JP4209413B2 - Customer trend collection system and customer trend collection program - Google Patents

Description

  The present invention relates to a customer trend collection system, a customer trend collection method, and a customer trend collection program for associating flow line information representing a travel route of a customer with purchase information in a store such as a department store or a supermarket.

  As an example of conventional technology related to a customer trend collection system, a customer trend collection method, and a customer trend collection program, for example, there is a technique disclosed in Patent Literature 1, and according to Patent Literature 1, a customer is not burdened. A system and method for efficiently and automatically collecting the flow line information is disclosed.

  The system disclosed in Patent Literature 1 includes an imaging unit, an image input unit, a person detection unit, a person extraction unit, a feature vector generation unit, an entrance information storage unit, a person verification unit, and a person determination unit. , Comprising a totaling unit and an output unit, the imaging means installed at each of a plurality of positions in the facility including the entrance and entrance, captures the image, detects the presence of the person from the captured image, extracts the person image, The extracted person image is subjected to image processing to generate a person-specific feature vector.

More specifically, a feature vector is generated from a person image captured by the imaging means when a person enters, and the generated feature vector is added with a person identification code, an entrance time, etc., and stored as entrance information. . Then, the feature vector generated from the person image captured by the imaging means at the time of moving within the facility and leaving is compared with the feature vector stored at the time of entry, and when the same person is identified, the person's identification code, shooting time, shooting position Are stored as flow line information, and the flow line information of the same identification code is extracted from the stored flow line information to collect the individual flow line information.
JP 2000-200377 A

  However, in the above-described prior art, although it is possible to obtain sales floor stop information from the obtained flow line information, the obtained flow line information is not related to purchase information. Therefore, even if the customer's sales floor can be grasped from the flow line information, it is difficult to continue the flow line measurement by individually identifying each customer with the camera image because the customers are densely packed in the checkout lane. . That is, with the above-described conventional technology, it is difficult to associate the flow line information up to the settlement and the purchase information at the settlement.

  The present invention has been made in consideration of the above-described problems, and even if customers are crowded in the checkout lane, the customer can collect the flow line information so far in association with the purchase information at the time of settlement. A trend collection system, a customer trend collection method, and a customer trend collection program are provided.

In order to solve the above-described problem, the customer trend collection system according to the present invention sequentially captures images captured by the imaging means installed in the sales floor at regular time intervals as described in claim 1, and captures the captured images. Image processing is performed to detect and store the customer imaged in the image, and by specifying the position where the stored customer is stored for each captured image, any of the customers after entering the sales floor receiving a flow line measuring means which forms and outputs a flow line information indicative of the route that has moved the sales floor until arranged in the checkout lane, the flow line information output from the flow line measuring means, the received stores allocate space for storing the flow line information, and flow line information storage means for outputting the stored addresses allocated to store said flow line information, which is outputted from the flow line information storage means the Moving along with receive憶番locations information, it issues a specific flow line index for the flow line information for each stored in the flow line information storing means, are identified from the issued flow line index and the flow line index The flow line index issuing means for associating the storage address of the line information and storing it in a predetermined storage area, and the payment information storage means that is obtained via the POS terminal at the time of payment each time the customer is settled. while reading has been purchase information indicating a purchase products customer has finished the settlement from the purchase information storage means to receive said flow line index information from the flow line index issuing means showing the flow line index those the customer, read is stored in the connection information storage means in association with the flow line index information received with the purchase information and a customer behavior coupling means, The customer trend linking means has a check-out lane unit queue storage unit for storing, in order of waiting, flow-line index information for each customer forming a queue in the check-out lane unit. When the notification of the end of the flow line measurement is received, the queue storage unit for writing the flow line index information of the customer to the checkout lane unit queue storage unit of the checkout lane in which the customer is lined up, and the queue at the end of the checkout And a customer trend linking unit that associates the customer flow information with the customer purchase information and updates the flow index information stored in the queue storage unit .

In order to solve the above-described problem, the customer trend collection program according to the present invention sequentially captures and captures images taken by the imaging means installed in the sales floor at regular time intervals, as described in claim 5. After the customer enters the sales floor by specifying the position where the customer is stored for each captured image, by detecting and storing the customer imaged in the image by image processing the image The flow line measuring means for forming and outputting the flow line information indicating the route moved through the sales floor before being arranged in any checkout lane, and receiving and receiving the flow line information output from the flow line measuring means A flow line information storage means for allocating and storing an area for storing the flow line information and outputting a storage address assigned for storing the flow line information; and the flow line information storage means The information on the storage address output from the flow line is received, a unique flow line index is issued for each flow line information stored in the flow line information storage means, and the issued flow line index and the flow line index are issued. The flow line index issuing means for associating the storage address of the flow line information identified from the data and storing it in a predetermined storage area, and the purchase that is acquired through the POS terminal at the time of payment every time the customer is settled While reading the purchase information indicating the purchased product of the customer who has finished the settlement stored in the information storage means from the purchase information storage means, the flow line index information indicating the flow line index of the customer from the flow line index issuing means Receiving and reading the purchase information and the received flow line index information in association with each other and storing them in the link information storage means And the customer trend linking means has a checkout lane unit queue storage unit for storing, in a waiting order, flow line index information for each customer forming a queue in the checkout lane unit, When receiving notification of the end of the flow measurement of the customer from the line measuring means, a queue storage unit for writing the flow line index information of the customer to the checkout lane unit queue storage unit of the checkout lane where the customer is lined up, A customer trend linking unit that associates the flow line index information of the customer at the head of the queue with the purchase information of the customer at the end of the checkout and updates the flow line index information stored in the queue storage unit; This is a program that functions as a system .

