Count system, count device, count method, count program, count display device, and count display program

Abstract

To provide a technique which can suitably count the number of particular persons who enter and exit a predetermined space.SOLUTION: A count system 1 includes: a person flow count sensor 20 which detects persons who enter and exit a predetermined space to calculate a person flow count number for determining the number of the persons who enter and exit the predetermined space; at least one RFID reader 10 which communicates with an RFID tag 4 situated at an entrance and exit gate of the predetermined space, the RFID tag being kept by each of first-type persons, to read ID information included in the RFID tag; and a count device 30 which, on the basis of the person flow count number and a result from the reading of the at least one RFID reader, calculates the number of second-type persons who enter and exit the predetermined space. The count device, when specified times of times at which the persons are detected by the person flow count sensor coincide with times at which the ID information is read by the at least one RFID reader, executes subtraction processing to subtract the number of the reading of the ID information at the specified times from the person flow count number to thereby calculate the number of second-type persons who enter and exit the predetermined space.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a counting system, a counting device, a counting method, a counting program, a counting display device, and a count display program, and in particular to a counting system, a counting device, a counting method, a counting program, a counting display device, and a count display program that count the number of specific people entering and leaving a specified space. [Background technology]

[0002] Conventionally, a system for counting the number of people entering and exiting a specified space is known. For example, JP 2022-72830 A (Patent Document 1) discloses a system for counting the number of people entering and exiting a specified space by reading ID information contained in an RFID (Radio Frequency Identification) tag by an RFID reader when a person carrying an RFID tag passes through an entrance or exit of the specified space. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2022-72830 Summary of the Invention [Problem to be solved by the invention]

[0004] According to the system disclosed in Patent Document 1, it is possible to count the number of people entering and exiting a given space using an RFID reader installed at the entrance and exit of the given space, but the people to be counted must carry an RFID tag. For example, in a store, if the number of customers entering and exiting the store is to be counted, all customers visiting the store must carry an RFID tag, which is practically impossible.

[0005] It is also possible to count only the number of customers by photographing people entering and leaving the store with a camera, analyzing the obtained images, and distinguishing between customers and store clerks. However, such a method requires time for image processing and consumes a lot of power for the computing device that executes the image processing. Furthermore, it is necessary to perform face authentication in advance for all store clerks, and face authentication cannot be performed for store clerks who suddenly join the work, which causes practical inconveniences.

[0006] The present disclosure has been made to solve such problems, and its purpose is to provide a technology that can appropriately count the number of specific people entering and leaving a specified space. [Means for solving the problem]

[0007] A counting system according to an aspect of the present disclosure includes a people flow count sensor that detects people, including a first person and a second person, entering and exiting a predetermined space and calculates a people flow count number to identify the number of people entering and exiting the predetermined space, at least one RFID reader that communicates with an RFID tag held by a first person located at an entrance or exit of the predetermined space and reads ID information contained in the RFID tag, and a counting device that calculates the number of second people entering and exiting the predetermined space based on the people flow count number and the reading result of the at least one RFID reader. If there is a specific time that coincides with the time when the ID information was read by the at least one RFID reader among the times when people were detected by the people flow count sensor, the counting device calculates the number of second people entering and exiting the predetermined space by performing a subtraction process of subtracting the number of ID information read at the specific time from the people flow count number.

[0008] A counting device according to another aspect of the present disclosure includes a calculation device and a storage device that stores a program executed by the calculation device. The calculation device acquires a people flow count number for identifying the number of people, including a first person and a second person, entering and exiting a predetermined space from a people flow count sensor, acquires from at least one RFID reader a reading result of ID information included in an RFID tag held by a first person positioned at an entrance or exit of the predetermined space, and, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which the ID information is read by at least one RFID reader, performs a subtraction process to subtract the number of ID information read at the specific time from the people flow count number to calculate the number of second people entering and exiting the predetermined space.

[0009] A counting method according to another aspect of the present disclosure includes, as processing executed by a computer, the steps of obtaining a people flow count number from a people flow count sensor to identify the number of people, including a first person and a second person, entering and exiting a specified space; obtaining from at least one RFID reader the reading results of ID information contained in an RFID tag held by a first person located at an entrance / exit of the specified space; and, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which the ID information was read by at least one RFID reader, calculating the number of second people entering and exiting the specified space by performing a subtraction process to subtract the number of ID information read at the specific time from the people flow count number.

[0010] A counting program according to another aspect of the present disclosure causes a computer to execute the steps of acquiring a people flow count number from a people flow count sensor to identify the number of people, including a first person and a second person, entering and exiting a specified space; acquiring from at least one RFID reader the reading results of ID information contained in an RFID tag held by a first person located at an entrance / exit of the specified space; and, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which the ID information was read by at least one RFID reader, calculating the number of second people entering and exiting the specified space by performing a subtraction process to subtract the number of ID information read at the specific time from the people flow count number. Effect of the Invention

[0011] According to the present disclosure, a people flow count number for identifying the number of people entering and exiting a specified space can be calculated using a people flow count sensor, and ID information of a first person entering and exiting the specified space can be read using an RFID reader, and further, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which the ID information is read by the RFID reader, the number of second people entering and exiting the specified space can be calculated by subtracting the number of ID information read at the specific time from the people flow count number. This allows the number of second people entering and exiting the specified space to be appropriately counted. [Brief description of the drawings]

[0012] [Figure 1] FIG. 2 is a diagram for explaining an application example of the counting system according to the first embodiment. [Diagram 2] FIG. 1 is a diagram showing a configuration of a counting system according to a first embodiment. [Diagram 3] 4 is a flowchart relating to a process executed by the RFID reader according to the first embodiment. [Figure 4] 4 is a flowchart relating to a process executed by the people flow count sensor according to the first embodiment. [Diagram 5]4 is a flowchart relating to a process executed by the counting device according to the first embodiment. [Figure 6] FIG. 11 is a diagram for explaining a counting system according to a second embodiment. [Figure 7] 11 is a diagram for explaining a process of estimating a person entering a store, performed by the counting device according to the second embodiment. FIG. [Figure 8] FIG. 11 is a diagram for explaining a counting system according to a third embodiment. [Figure 9] 13 is a diagram for explaining a process of estimating a person entering a store by the counting device according to the third embodiment. FIG. [Figure 10] FIG. 13 is a diagram for explaining a counting system according to a fourth embodiment. [Figure 11] FIG. 13 is a diagram for explaining a counting system according to embodiment 5. [Figure 12] 13 is a diagram for explaining an example of a process for estimating the behavior of a store clerk by the counting device according to the sixth embodiment. FIG. [Figure 13] 13 is a diagram for explaining an example of a process for estimating the behavior of a store clerk by the counting device according to the sixth embodiment. FIG. [Figure 14] FIG. 23 is a diagram for explaining an example of a process for determining an estimation result by the counting device according to the sixth embodiment. [Figure 15] FIG. 23 is a diagram for explaining an example of a process for estimating the behavior of a store clerk by the counting device according to the seventh embodiment. [Figure 16] FIG. 23 is a diagram for explaining an example of data processing of RFID information in the estimation process of the counting device according to the seventh embodiment. [Figure 17] FIG. 23 is a diagram for explaining an example of a process for inferring a store clerk's behavior by the counting device according to the eighth embodiment. [Figure 18] FIG. 23 is a diagram for explaining another example of the process of estimating the behavior of a store clerk by the counting device according to the eighth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference characters and their description will not be repeated.

[0014] <Embodiment 1> The counting system 1 according to the first embodiment will be described with reference to FIGS. 1 to 5. In the following, the "first person" will be exemplified as a "clerk" working in a store, and the "second person" will be exemplified as a "customer" visiting the store. The inside of the store will be referred to as the "inside of the store", and the outside of the store will be referred to as the "outside of the store". The "store" is an example of a "predetermined space". The "predetermined space" is not limited to the "store", but may be, for example, inside a building such as a library, or may be in other spaces. The "first person" is not limited to a clerk, but may be, for example, a staff member working in a building such as a library. Furthermore, the "second person" is not limited to a customer, but may be, for example, a person other than the staff member visiting a building such as a library.

[0015] [Example of application] FIG. 1 is a diagram for explaining an application example of the counting system 1 according to the first embodiment. As shown in FIG. 1, the counting system 1 is configured to count the number of people entering and leaving a store. Specifically, the counting system 1 is capable of counting the number of customers entering the store (number of store visitors). The counting system 1 may be configured to count not only the number of customers entering the store, but also the number of customers leaving the store. "Entering the store" and "leaving the store" are also collectively referred to as "entering and leaving the store." By using the counting system 1, a user can know the number of customers visiting the store during a given period (for example, one day from opening to closing).

[0016] As shown in FIG. 1, the counting system 1 according to the first embodiment includes at least one RFID reader 10, at least one people flow counting sensor 20, and a counting device 30 capable of communicating with each RFID reader 10 and each people flow counting sensor 20.

[0017] The RFID reader 10 is provided at least on one of the inside and outside of a store entrance. For example, a store is provided with at least one entrance 2 consisting of an automatic or manual door. In the example of FIG. 1, two entrances 2 are provided in the store, and the RFID reader 10 is provided near each entrance 2 outside the store. The RFID reader 10 communicates with an RFID tag 4 held by each clerk working in the store. The RFID tag 4 includes ID information for identifying each clerk working in the store. Each clerk is assigned one piece of ID information. The RFID reader 10 can read the ID information included in the RFID tag 4 by communicating with the RFID tag 4.

[0018] For example, when a store clerk passes through entrance / exit 2 and the RFID tag 4 held by the store clerk comes within the communication range of RFID reader 10, RFID reader 10 communicates with RFID tag 4 and reads the ID information stored in RFID tag 4. RFID reader 10 stores the read ID information in association with the time when the ID information was read (hereinafter also referred to as "read time"). RFID reader 10 transmits information including the stored ID information and the read time (hereinafter also referred to as "RFID information") to counting device 30 at a predetermined interval (for example, one second interval).

[0019] In this way, the counting system 1 is configured to use the RFID reader 10 to communicate with the RFID tag 4 located within the communication range of the RFID reader 10 to read the ID information of the store clerk. For this reason, the RFID reader 10 is provided near the entrance 2 so that the RFID reader 10 can communicate with the RFID tag 4 when the store clerk enters and leaves the store. Since the RFID reader 10 is configured to periodically communicate with the RFID tag 4 located within the communication range, the RFID reader 10 is not limited to communicating with the RFID tag 4 and reading the ID information when the store clerk enters the communication range from outside the communication range, but also communicates with the RFID tag 4 periodically to read the ID information even when the store clerk continues to stay within the communication range. For example, when a store clerk stays within the communication range, the RFID reader 10 reads the ID information every time it communicates with the RFID tag 4, so that the RFID reader 10 may read the ID information multiple times during the store clerk's stay.

[0020] Based on the RFID information acquired from the RFID reader 10, the counting device 30 can identify the ID information read by the RFID reader 10 and the time when the ID information was read by the RFID reader 10, but cannot identify whether the store clerk whose ID information was read by the RFID reader 10 entered the store or left the store.

[0021] The people flow count sensor 20 is provided near each entrance / exit 2 inside or outside the store. In the example of Fig. 1, the people flow count sensor 20 is attached to the ceiling inside the store, and by photographing and observing a predetermined detection area 3 near the entrance / exit 2 from the ceiling, it is possible to detect people passing through the detection area 3.

