Measuring the effectiveness of advertising

Abstract

To more accurately measure the effectiveness of advertisements placed in areas accessible to people. [Solution] The advertising analysis device determines whether a person is within the range in which the advertising display device can be seen. If a person is determined to be present, it identifies the advertisement that the advertising display device displayed to the person determined to be present. It then detects whether the person entered a store after the determination, and calculates a value indicating the effectiveness of the advertisement displayed on the advertising display device based on whether the store entered is associated with the advertisement identified to the person determined to be present.

Description

Technical Field

[0001] The present invention relates to measuring advertising effectiveness. Background Art

[0002] Attempts have been made to estimate the effectiveness of advertisements placed in locations where people pass through. Currently, the effectiveness of such advertisements is evaluated by obtaining an indicator of how many people view the advertisement.

[0003] Non-Patent Documents 1 and 5 disclose recognizing people or vehicles within camera images and analyzing how much the advertisement is viewed. Non-Patent Document 2 and Patent Document 1 describe acquiring attributes of pedestrians (age, gender, etc.) based on face recognition technology, and optimizing content based on the acquired information. Non-Patent Document 3 discloses estimating the number of people who viewed the advertisement based on the population of each area, which is derived from information indicating the position of mobile phones. Non-Patent Document 4 describes measuring advertising effectiveness based on traffic volume data.

[0004] The information set forth in this Background Art section is solely for the purpose of deepening understanding of the general background of the present disclosure, and should not be construed as an admission that such information is prior art known to those skilled in the art, nor should it be construed as implying any prior art known to those skilled in the art. Prior Art Documents Patent Documents

[0005] Patent Document 1 US Patent No. 11682037 Specification Non-Patent Documents

[0006] Non-Patent Document 1 Quividi, “Audience & traffic measurement,” https: / / quividi.com / , accessed: 2024-08-25 [Non-Patent Document 2] NEC, “Targeted Advertising Signage & Welcome Signage,” https: / / jpn.nec.com / d_signage / option / face_signage / index.html, accessed: 2024-09-12. [Non-Patent Document 3] NTT DOCOMO, INC., “Mobile spatial statistics (mobile kukan tokei),” https: / / www.docomo.ne.jp / corporate / disclosure / mobile_spatial_statistics / index.html, accessed: 2024-08-11. [Non-Patent Document 4] Arbeitsgemeinschaft Media-Analyse (agma), “ma out of home,” https: / / www.agma-mmc.de / media-analyse / ma-out-of-home, accessed: 2024- 08-25 [Non-Patent Document 5] AdMobilize, “People counting & audience measurement,” https: / / admobilize.com / people-counting-audience-measurement / , accessed: 2024-08-25. [Overview of the initiative] [Problems that the invention aims to solve]

[0007] Advertising exists to encourage actions such as purchasing. Traditionally, the effectiveness of advertising has been estimated based on the assumption that there is a uniform relationship between people seeing advertisements in places they pass by and making purchases. Therefore, it has not been possible to accurately estimate whether individual advertisements lead to purchases.

[0008] The purpose of this disclosure is to provide a technology for more accurately measuring the effectiveness of advertisements located in areas accessible to people. [Means for solving the problem]

[0009] The advertising analysis device includes an advertising identification means, a store entry detection means, and an effectiveness calculation means. The advertising identification means determines whether a person is present within a range in which the advertising display device can be seen, and if it is determined that a person is present, it identifies the advertisement that the advertising display device displayed to the person determined to be present. The store entry detection means detects that the person determined to be present entered a store after the determination. The effectiveness calculation means calculates a value indicating the effectiveness of the advertisement displayed on the advertising display device based on whether it is detected that the person determined to be present entered a store associated with the identified advertisement.

[0010] An advertising analysis method comprising: an advertising identification means determining whether a person is present within a range in which an advertising display device can be seen, and if such person is determined to be present, identifying the advertisement displayed by the advertising display device to the person determined to be present; a store entry detection means detecting that the person determined to be present entered a store after the determination; and an effect calculation means calculating a value indicating the effect of the advertisement displayed on the advertising display device based on whether the store that was detected to have been entered is associated with the advertisement identified to the person determined to be present. [Brief explanation of the drawing]

[0011] Features, aspects, and advantages of embodiments of this disclosure are shown below with reference to the accompanying drawings. The same reference numerals in the drawings represent the same elements. [Figure 1]It is a diagram illustrating elements related to the advertisement analysis system according to the first embodiment. [Figure 2] It is a diagram showing an example configuration of the advertisement analysis system. [Figure 3] It is a block diagram showing functions implemented by the advertisement analysis system. [Figure 4] It is a flow diagram showing an example process of the advertisement analysis system. [Figure 5] It is a diagram showing an example of information stored in a display device database. [Figure 6] It is a diagram showing an example of information stored in a displayed advertisement database. [Figure 7] It is a diagram showing an example of information stored in a store database. [Figure 8] It is a diagram showing an example of information stored in an attention history. [Figure 9] It is a diagram showing an example of information stored in a store entry / exit history. [Figure 10] It is a flow diagram showing an example of a process for calculating advertisement effectiveness. [Figure 11] It is a diagram showing an example of information stored in an in-store behavior history. [Figure 12] It is a diagram illustrating elements related to the advertisement analysis system according to the second embodiment. Mode for Carrying Out the Invention