  According to the customer trend collection system, the customer trend collection method, and the customer trend collection program according to the present invention, even if the customers are crowded around the checkout lane, the movement indicating the movement route from the customer entering the sales floor to the settlement is shown. Line information can be collected in association with purchase information at the time of checkout.

  In addition, using the collected information, it is possible to verify whether or not the product is actually purchased at the place where the customer visited after the fact, and analysis of the effects of in-store advertisements and sales floor layouts is more than before. It can be done easily.

  Embodiments of a customer trend collection system, a customer trend collection method, and a customer trend collection program according to the present invention will be described with reference to the accompanying drawings.

  In the present embodiment, in a store such as a department store or a supermarket, the customer's movement route, that is, the flow information indicating the flow of the customer and the purchase information indicating the information of the product purchased by the customer are associated with each other. An example of collecting customer trends is described below.

  FIG. 1 is a layout diagram schematically showing a layout in a store 1 to which a customer trend collection system, a customer trend collection method, and a customer trend collection program according to the present invention are applied.

  In the store 1 shown in FIG. 1, reference numerals 2a and 2b are entrances and exits for the customer 3 to enter and exit the store 1, reference numerals 4a to 4i are merchandise display shelves, and reference numerals 5a to 5d are POS terminals and the like. A checkout lane, 6 is a soccer stand for bagging purchased products. In addition, cameras 7a to 7i serving as imaging means for imaging the movement status of the customer 3 in the sales floor are installed above the passage formed by the respective merchandise display shelves 4a to 4i arranged on the floor.

  FIG. 2 is a block diagram schematically showing the system configuration of the customer trend collection system 10 according to the embodiment of the present invention.

  As shown in FIG. 2, the customer trend collection system 10 includes a flow line measuring unit 11 that forms flow line information indicating a movement route (flow line) in the sales floor of the customer 3, and each of the flow line measuring unit 11 formed. Flow line index issuing means 12 for issuing a flow line index unique to the flow line information of the flow line; flow line information storage means 13 for storing the flow line information formed by the flow line measuring means 11 for each flow line index; Point-of-sales information management means 14 for managing the purchase products of the customer 3, purchase information storage means 15 for storing information on the purchase products managed by the point-of-sales information management means 14 for each customer 3, and the flow of the same customer 3 Customer trend linking means 16 for associating information with purchase information, and linked information storage means for storing information (hereinafter referred to as linking information) in which customer trend linking means 16 associates flow line information and purchase information of the same customer 3 1 Comprising the door.

  Although omitted in FIG. 2, the customer trend collection system 10 includes an input interface, and can interrupt, resume, and end the processing operation in units of components.

  Then, each component 11-17 in the customer trend collection system 10 is demonstrated.

  FIG. 3 is a block diagram schematically showing the configuration of the flow line measuring means 11 in the customer trend collection system 10.

  The flow line measuring means 11 includes a camera 7a to 7i that captures the movement status of the customer 3 in the store 1 shown in FIG. 1, an image input unit 21 that captures images captured by the cameras 7a to 7i, and an image input unit 21. A person detection unit 22 that detects the presence of a person, that is, the customer 3 from the captured image, and a flow line formation unit 23 that obtains flow line information based on the position information and time information of the customer 3 detected by the person detection unit 22. Prepare.

  The image input unit 21 has a function of capturing images taken by the cameras 7a to 7i. Images captured by the cameras 7a to 7i are input to the image input unit 21 as needed. The image input unit 21 has current time information indicating the current time, and passes the current time information to the person detection unit 22.

  The person detection unit 22 has a function of reading the images input to the image input unit 21 from the cameras 7a to 7i and captured, for each of the cameras 7a to 7i. Further, the person detection unit 22 has a function of detecting a person existing in the image read from the image input unit 21, that is, the customer 3. Furthermore, the person detection unit 22 has a function of extracting, storing, collating, and erasing a person feature for the detected customer 3.

  The person detection unit 22 receives the current time information from the image input unit 21, and reads out the images of the cameras 7a to 7i captured in the image input unit 21 at regular time intervals (sampling periods). And it is detected whether the customer 3 exists in the read image. That is, the person detection unit 22 detects the presence / absence of the customer 3 in the image at the sampling period.

  The presence / absence of the customer 3 in the read image is detected by, for example, obtaining a difference value between the image without a person (background image) at each camera position captured in advance and the current image read from the image input unit 21. Made. Further, when detecting the customer 3, the person detecting unit 22 extracts and stores the characteristics of the detected customer (person) 3. Furthermore, when the person image of the customer 3 that has already been detected deviates from one camera image as the customer 3 moves, the person detection unit 22 tracks the person image of the customer 3 using another camera image. More specifically, the customer 3 detected by collating the characteristics of the customer 3 extracted from another camera image with the characteristics of the person already stored in the person detection unit 22 is tracked as the same person.

  On the other hand, when the person detection unit 22 determines whether or not the customer 3 is queued, and determines that the customer 3 is queued, the person detection unit 22 issues a request to end the flow line measurement of the customer 3 to the flow line. Send to forming unit 23.

  The flow line forming unit 23 has a function of converting the position of the customer 3 in the image where the person detecting unit 22 detects the customer 3 into position information in the store layout. Here, the position information in the store layout is, as shown in FIG. 1, how much from the reference point (origin O) on the xy plane representing the plane parallel to the floor of the store 1 in the x direction and the y direction. , That is, information indicating the coordinates (x, y) (hereinafter simply referred to as coordinate information).

  Further, the flow line forming unit 23 obtains position information in the store layout obtained by converting the position of the customer 3 moving in the store 1 and the current time information acquired from the image input unit 21 by the person detection unit 22. Connect. As a result, the flow line forming unit 23 obtains information indicating the movement route of the customer 3 moving in the store 1, that is, position information in the store layout of the customer 3 moving in the store 1, and a position based on the position information. The flow line information having the time information indicating the time when the customer 3 is present can be formed.

