Product number 902 that has not been processed
If there is, the process returns to step 1603.

In step 1609, all the action numbers 90
It is checked whether or not the process has been performed for the number 1, and if the process has been performed for all the activity numbers 901, the process ends. If there is an unprocessed action number 901, the process returns to step 1601.

According to the product selection behavior information calculation system of the present embodiment, there is an advantage that a product selected by a certain customer in a store can be specified accurately without manual intervention. Further, it is possible to specify whether or not a certain customer has purchased the selected product in the store.

Furthermore, the ability to track the continuous selection behavior 905 of the customer has the advantage that the attributes of a plurality of selected products can be classified without any manual intervention from what point of view the products have been examined and selected. is there.

As the information of the product selection action information table 1116 of the storage device 111, for example, a list of information on a product compared with a specific product or a list of information on a combination of products is read, and the obtained information is totaled and sold to a store salesperson. It can be provided as promotion information.

[0156] The sales clerk can utilize the information to appeal to a new customer who visits the store as a substitute product proposal for another product, which often takes a comparison action when a product is out of stock. Can be promoted. It is also effective for sales promotion, such as introducing products with many combination actions to customers as recommendations of product coordination, and encouraging bulk purchases to increase the unit price per customer.

For example, for a specific product, it is also possible to read out a list of compared products or information on a combination of products and provide the totalized information to a product planning department, which can be used as a product policy verification.

[0158] Generally, at the time of formulating a clothing product policy,
We are studying combinations (coordination) of products that promote sales more. In other words, it is a hypothesis on the planning side before sales, and by verifying it with information on actual customer behavior, it becomes possible to formulate a more effective product policy in the next product policy.

Also, by comparing a list of products that have performed poorly-sold products and performing a combination action, and information on whether or not the purchase flag 904 is set for the products, for example, the products to be compared can be compared. Analyzing the causes of poor sales, such as whether the products cannot be sold due to good sales or the appeal of the combination intended by the product planning side is weak, enables effective countermeasures to be taken in the future. .

As described above, according to the commodity selection behavior information calculation system of the present embodiment, the information on the commodity selected, studied or purchased by the customer is calculated by linking the customer movement information and the commodity movement information in the store. Therefore, it is possible to collect customer product selection behavior information from customer movement information and product movement information within a store.

[0161]

According to the present invention, the customer movement information and the goods movement information in the store are linked to calculate the information of the goods selected, examined or purchased by the customer, so that the customer movement information and the goods movement in the store are calculated. It is possible to collect customer product selection behavior information from the information.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a schematic configuration of a product selection behavior information calculation system according to an embodiment.

FIG. 2 is a transmission / reception device 10 attached to a product of the present embodiment.
4 is a diagram illustrating a configuration example of FIG.

FIG. 3 is a diagram illustrating a configuration example of a transmission / reception distance calculation processing device 105 according to the present embodiment.

FIG. 4 is a diagram illustrating a data configuration example of a product attribute information table 1111 according to the present embodiment.

FIG. 5 is a diagram illustrating a data configuration example of a customer movement information table 1112 according to the present embodiment.

FIG. 6 is a diagram illustrating a data configuration example of a product movement information table 1113 according to the present embodiment.

FIG. 7 is a diagram illustrating a data configuration example of a product purchase information table 1114 according to the embodiment.

FIG. 8 shows a selection behavior classification condition table 111 according to the embodiment.
FIG. 9 is a diagram showing a data configuration example of No. 5;

FIG. 9 is a product selection action information table 111 according to the embodiment.
6 is a diagram illustrating a data configuration example of No. 6. FIG.

FIG. 10 is a diagram showing an example of position information display in a store according to the embodiment.