[0012] Hereinafter, exemplary embodiments will be described in detail with reference to the accompanying drawings. While the present disclosure presents the drawings and description, it is not intended to be exhaustive, nor is it intended to limit the embodiments to the precise forms disclosed. Modifications and variations are possible in light of the present disclosure and may be obtained from the practice of the embodiments. Furthermore, one or more features or components of one embodiment may be incorporated into, or combined with, another embodiment (or may be incorporated into, or combined with, one or more features of another embodiment). Additionally, the flowcharts and descriptions relating to operations presented below relate to at least one embodiment of the present disclosure. It should be noted, however, that other embodiments may be devised that do not completely correspond to the flowcharts and their descriptions. It should also be understood that in other embodiments, one or more operations may be omitted (at least in part), one or more operations may be added, and one or more operations may be performed concurrently.

[0013] It is clear that the systems, methods, or both described herein may be implemented in various forms of hardware, software, or a combination of hardware and software. Furthermore, implementations should not be limited by the actual specific control hardware or software code used to implement these systems, methods, or both. Therefore, in the present specification, the operations and behavior of the systems, methods, or both are described without reference to specific software code. It should also be understood that based on the description herein, software and hardware may be designed to implement the systems, methods, or both.

[0014] Even if a particular combination of features is described in the claims, disclosed herein, or both, such particular combination is not intended to limit the disclosure of embodiments. Furthermore, many of these features may be combined in ways not specifically described in the claims, or not disclosed herein, or both. Also, even if a dependent claim is directly dependent on only one claim, the disclosure of embodiments may include combinations of that dependent claim and each of the claims described in the claims.

[0015] Elements, actions, or instructions used herein should not be construed as important or essential unless explicitly stated otherwise. Furthermore, nouns not referred to as plural (where the articles "a" and "an" are used with nouns in English) are intended to include cases where the noun is one or more in number and can be used synonymously with "one or more." In addition, terms such as "have," "possess," "include," and "contain" used herein are intended to be open-ended terms, meaning they may have or contain other elements, etc. Furthermore, the phrase "based on" is intended to mean "at least partially based on" unless otherwise specified. In addition, expressions such as "[A] and [B]," "[A] and / or [B]," or "[A] or [B]" should be understood to include A only, B only, or both A and B.

[0016] [First Embodiment] Figure 1 is a diagram illustrating elements related to an advertising system according to the first embodiment. The advertising system calculates a value indicating the effectiveness of the advertisement, more specifically, a value indicating the strength of the correlation between viewing the advertising display device 200 and taking action toward the store 400. The advertising system may further manage the display of advertisements on the advertising display device 200. Another device may manage the display of advertisements.

[0017] The advertising system may include an advertising management device 100, one or more advertising display devices 200, and a three-dimensional sensor 300. Store 400 is equipped with a store management device 410, a cash register 420, an entrance camera 430, a register camera 440, and an exit camera 450. The cash register 420, entrance camera 430, register camera 440, and exit camera 450 may be connected to the store management device 410. The advertising system analyzes the effectiveness of the advertisements using the information of the advertisements displayed on the advertising display devices 200 and the information acquired from the three-dimensional sensor 300. The advertising display devices 200 and the three-dimensional sensor 300 may be installed, for example, in a shopping mall or outlet mall that includes multiple stores 400.

[0018] Figure 2 shows an embodiment of the advertising management device 100. As shown in Figure 2, the device 100 includes a processor 110, a memory 120, a storage unit 130, an input unit 140, an output unit 150, a communication interface 160, and a bus 170.

[0019] As used herein, the processor 110 means any type of arithmetic circuit that may comprise hardware and software elements. The processor 110 may be implemented as a multi-core processor, a single-core processor, a combination of one or more multi-core processors, a combination of one or more single-core processors, or a combination of one or more multi-core processors and one or more single-core processors, or as a distributed processing system, etc. The processor 110 may also be a central processing unit (CPU), a graphics processing unit (GPU), an advanced processing unit (APU), an application-specific integrated circuit (ASIC), or any other type of processing unit.

[0020] Memory 120 includes a non-temporary computer-readable medium. Memory 120 includes random access memory (RAM), read-only memory (ROM), other types of dynamic or static storage devices (e.g., flash memory, magnetic memory, optical memory, or at least two of these), or at least two of these, for storing information, instructions, or both used by the processor 110. Memory 120 includes machine-readable instructions that the processor 110 can execute. When these machine-readable instructions are executed by the processor 110, they cause the processor 110 to perform one or more method steps of the embodiments described above.

[0021] The storage unit 130 stores information, software, or both related to the operation and use of the device 100. The storage unit 130 may include, for example, a hard disk (e.g., a magnetic disk, optical disk, magneto-optical disk, solid-state disk, or at least two of these), a CD (compact disc), a DVD (digital versatile disc), a floppy disk, a cartridge, a magnetic tape, other types of non-temporary computer-readable media, or at least two of these, along with a corresponding drive.