  The position information in the flow line information is not necessarily position information (coordinate information) obtained by capturing the position of the customer 3 with a point. For example, position information (hereinafter referred to as sales area information) that captures the position of the customer 3 based on individual divided areas (hereinafter referred to as sales areas) formed by dividing the store 1 shown in FIG. 1 into a plurality of parts is used. May be. Here, the sales floor area information is information indicating an identifier assigned to the sales floor area.

  In addition, when sales floor area information is applied as position information in the flow line information, information indicating the time at which the customer 3 is in the sales floor area is used as time information. In this case, the position is fixed at one point. Because there is no, there is no single time in the sales area. Therefore, it is necessary to define so that the time in the sales area is uniquely determined.

  The time in the sales floor area based on the sales floor area information is at least after the time of entering the sales floor area (hereinafter referred to as the entry time) and before the time of exiting the sales floor area (hereinafter referred to as the exit time). Any time that is uniquely defined among the times may be used. For example, any of the following times (a) to (c) is a time that is uniquely determined and can be applied as a time in the sales area.

(A) Time to enter the sales area (b) Time to exit from the sales area (c) Time in the middle between entering the sales area and exiting (hereinafter referred to as intermediate time)

  Here, the intermediate time is obtained by calculating the following Equation 1.

[Equation 1]
Intermediate time = entry time + (exit time-entry time) / 2 (1)
Applying sales floor area information as position information in the flow line information is advantageous in that the amount of information of the flow line information can be reduced as compared with the case of applying coordinate information.

  On the other hand, when forming the flow line information, the flow line forming unit 23 delivers the formed flow line information to the flow line index issuing unit 12 and the flow line information storage unit 13 described later.

  In addition, when the flow line forming unit 23 receives a request from the person detecting unit 22 to end the flow line measurement of the customer 3 determined to be queued, the flow line forming unit 23 ends the flow line measurement of the customer 3, A notification to the effect that the flow measurement of the customer 3 has been completed is sent to the customer trend linking means 16.

  In the flow line measuring means 11 configured as described above, when the person detection unit 22 reads an image captured by the image input unit 21 and detects the customer 3 in the read image, the flow line measurement is performed on the detected customer 3. To start. When the customer 3 who is the subject of the flow line measurement is in a waiting state for payment, the flow line measurement is finished. When the flow line measurement of the customer 3 is finished, the flow line measurement means 11 notifies the customer trend connection means 16 of the end of the flow line measurement of the customer 3.

  Here, with reference to FIG. 4, the timing which complete | finishes the flow line measurement of the customer 3 which the flow line measurement means 11 performs is demonstrated.

  FIG. 4 is a diagram schematically showing the settlement of customers 3a to 3i in the checkout lanes 3a to 3d shown in FIG. 1, and is an explanatory diagram for explaining the timing of the end of the flow line measurement.

  The customers 3a to 3d shown in FIG. 4 are currently paying out in the checkout lanes 5a to 5d, respectively. The customers 3e to 3h are customers that are arranged behind the customers 3a to 3d and are waiting for settlement in the checkout lanes 5a to 5d, respectively. In other words, the customers 3e to 3h form a waiting queue. Further, the customer 3i is about to line up in the checkout lane 5c.

  In the case shown in FIG. 4, for example, the timing of ending the flow measurement of the customer 3i will be described. At the point when the customer 3i is arranged at the tail of the queue waiting for payment in the checkout lane 5c, that is, behind the customer 3g. The flow line measuring means 11 considers that the customer 3i is in a payment waiting state and ends the flow line measurement.

  The flow line index issuing means 12 issues a unique flow line index to each flow line information measured by the flow line measuring means 11, flow line index information indicating the issued flow line index, and flow line information described later. The information is stored in the storage area (storage address) for storing the flow line information of the customer 3 received from the storage means 13.

  Next, referring to FIG. 5, the unique flow line index information issued for the flow line of the customer 3 and the storage address information of the flow line information storage means 13 for storing the flow line information of the customer 3 The association with will be described.

  As shown in FIG. 5, the flow line index issuing means 12 has a storage area 30 for associating the issued flow line index information with the information of the storage address storing the received flow line information. The flow line index issuing means 12 manages the flow line information of the customer 3 using this storage area 30.

  As shown in FIG. 5, the flow line index issuing means 12 includes at least one or more flow line index storage areas 31a to 35a for storing the issued flow line index, and a flow line information storage means 13 for storing the flow line information. It has at least one or more flow line information storage address storage areas 31b to 35b for storing information of storage addresses.

  The flow line information storage address storage areas 31b to 35b can be stored in association with the flow line index storage areas 31a to 35a, respectively. That is, the flow line index storage areas 31a to 35a and the flow line information storage address storage areas 31b to 35b have a one-to-one correspondence, and are therefore stored in the flow line index (flow line index storage areas 31a to 35a). Information) can be stored as a search key so that the storage address where one flow line information is stored (information stored in the flow line information storage address storage areas 31b to 35b) can be specified.

  For example, the flow line information storage address storage area 31b shown in FIG. 5 can be stored in association with the flow line index storage area 31a, and the flow line information storage address storage area 32b is associated with the flow line index storage area 32a. Can be remembered. The same applies to the other line information storage address storage areas 33b to 35b.

  The flow line index issuing means 12 having such a storage area 30 first issues a flow line index at the timing of moving to the in-store area for the customer 3 entering the store 1 through the entrances 2a and 2b. From the line information storage means 13, the information of the storage area (storage address) for storing the flow line information of the customer 3 moved to the in-store area is received. Here, the in-store area is an area located on the back (y-axis positive direction) side from the checkout lanes 5a to 5d shown in FIG.