[0022] For example, when a store clerk or customer passes through the entrance / exit 2 and enters the detection area 3, the people flow count sensor 20 detects the number of people who have entered the detection area 3 and the direction of movement of the people. Based on the number of people detected and the direction of movement, the people flow count sensor 20 counts the number of people who have entered the store and the number of people who have left the store, regardless of the type of customer or store clerk, separately and stores the counted number as a people flow count. The people flow count includes an entry count indicating the number of people who have entered the store and an exit count indicating the number of people who have left the store. The people flow count sensor 20 stores the people flow count in association with the time at which the person was detected (hereinafter also referred to as the "detection time"). The people flow count sensor 20 transmits information including the stored people flow count and the detection time (hereinafter also referred to as the "people flow count information") to the counting device 30 at a predetermined interval (for example, one second interval).

[0023] Based on the people flow count information obtained from people flow count sensor 20, counting device 30 can determine the number of people who entered the store (entry count), the number of people who left the store (exit count), and the detection time when the people were detected, but cannot determine whether the people detected by people flow count sensor 20 are store clerks or customers.

[0024] According to the counting system 1 configured in this manner, the counting device 30 can identify which store clerk was near the entrance / exit 2 at what time based on the RFID information. Furthermore, the counting device 30 can identify the number of people who entered the store and the number of people who left the store during a predetermined period (for example, one day from opening to closing) based on the people flow count information. However, when the counting device 30 uses the RFID information and the people flow count information individually, it is not possible to count the number of customers excluding the store clerk who enters the store or the number of customers excluding the store clerk who leaves the store. Therefore, the counting device 30 is configured to count the number of customers who enter the store or the number of customers who leave the store by combining the RFID information and the people flow count information.

[0025] Specifically, the counting device 30 compares the time when a person is detected by the people flow count sensor 20 in a predetermined period with the time when ID information is read by the RFID reader 10 in the predetermined period. If there is a specific time among the times when a person is detected by the people flow count sensor 20 in the predetermined period that matches the time when ID information was read by the RFID reader 10, the counting device 30 can estimate that the person detected by the people flow count sensor 20 at the specific time is a store clerk. Therefore, the counting device 30 can estimate that the number of ID information read by the RFID reader 10 at the specific time when the detection time and read time match is the number of store clerks entering the store in the predetermined period.

[0026] Therefore, the counting device 30 can calculate the number of customers entering the store during a specified period by performing a process (hereinafter also referred to as the "subtraction process") of subtracting the number of ID information read by the RFID reader 10 at a specific time when the detection time and reading time coincide (i.e., the number of times the ID information was read) from the number of people entering the store detected by the people flow count sensor 20 (entry count number).

[0027] For example, if the number of people entering a store detected by people flow counting sensor 20 during a specified period (entry count number) is "50" and the number of ID information read by RFID reader 10 at a particular time is "1," counting device 30 can subtract "1" from "50" to calculate "49" as the number of customers entering the store during the specified period.

[0028] When there are multiple specific times as described above within a specified period, the counting device 30 can calculate the number of customers entering the store during the specified period by performing the subtraction process described above for each of the multiple specific times.

[0029] Similarly, if there is a specific time during a specified period when people leaving the store are detected by people flow count sensor 20 and the specific time coincides with the time when ID information was read by RFID reader 10, counting device 30 can infer that the people leaving the store detected by people flow count sensor 20 at the specific time are store clerks. Therefore, counting device 30 can infer that the number of ID information read by RFID reader 10 at the specific time when the detection time and read time coincide is the number of store clerks leaving the store during the specified period.

[0030] Therefore, the counting device 30 can calculate the number of customers leaving the store in a specified period of time by performing a subtraction process to subtract the number of ID information read by the RFID reader 10 (i.e., the number of times the ID information was read) at a specific time when the detection time and reading time coincide from the number of people leaving the store detected by the people flow count sensor 20 (exit count number).

[0031] For example, if the number of people leaving the store detected by the people flow counting sensor 20 in a specified period of time (exit count) is "50" and the number of ID information read by the RFID reader 10 at a particular time is "1," the counting device 30 can subtract "1" from "50" to calculate "49" as the number of customers leaving the store in the specified period of time.

[0032] When there are multiple specific times as described above within a specified period, the counting device 30 can calculate the number of customers leaving the store within the specified period by performing the subtraction process described above for each of the multiple specific times.

[0033] In this way, the counting system 1 can use the people count sensor 20 to calculate a people count number for identifying the number of people entering and leaving the store, and can use the RFID reader 10 to read the ID information of the store clerks entering and leaving the store. Furthermore, if there is a specific time that coincides with the time that the RFID reader 10 reads ID information from the people count number during a specific period among the times at which people are detected by the people count sensor 20 during a specific period, the counting system 1 can calculate the number of customers entering and leaving the store during the specific period by subtracting the number of ID information read at the specific time from the people count number. This allows the counting system 1 to appropriately count the number of customers entering and leaving the store. Furthermore, the counting system 1 can count the number of customers entering and leaving the store by simply assigning one ID information to each store clerk, so there is no need to perform face authentication on all store clerks in advance.

[0034] In addition, since the people flow count sensor 20 is attached to the ceiling, it is possible to prevent information such as face or back shape that identifies an individual from being included in the captured image as much as possible, and the privacy of customers can be protected. In addition, when a person is photographed from the side, there is a possibility that the number of people counted may be incorrect when multiple people overlap. However, if the people flow count sensor 20 is installed on the ceiling, the number of customers entering and leaving the store can be properly counted even if multiple people are present in the detection area 3. In addition, since the people flow count sensor 20 tracks the movement of a person centered on the head and counts the number of people passing through the detection area 3, the number of people passing through the detection area 3 can be counted with high accuracy even if a person changes direction while passing through the detection area 3. Furthermore, since the people flow count sensor 20 is installed near all entrances and exits 2, the counting system 1 can count the number of customers entering and leaving the store without missing any.

[0035] [Counting system configuration] 2 is a diagram showing a configuration of the counting system 1 according to the embodiment 1. As shown in FIG. 2, the counting system 1 includes an RFID reader 10, a people flow count sensor 20, a counting device 30, and a count display device 40.

[0036] The RFID reader 10 includes a data processing IC (integrated circuit) 11, a memory 12, a communication interface 13, and an antenna 14.

[0037] The data processing IC 11 acquires the ID information contained in the RFID tag 4 read by the antenna 14, and stores the ID information together with the time at which the ID information was acquired (read time) in the memory 12. Furthermore, the data processing IC 11 transmits the RFID information including the ID information and the read time stored in the memory 12 to the counting device 30 via the communication interface 13.

[0038] The memory 12 includes a storage area for storing RFID information. Examples of the memory 12 include volatile memories such as DRAM (Dynamic Random Access Memory) and SRAM (Static Random Access Memory), and non-volatile memories such as ROM (Read Only Memory) and flash memory. Under the control of the data processing IC 11, the memory 12 stores the ID information read by the antenna 14 in association with the read time.

[0039] The communication interface 13 is an interface through which the data processing IC 11 communicates with the counting device 30. For example, the communication interface 13 may communicate data with the counting device 30 via a wired connection using an optical line or the like, or may communicate data with the counting device 30 via a wireless connection such as Wi-Fi (registered trademark). The communication interface 13 transmits the RFID information stored in the memory 12 to the counting device 30 under the control of the data processing IC 11.

[0040] The antenna 14 is provided near each entrance 2 outside the store, and communicates with the RFID tag 4 held by the store clerk to read the ID information contained in the RFID tag 4. The ID information obtained by the antenna 14 is acquired by the data processing IC 11.

[0041] The people flow counting sensor 20 includes a data processing IC 21, a memory 22, a communication interface 23, and a camera 24.

[0042] The data processing IC 21 analyzes image data of people passing through the detection area 3 acquired by the camera 24, and recognizes the number of people who have entered the detection area 3 and the direction of movement of the people. Based on the identified number of people and the direction of movement of the people, the data processing IC 21 calculates the number of people who have entered the store (entry count number) and the number of people who have left the store (exit count number) as people flow count numbers, and stores the people flow count numbers together with the detection times at which the people were detected in the memory 22. Furthermore, the data processing IC 11 transmits people flow count information including the people flow count number and detection times stored in the memory 22 to the counting device 30 via the communication interface 23.

[0043] The memory 22 includes a storage area for storing people flow count information. Examples of the memory 22 include volatile memories such as DRAM and SRAM, and non-volatile memories such as ROM and flash memory. Under the control of the data processing IC 21, the memory 22 stores the store entry count number and the store exit count number in association with the detection time.

[0044] The communication interface 23 is an interface for the data processing IC 21 to communicate with the counting device 30. For example, the communication interface 23 may communicate data with the counting device 30 via a wired connection using an optical line or the like, or may communicate data with the counting device 30 via a wireless connection such as Wi-Fi. The communication interface 23 transmits the people flow counting information stored in the memory 22 to the counting device 30 under the control of the data processing IC 21.

[0045] The camera 24 is provided near each entrance 2 inside or outside the store, and captures an image of the detection area 3 to detect a person passing through the detection area 3. Image data of the person captured by the camera 24 is acquired by the data processing IC 21.

[0046] The counting device 30 is configured, for example, as a server device managed within a store, or a cloud-type server device provided by the operating company of the counting system 1. The counting device 30 is not limited to a server device, and may be a desktop, laptop, or tablet PC (Personal Computer) configured integrally with a counting display device 40 described later, or a mobile terminal such as a smartphone. The counting device 30 includes a calculation device 31, a memory 32, a communication interface 33, and a storage device 34.

[0047] The arithmetic device 31 is a computing entity (computer) that executes a predetermined process. The arithmetic device 31 is composed of a processor such as a central processing unit (CPU), a micro-processing unit (MPU), a tensor processing unit (TPU), or a graphics processing unit (GPU). The processor, which is an example of the arithmetic device 31, has a function of executing a predetermined process by executing a predetermined program, but some or all of these functions may be implemented using a dedicated hardware circuit such as an application specific integrated circuit (ASIC) or a field-programmable gate array (FPGA). The term "processor" is not limited to a processor in the narrow sense that executes a process using a stored program method such as a CPU, an MPU, a TPU, or a GPU, but may include a hardwired circuit such as an ASIC or an FPGA. The arithmetic device 31 is not limited to a von Neumann type computer such as a CPU or a GPU, but may be composed of a non-von Neumann type computer such as a quantum computer or an optical computer. The arithmetic device 31 as described above can also be read as a processing circuitry that executes a predetermined process. The computing device 31 may be configured as one chip or multiple chips. Furthermore, the processor and related processing circuits may be configured as multiple computers interconnected by wire or wirelessly via a local area network or a wireless network. The processor and related processing circuits may be configured as a cloud computer that performs remote calculations based on input data and outputs the results of the calculations to other devices in remote locations.

[0048] The arithmetic device 31 acquires RFID information 342 from the RFID reader 10 via the communication interface 33 and stores it in the storage device 34, and also acquires people flow count information 343 from the people flow count sensor 20 and stores it in the storage device 34. The arithmetic device 31 also counts the number of customers entering and leaving the store based on the RFID information 342 and the people flow count information 343. Furthermore, the arithmetic device 31 generates image data for displaying the counted number of customers entering and leaving the store, and transmits it to the count display device 40 via the communication interface 33.

[0049] The memory 32 includes a storage area (for example, a working area) for storing program codes or work memories when the arithmetic unit 31 executes various programs. Examples of the memory 12 include volatile memories such as DRAM and SRAM, and non-volatile memories such as ROM and flash memory.

[0050] The storage device 34 stores various programs executed by the arithmetic device 31 or various data. For example, the storage device 34 stores a counting program 341 executed by the arithmetic device 31, RFID information 342 referred to by the arithmetic device 31, and people flow count information 343. The storage device 34 may be one or more non-transitory computer readable mediums, or one or more computer readable storage media. Examples of the storage device 34 include a hard disk drive (HDD) and a solid state drive (SSD).

