Information processing device and information processing method

Abstract

This reduces the workload when analyzing the interests of people captured by imaging devices. [Solution] The information processing device 1 includes: an identification unit 131 that identifies a degree of correlation indicating the strength of the relationship between the increase or decrease in the number of people appearing in a first image captured by a first imaging device 2A and the increase or decrease in the number of people appearing in a second image captured by a second imaging device 2B that captures a second imaging area within a predetermined range from the first imaging area which is the imaging area of ​​the first image; a setting unit 132 that sets a first attribute trend, which is the tendency of the attributes of people appearing in the first image, and a second attribute trend, which is the tendency of the attributes of people appearing in the second image; and an estimation unit 133 that estimates the attributes of multiple people appearing in the first image or the second image based on the degree of correlation identified by the identification unit 131 and the first and second attribute trends set by the setting unit 132.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus and an information processing method.

Background Art

[0002] Conventionally, analysis of the movement lines of people shown in images captured by an imaging device has been performed. For example, Patent Document 1 discloses creating a rectangle that includes an object indicating a person included in a captured image of a predetermined area, and creating a movement line by connecting the center points of the rectangle.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology, by creating the movement lines of people, it is possible to identify the destinations visited by people and the attributes of people, and analyze the interests of the people shown in the imaging device. However, there is a problem that when there are many people included in the imaging device, it is burdensome to perform the analysis.

[0005] Therefore, the present invention has been made in view of these points, and an object thereof is to reduce the load when analyzing the interests of people shown in an imaging device.

Means for Solving the Problems

[0006] An information processing device according to a first aspect of the present invention includes: a determination unit that determines the degree of correlation indicating the strength of the relationship between the increase or decrease in the number of people appearing in a first image captured by a first imaging device and the increase or decrease in the number of people appearing in a second image captured by a second imaging device that captures a second imaging area within a predetermined range from the first imaging area which is the imaging area of ​​the first image; a setting unit that sets a first attribute trend, which is the tendency of the attributes of the people appearing in the first image, and a second attribute trend, which is the tendency of the attributes of the people appearing in the second image; and an estimation unit that estimates the attributes of a plurality of people appearing in the first image or the second image based on the degree of correlation determined by the determination unit and the first and second attribute trends set by the setting unit.

[0007] The identifying unit may determine the degree of association based on the similarity between the increase or decrease in the number of people appearing in the first captured image and the increase or decrease in the number of people appearing in the second captured image. The identifying unit may determine the degree of relevance based on the distance from the first imaging area to the second imaging area, or the time it takes for a person to move from the first imaging area to the second imaging area.

[0008] The identifying unit may determine the degree of association based on the similarity between the increase or decrease in the number of people appearing in a plurality of first captured images taken during the first time period and the increase or decrease in the number of people appearing in a plurality of second captured images taken during the second time period, which is a predetermined time period after the first time period. The predetermined time may be determined based on the distance from the first imaging area to the second imaging area, or the time it takes for a person to move from the first imaging area to the second imaging area.

[0009] The identifying unit may further determine the degree of relevance based on the arrangement of pathways for people to walk around the first imaging area and the second imaging area. The identifying unit may determine the degree of relevance based on the positional relationship between the first imaging area and the second imaging area, and at least one of the orientation of the person and the imaging position of the person in the first or second imaging image.

[0010] The identification unit may identify the degree of relevance for each possible state of at least one of the elements of weather and temperature, time of day, date and time, and date classification in an area including at least one of the first imaging area and the second imaging area, and the estimation unit may estimate the attributes of a person appearing in the first or second imaging image based on the degree of relevance corresponding to each possible state of at least one of the elements of weather and temperature and time of day in an area including at least one of the first imaging area and the second imaging area.

[0011] The setting unit may set attribute tendencies of people appearing in the captured image corresponding to the imaging area based on attributes corresponding to the imaging area which is the area captured by the first imaging device or the second imaging device, or to one or more facilities in the vicinity of the imaging area.