  Subsequently, the flow line index issuing means 12 stores the issued flow line index in the flow line index storage areas 31a to 35a, and the storage address information in which the flow line information is stored in the flow line information storage address storage areas 31b to 35b. Remember. For example, the flow line index information of the customer 3a is stored in the flow line index storage area 31a, and the storage address information in which the flow line information of the customer 3a is stored is stored in the flow line information storage address storage area 31b.

  The flow line index issuing means 12 stores the flow line index and the flow line information corresponding to the flow line index in the storage area 30 so that the flow line information of the customer 3 can be managed by the flow line index. Become.

  The flow line index may be issued at the timing when the customer 3 scans the ID card at the entrances 2a and 2b of the store 1. In this case, a trigger generation unit (not shown) that generates a trigger at the timing when the ID card is scanned and outputs the trigger to the image input unit 21 is provided. Then, the flow line measurement unit 11 may be configured such that the image input unit 21 that has received the trigger output from the trigger generation unit starts capturing an image. The trigger generation unit may be provided either inside or outside the ID card scanner (card reader) device.

  FIG. 6 is a configuration diagram schematically showing the configuration of the flow line information storage means 13.

  As shown in FIG. 6, the flow line information storage means 13 has at least one storage area (hereinafter referred to as a flow line information storage area) that stores flow line information for each customer 3. The flow line information storage means 13 shown in FIG. 6 has flow line information storage areas 41 to 47.

  The flow line information storage areas 41 to 47 can store the flow line information of different customers 3, respectively. For example, in the case of the flow line information storage means 13 shown in FIG. 6, the flow line information of the customer 3a is in the flow line information storage area 41, the flow line information of the customer 3b is in the flow line information storage area 42, and the flow line of the customer 3c. Information is stored in the flow line information storage area 43, and flow line information of the customer 3 d is stored in the flow line information storage area 44.

  When the flow line information storage unit 13 receives the flow line information formed and output by the flow line forming unit 23 of the flow line measurement unit 11, the flow line information storage unit 13 starts storing the received flow line information for each customer 3. Then, the flow line information that changes due to the movement of the customer 3 is additionally stored in the storage areas 41 to 47 of the corresponding customer 3 as needed. For example, in the flow line information storage means 13 shown in FIG. 6, the flow line information of the customer 3a that changes due to the movement of the customer 3a is additionally stored in the flow line information storage area 41 as needed.

  Further, when the flow line information storage unit 13 starts to store the flow line information, the flow line information storage unit 13 hands over the information of the storage address assigned to store the flow line information to the flow line index issuing unit 12.

  The point-of-sale information management means 14 collects and manages information related to the number of sales and the sales time for each product from purchase information acquired from the POS terminal. As an example of the point-of-sale information management means 14, there is a so-called POS system control unit that collects information on the number of sales and sales time for each product from purchase information acquired from a POS terminal and manages the information by a computer.

  When the point-of-sale information management unit 14 finishes the settlement of the customer 3, the purchase information of the customer 3 is handed over to the purchase information storage unit 15, and the completion of the settlement is notified to the customer trend connection unit 16 described later.

  FIG. 7 is a configuration diagram schematically showing the configuration of the purchase information storage means 15.

  As shown in FIG. 7, the purchase information storage means 15 has the same number of storage areas as the checkout lane (hereinafter referred to as purchase information storage area). For example, the purchase information storage means 15 shown in FIG. 7 has the same number of four purchase information storage areas 51 to 54 as the checkout lanes 5a to 5d shown in FIG.

  The purchase information storage areas 51 to 54 of the purchase information storage means 15 store purchase information for each customer in different checkout lanes 5a to 5d, that is, information on products purchased for each checkout lane 5a to 5d. can do.

  For example, in the purchase information storage means 15 shown in FIG. 7, the purchase information of the customer 3a whose payment is completed in the checkout lane 5a is stored in the purchase information storage area 51, and the purchase information of the customer 3b whose payment is completed in the checkout lane 5b is stored. In the purchase information storage area 52, the purchase information of the customer 3c that has been settled in the checkout lane 5c is stored in the purchase information storage area 53, and the purchase information of the customer 3d in which the checkout lane 5d is completed is stored in the purchase information storage area 54. Remembered.

  As shown in FIG. 2, the customer trend linking means 16 has the same customer 3 as the queue storage unit 18 for storing a queue of customers 3 waiting for settlement in the checkout lanes 5a to 5d in units of the checkout lanes 5a to 5d. And a customer trend linking unit 19 that associates the flow line information with the purchase information.

  With reference to FIGS. 8 to 11, the queue storage unit 18 in the customer trend linking means 16 will be described.

  FIG. 8 is a configuration diagram schematically showing the configuration of the queue storage unit 18 in the customer trend linking means 16.

  The queue storage unit 18 is at least one or more, and a checkout lane unit queue storage that stores the flow index information of the customer 3 that forms a queue for each checkout lane that is the same number as the number of checkout lanes. For example, the queue storage unit 18 illustrated in FIG. 8 includes four checkout lane unit queue storage units 61 to 64.

  The checkout lane unit queue storage unit 61 shown in FIG. 8 stores the flow line index information of the customers 3a and 3e that form a queue in the checkout lane 5a shown in FIG. Similarly, the checkout lane unit queue storage unit 62 waits for the flow line index information of the customers 3b and 3f forming a queue in the checkout lane 5b, and the checkout lane unit queue storage unit 63 waits in the checkout lane 5c. The checkout lane unit queue storage unit 64 stores the flow line index information of the customers 3d, 3h forming the queue in the checkout lane 5d.