[0051] The counting program 341 includes a program that specifies a processing procedure for the arithmetic device 31 to count the number of customers entering and leaving the store based on the RFID information 342 and the people flow count information 343. The counting device 30 acquires the RFID information 342 from the RFID reader 10 via the communication interface 33. The counting device 30 acquires the people flow count information 343 from the people flow count sensor 20 via the communication interface 33. The counting device 30 may store the counting program 341 in advance in the storage device 34, or may store the counting program 341 acquired from the storage medium 50 by the media reading device 35 in the storage device 34.

[0052] The communication interface 33 is an interface for the arithmetic device 31 to communicate with other devices via a wired connection using an optical fiber line or a wireless connection such as Wi-Fi. For example, the communication interface 33 communicates with the RFID reader 10 under the control of the arithmetic device 31 to acquire RFID information 342 from the RFID reader 10. The communication interface 33 also communicates with the people flow count sensor 20 under the control of the arithmetic device 31 to acquire people flow count information 343 from the people flow count sensor 20. Furthermore, the communication interface 33 also communicates with the count display device 40 under the control of the arithmetic device 31 to transmit image data to the count display device 40 for displaying the number of customers entering and leaving the store.

[0053] The counting device 30 may include a media reading device. The media reading device receives a storage medium (not shown) and acquires various programs (e.g., the counting program 341) or various data from the storage medium. Examples of the storage medium include a CD (Compact Disc), a DVD (Digital Versatile Disc), a USB (Universal Serial Bus) memory, and an SD (Secure Digital) card.

[0054] The count display device 40 includes a communication interface 43 and a display 44 .

[0055] The communication interface 43 communicates with the counting device 30 via a wired connection using an optical fiber line or a wireless connection such as Wi-Fi, and obtains image data from the counting device 30 to display the number of customers entering and leaving the store.

[0056] The display 44 displays the number of customers entering and leaving the store based on image data acquired via the communication interface 43.

[0057] In the above example, the counting device 30 generates image data and transmits it to the counting display device 40, but the counting device 30 may transmit data indicating the calculated number of customers to the counting display device 40. Furthermore, the counting display device 40 may further include a calculation device, a memory, a storage device, etc., and may generate an image indicating the number of customers based on the data indicating the number of customers obtained from the counting device 30, and display it on the display 44.

[0058] [An example of counting system processing] An example of the process executed by the counting system 1 will be described with reference to FIGS.

[0059] Fig. 3 is a flowchart of the process executed by the RFID reader 10 according to the first embodiment. The data processing IC 11 of the RFID reader 10 executes a control program (not shown) to periodically and repeatedly execute the process of the flowchart shown in Fig. 3. In Figs. 3 to 5, "S" is used as an abbreviation for "STEP."

[0060] 3, the RFID reader 10 determines whether or not the antenna 14 has read the ID information contained in the RFID tag 4 (S11). If the RFID reader 10 has read the ID information (YES in S11), it identifies the time when the ID information was read (read time) (S12). The RFID reader 10 associates the ID information with the read time and stores it in the memory 12 as RFID information (S13).

[0061] If the RFID reader 10 has not read the ID information (NO in S11) or after the process of S13, it determines whether it is time to transmit the RFID information (S14). If it is time to transmit the RFID information (YES in S14), the RFID reader 10 transmits the RFID information stored in the memory 12 to the counting device 30 (S15).

[0062] On the other hand, if it is not the timing to transmit the RFID information (NO in S14), or after the process of S15, the RFID reader 10 ends the process of this flowchart.

[0063] In the flowchart shown in FIG. 3, the RFID reader 10 reads the ID information in S11 and then transmits the RFID information in S15. However, in parallel with repeatedly executing the process of reading the ID information, the RFID reader 10 may also execute a process of transmitting the RFID information at regular intervals.

[0064] 4 is a flowchart of a process executed by the people flow count sensor 20 according to the embodiment 1. The data processing IC 21 of the people flow count sensor 20 executes a control program (not shown) to periodically and repeatedly execute the process of the flowchart shown in FIG.

[0065] 4, the people flow count sensor 20 determines whether or not a person passing through the detection area 3 is detected based on image data acquired by the camera 24 (S21). When the people flow count sensor 20 detects a person passing through the detection area 3 (YES in S21), it identifies the movement direction of the detected person based on the image data acquired by the camera 24 (S22).

[0066] The people flow count sensor 20 identifies the time when the person was detected (detection time) (S23). The people flow count sensor 20 calculates the number of people who entered the store to update the store entry count number, calculates the number of people who left the store to update the store exit count number, and stores the people flow count information in memory 22 in association with the detection time when the person was detected (S24).

[0067] When the people flow count sensor 20 has not detected a person passing through the detection area 3 (NO in S21), or after the processing of S24, it determines whether or not it is time to transmit people flow count information (S24). When it is time to transmit people flow count information (YES in S25), the people flow count sensor 20 transmits the people flow count information stored in the memory 22 to the counting device 30 and clears the people flow count information to set the people flow count number to 0 (S26).

[0068] On the other hand, if it is not the timing to transmit people flow count information (NO in S25), or after the process of S26, the people flow count sensor 20 ends the process of this flowchart.

[0069] In the flowchart shown in Figure 4, the people flow count sensor 20 detects a person in S21 and then transmits people flow count information in S26, but it may also execute a process of transmitting people flow count information at regular intervals in parallel with repeatedly executing the process of detecting people.

[0070] Fig. 5 is a flowchart of the process executed by the counting device 30 according to the first embodiment. The arithmetic device 31 of the counting device 30 executes the counting program 341 to execute the process of the flowchart shown in Fig. 5. The arithmetic device 31 may periodically and repeatedly execute the process of the flowchart shown in Fig. 5, may execute the process when RFID information or people flow count information is acquired, may execute the process when the amount of data of RFID information or people flow count information stored in the storage device 34 exceeds a threshold, or may execute the process at a timing set by the user (for example, when the store is closing).

[0071] 5, the counting device 30 acquires people flow count information (S31). In the process of S31, the counting device 30 may acquire people flow count information in real time from the people flow count sensor 20, or may acquire from the storage device 34 RFID information that has already been acquired from the people flow count sensor 20 and stored in the storage device 34.

[0072] The counting device 30 acquires the RFID information (S32). In the process of S32, the counting device 30 may acquire the RFID information in real time from the RFID reader 10, or may acquire from the storage device 34 the RFID information that has already been acquired from the RFID reader 10 and stored in the storage device 34.

[0073] The counting device 30 determines the time when a person was detected by the people flow count sensor 20 during a specified period (for example, one day from opening to closing) based on the people flow count information, and determines the time when the ID information was read by the RFID reader 10 during the specified period (for example, one day from opening to closing) based on the RFID information, and determines whether or not there is a specific time among the times when a person was detected by the people flow count sensor 20 that matches the time when the ID information was read by the RFID reader 10 (S33).

[0074] If there is a specific time among the times when a person was detected by the people flow counting sensor 20 during a specified period that coincides with the time when ID information was read by the RFID reader 10 (YES in S33), the counting device 30 executes a subtraction process to subtract the number of ID information read by the RFID reader 10 at the specific time from the people flow count number (S34).

[0075] For example, when counting the number of customers entering a store in a specified period of time, the counting device 30 may subtract the number of ID information read by the RFID reader 10 at a specific time from the store entry count included in the people flow count. On the other hand, when counting the number of customers leaving a store in a specified period of time, the counting device 30 may subtract the number of ID information read by the RFID reader 10 at a specific time from the store exit count included in the people flow count.

[0076] If there is no specific time among the times when a person was detected by the people flow counting sensor 20 that matches the time when the ID information was read by the RFID reader 10 (NO in S33), or after processing S34, the counting device 30 ends the processing of this flowchart.

[0077] The counting device 30 executes the process of the flowchart shown in Fig. 5 for each of the multiple ID information read by the RFID reader 10. For example, in S32, the counting device 30 acquires the specific ID information read by the RFID reader 10 and the reading time of the specific ID information. In S33, the counting device 30 determines whether or not there is a specific time that coincides with the time that the specific ID information was read by the RFID reader 10 among the times at which people were detected by the people flow count sensor 20 during a predetermined period, and if there is a specific time, subtracts the number of specific ID information read by the RFID reader 10 at the specific time from the people flow count number.

[0078] As described above, the counting device 30 can use the people flow count sensor 20 to calculate a people flow count number for identifying the number of people entering and leaving the store, and can read the ID information of store staff entering and leaving the store using the RFID reader 10. Furthermore, if there is a specific time among the times when people are detected by the people flow count sensor 20 that coincides with the time when the ID information was read by the RFID reader 10, the counting device 30 can calculate the number of customers entering and leaving the store by subtracting the number of ID information read at the specific time from the people flow count number. This allows the counting device 30 to appropriately count the number of customers entering and leaving the store.

[0079] <Embodiment 2> The counting system 1 according to the second embodiment will be described with reference to Figures 6 and 7. Only the parts of the counting system 1 according to the second embodiment that are different from the counting system 1 according to the first embodiment will be described below.

[0080] In the counting system 1 according to the first embodiment described above, the RFID reader 10 is provided only near the entrance 2 outside the store. The counting device 30 can identify the ID information read by the RFID reader 10 and the time when the ID information was read by the RFID reader 10 based on the RFID information acquired from the RFID reader 10, but cannot identify whether the store clerk whose ID information was read by the RFID reader 10 entered the store, whether the store clerk left the store, or whether the store clerk stayed within the communication range. For this reason, for example, if a customer passes through the entrance 2 and enters the store while a store clerk is working near the entrance 2 outside the store, the time when the customer entering the store is detected by the people flow count sensor 20 and the time when the ID information of the store clerk staying outside the store is read by the RFID reader 10 may coincidentally match. In such a case, the counting device 30 may mistakenly assume that the person entering the store detected by the people flow counting sensor 20 is a store clerk, even though the store clerk is actually outside the store, and perform a subtraction process.

[0081] In view of the above, the counting system 1 of embodiment 2 is further equipped with an RFID reader 10 not only outside the store but also inside the store, and is configured to estimate whether a store clerk has entered, left, or stayed inside the store using the two RFID readers 10 installed both inside and outside the store.

[0082] Fig. 6 is a diagram for explaining the counting system 1 according to the embodiment 2. As shown in Fig. 6, the counting system 1 according to the embodiment 2 includes, at each entrance 2, one RFID reader 10A located outside the store and one RFID reader 10B located inside the store.

[0083] The RFID reader 10A is an example of a "first RFID reader." When the RFID tag 4 held by a store clerk enters the communication range of the RFID reader 10A provided outside the store, the RFID reader 10A communicates with the RFID tag 4 to read the ID information. The RFID reader 10A may also communicate with the RFID tag 4 to read the ID information when the RFID tag 4 held by a store clerk enters the communication range of the RFID reader 10B provided outside the store. For example, when a store clerk enters the store from outside, the ID information is read first by the RFID reader 10A, and then by the RFID reader 10B, and when a store clerk leaves the store, the ID information is read first by the RFID reader 10B, and then by the RFID reader 10A.

[0084] The RFID reader 10B is an example of a "second RFID reader." When the RFID tag 4 held by the store clerk enters the communication range of the RFID reader 10B installed inside the store, the RFID reader 10B communicates with the RFID tag 4 to read the ID information. Note that the RFID reader 10B may also communicate with the RFID tag 4 to read the ID information when the RFID tag 4 held by the store clerk enters the communication range of the RFID reader 10A installed outside the store. For example, when the store clerk enters the store from outside, the ID information is read first by the RFID reader 10A, and then by the RFID reader 10B, and when the store clerk leaves the store, the ID information is read first by the RFID reader 10B, and then by the RFID reader 10A.