[0022] The input unit 140 is configured to receive information such as user input. The input unit 140 may include, but is not limited to, a touchscreen display, keyboard, keypad, mouse, buttons, switches, microphone, or at least two of these. Furthermore, or alternatively, the input unit 140 may include sensors for detecting information (e.g., GPS (global positioning system), accelerometer, gyroscope, actuator, or at least two of these).

[0023] The output unit 150 is configured to provide output information from the device 100. The output unit 150 may, but is not limited to, a display, a speaker, a command device for an external device, one or more light-emitting diodes (LEDs), or a combination of at least two of these.

[0024] The communication interface 160 is an interface that provides communication connections with other devices, such as external devices and internal devices. The connection via the communication interface 160 may be a wired connection, a wireless connection, or a combination of wired and wireless connections, and may be a direct or indirect connection via a communication network existing between device 100 and other devices. In other words, the specifications of the communication interface 160 are not limited.

[0025] Bus 170 interconnects the processor 110, memory 120, storage unit 130, input unit 140, output unit 150, and communication interface 160 of device 100. Bus 170 may include wired or wireless interconnections.

[0026] The number and arrangement of components shown in Figure 2 are provided as an example. In practice, the device 100 may include additional components, fewer components, different components, or components in different arrangements compared to the components shown in Figure 2. Furthermore, or alternatively, a set of components of the device 100 (e.g., one or more components) may perform one or more functions as described above, which are performed by other sets of components of the device 100. Furthermore, one or more method steps described in any embodiment may be performed using multiple devices 100 communicating with one another.

[0027] The advertising management device 100 is a platform including at least one or more processors 110, memory 120, storage unit 130, communication interface 160, and bus 170, and may reside on a virtual server or container-based platform. The machine-readable instructions executed to implement the processing on the advertising management device 100, as described below, may be executed on these platforms. Alternatively, a system consisting of one or more computers may provide the same functionality as the advertising management device 100.

[0028] The advertising display device 200 constitutes a so-called digital signage system. The advertising display device 200 acquires information on the images (including video) of the advertisements to be displayed from the advertising management device 100 and displays the images of the advertisements. It also switches the displayed advertisements based on the control of the advertising management device 100.

[0029] The 3D sensor 300 is, for example, a LiDAR (light detection and ranging) sensor, which acquires the 3D position and shape of surrounding objects. The 3D sensor 300 may be any other type of sensor, such as a stereo camera, as long as it has the capability to detect the 3D position of a person. The 3D sensor 300 may output, for example, time-series data of a point cloud.

[0030] The entrance camera 430 faces outwards from the entrance of the store 400 and photographs customers entering the store 400. The cash register 420 processes payments for customers' goods within the store 400. The register camera 440 faces near the cash register 420 and photographs customers purchasing goods in the store. The exit camera 450 faces near the exit of the store 400 and photographs customers leaving the store. The entrance camera 430, register camera 440, and exit camera 450 may be adjusted to photograph the faces of the customers they wish to photograph. The store management device 410 acquires images from the entrance camera 430, register camera 440, and exit camera 450, and purchase information from the cash register 420. The store management device 410 may analyze this information or transmit this information or the analysis results to an external system, including the advertising management device 100. The transmission and reception of information may be carried out via a cloud system such as storage on the internet.

[0031] Next, the processing of the advertising system will be explained. Figure 3 is a block diagram showing the functions realized by the advertising system. Unless otherwise specified, Figure 3 shows the functions realized by the advertising management device 100. The advertising system includes a sensor data acquisition unit 510, a tracking unit 520, a focus advertisement determination unit 530, a store entry detection unit 540, an advertising effectiveness calculation unit 550, a display device database 610, a displayed advertisement database 620, a store database 630, a focus history 640, a store entry / exit history 650, and a store behavior history 660. These are realized by one or more processors 110 executing programs (more precisely, machine-readable instructions) stored in memory 120 and controlling the storage unit 130, communication interface 160, etc. The display device database 610, the displayed advertisement database 620, the store database 630, the focus history 640, the store entry / exit history 650, and the store behavior history 660 are mainly composed of the storage unit 130. These may be implemented by the processor 110 executing a database management system program stored in memory 120 and reading and writing data to the storage unit 130. Furthermore, these may be located on servers other than the advertising management device 100.

[0032] The display device database 610 stores information about the advertising display devices 200. The information about the advertising display devices 200 may include information indicating the location and visible range for each of the one or more advertising display devices 200. The displayed advertisement database 620 stores information about the advertisements displayed on the advertising display devices 200. The advertisement information may include information identifying the advertisement, the period during which the advertisement was displayed, information identifying the advertising display device 200 that displayed the advertisement, and information identifying the store 400 associated with the advertisement. The store database 630 stores information about the store 400. The information about the store 400 may include information indicating the location, orientation, and size of the store's entrance, and information indicating the location, orientation, and size of the store's exit.

[0033] The sensor data acquisition unit 510 acquires 3D data measured by the 3D sensor 300 from the 3D sensor 300.