  The flow line index information of the customer 3 stored in the queue storage unit 18 can be used as queues of the checkout lanes 5a to 5d, that is, queue information indicating the customers 3 and their arrangement order. . Therefore, by grasping the queue information, the queue states of the checkout lanes 5a to 5d can be grasped.

  The customer trend linking unit 19 manages the stored contents of the checkout lane unit queue storage units 61 to 64 of the queue storage unit 18 and grasps the queue states of the checkout lanes 5a to 5d. Further, when receiving the purchase information of the customer 3, the customer trend linking unit 19 delivers the flow line information for the same customer 3 together with the purchase information received to the connection information storage means 17 described later.

  With reference to FIG. 9 to FIG. 11, a method for managing the storage contents of the checkout lane unit queue storage units 61 to 64 of the queue storage unit 18 will be described.

  9, 10, and 11 show changes in the storage contents of the checkout lane unit queue storage unit 63 that is an example of the checkout lane unit queue storage units 61 to 64 of the queue storage unit 18 over time. It is explanatory drawing demonstrated. Each figure is a block diagram schematically showing the configuration of the checkout lane unit queue storage unit 63, and the order of passage of time is the order of FIG. 9, FIG. 10, and FIG.

  The checkout lane unit queue storage unit 63 includes, for example, flow line index information storage areas 71 to 77 for storing at least one flow line index as shown in FIG.

  In the checkout lane unit queue storage unit 63, the flow line index of the customer 3 is sequentially stored from the upper storage area shown in FIGS. That is, the flow line index information storage area 71, the flow line index information storage area 72, the flow line index information storage area 73, ..., the flow line index information storage area 77 are sequentially stored in the order of the queue.

  Next, a change in the contents stored in the checkout lane unit queue storage unit 63 with the passage of time will be described.

  In the checkout lane unit queue storage unit 63 shown in FIG. 9, the flow line indexes of the customer 3c and the second customer 3g that are arranged at the head of the queue are stored in the flow line index information storage areas 71 and 72, respectively. Yes.

  In the state shown in FIG. 9, when the customer 3i is arranged after the customer 3g, the customer trend linking unit 19 displays the flow line index of the customer 3i arranged third from the head in the checkout lane unit queue storage unit 63. This is stored in the flow line index information storage area 73.

  That is, as shown in FIG. 10, the flow line index information storage area 71 of the checkout lane unit queue storage unit 63 contains the flow line index of the customer 3c, and the flow line index information storage area 72 contains the flow line of the customer 3g. In the flow line index information storage area 73, the flow line index of the customer 3i is stored.

  Thereafter, when the settlement of the first customer 3c is completed in the state shown in FIG. 10, the customer trend linking unit 19 delivers the flow line index information of the customer 3c together with the purchase information of the customer 3c, and also checks the queue of checkout lanes. The stored contents of the storage unit 63 are updated. More specifically, the flow line index information stored in the flow line index information storage areas 71 to 77 is shifted upward one by one.

  That is, as shown in FIG. 11, the flow line index information of the customer 3g stored in the flow line index information storage area 72 is used as the flow line index of the customer 3g at the head of the queue. 71 is stored. Further, the flow line index information of the customer 3i stored in the flow line index information storage area 73 is stored in the flow line index information storage area 72 as the flow line index of the customer 3i that is second in the queue.

  On the other hand, in the flow line index information storage area 73, the information stored in the flow line index information storage area 74 is stored. However, when the customer 3c has completed the checkout, there is no fourth queue. The flow line index of the customer 3i stored in the flow line index information storage area 73 is deleted, and nothing is stored.

  As described above, the customer trend linking unit 19 stores and deletes the flow line index of the customer 3 forming a queue in each check-out lane 5a to 5d in the flow line index information storage areas 71 to 77, so that each check-out is performed. The current queue status in lanes 5a-5d can be managed.

FIG. 12 is a configuration diagram schematically showing the configuration of the connection information storage means 17.

  As shown in FIG. 12, the connection information storage unit 17 stores at least one or more flow line index information storage areas 81 a to 85 a that store the flow line index information of the customer 3, and at least one or more that stores the purchase information of the customer 3. Purchase information storage areas 81b to 85b.

  The connection information storage means 17 is basically configured in the same manner as the storage area 30 of the flow line index issuing means 12 shown in FIG. 5, and the flow line index information and purchase information received from the customer trend connection means 16 Can be concatenated and stored. That is, the information stored in the flow line index information storage areas 81a to 85a and the information stored in the purchase information storage areas 81b to 85b can be stored in association with each other.

  As described above, the customer trend collection system 10 according to the present embodiment is a motion that shows the movement path from the customer entering the sales floor to the settlement, even if the customers 3 are dense around the checkout lanes 5a to 5d. Line information can be collected in association with purchase information at the time of checkout.

  The customer trend collection system 10 includes a flow line information storage unit 13, a purchase information storage unit 15, and a linked information storage unit 17 as means for storing information according to the use, but is necessary for executing the processing. As long as all information can be stored, the storage means 13, 15, and 17 may be combined into one or two storage means.

  Next, with reference to FIG. 13 and FIG. 14, the customer trend collection method according to the present embodiment will be described.

  The customer trend collection method according to the present embodiment generates a flow information measuring step for individually measuring a flow line of the customer 3 entering the sales floor, and generates linked information obtained by connecting the flow line information and purchase information of the same customer. The link information generation step is performed, and the link information generation step is performed in parallel with the flow line measurement step to be measured.

  FIG. 13 is a process flow diagram illustrating a process procedure of a flow line measurement step in the customer trend collection method.

  As shown in FIG. 13, the flow line measurement step includes the processing steps of Step S <b> 1 to Step S <b> 16, and when the customer trend collection system 10 has a request to start the flow line measurement, the flow line measurement step starts. (START).