[0085] By using RFID reader 10A and RFID reader 10B installed inside and outside the store, counting device 30 can estimate whether the store clerk, who is the owner of the ID information read by RFID reader 10A and RFID reader 10B, entered, left, or stayed in the store. Specifically, counting device 30 can estimate whether the store clerk entered (entered the store), left (left the store), or stayed in the store, depending on which of RFID reader 10A and RFID reader 10B read the ID information first.

[0086] Fig. 7 is a diagram for explaining the estimation process of a person entering a store by the counting device 30 according to the embodiment 2. In Fig. 7, information indicating whether or not the ID information has been read for each of the RFID reader 10A and the RFID reader 10B, information indicating which of the RFID reader 10A and the RFID reader 10B read the ID information first, and information indicating whether the person entering the store detected by the people flow count sensor 20 at the detection time that coincides with the reading time of the ID information is a customer or a store clerk are collected for each pattern.

[0087] As shown in Fig. 7, pattern A shows a situation in which the ID information is first read by RFID reader 10B in the store, and then the ID information is read again by RFID reader 10B. In this case, counting device 30 can assume that the ID information is read two or more times in succession by RFID reader 10B while a store clerk is working and staying near entrance / exit 2 in the store. Therefore, counting device 30 can assume that the person entering the store detected by people flow count sensor 20 when the ID information is read by RFID reader 10B is a customer, not a store clerk.

[0088] Therefore, in the case of pattern A, even if the time when a person entering the store is detected by the people flow counting sensor 20 coincides with the time when the ID information is read by the RFID reader 10B, the counting device 30 does not subtract the number of ID information read by the RFID reader 10B (in this case, "1") from the people flow count (entry count).

[0089] Pattern B shows a situation where the ID information is first read by RFID reader 10A outside the store, and then the ID information is read again by RFID reader 10A. In this case, counting device 30 can assume that the ID information is read two or more times in succession by RFID reader 10A while the store clerk is working near entrance / exit 2 outside the store. Therefore, counting device 30 can assume that the person entering the store detected by people flow count sensor 20 when the ID information is read by RFID reader 10A is a customer, not a store clerk.

[0090] Therefore, in the case of pattern B, even if the time when a person entering the store is detected by the people flow counting sensor 20 coincides with the time when the ID information is read by the RFID reader 10A, the counting device 30 does not subtract the number of ID information read by the RFID reader 10A (in this case, "1") from the people flow count (entry count).

[0091] Pattern C shows a situation where the ID information is first read by RFID reader 10A outside the store, and then read by RFID reader 10B inside the store. In this case, counting device 30 can assume that the ID information is read by RFID reader 10A and RFID reader 10B when the store clerk enters the store from outside. Therefore, counting device 30 can assume that one of the people entering the store detected by people flow counting sensor 20 when the ID information is read by RFID reader 10A or RFID reader 10B is the store clerk.

[0092] Therefore, in the case of pattern C, when the time when a person entering the store is detected by people flow counting sensor 20 coincides with the time when ID information is read by RFID reader 10A or RFID reader 10B, counting device 30 subtracts the number of ID information read by RFID reader 10A or RFID reader 10B (in this case, "1") from the people flow count (entry count).

[0093] Pattern D shows a situation where the ID information is first read by RFID reader 10B inside the store, and then read by RFID reader 10A outside the store. In this case, counting device 30 can assume that the ID information is read by RFID reader 10A and RFID reader 10B when the store clerk leaves the store. Therefore, counting device 30 can assume that the person entering the store detected by people flow count sensor 20 when the ID information is read by RFID reader 10A or RFID reader 10B is a customer, not a store clerk. That is, an example of pattern D is an example where a customer enters the store to replace the store clerk.

[0094] Therefore, in the case of pattern D, even if the time when a person entering the store is detected by people flow counting sensor 20 coincides with the time when ID information is read by RFID reader 10A or RFID reader 10B, counting device 30 does not subtract the number of ID information read by RFID reader 10A or RFID reader 10B (in this case, "1") from the people flow count (entry count).

[0095] Note that the example in Figure 7 shows a case where the counting device 30 calculates the number of customers entering the store, but even when the counting device 30 calculates the number of customers leaving the store, it can estimate whether a store clerk has entered the store (entered the store), exited the store (left the store), or stayed there, depending on which of RFID reader 10A and RFID reader 10B read the ID information first.

[0096] For example, if the ID information is first read by RFID reader 10A outside the store and then by RFID reader 10B inside the store, counting device 30 can assume that the ID information was read by RFID reader 10A and RFID reader 10B when the store clerk entered the store from outside. In this case, counting device 30 assumes that the person exiting the store detected by people flow count sensor 20 when the ID information was read by RFID reader 10A or RFID reader 10B is a customer and not a store clerk, and does not perform the subtraction process.

[0097] On the other hand, if the ID information is first read by RFID reader 10B inside the store and then by RFID reader 10A outside the store, counting device 30 can assume that the ID information was read by RFID reader 10A and RFID reader 10B when the store clerk leaves the store. In this case, counting device 30 can assume that the person exiting the store detected by people flow count sensor 20 when the ID information is read by RFID reader 10A or RFID reader 10B is a store clerk and perform subtraction processing.

[0098] As described above, counting device 30 selects whether to execute the subtraction process depending on which of RFID reader 10A and RFID reader 10B reads the ID information first. This allows counting device 30 to exclude from subtraction process cases where ID information is acquired by RFID reader 10 while a store clerk is inside or outside the store, or cases where a customer enters and leaves the store to replace the store clerk, and therefore allows counting the number of customers entering and leaving the store with higher accuracy.

[0099] <Embodiment 3> The counting system 1 according to the third embodiment will be described with reference to Fig. 8 and Fig. 9. In the following, only the parts of the counting system 1 according to the third embodiment that are different from the counting systems 1 according to the first and second embodiments will be described.

[0100] In the counting system 1 of the above-mentioned embodiment 2, the counting device 30 uses two RFID readers 10 installed both inside and outside the store, whereas in the counting system 1 of embodiment 3, a single RFID reader 10C installed outside the store is used to estimate whether the store clerk who is the owner of the ID information read by the RFID reader 10C has entered, left, or stayed inside the store.

[0101] FIG. 8 is a diagram for explaining the counting system 1 according to the third embodiment. As shown in FIG. 8, the counting system 1 according to the third embodiment includes one RFID reader 10C located outside the store at each entrance 2. When the RFID tag 4 held by a store clerk enters the communication range of the RFID reader 10C, the RFID reader 10C communicates with the RFID tag 4 to read the ID information stored in the RFID tag 4, and can also detect the reception strength in communication with the RFID tag 4. The communication range of the RFID reader 10C is wide, and the ID information stored in the RFID tag 4 can be read not only outside the store but also when the store clerk is inside the store. When the store clerk is outside the store, the RFID reader 10C detects that the reception strength in communication with the RFID tag 4 is strong, and when the store clerk is inside the store, the RFID reader 10C detects that the reception strength in communication with the RFID tag 4 is weak.

[0102] Depending on the change in reception strength in communication with the RFID tag 4 detected by the RFID reader 10C, the counting device 30 can estimate whether the store clerk who is the owner of the ID information read by the RFID reader 10C has entered, left, or stayed in the store.

[0103] Fig. 9 is a diagram for explaining the estimation process of a person entering a store by the counting device 30 according to the embodiment 3. In Fig. 9, information indicating the reception strength when ID information is read by the RFID reader 10C installed outside the store and information indicating whether a person entering the store detected by the people flow count sensor 20 at the detection time that coincides with the reading time of the ID information is a customer or a store clerk are collected for each pattern.

[0104] 9, pattern A shows a situation where the reception strength in the communication between the RFID reader 10C and the RFID tag 4 is strong first, and the reception strength in the communication between the RFID reader 10C and the RFID tag 4 is also strong thereafter. In this case, the counting device 30 can assume that the ID information is read twice in succession by the RFID reader 10C while the store clerk is working near the entrance / exit 2 outside the store. Therefore, the counting device 30 can assume that the person entering the store detected by the people flow counting sensor 20 when the ID information is read by the RFID reader 10C is a customer, not a store clerk.

[0105] Therefore, in the case of pattern A, even if the time when a person entering the store is detected by the people flow counting sensor 20 coincides with the time when the ID information is read by the RFID reader 10C, the counting device 30 does not subtract the number of ID information read by the RFID reader 10C (in this case, "1") from the people flow count (entry count).

[0106] Pattern B shows a situation where the reception strength in the communication between the RFID reader 10C and the RFID tag 4 is weak at first, and continues to be weak thereafter. In this case, the counting device 30 can assume that the ID information is read twice in succession by the RFID reader 10C while the store clerk is working near the entrance / exit 2 in the store. Therefore, the counting device 30 can assume that the person entering the store detected by the people flow count sensor 20 when the ID information is read by the RFID reader 10C is a customer, not a store clerk.

[0107] Therefore, in the case of pattern B, even if the time when a person entering the store is detected by the people flow counting sensor 20 coincides with the time when the ID information is read by the RFID reader 10C, the counting device 30 does not subtract the number of ID information read by the RFID reader 10C (in this case, "1") from the people flow count (entry count).

[0108] Pattern C shows a situation where the reception strength in the communication between the RFID reader 10C and the RFID tag 4 is strong first, and then the reception strength in the communication between the RFID reader 10C and the RFID tag 4 becomes weak. In this case, the counting device 30 can assume that the ID information is read twice in succession by the RFID reader 10C when the store clerk enters the store from outside. Therefore, the counting device 30 can assume that one of the people entering the store detected by the people flow counting sensor 20 when the ID information is read by the RFID reader 10C is the store clerk.

[0109] Therefore, in the case of pattern C, when the time when a person entering the store is detected by the people flow counting sensor 20 coincides with the time when the ID information is read by the RFID reader 10C, the counting device 30 subtracts the number of ID information read by the RFID reader 10C (in this case, "1") from the people flow count (entry count).

[0110] Pattern D shows a situation where the reception strength in the communication between the RFID reader 10C and the RFID tag 4 is weak at first, and then the reception strength in the communication between the RFID reader 10C and the RFID tag 4 becomes strong. In this case, the counting device 30 can assume that the ID information is read twice in succession by the RFID reader 10C when the store clerk leaves the store. Therefore, the counting device 30 can assume that the person entering the store detected by the people flow counting sensor 20 when the ID information is read by the RFID reader 10C is a customer, not a store clerk. That is, the example of pattern D is an example where a customer enters the store to replace the store clerk.

[0111] Therefore, in the case of pattern D, even if the time when a person entering the store is detected by the people flow counting sensor 20 coincides with the time when the ID information is read by the RFID reader 10C, the counting device 30 does not subtract the number of ID information read by the RFID reader 10C (in this case, "1") from the people flow count (entry count).

[0112] Note that the example in Figure 9 shows a case where the counting device 30 calculates the number of customers entering the store, but even when the counting device 30 calculates the number of customers leaving the store, it can estimate whether a store clerk has entered the store (entered the store), exited the store (left the store), or stayed there, depending on the change in reception strength in communication with the RFID tag 4 detected by the RFID reader 10C.