[0012] The setting unit may set attribute trends of people appearing in the captured images corresponding to the imaging area, for each weather and temperature, time of day, and date category in the imaging area which is the area captured by the first imaging device or the second imaging device.

[0013] A second aspect of the present invention relates to an information processing method, which includes the steps of: determining a degree of correlation, which is performed by a computer, that indicates the strength of the relationship between the increase or decrease in the number of people appearing in a first image captured by a first imaging device and the increase or decrease in the number of people appearing in a second image captured by a second imaging device that captures a second imaging area within a predetermined range from the first imaging area, which is the imaging area of ​​the first image; setting a first attribute trend, which is the tendency of the attributes of the people appearing in the first image, and a second attribute trend, which is the tendency of the attributes of the people appearing in the second image; and estimating the attributes of a plurality of people appearing in the first image or the second image based on the determined degree of correlation and the set first and second attribute trends. [Effects of the Invention]

[0014] According to the present invention, the burden on the system when analyzing the interests of a person captured by an imaging device can be reduced. [Brief explanation of the drawing]

[0015] [Figure 1] This is a diagram illustrating the overview of an information processing device. [Figure 2] This diagram shows the functional configuration of an information processing device. [Figure 3] This figure shows an example of the analysis results screen. [Figure 4] This is a flowchart showing the processing flow in an information processing device. [Modes for carrying out the invention]

[0016] [Overview of Information Processing Device 1] Figure 1 shows an overview of the information processing device 1. The information processing device 1 is a computer that estimates the attributes of multiple people appearing in the captured image taken by the imaging device 2. As shown in Figure 1, the information processing device 1 is communicated to the imaging device 2. Multiple imaging devices 2 are provided, and each captures an imaging area, which is the area where the attributes of the people are analyzed. In this embodiment, an example is described in which multiple imaging devices 2 are provided, including a first imaging device 2A that captures a first imaging area and a second imaging device 2B that captures a second imaging area within a predetermined range from the first imaging area. In the following description, when the first imaging device 2A and the second imaging device 2B are not distinguished, they will simply be referred to as imaging device 2.

[0017] In order to estimate the attributes of multiple people appearing in the captured image, the information processing device 1 acquires multiple first captured images captured by the first imaging device 2A and multiple second captured images captured by the second imaging device 2B (Figure 1 (1)).

[0018] The information processing apparatus 1 identifies the degree of relevance indicating the strength of the relationship between the increase and decrease in the number of people reflected in a plurality of first captured images that capture the first imaging area and the increase and decrease in the number of people reflected in a plurality of second captured images that capture the second imaging area ((2) in FIG. 1). For example, in the example shown in FIG. 1, it is assumed that there is a strong correlation between the increase and decrease in the number of people reflected in a plurality of first captured images that capture the first imaging area and the increase and decrease in the number of people reflected in a plurality of second captured images that capture the second imaging area, and the degree of relevance is identified as 0.8.

[0019] The information processing apparatus 1 sets a first attribute tendency that is a tendency of the attributes of the people reflected in the first captured image and a second attribute tendency that is a tendency of the attributes of the people reflected in the second captured image ((3) in FIG. 1). For example, in the example shown in FIG. 1, it is assumed that as the first attribute tendency, 90% of the people reflected in the first captured image are set as employees of Company A shown in black in FIG. 1, and as the second attribute tendency, 70% of the people reflected in the second captured image are set as users of Restaurant B Spot.

[0020] The information processing apparatus 1 estimates the attributes of a plurality of people reflected in the first captured image or the second captured image based on the identified degree of relevance and the set first attribute tendency and second attribute tendency ((4) in FIG. 1).

[0021] For example, in the example shown in FIG. 1, the information processing apparatus 1 multiplies the identified degree of relevance 0.8, the probability of 90% including employees of Company A set as the first attribute tendency, and the probability of 70% including users of Restaurant B Spot set as the second attribute tendency, and estimates that approximately 50% of the people reflected in the second captured image are employees of Company A who use Restaurant B Spot.