  In step S1, the image input unit 21 in the flow line measuring unit 11 captures an image from each of the cameras 7a to 7i (image capture step: step S1).

  Subsequently, in step S2, it is determined whether a predetermined time (sampling period) has elapsed based on the information on the current time held by the image input unit 21 in the flow line measurement unit 11 (person detection execution timing determination step: Step S2).

  If the image input unit 21 determines that the predetermined time has elapsed in step S2 (YES in step S2), the process proceeds to step S3, and the person detection unit 22 is captured by the image input unit 21. An image is read for each of the cameras 7a to 7i, and the presence of a person (customer 3) is determined in the read image (customer existence determination step: step S3).

  When the customer 3 is detected in the image in step S3 (YES in step S3), the process proceeds to step S4, and whether or not the cameras 7a to 7i in which the person detection unit 22 images the customer 3 is the camera 7g or the camera 7i. (New customer determination step: step S4).

  If the cameras 7a to 7i that image the customer 3 detected in step S3 are the entrance / exit area imaging camera 7g or the camera 7i (YES in step S4), the process proceeds to step S5, and the customer detected by the person detection unit 22 3 features are stored (customer feature storage step: step S5).

  Subsequently, in step S6, the flow line forming unit 23 forms flow line information indicating the movement path of the customer 3 detected in step 3 (flow line information forming step: step S6).

  In step S7, the flow line index issuing means 12 issues a flow line index to the flow line information of the customer 3 detected in step S3 (flow line index issuing step: step S7). In subsequent step S8, the flow line information storage means 13 stores the flow line information formed in step S6 (flow line information storage step: step S8).

  In step S9 (end request presence / absence confirmation step), when there is a request to end the flow line measurement for all customers 3 (YES in step S9), the flow line measurement step ends (END). .

  On the other hand, when the image input unit 21 determines in step S2 that the predetermined time (sampling period) has not elapsed (NO in step S2), the process returns to step S2.

  If no person (customer 3) is detected in the image in step S3 (NO in step S3), the process returns to step S2.

  Furthermore, in Step S4, when the cameras 7a to 7i capturing the customer 3 detected in Step S3 are other than the cameras 7g and 7i, that is, the cameras 7a to 7f and 7h (in the case of NO in Step S4), Proceeding to step S10, the feature of the customer 3 detected by the person detection unit 22 is matched with the feature of the customer 3 stored in step S5 (customer feature matching step: step S10).

  Subsequently, in step S11, the flow line forming unit 23 forms flow line information indicating the movement path of the customer 3 detected in step 3 (flow line information forming step: step S11). The flow line information forming step in step S11 is substantially the same processing step as the flow line information forming step in step S6.

  In step S12, the flow line information storage means 13 adds the flow line information formed in step S11 as the flow line information of the same person as the customer 3 collated in step S10 (flow line information storage step: step S12). Note that the flow line information storage step in step S12 differs from the flow line information storage step in step S8 in that there is already flow line information, but it is substantially the same processing step. .

  In step S13, the person detection unit 22 determines whether the cameras 7a to 7i currently detected for the customer 3 are the cameras 7h. When the person detection unit 22 determines that the cameras 7a to 7i detected for the customer 3 are cameras 7h (YES in step S13), the process proceeds to step S14, and the customer currently detected by the camera 7h. It is determined whether or not 3 is queued (payment waiting customer determination step: step S13, step S14).

  In step S14, when the person detection unit 22 determines that the customer 3 currently detected is queued (YES in step S14), the process proceeds to step S15, where the flow line formation unit 23 is queued. The formation of the flow line information of the customers 3 lined up in (1) is finished and the fact that it has been finished is notified to the customer trend connection means 16 (customer flow line information formation completion notification step: step S15).

  Subsequently, in step S16, among the features of the customer 3 stored in the person detection unit 22, the features of the customer 3 whose flow line measurement has been completed in step S15 are deleted (customer feature deletion step: step S16). When the processing step of step S16 is completed, the process proceeds to step S9, and the processing steps after step S9 are performed.

  On the other hand, when the person detection unit 22 determines in step S13 that the cameras 7a to 7i that are currently detecting the customer 3 are other than the camera 7h, that is, the cameras 7a to 7g and 7i (NO in step S13). ), The process proceeds to step S9.

  In step S14, if the person detection unit 22 determines that the customer 3 currently detected is not queued (NO in step S14), the process proceeds to step S9.

  FIG. 14 is a process flow diagram showing the processing procedure of the link information generation step in the customer trend collection method.

  As shown in FIG. 14, the linked information generation step includes processing steps from step S21 to step S27. Further, since the connection information generation step is performed in parallel with the above-described flow line measurement step, the connection information generation step processing is also started at the timing when the flow line measurement step processing is started (START).

  In step S21, the customer trend linking means 16 confirms whether or not the flow line measurement means 11 has notified the end of the flow line measurement, that is, the end of the formation of the flow line information (flow line measurement end confirmation step: step S21).

  If it is confirmed in step S21 that the customer trend linking unit 19 of the customer trend linking unit 16 has notified that the flow line information has been formed (YES in step S21), the process proceeds to step S22. The trend linking unit 19 reads out the flow line index information of the customer 3 who has finished the flow line measurement from the flow line information storage unit 13 and registers it in the flow line index information storage areas 71 to 77 of the queue storage unit 18 (queue) Information update step: Step S22).

  Subsequently, in step S23 (end request presence / absence confirmation step), when there is a request to end the flow line measurement for all the customers 3 (in the case of YES in step S23), the link information generation step ends (END). ).

  On the other hand, when the customer trend linking unit 19 confirms that there is no notification of the end of the flow line measurement in step S21 (NO in step S21), the process proceeds to step S24, and the point-of-sales information management unit 14 determines that one customer It is confirmed whether or not payment completion is notified (checkout completion confirmation step: step S24).