[0113] For example, if the reception strength in the communication between the RFID reader 10C and the RFID tag 4 is strong first and then becomes weak, the counting device 30 can assume that the ID information is read twice in succession by the RFID reader 10C when a store clerk enters the store from outside. In this case, the counting device 30 assumes that the person exiting the store detected by the people flow count sensor 20 when the ID information is read by the RFID reader 10C is a customer and not a store clerk, and does not perform the subtraction process.

[0114] On the other hand, if the reception strength in the communication between RFID reader 10C and RFID tag 4 is weak at first and then becomes strong, counting device 30 can assume that the ID information has been read twice in succession by RFID reader 10C when the store clerk leaves the store. In this case, counting device 30 can assume that the person exiting the store detected by people flow count sensor 20 when the ID information is read by RFID reader 10C is the store clerk and executes the subtraction process.

[0115] As described above, the counting device 30 may select whether or not to execute the subtraction process depending on the change in the reception strength detected by the RFID reader 10C in communication with the RFID tag 4. This allows the counting device 30 to exclude from the subtraction process cases where the ID information is acquired by the RFID reader 10C while a store clerk is inside or outside the store, or cases where a customer enters or leaves the store to replace the store clerk, and therefore allows the number of customers entering and leaving the store to be counted with higher accuracy.

[0116] In the counting system 1 according to the above-mentioned embodiment 3, the RFID reader 10C is configured to detect the reception strength in communication with the RFID tag 4. However, as shown in FIG. 8, the RFID reader 10C may have directionality in communication with the RFID tag 4.

[0117] For example, only one RFID reader 10C is provided inside or outside the store, and may be capable of communicating with an RFID tag 4 held by a store clerk located near the entrance / exit 2 inside the store, and also with an RFID tag 4 held by a store clerk located near the entrance / exit 2 outside the store. Furthermore, RFID reader 10C may transmit ID information to counting device 30 in a manner that allows the user to distinguish whether the ID information has been acquired from an RFID tag 4 located inside the store or from an RFID tag 4 located outside the store.

[0118] This allows the counting device 30 to specify whether it has read the ID information from the RFID tag 4 of a store clerk located inside the store, or from the RFID tag 4 of a store clerk located outside the store, using one RFID reader 10C installed inside or outside the store. Therefore, using this information, the counting device 30 can estimate whether a person entering or leaving the store is a store clerk or a customer, based on patterns A to D as shown in Fig. 7, and can select whether or not to perform a subtraction process depending on the estimation result.

[0119] <Fourth embodiment> A counting system 1 according to embodiment 4 will be described with reference to Fig. 10. Only the parts of the counting system 1 according to embodiment 4 that are different from the counting systems 1 according to embodiments 1 to 3 will be described below. Fig. 10 is a diagram for explaining the counting system 1 according to embodiment 4.

[0120] In the counting systems 1 according to each of the first to third embodiments described above, the RFID reader 10 was able to communicate with the RFID tag 4 and read the ID information simply when the RFID tag 4 was located within the communication range, but as shown in Fig. 10, the counting system 1 according to the fourth embodiment may be configured so that the RFID reader 10D communicates with the RFID tag 4 and reads the ID information when a store clerk touches the RFID reader 10D with a card including the RFID tag 4. Alternatively, the counting system 1 may be provided with a button that can only be pressed by the store clerk, near the entrance 2 inside or outside the store, instead of the RFID reader 10D.

[0121] In this way, when a store clerk enters or leaves the store, the counting device 30 can count the number of store clerks entering and leaving the store by touching the RFID tag 4 to the RFID reader 10D or pressing a button, and subtract the number of counted store clerks from the people flow count number by subtraction processing. As a result, the counting system 1 does not need to read ID information from the RFID tag 4 owned by a store clerk working near the entrance / exit 2, for example, because the RFID reader 10D does not need to communicate with the RFID tag 4 that is several meters away, and it can count the number of customers entering the store with high accuracy by simply subtracting the number of store clerks entering the store from the people flow count information based on the touched ID information.

[0122] <Embodiment 5> A counting system 1 according to embodiment 5 will be described with reference to Fig. 11. Only the parts of counting system 1 according to embodiment 5 that are different from counting systems 1 according to each of embodiments 1 to 4 will be described below. Fig. 11 is a diagram for explaining counting system 1 according to embodiment 5.

[0123] As shown in Fig. 11, in the counting system 1 according to the fifth embodiment, the RFID reader 10E may be provided inside the store at each entrance 2, and may be configured to communicate with the RFID tag 4 located outside the store from inside the store by directing the radiation direction of the antenna 14 from inside the store to outside the store, thereby reading ID information from the RFID tag 4. Furthermore, the RFID reader 10E may be provided on the ceiling inside the store at each entrance 2, and may be configured to communicate with the RFID tag 4 located outside the store from the inside ceiling of the store, thereby reading ID information from the RFID tag 4. In this case, the RFID reader 10E does not need to detect the location, but is configured to have directivity to emit radio waves only in the direction of the arrow in Fig. 11, and thus a highly accurate counting system 1 can be configured by combining it with a detection process of reception strength like the counting system 1 according to the third embodiment.

[0124] This allows the counting system 1 to communicate with the RFID tag 4 held by a store clerk located outside the store, while the RFID reader 10 is installed inside the store.

[0125] <Sixth embodiment> 12 to 14, the counting system 1 according to the sixth embodiment will be described. In the following, only the parts of the counting system 1 according to the sixth embodiment that are different from the counting systems 1 according to the first to fifth embodiments will be described.

[0126] In the counting system 1 of embodiment 6, the counting device 30 is configured to estimate, based on the time when the ID information is read by the RFID reader 10, whether the store clerk who is the owner of the ID information read by the RFID reader 10 has entered, left, or stayed in the store.

[0127] 12 and 13 are diagrams for explaining an example of the estimation process of the behavior of a store clerk by the counting device 30 according to the sixth embodiment. As shown in FIG. 12, a specific ID information included in an RFID tag 4 held by a specific store clerk is assumed to be "ID_0". RFID information including ID information (ID_0) read during a predetermined period (for example, one day from opening to closing) and the read time (timestamp) of the ID information is configured as shown in FIG. 12. For example, the counting device 30 acquires RFID information of "RFID:ID_0,t:timestamp_00" in the first reading by the RFID reader 10, acquires RFID information of "RFID:ID_0,t:timestamp_01" in the second reading by the RFID reader 10, and acquires RFID information of "RFID:ID_0,t:timestamp_n+1" in the nth reading by the RFID reader 10. Here, "n" is an integer equal to or greater than 0.

[0128] The counting device 30 calculates the time difference between the time when the ID information (ID_0) is read first and the time when the ID information (ID_0) is read later. For example, the counting device 30 calculates "d_timestamp_00" as the time difference between them by subtracting "timestamp_00" from "timestamp_01". The counting device 30 also calculates "d_timestamp_n" as the time difference between them by subtracting "timestamp_n" from "timestamp_n+1".

[0129] The counting device 30 compares the time difference "d_timestamp_n" (n=0, 1, 2, ...) with a predetermined threshold, and estimates whether the store clerk with ID information (ID_0) entered, left, or stayed in the store depending on whether the time difference "d_timestamp_n" exceeds the threshold.

[0130] For example, if "d_timestamp_n" is less than a first threshold value (e.g., 10 seconds), the time difference between two successive reading times ("timestamp_n+1", "timestamp_n") is relatively short, so the counting device 30 can estimate that a store clerk was inside the store or near the entrance / exit 2 from the earlier reading time ("timestamp_n") to the later reading time ("timestamp_n+1").

[0131] If "d_timestamp_n" exceeds a second threshold (for example, 600 seconds), the time difference between two consecutive reading times ("timestamp_n+1", "timestamp_n") is relatively long, and the counting device 30 can infer that the store clerk was away from the store, i.e., had left the store, from the earlier reading time ("timestamp_n") to the later reading time ("timestamp_n+1").

[0132] If "d_timestamp_n" is equal to or greater than the first threshold and equal to or less than the second threshold, the counting device 30 can estimate that a store clerk entered the store at the previous reading time ("timestamp_n").

[0133] The counting device 30 may use one threshold, not just two first and second thresholds, to estimate the presence of a store clerk when "d_timestamp_n" is less than the threshold, and estimate the departure of the store clerk when "d_timestamp_n" is equal to or greater than the threshold. In this case, the counting device 30 may estimate the entrance of a store clerk at the timing when "d_timestamp_n" goes from equal to or greater than the threshold to less than the threshold. The first and second thresholds may be set in advance by a user, or may be calculated by machine learning or a probability model. The first threshold may be set to a value such that "d_timestamp_n" is less than the first threshold if a store clerk was present at the previous reading time ("timestamp_n"). The second threshold may be set to a value such that "d_timestamp_n" exceeds the second threshold if a store clerk was leaving at the previous reading time ("timestamp_n").

[0134] For example, as shown in Figure 13, if a histogram is created by plotting the time difference between read times "d_timestamp_n" on the horizontal axis and the number of data on the vertical axis, it can be seen that when "d_timestamp_n" is equal to or greater than the second threshold, the number of "d_timestamp_n" is smaller than when "d_timestamp_n" is less than the second threshold.

[0135] If the time when a person entering the store is detected by the people flow counting sensor 20 and the time when specific ID information (ID_0) is read by the RFID reader 10 coincide at a specific time, the counting device 30 calculates the time difference "d_timestamp_n" (n = 0, 1, 2, ...) for the specific ID information (ID_0), and if it estimates that the store clerk entered the store at the specific time based on the time difference "d_timestamp_n", it subtracts the number of specific ID information (ID_0) read by the RFID reader 10 at the specific time (in this case, "1") from the people flow count (entry count).

[0136] As described above, counting device 30 calculates the time difference between the time when specific ID information was read by RFID reader 10 and the time after specific ID information was read by RFID reader 10, estimates whether a specific store clerk entered, left, or stayed in the store based on the time difference, and selects whether to perform subtraction processing depending on the estimation result. In this way, counting device 30 can estimate the behavior of the store clerk using the reading time of RFID reader 10, and can count the number of customers entering and leaving the store with higher accuracy.

[0137] Furthermore, the counting device 30 may compare the estimation result of the behavior of the store clerk based on the RFID information as described above with the people flow count number to determine whether the estimation result is correct or not. Fig. 14 is a diagram for explaining an example of the estimation result determination process by the counting device 30 according to the sixth embodiment.

[0138] 14, when the counting device 30 estimates based on the RFID information that a store clerk with specific ID information entered the store at a specific read time, if the store entry count number detected by the people flow count sensor 20 at the detection time that coincides with the specific read time is 1 or more, there is a possibility that the store clerk has entered the store, and therefore it can be determined that there is a high probability that the store clerk with the specific ID information entered the store at the specific read time. In this case, the counting device 30 subtracts the number of specific ID information read by the RFID reader 10 at the specific read time (in this case, "1") from the people flow count number (store entry count number).

[0139] On the other hand, if the counting device 30 estimates based on the RFID information that a store clerk with specific ID information entered the store at a specific read time, if the entry count number detected by the people flow count sensor 20 at the detection time that coincides with the specific read time is 0, there is no possibility that a store clerk entered the store, and therefore it can determine that there is a high probability that the store clerk with the specific ID information did not enter the store at the specific read time.

[0140] In addition, if the counting device 30 estimates based on the RFID information that a store clerk with specific ID information has left the store at a specific read time, if the exit count number detected by the people flow count sensor 20 at a detection time that coincides with the specific read time is 1 or more, it is possible that the store clerk has left the store, and therefore it can determine that there is a high probability that the store clerk with the specific ID information left the store at the specific read time.