[0022] In this way, by estimating the attributes of people based on the identified degree of relevance and the first attribute tendency and the second attribute tendency, compared to the case of generating the movement lines of the people reflected in the captured images and analyzing the interests of the people reflected in the imaging device, the load in the case of analyzing the interests of the people reflected in the imaging device can be reduced.

[0023] [Functional Configuration of Information Processing Apparatus 1] Next, the functions of the information processing device 1 will be explained. Figure 2 is a diagram showing the functional configuration of the information processing device 1. The information processing device 1 includes a communication unit 11, a storage unit 12, and a control unit 13.

[0024] The communication unit 11 is a communication interface for sending and receiving data with external devices such as the imaging device 2 via a communication network such as the Internet. The memory unit 12 is a storage medium for storing various types of data, and includes ROM (Read Only Memory), RAM (Random Access Memory), and hard disks. The memory unit 12 stores programs to be executed by the control unit 13. The memory unit 12 stores programs that cause the control unit 13 to function as a specific unit 131, a setting unit 132, an estimation unit 133, and an output unit 134.

[0025] The control unit 13 is, for example, a CPU (Central Processing Unit). The control unit 13 functions as a specific unit 131, a setting unit 132, an estimation unit 133, and an output unit 134 by executing a program stored in the memory unit 12.

[0026] The identification unit 131 identifies a degree of correlation that indicates the strength of the relationship between the increase or decrease in the number of people appearing in the first image captured by the first imaging device 2A and the increase or decrease in the number of people appearing in the second image captured by the second imaging device 2B, which captures a second imaging area within a predetermined range from the first imaging area, which is the imaging area of ​​the first image.

[0027] First, the identification unit 131 acquires multiple first images captured by the first imaging device 2A at each of multiple time points, and acquires multiple second images captured by the second imaging device 2B at each of multiple time points, from the second imaging device 2B. The identification unit 131 determines the degree of relevance based on the similarity between the increase or decrease in the number of people appearing in the multiple first images and the increase or decrease in the number of people appearing in the multiple second images. The identification unit 131 also determines the degree of relevance based on the distance from the first imaging area to the second imaging area, or the time it takes for people to move from the first imaging area to the second imaging area.

[0028] Here, the relevance is a numerical value between 0 and 1. If the relevance is 1, it means that the person in the first image and the person in the second image are exactly the same, and all the people in the second image came from the first imaging area. If the relevance is 0, it means that the person in the first image and the person in the second image are not the same at all, and none of the people in the second image came from the first imaging area.

[0029] For example, the identification unit 131 identifies a first relevance, which is a degree of relevance indicating the proportion of people who flowed into the second imaging area from the first imaging area, based on the similarity between the increase or decrease in the number of people appearing in a plurality of first imaging images taken in the first time period and the increase or decrease in the number of people appearing in a plurality of second imaging images taken in the second time period, which is a predetermined time after the first time period. The predetermined time is determined based on at least one of the distance from the first imaging area to the second imaging area and the travel time of people from the first imaging area to the second imaging area.

[0030] First, the identification unit 131 identifies the number of people appearing in each of the multiple first captured images taken during the first time period, and identifies the amount of change in the number of people over time, as an increase or decrease in the number of people appearing in each of the multiple first captured images during the first time period. Similarly, the identification unit 131 identifies the number of people appearing in each of the multiple second captured images taken during the second time period, and identifies the amount of change in the number of people over time, as an increase or decrease in the number of people appearing in each of the multiple second captured images during the second time period. Then, the identification unit 131 calculates the correlation between the increase or decrease in the number of people appearing in each of the multiple first captured images and the increase or decrease in the number of people appearing in each of the multiple second captured images, and uses this correlation as the first correlation degree, which indicates the strength of the relationship between the increase or decrease in the number of people appearing in the first captured images and the increase or decrease in the number of people appearing in the second captured images.