  Subsequently, in step S25, the customer trend linking unit 19 of the customer trend linking unit 16 reads and reads the flow line index information of the customer 3 that has been settled from the flow line index information storage area 71 of the queue storage unit 18. Information is delivered to the linked information storage means 17. Then, the connection information storage unit 17 writes the received flow line index information in the flow line index information storage areas 81a to 85a. Note that “R / W” shown in step S25 and step S26 means read / write (reading and writing).

  In step S26, the customer trend linking unit 19 of the customer trend linking unit 16 reads out the purchase information of the customer 3 who has completed the settlement stored in the purchase information storage areas 51 to 54 of the purchase information storage unit 15, and reads the information. Is transferred to the connection information storage means 17. Then, the linked information storage means 17 writes the received purchase information in the purchase information storage areas 81b to 85b corresponding to the flow line index information storage areas 81a to 85a written in step S25 (linked information storage step: step). S25, step S26).

  The purchase information storage areas 81b to 85b corresponding to the flow line index information storage areas 81a to 85a are purchase information storage areas to which the same numerals as the signs of the flow line index information storage areas are attached. For example, the purchase information storage area 81b corresponds to the flow line index information storage area 81a, and the purchase information storage area 85b corresponds to the flow line index information storage area 85a.

  Subsequently, in step S27, the customer trend linking unit 19 deletes the flow line index information read in step S25 from the flow line index information storage area 71, and sequentially shifts the registered flow line index information one by one. (Queue information update step: Step S27).

  More specifically, step S27 will be described. When the customer trend linking unit 19 executes the processing step of step S27 under the situation shown in FIG. 10, the stored contents of the checkout lane unit queue storage unit 63 are as shown in FIG. Updated to the state shown. When the processing step of step S27 is completed, the process proceeds to step S23, and the processing steps after step S23 are performed.

  In the connection information generation step shown in FIG. 14, the R / W of the flow line index information (Step S25) is performed and then the R / W of the purchase information (Step S26) is performed. The order of R / W and purchase information R / W may be reversed.

  As described above, according to the customer trend collecting method according to the present embodiment, even if the customers 3 are crowded around the check-out lanes 5a to 5d, the movement path from the entry to the sales floor for the customer 3 to the settlement is shown. The flow line information can be collected in association with purchase information at the time of checkout.

  As described above, according to the customer trend collection system, the customer trend collection method, and the customer trend collection program according to the present embodiment, even if the customers 3 are crowded around the checkout lanes 5a to 5d, the customer 3 enters the sales floor. The flow line information indicating the travel route from the checkout to the checkout can be collected in association with the purchase information at the time of checkout.

  Further, if the collected information is used afterwards, it is possible to verify whether or not the product is actually purchased at the place where the customer 3 visited. In this way, it is possible to verify whether or not a product is actually purchased at a place where the customer 3 has visited, so that it is easier for store managers to analyze the effects of in-store advertising and sales floor layout than before. Can be done.

  In this embodiment, the case has been described in which the function for carrying out the invention is recorded (installed) in advance in the apparatus. However, the present invention is not limited to this, and the same function may be downloaded from the network to the apparatus. Those having these functions stored in a recording medium may be installed in the apparatus. As long as it is a recording medium, the form may be any form. Further, the function may be realized in cooperation with an OS (Operating System) or the like.

The layout figure which showed schematically the layout in the store which applies the customer trend collection system, customer trend collection method, and customer trend collection program which concern on this invention. 1 is a block diagram schematically showing the system configuration of a customer trend collection system according to the present invention. The block diagram which showed roughly the structure of the flow line measurement means in the customer trend collection system which concerns on this invention. Explanatory drawing explaining the timing which complete | finishes the customer flow line measurement in the store which applies the customer trend collection system customer trend collection method and customer trend collection program which concern on this invention. The block diagram which showed roughly the memory area which the flow line index issuing means in the customer trend collection system which concerns on this invention has. The block diagram which showed schematically the structure of the flow line information storage means in the customer trend collection system which concerns on this invention. The block diagram which showed schematically the structure of the purchasing information storage means in the customer trend collection system which concerns on this invention. The block diagram which showed schematically the structure of the queue memory | storage part of the customer trend connection means in the customer trend collection system which concerns on this invention. It is the block diagram which showed schematically the structure of the checkout lane unit queue memory | storage part of the queue memory | storage part of the customer trend connection means in the customer trend collection system which concerns on this invention, and the memory | storage of a checkout lane unit queue memory | storage part Explanatory drawing which shows the content. FIG. 10 is a configuration diagram schematically showing a configuration of a check-out lane unit queue storage unit of a queue storage unit of a customer trend connection unit in the customer trend collection system according to the present invention, and one more customer from the state shown in FIG. Explanatory drawing which shows the memory content of the checkout lane unit queue memory | storage part after arranging. It is the block diagram which showed roughly the structure of the checkout lane unit queue memory | storage part of the queue memory | storage part of the customer trend connection means in the customer trend collection system which concerns on this invention, and shows the top customer from the state shown in FIG. Explanatory drawing which shows the memory content of the checkout lane unit queue memory | storage part after completion of payment. The block diagram which showed schematically the structure of the connection information storage means in the customer trend collection system which concerns on this invention. The processing flowchart which showed the process sequence of the flow line measurement step in the customer trend collection method which concerns on this invention. The processing flowchart which showed the process sequence of the connection information production | generation step in the customer trend collection method which concerns on this invention.