[0141] On the other hand, if the counting device 30 estimates based on the RFID information that a store clerk with specific ID information has left the store at a specific read time, if the exit count number detected by the people flow count sensor 20 at the detection time that coincides with the specific read time is 0, there is no possibility that the clerk has left the store, and it can determine that there is a high probability that the clerk with the specific ID information has not left the store at the specific read time.

[0142] In addition, when the counting device 30 estimates based on the RFID information that a store clerk with specific ID information was in the store at a specific read time, if the entry count number and exit count number detected by the people flow count sensor 20 at the detection time that coincides with the specific read time are both 0, there is no possibility that the clerk entered or left the store, and therefore it can determine that there is a high probability that a store clerk with specific ID information was in the store at the specific read time.

[0143] On the other hand, if the counting device 30 estimates based on the RFID information that a store clerk with specific ID information was in the store at a specific read time, if either the entry count number or the exit count number detected by the people flow count sensor 20 at the detection time that coincides with the specific read time is 1, it is possible that a store clerk has entered or left the store, and therefore it can be determined that there is a high probability that the store clerk with the specific ID information was not in the store at the specific read time.

[0144] As described above, the counting device 30 can determine whether the inference results of the behavior of store staff based on RFID information are correct based on the people flow count number, thereby enabling the number of customers entering and leaving the store to be counted with greater accuracy.

[0145] <Embodiment 7> 15 and 16, the counting system 1 according to the seventh embodiment will be described. In the following, only the parts of the counting system 1 according to the seventh embodiment that are different from the counting systems 1 according to the first to sixth embodiments will be described.

[0146] In the counting system 1 of embodiment 7, the counting device 30 is configured to estimate, based on the reading result of the RFID reader 10 and the people flow count calculated by the people flow count sensor 20, whether the store clerk who is the owner of the ID information read by the RFID reader 10 has entered or left the store (entered or left the store) or stayed there.

[0147] FIG. 15 is a diagram for explaining an example of a process of estimating the behavior of a store clerk by the counting device 30 according to the seventh embodiment. As shown in FIG. 15, a specific ID information included in an RFID tag 4 held by a specific store clerk is assumed to be "ID_0". RFID information including ID information (ID_0) read during a predetermined period (for example, 1800 seconds) and the read time of the ID information is configured as shown in FIG. 15. For example, the counting device 30 acquires RFID information of "RFID:ID_0,t:35" in the first reading by the RFID reader 10 at 35 seconds, acquires RFID information of "RFID:ID_0,t:467" in the second reading by the RFID reader 10 at 467 seconds, and acquires RFID information of "RFID:ID_0,t:611" in the third reading by the RFID reader 10 at 611 seconds.

[0148] The counting device 30 processes the acquired RFID information and divides the data into the number of times the ID information is read for every 60 seconds. For example, Fig. 16 is a diagram for explaining an example of data processing of the RFID information in the estimation process of the counting device 30 according to the seventh embodiment.

[0149] 16, the counting device 30 assigns an identifier ("1" in this example) to the time when specific ID information (ID_0) is read by the RFID reader 10. In the example of Fig. 16, the counting device 30 assigns an identifier ("1") to each of the following times: 35 seconds, 467 seconds, 611 seconds, 612 seconds, 918 seconds, and 1708 seconds.

[0150] The counting device 30 divides 1800 seconds into time segments of 60 seconds each, t1 to t30. The counting device 30 generates vector data that can identify the presence or absence of an identifier ("1") for each of t1 to t30. For example, at t1, which corresponds to 1 to 60 seconds, an identifier ("1") is assigned at 35 seconds. At t11, which corresponds to 601 to 660 seconds, an identifier ("1") is assigned at 611 seconds and 612 seconds. At t29, which corresponds to 1681 to 1740 seconds, an identifier ("1") is assigned at 1708 seconds.

[0151] By generating 30 pieces of vector data t1 to t30 as described above, the counting device 30 can generate processed RFID information indicating whether or not specific ID information (ID_0) has been read every 60 seconds.

[0152] Returning to Fig. 15, the counting device 30 acquires people flow count information. The people flow count information includes the store entry count number (IN in the figure) and the store exit count number (OUT in the figure) for each of t1 to t30 in each 60 second period.

[0153] The counting device 30 processes the acquired people flow count information and converts it into data including the number of people entering and leaving the store (INOUT in the figure) and the number of people staying there (STAY in the figure) every 60 seconds. For example, in the case of the people flow count information of t1, the number of people entering the store (IN) is "1" and the number of people leaving the store (OUT) is "0", so the counting device 30 converts the number of people entering and leaving the store (INOUT) to "1" and the number of people staying there (STAY) to "0" in the people flow count information of t1 after processing. In the case of the people flow count information of t2, the number of people entering the store (IN) is "0" and the number of people leaving the store (OUT) is "0", so the counting device 30 converts the number of people entering and leaving the store (INOUT) to "0" and the number of people staying there (STAY) to "1" in the people flow count information of t2 after processing.

[0154] The counting device 30 includes an estimation model 36. For example, the estimation model 36 includes a neural network and parameters (not shown). For example, in the case of supervised learning, when processed RFID information is input, the estimation model 36 estimates, using a neural network, whether a store clerk with specific ID information (ID_0) stayed or entered or left the store during each 60-second time period based on the processed RFID information. The estimation model 36 extracts only the time periods determined to be stays from the estimation results, and determines whether the extracted time periods match the time periods in which the stay count number (STAY) included in the processed people flow count information is "1". If the two match, the estimation model 36 does not update the parameters, but if they do not match, updates the parameters so that the two match, thereby optimizing the parameters.

[0155] Using the estimation model 36 trained by machine learning as described above, the counting device 30 estimates whether a store clerk with specific ID information (ID_0) stayed or entered or left the store during each 60-second time period based on the processed RFID information. The counting device 30 determines the time when the store clerk with specific ID information (ID_0) switched from entering or leaving the store to staying as the specific time when the store clerk entered the store, and subtracts the number of pieces of ID information read by the RFID reader 10 at the specific time (in this case, "1") from the number of people who entered the store included in the people flow count (entry count).

[0156] As described above, the counting device 30 estimates whether the store clerk whose ID information is read by the RFID reader 10 has entered, left, or stayed in the store based on the RFID information and the people flow count number, and selects whether to execute a subtraction process depending on the estimation result. This allows the counting device 30 to estimate the behavior of the store clerk using the RFID information and the people flow count information, and therefore allows the number of customers entering and leaving the store to be counted with higher accuracy.

[0157] <Embodiment 8> 17 and 18, the counting system 1 according to the eighth embodiment will be described. In the following, only the parts of the counting system 1 according to the eighth embodiment that are different from the counting systems 1 according to the first to seventh embodiments will be described.

[0158] In the counting system 1 of embodiment 8, the counting device 30 is configured to estimate, based on the reading result of the RFID reader 10 and the people flow count calculated by the people flow count sensor 20, whether the store clerk who is the owner of the ID information read by the RFID reader 10 has entered, left, or stayed in the store.

[0159] Fig. 17 is a diagram for explaining an example of a process of estimating the behavior of store staff by the counting device 30 according to the eighth embodiment. As shown in Fig. 17, the counting device 30 acquires RFID information and people flow count information at the time when specific ID information is read by the RFID reader 10 during a predetermined period (for example, one day from opening to closing). The people flow count information includes the number of people entering the store (IN in the figure) and the number of people leaving the store (OUT in the figure).

[0160] In STEP 1, the counting device 30 estimates that a store clerk was present for people flow count information where the entry count number and exit count number are both "0." In the example of Fig. 17, the counting device 30 estimates that a store clerk was present for people flow count information corresponding to each of t1, t3, and t11.

[0161] In STEP 2, the counting device 30 estimates that a store clerk has entered the store for people flow count information whose first entry count number is "1" or more in a predetermined period. In the example of FIG. 17, the counting device 30 estimates that a store clerk has entered the store for people flow count information corresponding to t2. The counting device 30 also estimates that a store clerk has left the store for people flow count information whose last exit count number is "1" or more in a predetermined period. In the example of FIG. 17, the counting device 30 estimates that a store clerk has left the store for people flow count information whose last exit count number is "1" or more in a predetermined period.

[0162] In STEP 3, the counting device 30 extracts people flow count information in which one of the entry count number and the exit count number is "0" and the other is "1" or more, from people flow count information other than the people flow count information that was the estimation target in STEP 1 and STEP 2 (people flow count information that has not yet been the estimation target). In the example of Fig. 17, the counting device 30 extracts people flow count information corresponding to t5 and t9.

[0163] In STEP 4, the counting device 30 estimates whether the store clerk entered, left, or stayed in the store for the people flow count information extracted in STEP 3. In the example of FIG. 17, the counting device 30 estimates that the store clerk left the store for the people flow count information corresponding to t5 because a value of "1" or more is input only to the exit count number in the people flow count information corresponding to t5. Also, the counting device 30 estimates that the store clerk entered the store for the people flow count information corresponding to t9 because a value of "1" or more is input only to the entry count number in the people flow count information corresponding to t9. Note that the counting device 30 may estimate whether the store clerk entered, left, or stayed in the store by a method executed by the counting device 30 according to the sixth embodiment in FIG. 12 to FIG. 14 or the counting device 30 according to the seventh embodiment in FIG. 15 and FIG. 16.

[0164] In STEP 5, the counting device 30 extracts people flow count information other than the people flow count information that was the estimation target in STEP 1 to STEP 4 (people flow count information that has not yet been the estimation target) as an estimation target, and assigns one of groups A to D shown below to the people flow count information of the estimation target.

[0165] Specifically, when people flow count information corresponding to a time earlier than the people flow count information to be estimated (hereinafter also referred to as "early people flow count information") and people flow count information corresponding to a time later than the people flow count information to be estimated (hereinafter also referred to as "later people flow count information") are both estimated to be entering or leaving the store, and the entering / leaving status differs between the earlier people flow count information and the later people flow count information, the counting device 30 assigns group A to the people flow count information to be estimated.

[0166] For example, when the estimation results are arranged in chronological order of entering, estimated target, and leaving, the counting device 30 assigns group A to the people flow count information of the estimated target. Also, when the estimation results are arranged in chronological order of leaving, estimated target, and entering, the counting device 30 assigns group A to the people flow count information of the estimated target.

[0167] When the earlier and later people flow count information both estimate that a store employee has entered or left the store, and the earlier and later people flow count information indicate the same entry and exit status, the counting device 30 assigns group B to the people flow count information that is the subject of estimation.

[0168] For example, when the estimation results are arranged in chronological order of "enter store," "estimated target," and "enter store," the counting device 30 assigns group B to the people flow count information of the estimated target. When the estimation results are arranged in chronological order of "leave store," "estimated target," and "leave store," the counting device 30 assigns group B to the people flow count information of the estimated target.

[0169] If the earlier or later people flow count information estimates the people flow count information or stay of another estimation target extracted in STEP 5, and the estimated earlier people flow count information and the later people flow count information at the time closest to the time of the people flow count information of the estimation target have different entry and exit states, the counting device 30 assigns group C to the people flow count information of the estimation target.

[0170] For example, when the estimation results are arranged in chronological order of entering the store, the estimated target, staying or another estimated target, and leaving the store, the counting device 30 assigns group C to the people flow count information of the estimated target. Also, when the estimation results are arranged in chronological order of leaving the store, the estimated target, staying or another estimated target, and entering the store, the counting device 30 assigns group C to the people flow count information of the estimated target.

[0171] If the earlier or later people flow count information estimates the people flow count information or stay of another estimation target extracted in STEP 5, and the estimated earlier people flow count information and the later people flow count information have the same entry and exit status at the time closest to the time of the people flow count information of the estimation target, the counting device 30 assigns group D to the people flow count information of the estimation target.