[0034] Based on the acquired 3D data, the tracking unit 520 recognizes a person within the range measurable by the 3D sensor 300 and tracks their movement.

[0035] The attention ad determination unit 530 determines, based on the recognition result of the tracking unit 520, whether a person is within sight of the ad display device 200. The attention ad determination unit 530 may further determine whether a person is paying attention to the ad display device 200 within sight of the ad display device 200. The attention ad determination unit 530 then identifies the ad displayed by the ad display device 200 to the person who is determined to be present. Information indicating the identified ad may be stored in the attention history 640 in association with information identifying the person and the time of their presence.

[0036] The store entry detection unit 540 detects that a person determined to be within sight of the advertising display device 200 has entered the store 400 after that determination. The store entry detection unit 540 may detect that the person has entered one of several stores 400 based on tracking of the person based on the output of the 3D sensor. If the store entry detection unit 540 detects that the person has entered the store 400, it may store information identifying the person, the time of entry, and information identifying the store 400 in association with each other in the store entry / exit history 650. The store entry detection unit 540 may detect that a person determined to be within sight of the advertising display device 200 has reached the entrance of the store 400 after that determination. The store entry detection unit 540 may detect that the person has reached the entrance of the store 400 as having entered the store 400.

[0037] The advertising effectiveness calculation unit 550 calculates a value indicating the effectiveness of the advertisement displayed on the advertising display device 200 based on whether the determined person entered the store 400 associated with the advertisement identified for that person. The advertising effectiveness calculation unit 550 may further calculate a value indicating the effectiveness of the advertisement based on the determined person's purchase record at the store 400. The store activity history 660 stores the purchase record of people at the store 400. Here, the advertising effectiveness calculation unit 550 may associate the detected person with the customer identified by the output of the entrance camera 430 based on the time when it was detected that the determined person entered the store 400. Then, the advertising effectiveness calculation unit 550 calculates a value indicating the effectiveness of the advertisement displayed on the advertising display device 200 based on the purchase record of the customer associated with the detected person.

[0038] The advertising effectiveness calculation unit 550 may further associate a person detected entering the store 400 with a person determined to have exited the store 400 through its exit, based on the output of cameras installed at the store's entrance and exit. This store 400 may be different from the store 400 associated with the specific advertisement for that person. Furthermore, the advertising effectiveness calculation unit 550 may calculate a value indicating the effectiveness of the advertisement displayed on the advertising display device 200 based on whether or not the person determined to have exited the store 400 through its exit subsequently entered the store 400 associated with the specific advertisement.

[0039] Next, the processing will be explained in more detail. Figure 4 is a flowchart showing an example of the processing of the advertising management device 100. Here, it is assumed that time-series data measured by the 3D sensor 300 has been acquired in advance and stored in the storage unit 130, and that information on advertisements displayed during the measured period has been stored in the display advertisement database 620. The initial value of the target time may be the time range in which the data from the 3D sensor 300 exists and the beginning of the time range for calculating the advertising effect. The advertising management device 100 may also acquire information on advertisements displayed on the 3D sensor 300 and the advertising display device 200 in real time and execute the processing shown in Figure 4.

[0040] First, the sensor data acquisition unit 510 acquires data from the 3D sensor 300 during the target time period (S101). The tracking unit 520 performs person recognition and tracking based on the acquired data (S102). The method for recognizing a person from the data of the 3D sensor 300 and the method for tracking the recognized person are publicly known, so a detailed explanation is omitted. A tracking ID is assigned to the person recognized at this stage. If the tracking determines that the person is the same as a person previously recognized, the same tracking ID is used and it is used as one of the pieces of information to identify the person.

[0041] The tracking unit 520 may acquire the position and the direction the body is facing (direction of gaze) for each recognized person. Here, the position and the direction the body is facing may be converted into a coordinate system of a virtual space (hereinafter referred to as the "virtual coordinate system") in which the store 400, the 3D sensor 300, and the advertising display device 200 are placed at positions corresponding to the real space.

[0042] The attention ad determination unit 530 determines whether the recognized person is paying attention to the advertising display device 200 (S103). More specifically, the attention ad determination unit 530 extracts recognized people who are within the visible range for each of the advertising display devices 200. The attention ad determination unit 530 may further determine that the person is paying attention to the advertising display device 200 if the difference between the orientation of the extracted person towards the advertising display device 200 and the direction of their gaze falls within a predetermined range (for example, ±120°).

[0043] Figure 5 shows an example of information stored in the display device database 610. The display device database 610 stores data for each advertising display device 200. The data for one advertising display device 200 includes the display device ID, coordinates, and visible range. The display device ID is information that identifies the advertising display device 200, and the coordinates indicate the position of the advertising display device 200 in a virtual coordinate system. The visible range is information that indicates the area in the virtual coordinate system where the advertising display device 200 can be seen. In the example in Figure 5, this area is shown by the angle and distance from the advertising display device 200. The attention advertisement determination unit 530 may, for example, determine that a recognized person is within the visible range and extract that person if the angle and distance of the recognized person to the advertising display device 200 are within the visible range.