[0031] Similarly, the identification unit 131 identifies a second relevance, which is a degree of relevance indicating the proportion of people who flowed into the first imaging area from the second imaging area, based on the similarity between the increase or decrease in the number of people appearing in multiple second imaging images taken in the first time period and the increase or decrease in the number of people appearing in multiple first imaging images taken in the second time period, which is a predetermined time after the first time period.

[0032] Furthermore, even if there is a high correlation between the increase or decrease in the number of people appearing in each of the multiple first image captures and the increase or decrease in the number of people appearing in each of the multiple second image captures, there may be other factors besides people moving from the first imaging area into the second imaging area that result in a high correlation between the increase or decrease in the number of people appearing in each of the multiple first image captures and the increase or decrease in the number of people appearing in each of the multiple second image captures. For this reason, the identification unit 131 may adjust the correlation when it is thought that the correlation is high due to other factors.

[0033] For example, the identification unit 131 may correct the degree of relevance based on the distance or path between the first imaging area and the second imaging area. For example, the identification unit 131 sets a correction coefficient to correct the correction value based on the distance or path between the first imaging area and the second imaging area. Here, the correction value is assumed to be between 0 and 1. The identification unit 131 sets the value of the correction coefficient lower as the distance or path between the first imaging area and the second imaging area increases. Then, the identification unit 131 determines the degree of relevance that takes into account the distance between the first imaging area and the second imaging area by multiplying the degree of relevance by the correction coefficient.

[0034] In this way, the information processing device 1 can reduce the correlation between the increase or decrease in the number of people appearing in each of the multiple first imaging images and the increase or decrease in the number of people appearing in each of the multiple second imaging images, in cases where the distance or path between the first imaging area and the second imaging area is large, and the correlation is considered to be high due to factors other than people from the first imaging area flowing into the second imaging area.

[0035] Furthermore, the identification unit 131 may determine the degree of relevance based on the arrangement of pathways for people to walk in the vicinity of the first imaging area and the second imaging area. For example, the identification unit 131 analyzes map information or images captured by the imaging device 2 to determine whether or not there are pedestrian pathways passing through the first imaging area and the second imaging area. If the identification unit 131 determines that there are pedestrian pathways, it sets a correction coefficient greater than 1 and multiplies the identified degree of relevance by the correction coefficient to determine a degree of relevance that takes pedestrian pathways into account.

[0036] The identification unit 131 may determine the width of the pedestrian walkway based on map information or captured images, and may change the correction coefficient based on the determined width of the pedestrian walkway. For example, the identification unit 131 may increase the correction coefficient as the pedestrian walkway widens.

[0037] Furthermore, the identification unit 131 may determine the degree of association based on the positional relationship between the first imaging area and the second imaging area, and at least one of the orientation of a person and the imaging position of the person in the first or second imaging image. For example, the identification unit 131 may determine the direction of the second imaging area in the first imaging image and the neighborhood area, which is a part of the multiple sub-areas included in the first imaging area that is close to the second imaging area. The identification unit 131 calculates the proportion of people facing a predetermined area among the multiple people captured in each of the multiple first imaging images captured by the first imaging device 2A. The identification unit 131 also calculates the number of people captured in the neighborhood area included in each of the multiple first imaging images captured by the first imaging device 2A.

[0038] The identification unit 131 then sets a correction coefficient greater than 1 based on the proportion of people facing the predetermined area and the proportion of people facing the predetermined area. For example, the identification unit 131 sets a higher correction coefficient the higher the proportion of people facing the predetermined area, and also sets a higher correction coefficient the more people are visible in the surrounding area.

[0039] In this way, the information processing device 1 can increase the relevance when there are many people in the first imaging device 2A who are facing the second imaging area or are located near the second imaging area, and therefore the probability of them moving to the second imaging area is considered high.