[0172] For example, when the estimation results are arranged in chronological order of entering the store, estimated target, staying or other estimated target, entering the store, the counting device 30 assigns group D to the people flow count information of the estimated target. When the estimation results are arranged in chronological order of leaving the store, estimated target, staying or other estimated target, leaving the store, the counting device 30 assigns group D to the people flow count information of the estimated target.

[0173] In the example of FIG. 17, the counting device 30 assigns group C to people flow count information corresponding to each of t4, t6, t7, and t8.

[0174] In STEP 6, the counting device 30 estimates whether the store clerk entered, left, or stayed in the store based on the people flow count information of the estimation target according to the group assigned in STEP 5.

[0175] Specifically, the counting device 30 estimates that a store clerk was present for the people flow count information assigned to group A. The counting device 30 estimates that the people flow count information assigned to group B is a state different from the earlier people flow count information and the later people flow count information.

[0176] The counting device 30 estimates whether the store clerk entered / left the store or stayed there, for the people flow count information assigned to group C, using a method executed by the counting device 30 according to the sixth embodiment in Figs. 12 to 14 or the counting device 30 according to the seventh embodiment in Figs. 15 and 16. In the example of Fig. 17, the counting device 30 estimates that the store clerk entered / left the store for the people flow count information corresponding to t4, t7, and t8, and estimates that the store clerk stayed there for the people flow count information corresponding to t6.

[0177] Furthermore, when the number of people flow count information sandwiched between estimated people flow count information pieces for which a store clerk has entered or left the store is even, the counting device 30 assigns entries or exits so that the entries and exits are repeated. In the example of Fig. 17, the counting device 30 estimates that a store clerk has entered the store for the people flow count information corresponding to t7, and estimates that the store clerk has left the store for the people flow count information corresponding to t8.

[0178] On the other hand, when the number of people flow count information sandwiched between estimated people flow count information for which a store clerk has entered and left the store is odd, the counting device 30 assigns the people flow count information with the highest probability of being estimated as staying to staying, and assigns the remaining people flow count information as entering or leaving so that entering and leaving are repeated. In the example of Figure 17, the counting device 30 estimates that a store clerk has stayed for the people flow count information corresponding to t4.

[0179] For the people flow count information assigned to group D, the counting device 30 estimates whether a store clerk entered, left, or stayed in the store using a method performed by the counting device 30 relating to embodiment 6 in Figures 12 to 14, or the counting device 30 relating to embodiment 7 in Figures 15 and 16.

[0180] Furthermore, when the number of people flow count information sandwiched between estimated people flow count information pieces for people flow count information that estimates that a store clerk has entered or left the store is odd, the counting device 30 assigns entry or exit so that the clerk repeats entering and leaving the store.

[0181] On the other hand, when the number of people flow count information sandwiched between estimated people flow count information pieces is an even number, the counting device 30 assigns "stay" to the people flow count information that is estimated to be "stay" most likely to be "stay" for the people flow count information estimated to be "stay" for the remaining people flow count information, and assigns "enter" or "leave" to the remaining people flow count information so that the store repeatedly enters and leaves.

[0182] As a result of the above estimation, the counting device 30 subtracts the number of ID information read by the RFID reader 10 at the time when it is estimated that the store clerk entered the store from the number of people who entered the store included in the people flow count (store entry count). In the example of Fig. 17, the counting device 30 subtracts the number of ID information read by the RFID reader 10 at t2, t7, and t9 (in this case, "3") from the number of people who entered the store included in the people flow count (store entry count).

[0183] As described above, the counting device 30 estimates whether the store clerk whose ID information is read by the RFID reader 10 has entered, left, or stayed in the store based on the RFID information and the people flow count number, and selects whether to execute a subtraction process depending on the estimation result. This allows the counting device 30 to estimate the behavior of the store clerk using the RFID information and the people flow count information, and therefore allows the number of customers entering and leaving the store to be counted with higher accuracy.

[0184] FIG. 18 is a diagram for explaining another example of the process of estimating the behavior of a store clerk by the counting device according to the eighth embodiment.

[0185] In STEP 1, the counting device 30 estimates that a store clerk was present for people flow count information in which the entry count number and exit count number are both "0." In the example of Fig. 18, the counting device 30 estimates that a store clerk was present for people flow count information corresponding to t1.

[0186] In STEP 2, the counting device 30 estimates that a store clerk has entered the store for people flow count information whose entry count number is the first to be "1" or more in a predetermined period. In the example of FIG. 18, the counting device 30 estimates that a store clerk has entered the store for people flow count information corresponding to t2. The counting device 30 also estimates that a store clerk has left the store for people flow count information whose exit count number is the last to be "1" or more in a predetermined period. In the example of FIG. 18, the counting device 30 estimates that a store clerk has left the store for people flow count information whose exit count number is the last to be "1" or more in a predetermined period.

[0187] In STEP 3, the counting device 30 extracts people flow count information in which one of the entry count number and the exit count number is "0" and the other is "1" or more, from people flow count information other than the people flow count information that was the estimation target in STEP 1 and STEP 2 (people flow count information that has not yet been the estimation target). In the example of Fig. 18, the counting device 30 extracts people flow count information corresponding to t3, t5, t9, and t10, respectively.

[0188] In STEP 4, the counting device 30 estimates whether a store clerk entered, left, or stayed in the store for the people flow count information extracted in STEP 3. In the example of Fig. 18, the counting device 30 estimates that a store clerk entered the store for the people flow count information corresponding to each of t3 and t9. The counting device 30 also estimates that a store clerk left the store for the people flow count information corresponding to each of t5 and t10.

[0189] In STEP 5, the counting device 30 extracts people flow count information other than the people flow count information that has been subject to estimation in STEPs 1 to 4 (people flow count information that has not yet been subject to estimation) as a subject of estimation, and assigns one of groups A to D to the people flow count information that is subject to estimation. In the example of Fig. 18, the counting device 30 assigns group C to the people flow count information corresponding to each of t4, t6, t7, and t8.

[0190] In STEP 6, the counting device 30 estimates whether a store clerk entered, left, or stayed in the store for the people flow count information to be estimated according to the group assigned in STEP 5. In the example of Fig. 18, the counting device 30 estimates that a store clerk entered and left the store for the people flow count information corresponding to t4, t7, and t8, and estimates that a store clerk stayed in the store for the people flow count information corresponding to t6.

[0191] Furthermore, when the number of people flow count information sandwiched between estimated people flow count information pieces for which a store clerk has entered or left the store is even, the counting device 30 assigns entries or exits so that the entries and exits are repeated. In the example of Fig. 18, the counting device 30 estimates that a store clerk has entered the store for the people flow count information corresponding to t7, and estimates that the store clerk has left the store for the people flow count information corresponding to t8.

[0192] On the other hand, when the number of people flow count information sandwiched between estimated people flow count information for which a store clerk has entered and left the store is odd, the counting device 30 assigns the people flow count information with the highest probability of being estimated as staying to staying, and assigns the remaining people flow count information as entering or leaving so that entering and leaving are repeated. In the example of Fig. 18, the counting device 30 estimates that a store clerk has stayed for the people flow count information corresponding to t4.

[0193] In STEP 7, when there is people flow count information in which consecutive entry into the store is estimated among all the people flow count information, the counting device 30 assigns a stay to people flow count information corresponding to a later time in the time series, so that the people flow count information repeats entering and leaving the store in the time series. In the example of Fig. 18, since the people flow count information corresponding to the later time t3 is estimated to be a store clerk's stay, since the people flow count information corresponding to the later time t2 and t3 is estimated to be entering the store in succession.

[0194] Furthermore, when there is people flow count information in which store exit is estimated to occur consecutively in time series among all the people flow count information, the counting device 30 assigns the stay to the people flow count information corresponding to the earlier time in time series so that the people flow count information repeats entering and leaving the store in time series. In the example of Fig. 18, since store exit is estimated to occur consecutively at t10 and t11, the counting device 30 estimates that a store clerk has stayed for the people flow count information corresponding to the earlier time t10.

[0195] As a result of the above estimation, the counting device 30 subtracts the number of ID information read by the RFID reader 10 at the time when it is estimated that the store clerk entered the store from the number of people who entered the store included in the people flow count (store entry count). In the example of Fig. 18, the counting device 30 subtracts the number of ID information read by the RFID reader 10 at t2, t7, and t9 (in this case, "3") from the number of people who entered the store included in the people flow count (store entry count).

[0196] As described above, the counting device 30 estimates whether the store clerk whose ID information is read by the RFID reader 10 has entered, left, or stayed in the store based on the RFID information and the people flow count number, and selects whether to execute a subtraction process depending on the estimation result. This allows the counting device 30 to estimate the behavior of the store clerk using the RFID information and the people flow count information, and therefore allows the number of customers entering and leaving the store to be counted with higher accuracy.

[0197] <Embodiment 9> A description will be given of the counting system 1 according to the embodiment 9. In the following, only the parts of the counting system 1 according to the embodiment 9 that are different from the counting systems 1 according to the embodiments 1 to 8 will be described.

[0198] In the counting system 1 according to the ninth embodiment, the count display device 40 may have the configuration of the count device 30. Specifically, the count display device 40 includes a calculation device 31, a storage device 34, and a display 44. The calculation device 31 obtains the people flow count number from the people flow count sensor 20, obtains the reading result of the ID information included in the RFID tag 4 held by the store clerk from at least one RFID reader 10, and, if there is a specific time among the times when a person is detected by the people flow count sensor 20 that coincides with the time when the ID information is read by at least one RFID reader 10, performs a subtraction process to subtract the number of ID information read at the specific time from the people flow count number, thereby calculating the number of customers entering and leaving the store. Furthermore, the calculation device 31 may display the calculated number of customers on the display 44. The above-mentioned process is realized by the calculation device 31 (computer) executing a count display program. The count display program is a program in which a step of displaying the calculated number of customers on the display 44 is added to the processing of each step defined in the count program 341.

[0199] <Aspects> (1) A counting system (1) according to one embodiment includes a people flow count sensor (20) that detects people, including a first person and a second person, entering and exiting a predetermined space and calculates a people flow count number for identifying the number of people entering and exiting the predetermined space, at least one RFID reader (10) that communicates with an RFID tag (4) held by a first person located at an entrance or exit of the predetermined space and reads ID information contained in the RFID tag, and a counting device (30) that calculates the number of second people entering and exiting the predetermined space based on the people flow count number and the reading result of the at least one RFID reader. When there is a specific time that coincides with the time when ID information is read by at least one RFID reader among the times when people are detected by the people flow count sensor, the counting device calculates the number of second people entering and exiting the predetermined space by performing a subtraction process of subtracting the number of ID information read at the specific time from the people flow count number.

[0200] (2) In the counting system described in (1), the people flow count sensor calculates the number of people entering a predetermined space as a people flow count number. The counting device calculates the number of second people entering the predetermined space by performing a subtraction process.

[0201] (3) In the counting system according to the first or second aspect, the people flow count sensor calculates the number of people leaving the predetermined space as the people flow count number. The counting device calculates the number of second people leaving the predetermined space by performing a subtraction process.

[0202] (4) In the counting system according to any one of 1 to 3, at least one RFID reader is provided at least one of the predetermined space side at the entrance and the outside of the predetermined space.

[0203] (Clause 5) In the counting system described in any one of clauses 1 to 4, at least one RFID reader is provided on the designated space side of the entrance / exit, and communicates with an RFID tag located outside the designated space from the designated space side to read ID information.