[0044] If the recognized person is paying attention to the advertisement display device 200 (Y in S103), the attention advertisement determination unit 530 stores the advertisement ID of the advertisement displayed during the target time and the recognized person in the attention history 640 (S104).

[0045] This process uses the display advertisement database 620. Figure 6 shows an example of the information stored in the display advertisement database 620. The display advertisement database 620 stores information about advertisements displayed on the advertisement display device 200 (advertisement information). In the example in Figure 6, the advertisement information includes the display device ID, start time, end time, advertisement ID, and store ID. The advertisement ID and store ID are information that identifies the advertisement and store 400, respectively. The start time and end time are the time when the advertisement display device 200 starts and ends displaying the advertisement. Note that the store ID is not stored in the display advertisement database 620 and may be associated separately with the advertisement ID.

[0046] In S104, the attention advertisement determination unit 530 acquires advertisement information that has the display device ID of the attention advertisement display device 200 and whose target time is between the start time and the end time, obtains the advertisement ID from that advertisement information, and stores the attention information including these in the attention history 640.

[0047] Figure 7 shows an example of information stored in the attention history 640. The attention information includes the attention time, tracking ID, and advertisement ID. The attention time is set to the time period during which processing is taking place. Note that the attention advertisement determination unit 530 does not need to store new attention information if attention information already exists for the same tracking ID and advertisement ID. Only the oldest attention information for the same tracking ID and advertisement ID may exist.

[0048] If the recognized person is not paying attention to the advertising display device 200 (N in S103), S104 is skipped. If there are multiple recognized people, the processes in S103 and S104 are performed for each person.

[0049] The store entry detection unit 540 also determines whether there is a person entering the store 400 (S105). This determination may be made based on whether or not a recognized person is present at the entrance of the store 400.

[0050] Information stored in the store database 630 is used for this determination. Figure 8 shows an example of the information stored in the store database 630. The information for store 400 includes the store ID, entrance information, and exit information. The entrance information includes information indicating the position, orientation, and width of the entrance to store 400 in a virtual coordinate system, and the exit information includes information indicating the position, orientation, and width of the exit to store 400 in a virtual coordinate system. Here, the shapes of the entrance and exit are approximated as line segments, and the orientation of the entrance and exit corresponds to the normal of the line segment, corresponding to the direction in which customers enter and exit.

[0051] The store entry detection unit 540 may determine that a person is entering the store 400 if the distance between the recognized person's location and the entrance of the store 400 is below a threshold, or it may determine that a person is entering the store 400 if that distance was previously recognized as below a threshold but the person is not currently recognized.

[0052] If it is determined that someone is entering store 400 (Y in S105), the store ID, information indicating the person entering store 400, and the time of entry are stored in the entry / exit history 650 (S106).

[0053] Figure 9 shows an example of information stored in the entry / exit history 650. The entry / exit history 650 stores at least one of the entry time and exit time, associated with a combination of tracking ID and store ID. The entry time may be the target time at the time of storage. The combination of tracking ID and store ID characterizes the person and the store 400. The exit time is the target time when it is determined that the person with that tracking ID has left the store. Although not shown in Figure 4, for example, in S102, if the distance between the person who is first recognized and the exit is below a threshold, the tracking unit 520 may store the store ID, the tracking ID indicating the person leaving the store 400, and the exit time (target time) in the entry / exit history 650.

[0054] If it is determined that there are no people entering store 400 (N in S105), S106 is skipped. Processing S105 and S106 is performed for each person to be targeted (including multiple recognized people).

[0055] Then, if the termination condition is not met (N in S107), the sensor data acquisition unit 510 advances the target time (S108) and repeats the processing from S101 onwards. The termination condition may be that the target time is the end time of the advertising effect measurement.

[0056] If the termination condition is met (Y in S107), the advertising effectiveness calculation unit 550 calculates a value indicating the advertising effectiveness based on the attention history 640 and the store entry / exit history 650 (S109).

[0057] Figure 10 is a flowchart showing an example of the process for calculating advertising effectiveness. The process shown in Figure 10 is performed for a target advertisement that has a certain advertising ID. The advertising effectiveness calculation unit 550 may obtain the store ID of the store 400 corresponding to the target advertisement in advance from the displayed advertisement database 620 or the like. When obtaining statistical values ​​for multiple advertisements, the process shown in Figure 10 can be performed for each advertisement, and statistical processing can be performed from the results for each advertisement as needed.

[0058] The advertising effectiveness calculation unit 550 identifies people who entered the store 400 associated with the advertisement after viewing it. This is mainly done in the processes S201, S204, S206, and S207. In S205 and S209, the advertising effectiveness calculation unit 550 calculates statistical values ​​of the advertising effectiveness based on this identification.

[0059] In S201, the advertising effectiveness calculation unit 550 extracts attention information from the attention history 640 of people who viewed the target advertisement. More specifically, the advertising effectiveness calculation unit 550 extracts attention information, including the tracking ID and the time of viewing, which is associated with the advertising ID of the target advertisement. The number of attention information entries extracted may be one or more.