FIG. 11 is a customer movement information calculation processing unit according to the embodiment;
3 is a flowchart illustrating a processing procedure of No. 1;

FIG. 12 is a product movement information calculation processing unit 108 according to the embodiment.
6 is a flowchart illustrating a processing procedure of No. 2;

FIG. 13 is a merchandise item selection behavior calculation processing unit 108 of the embodiment.
13 is a flowchart illustrating a processing procedure of No. 3;

FIG. 14 is a flowchart illustrating a procedure of a process of calculating merchandise selection action information according to the embodiment.

FIG. 15 is a flowchart illustrating a procedure of a process of calculating merchandise purchase information according to the embodiment;

FIG. 16 is a flowchart illustrating a processing procedure of a classification process of a customer's product selection behavior according to the present embodiment.

[Explanation of symbols]

101 furniture sensor, 102 payment machine, 103 image input device, 104 receiving / transmitting device, 105 receiving / transmitting distance calculation processing device, 106 communication device, 107 input device, 1
08 ... processing device, 109 ... screen display device, 110 ... output device, 111 ... storage device, 1081 ... customer movement information calculation processing unit, 1082 ... product movement information calculation processing unit, 1083
... Commodity selection behavior calculation processing section, 1111 .commodity attribute information table, 1112.
Commodity transfer information table, 1114 ... Commodity purchase information table, 1115 ... Selection action classification condition table, 1116 ...
Product selection action information table, 10 ... store layout example,
11: furniture, 201: receiving / transmitting device identification ID, 202: transmission status information, 203: control device, 301: clock unit, 30
2 ... ID, 303 ... Receiver / transmitter ID recognition unit, 3
04: reception / transmission distance calculation processing unit, 305: communication control unit, 3
06 communication control unit, 401 product number, 402 product classification, 403 product name, 404 color, 405 size,
406 unit price, 501 action number, 502 measurement time,
503: position information, 601: product number, 602: measurement time, 603: product position information, 701: purchase number, 702
... Payment machine position information, 703 ... Purchase time, 704 ... Product number, 705 ... Unit price, 706 ... Quantity, 707 ... Total price, 801 ... Selection action, 802 ... Classification condition, 901 ... Action number, 902 ... Product number, 903: Product classification, 904
... purchase flag, 905 ... selection action, 906 ... product number,
1001 start point, 1002 x-axis, 1003 y-axis, 1
004: An example of customer location information, 1005: An example of product location information.

Claims (4)

[Claims] 1. A product selection behavior information calculation method for collecting behavior information when a customer selects a product, wherein position information of the customer in the store is calculated from image information in the store at specific time intervals, and the customer movement information is calculated. And collects the product movement information by calculating the position information of the product at specific time intervals based on the information from the receiving / transmitting device attached to the product, and collects the customer movement information and the product movement information collected by the customer. By associating the purchased product purchase information with the purchase information of a particular customer from the time the customer enters the store until the customer leaves the store, the information on the purchased product is calculated and collected as the product selection behavior information of the customer. Characteristic product selection behavior information calculation method. 2. The information of the selected, considered or purchased merchandise is classified according to a predetermined classification condition from the time when the specific customer enters the store until the specific customer enters the store, and the merchandise is selected by any selection action. 2. The method according to claim 1, wherein information indicating whether or not the item has been selected is collected. 3. A product selection behavior information calculation system for collecting behavior information when a customer selects a product, wherein customer position information in a store is calculated at a specific time interval from image information in the store, and customer movement information is calculated. A customer movement information calculation processing unit that collects product movement information, and a product movement information calculation processing unit that calculates product position information at specific time intervals based on information from a transmission / reception device attached to the product and collects product movement information, Correlate the collected customer movement information with the product movement information and the product purchase information purchased by the customer, and calculate information on the products selected, considered, or purchased between the time the particular customer enters the store and the time the customer leaves. And a product selection behavior calculation processing unit that collects the product selection behavior information of the customer. 4. The product selection behavior calculation processing section classifies information of a product selected, studied or purchased between the time the specific customer enters the store and the time the customer leaves the store in accordance with a preset classification condition, and 4. The product selection behavior information calculation system according to claim 3, wherein information indicating what selection behavior was selected is collected.