[0040] Furthermore, the identification unit 131 may acquire actual values ​​of the number of visits to the second imaging area by multiple people appearing in the first image captured by the imaging device 2, and identify the degree of relevance based on the acquired actual values. For example, the identification unit 131 may identify the feature quantities of each of the multiple people appearing in the first image captured by the first imaging device 2A, and also identify the feature quantities of each of the multiple people appearing in the second image captured by the second imaging device 2B.

[0041] The identification unit 131 identifies the same person based on the feature quantities of each person in the first captured image and the feature quantities of each person in the second captured image. The identification unit 131 identifies the identified same person as a person who visited the second captured area, as seen in the second captured image, thereby determining the number of people who visited the second captured area from among the multiple people in the first captured image. The identification unit 131 then determines the degree of association based on the number of people in the second captured image and the number of identified visiting people. In this way, the information processing device 1 can accurately determine the degree of association based on the actual actions of the multiple people in the first and second captured images.

[0042] Furthermore, the identification unit 131 may identify the degree of correlation for each possible state of weather, temperature, and time of day in the surrounding area which includes at least one of the first imaging area and the second imaging area.

[0043] In this case, for example, the identification unit 131 identifies the degree of similarity between the increase or decrease in the number of people appearing in multiple first captured images taken in the first time period and the increase or decrease in the number of people appearing in multiple second captured images taken in the second time period, a predetermined time after the first time period, for each weather and temperature of the surrounding area, time of day, and date category. The date category is information for classifying dates, such as weekdays, holidays, and public holidays. The identification unit 131 then identifies the degree of relevance based on the identified similarity for each weather and temperature of the surrounding area, time of day, and date category.

[0044] The setting unit 132 sets a first attribute trend, which is the tendency of the attributes of people appearing in the first captured image, and a second attribute trend, which is the tendency of the attributes of people appearing in the second captured image. For example, the setting unit 132 sets an attribute trend as a combination of the attributes of people appearing in the captured image and the probability that the people appearing in the captured image include people with those attributes.

[0045] In other words, the setting unit 132 sets a combination of a first attribute, which is the attribute of a person appearing in the first captured image, and a first attribute probability, which is the probability that a person with the first attribute is included in the person appearing in the first captured image, as the first attribute tendency. Similarly, the setting unit 132 sets a combination of a second attribute, which is the attribute of a person appearing in the second captured image, and a second attribute probability, which is the probability that a person with the second attribute is included in the person appearing in the second captured image, as the second attribute tendency.

[0046] For example, the setting unit 132 sets attribute tendencies of people appearing in the captured image corresponding to the imaging area based on attributes corresponding to the imaging area, which is the area imaged by the first imaging device 2A or the second imaging device 2B, or attributes corresponding to one or more facilities in the vicinity of the imaging area. For example, if the imaging area is assigned the attribute of a plaza or a popular spot, the setting unit 132 sets the attribute of people appearing in the captured image to be visitors to the plaza or spot. The setting unit 132 then sets the probability that people with that attribute are included in the captured image corresponding to the imaging area based on at least one of the size and fame of the plaza or spot indicated by the imaging area. Note that at least one of the size and fame is also referred to as size, etc.

[0047] Furthermore, if there are multiple facilities around the imaging area, the setting unit 132 sets the attributes of the people appearing in the image to be visitors to each of the multiple facilities, such as visitors to the first facility and visitors to the second facility. Then, based on the ratio of the size of the multiple facilities, the setting unit 132 sets the probability that the people appearing in the image corresponding to the imaging area will include people with each of these attributes.

[0048] Furthermore, the setting unit 132 may set attribute tendencies of people appearing in the captured images corresponding to the imaging area for each weather and temperature, time of day, and date category of the imaging area. In this way, the information processing device 1 can appropriately set attribute tendencies in response to cases where visitor tendencies to facilities differ depending on the weather and temperature, time of day, and date category of the imaging area.