[0204] (Clause 6) In the counting system described in any one of clauses 1 to 4, the at least one RFID reader includes a first RFID reader provided outside the predetermined space at the entrance and a second RFID reader provided on the predetermined space side of the entrance. The counting device selects whether to execute a subtraction process depending on which of the first RFID reader and the second RFID reader reads the ID information first.

[0205] (Clause 7) In the counting system described in any one of clauses 1 to 4, at least one RFID reader is capable of detecting a reception strength in communication with an RFID tag. The counting device selects whether or not to execute a subtraction process according to a change in the reception strength in the at least one RFID reader.

[0206] (Clause 8) In the counting system described in any one of clauses 1 to 7, the counting device calculates a time difference between a first time when ID information is read by at least one RFID reader and a second time when ID information is read by at least one RFID reader, and estimates, based on the time difference, whether the first person has entered the specified space, whether the first person has exited the specified space, or whether the first person has stayed in the specified space, and selects whether to perform a subtraction process depending on the estimation result.

[0207] (Item 9) In the counting system described in item 8, the counting device determines whether or not the estimation result is correct based on the number of people flow counted.

[0208] (Clause 10) In the counting system described in any one of clauses 1 to 9, the counting device estimates whether the first person has entered or exited a specified space, or whether the first person has stayed in the specified space, based on the reading result of at least one RFID reader and the people flow count calculated by the people flow count sensor, and selects whether to perform a subtraction process depending on the estimation result.

[0209] (Item 11) In the counting system described in any one of items 1 to 10, the predetermined space is inside a store. The first person is a store clerk. The second person is a customer.

[0210] (Item 12) In the counting system described in any one of items 1 to 11, the counting system further includes a display device (40) that displays the number of second persons entering and leaving the specified space calculated by the counting device.

[0211] (Clause 13) A counting device (10) according to one embodiment includes a calculation device (31) and a storage device (34) that stores a program (341) executed by the calculation device. The calculation device acquires a people flow count number for identifying the number of people, including a first person and a second person, entering and exiting a predetermined space from a people flow count sensor (20), acquires from at least one RFID reader (10) a reading result of ID information included in an RFID tag (4) held by a first person located at an entrance or exit of the predetermined space, and, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which ID information is read by at least one RFID reader, performs a subtraction process to subtract the number of ID information read at the specific time from the people flow count number, thereby calculating the number of second people entering and exiting the predetermined space.

[0212] (Clause 14) A counting method according to one embodiment includes, as processing executed by a computer (11), a step (S31) of acquiring from a people flow count sensor (20) a people flow count number for identifying the number of people, including a first person and a second person, entering and exiting a specified space; a step (S32) of acquiring from at least one RFID reader (10) the reading results of ID information contained in an RFID tag (4) held by a first person located at an entrance / exit of the specified space; and, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which the ID information was read by at least one RFID reader, a step (S34) of calculating the number of second people entering and exiting the specified space by executing a subtraction process of subtracting the number of ID information read at the specific time from the people flow count number.

[0213] (Clause 15) A counting program (341) according to one embodiment causes a computer (31) to execute the steps of: acquiring from a people flow count sensor (20) a people flow count number for identifying the number of people, including a first person and a second person, entering and exiting a specified space; acquiring from at least one RFID reader (10) a reading result of ID information contained in an RFID tag (4) held by a first person located at an entrance / exit of the specified space; and, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which the ID information was read by at least one RFID reader, calculating the number of second people entering and exiting the specified space by executing a subtraction process to subtract the number of ID information read at the specific time from the people flow count number (S34).

[0214] (16th Item) A count display device (40) according to one embodiment includes a calculation device (31), a storage device (34) that stores a program (341) executed by the calculation device, and a display (44). The calculation device acquires a people flow count number for identifying the number of people, including a first person and a second person, entering and exiting a predetermined space from a people flow count sensor (20), acquires from at least one RFID reader (10) a reading result of ID information included in an RFID tag (4) held by a first person located at an entrance or exit of the predetermined space, and, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which the ID information is read by at least one RFID reader, executes a subtraction process to subtract the number of ID information read at the specific time from the people flow count number, thereby calculating the number of second people entering and exiting the predetermined space, and displays the calculated number of second people on the display.

[0215] (Clause 17) A count display program according to one embodiment causes a computer (31) to execute the steps of: acquiring from a people flow count sensor a people flow count number for identifying the number of people, including a first person and a second person, entering and exiting a specified space; acquiring from at least one RFID reader (10) a reading result of ID information contained in an RFID tag (4) held by a first person located at an entrance / exit of the specified space; and, if there is a specific time among the times at which a person is detected by the people flow count sensor that coincides with the time at which the ID information was read by at least one RFID reader, calculating the number of second people entering and exiting the specified space by performing a subtraction process to subtract the number of ID information read at the specific time from the people flow count number; and displaying the number of second people calculated by the calculating step on a display (S34).

[0216] Although a number of embodiments have been described above, the features of each of these embodiments can be appropriately combined to the extent that no contradiction occurs.

[0217] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is defined by the claims, not by the description of the embodiments described above, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0218] 1 Counting system, 2 Entrance / exit, 3 Detection area, 4 RFID tag, 10, 10A, 10B, 10C, 10D, 10E RFID reader, 12, 22, 32 Memory, 13, 23, 33, 43 Communication interface, 14 Antenna, 20 People flow counting sensor, 24 Camera, 30 Counting device, 31 Calculation device, 34 Storage device, 36 Estimation model, 40 Counting display device, 44 Display, 50 Storage medium, 341 Counting program, 342 RFID information, 343 People flow counting information.

Claims

1. 1. A counting system comprising: A people flow count sensor that detects people including a first person and a second person entering and exiting a predetermined space and calculates a people flow count number to identify the number of the people entering and exiting the predetermined space; At least one RFID reader that communicates with an RFID tag held by the first person located at an entrance / exit of the predetermined space and reads ID information included in the RFID tag; a counting device that calculates the number of the second people entering and leaving the specified space based on the people flow count number and the reading result of the at least one RFID reader; The counting system wherein, when there is a specific time among the times at which the person is detected by the people flow count sensor that coincides with the time at which the ID information is read by the at least one RFID reader, the counting device calculates the number of the second people entering and leaving the specified space by performing a subtraction process to subtract the number of the ID information read at the specific time from the people flow count number.

2. The people flow count sensor calculates the number of people entering the specified space as the people flow count number, The counting system according to claim 1 , wherein the counting device calculates the number of the second persons entering the predetermined space by performing the subtraction process.

3. The people flow count sensor calculates the number of people leaving the specified space as the people flow count number, The counting system according to claim 1 , wherein the counting device calculates the number of the second persons exiting the predetermined space by executing the subtraction process.

4. The counting system according to any one of claims 1 to 3, wherein the at least one RFID reader is provided at least one of the side of the entrance on which the predetermined space is located and the outside of the predetermined space.

5. The counting system described in any one of claims 1 to 3, wherein the at least one RFID reader is provided on the specified space side of the entrance / exit, and communicates with the RFID tag located outside the specified space from the specified space side to read the ID information.

6. the at least one RFID reader includes a first RFID reader provided outside the predetermined space at the entrance and a second RFID reader provided on a side of the predetermined space at the entrance, The counting system according to any one of claims 1 to 3, wherein the counting device selects whether or not to execute the subtraction process depending on which of the first RFID reader and the second RFID reader reads the ID information first.

7. the at least one RFID reader is capable of detecting a reception strength in communication with the RFID tag; 4. The counting system according to claim 1, wherein the counting device selects whether or not to execute the subtraction process depending on a change in the reception strength in the at least one RFID reader.

8. The counting device is calculating a time difference between a first time when the ID information is read by the at least one RFID reader and a second time when the ID information is read by the at least one RFID reader after the first time; Based on the time difference, estimating any one of an entry by the first person into the predetermined space, an exit by the first person from the predetermined space, and a stay by the first person in the predetermined space; 4. The counting system according to claim 1, further comprising a step of selecting whether or not to execute the subtraction process depending on a result of the estimation.

9. The counting system according to claim 8 , wherein the counting device determines whether the estimation result is correct based on the number of people counted.

10. The counting device estimates, based on the reading result of the at least one RFID reader and the people flow count calculated by the people flow count sensor, whether the first person has entered or left the specified space, or whether the first person has stayed in the specified space; 4. The counting system according to claim 1, further comprising a step of selecting whether or not to execute the subtraction process depending on a result of the estimation.

11. The predetermined space is within a store, the first person is a store clerk; The counting system according to any one of claims 1 to 3, wherein the second person is a customer.

12. The counting system according to any one of claims 1 to 3, further comprising a display device that displays the number of said second persons entering and leaving said specified space calculated by said counting device.

13. 1. A counting device, comprising: A computing device; a storage device for storing a program executed by the arithmetic unit; The computing device includes: Obtaining a people flow count from a people flow count sensor to identify the number of people including the first person and the second person entering and exiting the predetermined space; acquiring, from at least one RFID reader, a reading result of ID information included in an RFID tag held by the first person located at an entrance / exit of the predetermined space; A counting device that, when a specific time among the times at which the person is detected by the people flow count sensor coincides with the time at which the ID information is read by the at least one RFID reader, calculates the number of the second people entering and leaving the specified space by performing a subtraction process to subtract the number of the ID information read at the specific time from the people flow count number.

14. A counting method comprising the steps of: The process that the computer executes is as follows: obtaining a people flow count from the people flow count sensor to identify a number of people, including a first person and a second person, entering and exiting a predetermined space; acquiring, from at least one RFID reader, a reading result of ID information included in an RFID tag held by the first person located at an entrance / exit of the predetermined space; and when there is a specific time among the times at which the person is detected by the people flow count sensor that coincides with the time at which the ID information is read by the at least one RFID reader, calculating the number of the second people entering and leaving the specified space by performing a subtraction process to subtract the number of the ID information read at the specific time from the people flow count number.

15. 1. A counting program comprising: On the computer, obtaining a people flow count from the people flow count sensor to identify a number of people, including a first person and a second person, entering and exiting a predetermined space; acquiring, from at least one RFID reader, a reading result of ID information included in an RFID tag held by the first person located at an entrance / exit of the predetermined space; and if there is a specific time among the times at which the person is detected by the people flow count sensor that coincides with the time at which the ID information is read by the at least one RFID reader, a subtraction process is performed to subtract the number of the ID information read at the specific time from the people flow count number, thereby calculating the number of the second people entering and leaving the specified space.

16. A count display device, A computing device; A storage device that stores a program executed by the arithmetic unit; A display. The computing device includes: Obtaining a people flow count from a people flow count sensor to identify the number of people including the first person and the second person entering and exiting the predetermined space; acquiring, from at least one RFID reader, a reading result of ID information included in an RFID tag held by the first person located at an entrance / exit of the predetermined space; When a specific time coincides with a time when the ID information is read by the at least one RFID reader among the times when the person is detected by the people flow count sensor, a subtraction process is performed to subtract the number of the ID information read at the specific time from the people flow count number, thereby calculating the number of the second people entering and leaving the specified space; A count display device that displays the calculated number of the second persons on the display.

17. A count display program, On the computer, obtaining a people flow count from the people flow count sensor to identify a number of people, including a first person and a second person, entering and exiting a predetermined space; acquiring, from at least one RFID reader, a reading result of ID information included in an RFID tag held by the first person located at an entrance / exit of the predetermined space; a step of calculating the number of the second people entering and leaving the specified space by performing a subtraction process in which, when a specific time coincides with a time when the ID information is read by the at least one RFID reader among the times when the person is detected by the people flow count sensor, the number of the ID information read at the specific time is subtracted from the people flow count number; and a step of displaying the number of the second persons calculated by the calculating step on a display.

Images