[0060] The advertising effectiveness calculation unit 550 calculates attention statistics based on the extracted attention information (S202). The attention statistics may simply be a count of the number of people. This indicates the number of times the target advertisement was noticed or the number of people who noticed the advertisement.

[0061] Next, the advertising effectiveness calculation unit 550 selects one of the extracted and unprocessed pieces of attention information (S203). The advertising effectiveness calculation unit 550 determines whether there is a record in the entry / exit history 650 of someone entering the target store 400 after the time of attention (S204). More specifically, the advertising effectiveness calculation unit 550 searches the entry / exit history 650 for information that has the store ID of the target store 400 and the tracking ID included in the selected piece of attention information, and whose entry time is newer than the time of attention. If such information exists, the advertising effectiveness calculation unit 550 determines that there is a record of someone entering the store.

[0062] If it is determined that there is a record of entering the store (Y in S204), the advertising effectiveness calculation unit 550 calculates an effectiveness element value based on the result of comparing the store behavior history 660 with the record of entering the store, and stores it in the effectiveness list (S205).

[0063] Figure 11 shows an example of information stored in the store activity history 660. The store activity history 660 stores activity information for each customer who enters a store 400. In the example in Figure 11, the activity information includes the store ID, customer ID, entry time, exit time, and purchase record. In the example in Figure 11, activity information is acquired in advance from the store management devices 410 of multiple stores 400 and stored in the store activity history 660. If activity information is stored separately for each store 400, the activity information included in the store activity history 660 does not need to include the store ID. The purchase record is information indicating whether or not a customer who entered store 400 purchased goods, and may include not only whether or not a purchase occurred but also information on the purchase amount.

[0064] The store management device 410 acquires facial feature quantities of customers from images captured by the entrance camera 430, the cash register camera 440, and the exit camera 450, and identifies customers based on these feature quantities. A customer ID is uniquely assigned to each person identified as the same customer. Alternatively, customers may be identified using camera-based tracking technology. Behavioral information may further include facial feature quantities. Behavioral information may be stored for each action, including entering the store, leaving the store, and making a purchase. In this case, it may include the type of action, the time of the action, and information that identifies the customer (customer ID or feature quantities).

[0065] In S205, the advertising effectiveness calculation unit 550 searches the store behavior history 660 for customer behavior information that includes the time of entry, which corresponds to the time of entry obtained from the store entry and exit history 650 at the target store 400. The behavior information may include the direction in which the customer entered the store 400. In this case, the advertising effectiveness calculation unit 550 may search for customer behavior information that corresponds not only to the time of entry, but also to the direction of entry, which corresponds to the position obtained from the output of the 3D sensor 300 at the time of entry.

[0066] When corresponding customer behavior information is found, the advertising effectiveness calculation unit 550 calculates an effectiveness element value based on the purchase record included in that behavior information. For example, the advertising effectiveness calculation unit 550 may set a first value for the effectiveness element value if there is a purchase record, and a second value smaller than the first value for the effectiveness element value if there is a purchase record. The advertising effectiveness calculation unit 550 may also simply increment the count if there is a record of the customer entering the store.

[0067] On the other hand, if it is determined that there is no record of entry into the store (N in S204), S206 and S207 determine whether the user entered the target store 400 via another store. This process addresses the issue that, in the case of a low-resolution 3D sensor 300, if the user leaves the sensing range and then re-enters it, a different tracking ID will be assigned.

[0068] In S206, the system determines whether there is a record in the entry / exit history 650 of the selected attention information entering another store 400 after the attention time (S204). More specifically, the advertising effectiveness calculation unit 550 searches the entry / exit history 650 for information that has a tracking ID included in the selected attention information and whose entry time is newer than the attention time. If such information exists, the advertising effectiveness calculation unit 550 determines that there is a record of the person entering another store 400.

[0069] If it is determined that there is a record of the customer entering another store 400 (Y in S206), the advertising effectiveness calculation unit 550 obtains other tracking IDs of the same person based on the departure time from the other store 400, and determines whether there is a record of the customer entering the target store 400 in the entry / exit history 650 for that tracking ID (S207). More specifically, the advertising effectiveness calculation unit 550 obtains the behavior information of the corresponding customer of the other store 400 from the store behavior history 660 based on the entry time, and also obtains the departure time of that customer from that behavior information. Then, from the entry / exit history 650, it obtains a tracking ID that is stored in association with the information having a departure time corresponding to the departure time obtained for that store 400. Then, the same processing as in S204 is performed on that tracking ID.

[0070] If there is a record of entering the store (Y in S207), the process in S205 is executed. If there is no record of entering any other store 400 (N in S206), or if there is no record of entering the target store 400 (N in S207), the process in S205 is skipped.

[0071] Once the above processing is complete, the advertising effectiveness calculation unit 550 determines whether there is any unprocessed attention information (S208). If there is any unprocessed attention information (Y in S208), the processing from S203 onwards is repeated.

[0072] If there is no unprocessed information of interest (N in S208), the advertising effectiveness calculation unit 550 calculates a statistical value of advertising effectiveness based on the effect element values ​​stored in the effectiveness list (S209). The statistical value of advertising effectiveness may simply be the sum of the effect element values, or it may be the sum divided by the statistical value of interest.