Explanation of symbols

1 store 2a, 2b doorway 3, 3a-3i customer 4a-4i product display shelf 5a-5d checkout lane 6 soccer table 7a-7i camera (imaging means)
DESCRIPTION OF SYMBOLS 10 Customer trend collection system 11 Flow line measurement means 12 Flow line index issue means 13 Flow line information storage means 14 Point-of-sale information management means 15 Purchase information storage means 16 Customer trend connection means 17 Connection information storage means 18 Queue storage part 19 Customer Trend linking unit 21 Image input unit 22 Human detection unit 23 Flow line forming unit 30 Storage areas 31a to 35a (flow line index issuing means) Flow line index storage areas 31b to 35b Flow line information storage address storage areas 41 to 47 Information storage areas 51-54 Purchasing information storage areas 61-64 Checkout lane unit queue storage units 71-77 Flow line index information storage areas 81a-85a Flow line index information storage areas 81b-85b Flow line information storage address storage areas

Claims (8)

Images taken by the image pickup means installed at the sales floor are sequentially captured at regular time intervals, and the captured images are processed to detect and store customers captured in the images, and stored sequentially for each captured image. by identifying the customer is present position, the customer, and outputs the formed flow line information indicative of the route that has moved the sales during the period from enters the shop until arranged in one of the check-out lane Flow line measuring means;
Flow line information that receives the flow line information output from the flow line measuring means, allocates and stores an area for storing the received flow line information, and outputs a storage address assigned to store the flow line information Storage means;
The information on the storage address output from the flow line information storage means is received, a unique flow line index is issued for each flow line information stored in the flow line information storage means, and the issued flow line A flow line index issuing means for associating the index and the storage address of the flow line information specified from the flow line index in a predetermined storage area ;
Each time the checkout for each of the customers is completed, the purchase information indicating the purchase product of the customer who has finished the checkout, which is acquired via the POS terminal at the time of checkout and stored in the purchase information storage means, is read from the purchase information storage means. on the other hand, it receives the flow line index information from the flow line index issuing means showing the flow line index those the customer, customer behavior linked to be stored in the connection information storage means in association with the flow line index information received and purchase information read Means ,
The customer trend linking means includes a checkout lane unit queue storage unit that stores flow order index information for each customer that forms a queue in the checkout lane unit in a waiting order, and from the flow line measurement unit to the customer When the notification of the end of the flow line measurement is received, the queue storage unit that writes the flow line index information of the customer to the checkout lane unit queue storage unit of the checkout lane in which the customer is lined up,
A customer trend linking unit that associates the flow line index information of the customer at the head of the queue with the purchase information of the customer at the end of the checkout and updates the flow line index information stored in the queue storage unit; Customer trend collection system characterized by that. 2. The customer trend collection system according to claim 1, wherein the flow line information includes position information indicating a position of the moved customer and time information indicating a time at a position based on the position information. . The flow line information, and sales area information indicating the flies rear sale to the customer has passed out of the plurality of sales areas set by dividing the previously said sales floor, the customer sales area based on the sales floor area information The customer trend collection system according to claim 1 , further comprising: an approach time that has entered, an exit time that has exited, and time information that indicates one of an intermediate time between the entry time and the exit time . The flow line measuring means includes a person detection section that detects and stores the existence of a customer to be imaged in the image by image processing the image acquired from the imaging unit,
Customer according to claim 1, characterized in that it comprises a flow line forming unit for outputting the person detection section forming said flow line information to customers stored in 顧 each customer to the flow line measuring means Trend collection system. Computer
Images taken by the image pickup means installed at the sales floor are sequentially captured at regular time intervals, and the captured images are processed to detect and store customers captured in the images, and stored sequentially for each captured image. By identifying the location of the customer, the flow line that forms and outputs the flow line information indicating the route that the customer moved to the checkout lane after entering the sales floor Measuring means;
Flow line information that receives the flow line information output from the flow line measuring means, allocates and stores an area for storing the received flow line information, and outputs a storage address assigned to store the flow line information Storage means;
The information on the storage address output from the flow line information storage means is received, a unique flow line index is issued for each flow line information stored in the flow line information storage means, and the issued flow line A flow line index issuing means for associating an index and a storage address of flow line information specified from the flow line index in a predetermined storage area;
Each time the checkout for each of the customers is completed, the purchase information indicating the purchased product of the customer who has finished the checkout and is stored in the purchase information storage means acquired through the POS terminal at the time of checkout is read from the purchase information storage means. On the other hand, customer trend connection means for receiving flow line index information indicating the flow line index of the customer from the flow line index issuing means, and storing the read purchase information and the received flow line index information in association information storage means And
The customer trend linking means includes a checkout lane unit queue storage unit that stores flow order index information for each customer that forms a queue in the checkout lane unit in a waiting order, and from the flow line measurement unit to the customer When the notification of the end of the flow line measurement is received, the queue storage unit for writing the flow line index information of the customer to the checkout lane unit queue storage unit of the checkout lane in which the customer is lined up, and the queue at the end of the checkout Customer trends that function as a system comprising a customer trend link unit that associates the customer flow index information with the customer purchase information and updates the flow index information stored in the queue storage unit. Collection program. 6. The customer trend collection program according to claim 5 , wherein the flow line information includes position information indicating a position of the moved customer and time information indicating a time at a position based on the position information . The flow line information includes sales floor area information indicating a sales floor area through which the customer has passed among a plurality of sales floor areas set in advance by dividing the sales floor, and the customer has entered the sales floor area based on the sales floor area information. 6. The customer trend collection program according to claim 5, further comprising time information indicating an entry time, an exit time at which the user has exited, and an intermediate time between the entry time and the exit time . The flow line measurement means detects and stores the presence of a customer imaged in the image by performing image processing on the image captured from the imaging means,
The customer trend collection according to claim 5, further comprising: a flow line forming unit that forms the flow line information for each customer stored in the person detection unit and outputs the flow line information to the flow line measuring unit. program.

Images