[0049] The estimation unit 133 estimates the attributes of multiple people appearing in the first or second captured image based on the degree of relevance identified by the identification unit 131 and the first attribute trend and second attribute trend set by the setting unit 132.

[0050] For example, the estimation unit 133 estimates the probability that the second image contains multiple people who possess both the first and second attributes by multiplying the probability of the first attribute, the probability of the second attribute, and the degree of association, thereby estimating these as attributes of the multiple people in the second image.

[0051] Furthermore, the estimation unit 133 estimates the probability that the multiple people captured in the first captured image include individuals possessing the first and second attributes by multiplying the probability of the first attribute, the probability of the second attribute, and the degree of association, thereby estimating these as attributes of the multiple people captured in the first captured image.

[0052] Furthermore, when the identification unit 131 identifies the degree of relevance for each possible state of at least one of the elements of weather and temperature in the surrounding area, time of day, date and time, and date category, the estimation unit 133 estimates the attributes of the person captured in the first or second captured image based on the degree of relevance corresponding to each possible state of at least one of the elements of weather and temperature in the surrounding area, time of day, date and time, and date category.

[0053] Furthermore, if the identification unit 131 identifies a first relevance, which indicates the proportion of people flowing from the first imaging area to the second imaging area, the estimation unit 133 estimates the probability that the multiple people captured in the second image include individuals with the first and second attributes by multiplying the probability of the first attribute, the probability of the second attribute, and the first relevance, thereby estimating the attributes of the multiple people captured in the second image.

[0054] Similarly, if the identification unit 131 identifies a second relevance, which indicates the proportion of people flowing from the second imaging area to the first imaging area, the estimation unit 133 estimates the probability that the multiple people in the first imaging image include individuals with the first and second attributes by multiplying the probability of the first attribute, the probability of the second attribute, and the second relevance, thereby estimating the attributes of the multiple people in the first imaging image.

[0055] The output unit 134 outputs information relating to the attributes of multiple people captured in the first or second captured image. For example, it outputs an analysis result screen showing the proportion of attributes of multiple people captured in the first or second captured image, estimated by the estimation unit 133, to an analysis device (not shown) that performs pedestrian flow analysis in the first and second imaging areas. For example, the output unit 134 accepts a request to display either an analysis result screen corresponding to the first imaging area or an analysis result screen corresponding to the second imaging area, and displays the analysis result screen corresponding to the accepted imaging area. Alternatively, the output unit 134 may output one analysis result screen corresponding to each of the first and second imaging areas to the analysis device.

[0056] Figure 3 shows an example of an analysis results screen. For example, suppose the first attribute of the first imaging area (Area A) is set to be an employee of Company A, with a probability of 80%, and the second attribute of the second imaging area (Area B) is set to be a visitor to Restaurant C, with a probability of 30%. Suppose the correlation between Area A and Area B, that is, the proportion of people who flowed into Area B from Area A, is identified as 70%. Figure 3 shows the analysis results screen corresponding to the second imaging area.

[0057] The analysis results screen shown in Figure 3 displays a Venn diagram, confirming that 70% of the people in Area B came from Area A. Figure 3 also shows that the percentage of A Company employees in Area B who came from Area A is the product of 70% and 80%, i.e., 56%. Furthermore, it can be seen that 30% of the people in Area B are visitors to Restaurant C, and that in Area B, there is a 21% probability that a visitor to Restaurant C came from Area A, and a 16% probability that a visitor to Restaurant C is both an A Company employee and a visitor to Restaurant C.

[0058] [flowchart] Next, we will explain the processing flow of the information processing device 1. Figure 4 is a flowchart showing the processing flow in the information processing device 1. First, the identification unit 131 identifies the degree of correlation, which indicates the strength of the relationship between the increase or decrease in the number of people appearing in the first captured image and the increase or decrease in the number of people appearing in the second captured image captured by the second imaging device 2B (S1).