[0073] These processes allow the effectiveness of an advertisement to be calculated based on the actions taken at the corresponding store 400 after paying attention to an advertisement displayed on the advertisement display device 200. Note that matching with the store behavior history 660 in S205 is not required. In this case, a value indicating the strength of the relationship between attention to the advertisement and entering the store 400 is calculated as a value indicating the advertising effectiveness.

[0074] [Second Embodiment] The second embodiment differs from the first embodiment mainly in that it calculates a value indicating advertising effectiveness without using the 3D sensor 300. The differences from the first embodiment will be explained below.

[0075] Figure 12 is a diagram illustrating elements related to the advertising system according to the second embodiment. The advertising system may include an advertising management device 100, one or more advertising display devices 200. Each of the one or more advertising display devices 200 is equipped with a front camera 210 that captures an area within view of the advertising display device 200. The store 400 is equipped with a store management device 410, a cash register 420, an entrance camera 430, a register camera 440, and an exit camera 450. The advertising system analyzes the effectiveness of the advertisement using the information of the advertisement displayed on the advertising display device 200, images of people of interest captured by the front camera 210, and images of people entering the store captured by the entrance camera 420.

[0076] In the second embodiment, tracking technology and tracking IDs are not required for person identification. Instead, known facial recognition technology may be applied to images captured by any of the front camera 210, entrance camera 430, cashier camera 440, or exit camera 450, and people may be identified using facial features obtained from the person's face.

[0077] The functions realized by the advertising system will be explained by describing the differences from the first embodiment. In the second embodiment, unlike the functions shown in Figure 3, the tracking unit 520 does not need to be present. The following will mainly describe the functions that differ in processing.

[0078] The attention ad determination unit 530 determines whether a person is present within the range from which the advertisement display device 200 can be seen, based on the image output from the front camera 210 located on the advertisement display device 200. For example, the attention ad determination unit 530 determines whether a person is present by detecting a person from the image based on known facial recognition technology. The attention ad determination unit 530 also generates feature quantities of the detected person based on known facial recognition technology. The attention ad determination unit 530 identifies the advertisement that the advertisement display device 200 displayed to the detected person, and stores the information indicating the identified advertisement, the person's feature quantities, and the time period in which the person was determined to be present in the attention history 640. In the second embodiment, feature quantities are used as information to identify a person instead of an identification ID or customer ID.

[0079] The store entry detection unit 540 detects that a person who is determined to be within sight of the advertising display device 200 based on the image output from the entrance camera 430 of the store 400 has entered the store after that determination. More specifically, the store entry detection unit 540 may acquire the person's feature quantities generated by applying facial recognition technology to the image output from the entrance camera 430, and then find the attention information stored in the attention history 640, which is associated with the feature quantities corresponding to the acquired feature quantities and the time of attention prior to the time the image was taken. Note that the information in the store database 630 shown in Figure 5 is not necessary. If the store entry detection unit 540 detects that a person has entered the store 400, it may store the feature quantities that identify the person, the time of entry, and the information that identifies the store 400 in the entry / exit history 650, associating them with each other.

[0080] The attention ad determination unit 530 performs the following different processing from that shown in Figure 10: Instead of a tracking ID, it obtains feature quantities, and the feature quantities in the attention history 640, the feature quantities in the store entry / exit history 650, and the feature quantities stored in the store behavior history 660 are associated and matched based on whether they indicate the same person. Note that since the feature quantities do not change even when visiting other stores, the processing corresponding to S206 and S207 is not performed.

[0081] Thus, even without using the 3D sensor 300, it is possible to calculate values ​​that indicate the effectiveness of advertising.

[0082] Furthermore, the 3D sensor 300 and the front camera 210 may be combined. The accuracy of the store entry detection unit 540 can be improved by confirming, not only by tracking ID but also by feature quantities, whether the person who entered the store is the same person who was determined to be within the range where the advertising display device 200 can be seen.

[0083] [Note] This specification discloses a variety of technical ideas, including the following disclosures: (1) An advertising analysis device comprising: an advertising identification means for determining whether a person is present within a range in which an advertising display device can be seen, and if such person is determined to be present, an advertising display device for determining whether an advertising display device has displayed an advertising to the person determined to be present; an entry detection means for detecting whether the person determined to be present has entered a store after the determination; and an effect calculation means for calculating a value indicating the effect of an advertising display device based on whether or not it has been detected that the person determined to be present has entered a store associated with the identified advertising.

[0084] (2)(1) The advertising analysis device wherein the advertising identification means determines, based on the output of a three-dimensional sensor, whether a person is present within a range in which the advertising display device can be seen, and the store entry detection means detects, based on tracking the position of the determined person based on the output of the three-dimensional sensor, that the determined person entered the store after the determination.

[0085] (3)(2) The advertising analysis device in which the three-dimensional sensor is a LiDAR (Light Detection And Ranging) sensor.

[0086] (4)(2) or (3), the effect calculation means calculates a value indicating the effect of the advertisement displayed on the advertisement display device based on whether or not it is detected that the determined person entered the store associated with the identified advertisement and the purchase record of the determined person at the store.

[0087] (5)(4) The store entry detection means detects, based on the tracking, that the person who was determined enters the store after the determination; the effect calculation means associates the detected person with a customer identified by the output of a camera installed at the entrance, based on the time of detection; and the effect calculation means calculates a value indicating the effect of the advertisement displayed on the advertisement display device based on the purchase record of the customer associated with the detected person.

[0088] (6)(5) The store entry detection means detects that the person determined entered another store after the determination based on tracking of the location of the person determined based on the output of the three-dimensional sensor; the effect calculation means associates the detected person with a person determined to have left the other store from the exit based on the output of the three-dimensional sensor, based on the output of cameras installed at the entrance and exit of the other store; the store entry detection means detects that a person determined to have left the other store from the exit enters the store after the determination, based on the tracking; and the effect calculation means calculates a value indicating the effect of the advertisement displayed on the advertisement display device based on whether or not it is detected that a person determined to have left the other store from the exit entered a store associated with the identified advertisement after the other store.

[0089] (7)(1) The advertising identification means determines whether a person is within the range in which the advertising display device can be seen by using an image output from a first camera that photographs a location in which the advertising display device can be seen, and the store entry detection means detects whether the person determined entered the store after the determination by using an image output from a second camera installed at the entrance of the store.

[0090] (8) An advertising analysis method comprising: an advertising identification means determining whether a person is present within a range in which an advertising display device can be seen, and if it is determined that a person is present, identifying an advertisement that the advertising display device displayed to the person determined to be present; a store entry detection means detecting that the person determined to be present entered a store after the determination; and an effect calculation means calculating a value indicating the effect of the advertisement displayed on the advertising display device based on whether the store that was detected to have been entered is associated with the advertisement identified to the person determined to be present. [Explanation of symbols]

[0091] 100 Advertising management device, 110 Processor, 120 Memory, 130 Storage unit, 140 Input unit, 150 Output unit, 160 Communication interface, 170 Bus, 200 Advertising display device, 210 Front camera, 300 3D sensor, 400 Store, 410 Store management device, 420 Cash register, 430 Entrance camera, 440 Register camera, 450 Exit camera, 510 Sensor data acquisition unit, 520 Tracking unit, 530 Attention advertisement determination unit, 540 Store entry detection unit, 550 Advertising effectiveness calculation unit, 610 Display device database, 620 Display advertisement database, 630 Store database, 640 Attention history, 650 Store entry / exit history, 660 Store behavior history.

Claims

1. A process to determine if a person is present within the range where the advertising display device can be seen, If it is determined that the person exists, the process of identifying the advertisement displayed by the advertisement display device to the person determined to exist, A process to identify the first store corresponding to the advertisement from among a predetermined number of stores, A process to detect that the person determined entered the first store after the determination, A process to calculate a value indicating the effectiveness of the advertisement displayed on the advertisement display device based on whether or not it is detected that the person determined to have entered the first store, An advertising analysis device that performs this task.

2. The process of determining whether a person is present within the range in which the advertising display device can be seen is performed based on the output of the 3D sensor, Based on the data indicating the time-series position of the determined person based on the output of the three-dimensional sensor, a process is further executed to detect the movement of the determined person from the range in which the advertisement display device can be seen. Based on the detection that the person determined to have moved to the first store, it is detected that the person determined to have entered the first store. The advertising analysis apparatus according to claim 1.

3. The aforementioned three-dimensional sensor is a LiDAR (Light Detection And Ranging) sensor. The advertising analysis apparatus according to claim 2.

4. Based on the purchase records of the person determined in the first store, a value indicating the effectiveness of the advertisement displayed on the advertisement display device is calculated. The advertising analysis apparatus according to claim 1.

5. Based on the detection that the person determined has moved to the first store and a first photograph showing a customer in the first store, it is detected that the person determined has entered the first store. The advertising analysis apparatus according to claim 2.

6. Further process is performed to identify a customer who corresponds to the person determined to be the customer from among the customers of the first store captured in the first image, Based on the purchase records of identified customers at the first store, a value indicating the effectiveness of the advertisement displayed on the advertisement display device is calculated. The advertising analysis apparatus according to claim 5.

7. When it is detected that the person determined has moved to a second store different from the first store, the process of identifying a customer who corresponds to the person determined from among the customers of the second store captured in the second image is further performed. Further processing is performed to detect whether the customer of the second store identified by the second image is also visible in the first image taken of the first store, and based on this detection, it is detected that the person determined to have entered the first store. The advertising analysis apparatus according to claim 2.

8. A step of determining whether a person is present within a range in which the advertising display device can be seen. If it is determined that the person exists, the step is to identify the advertisement that the advertisement display device showed to the person determined to exist, The steps include identifying the first store corresponding to the advertisement from among a predetermined number of stores, A step of detecting that the person who was determined entered the first store after the determination, A step of calculating a value indicating the effectiveness of the advertisement displayed on the advertisement display device based on whether or not it has been detected that the person determined to have entered the first store, An advertising analysis method performed by one or more computers, including [the specified method].

Images