[0059] The setting unit 132 sets a first attribute trend, which is the tendency of the attributes of the people captured in the first captured image, and a second attribute trend, which is the tendency of the attributes of the people captured in the second captured image (S2). Next, the estimation unit 133 estimates the attributes of multiple people captured in the first or second captured image based on the correlation degree identified by the identification unit 131 and the first and second attribute trends set by the setting unit 132 (S3). Subsequently, the output unit 134 outputs an analysis result screen showing information related to the attributes of multiple people captured in the first or second captured image (S4).

[0060] [Effects of Information Processing Device 1] As described above, the information processing device 1 according to this embodiment identifies a correlation degree that indicates the strength of the relationship between the increase or decrease in the number of people appearing in the first image captured by the first imaging device 2A and the increase or decrease in the number of people appearing in the second image captured by the second imaging device 2B. It sets a first attribute trend, which is the tendency of the attributes of the people appearing in the first image, and a second attribute trend, which is the tendency of the attributes of the people appearing in the second image. Based on the identified correlation degree and the set first and second attribute trends, it estimates the attributes of multiple people appearing in the first or second image. In this way, the information processing device 1 can reduce the load when analyzing the interests of people appearing in the imaging devices.

[0061] Furthermore, this invention will make it possible to contribute to Goal 9 of the United Nations-led Sustainable Development Goals (SDGs), "Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation."

[0062] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of its gist. For example, all or part of the apparatus can be configured by functionally or physically distributing and integrating in any unit. Furthermore, new embodiments resulting from any combination of multiple embodiments are also included in the embodiments of the present invention. The effects of the new embodiments resulting from the combinations are combined with the effects of the original embodiments. [Explanation of Symbols]

[0063] 1. Information Processing Device 2. Imaging device 11 Communications Department 12 Storage section 13 Control Unit 131 Specific part 132 Acquisition Department 133 Estimation Department 134 Output section

Claims

1. A specific unit for identifying a degree of correlation that indicates the strength of the relationship between the increase or decrease in the number of people appearing in a first image captured by a first imaging device and the increase or decrease in the number of people appearing in a second image captured by a second imaging device that captures a second imaging area within a predetermined range from the first imaging area, which is the imaging area of ​​the first image, wherein the specific unit identifies the degree of correlation based on the distance from the first imaging area to the second imaging area, or the time it takes for people to move from the first imaging area to the second imaging area, and the similarity between the increase or decrease in the number of people appearing in a plurality of first images captured in a first time period and the increase or decrease in the number of people appearing in a plurality of second images captured in a second time period. A setting unit sets a first attribute probability, which is the probability that a person appearing in the first captured image has a first attribute corresponding to one or more facilities in the first imaging area or the surrounding area, as the first attribute trend, which is the tendency of the attributes of the person appearing in the first captured image, and a second attribute probability, which is the probability that a person appearing in the second captured image has a second attribute corresponding to one or more facilities in the second imaging area or the surrounding area, as the second attribute trend, which is the tendency of the attributes of the person appearing in the second captured image. An estimation unit estimates the probability that a person possessing the first attribute and the second attribute is included among a plurality of people captured in the first or second captured image by multiplying the degree of relevance identified by the identification unit by the first attribute probability as a first attribute tendency and the second attribute probability as a second attribute tendency set by the setting unit. An information processing device having

2. The identifying unit determines the degree of association based on the similarity between the increase or decrease in the number of people appearing in the first captured image and the increase or decrease in the number of people appearing in the second captured image. The information processing apparatus according to claim 1.

3. The identifying unit determines the degree of association based on the similarity between the increase or decrease in the number of people appearing in a plurality of first captured images taken during the first time period and the increase or decrease in the number of people appearing in a plurality of second captured images taken during the second time period, which is a time period predetermined to be later than the first time period. The information processing apparatus according to claim 1.

4. The predetermined time is determined based on the distance from the first imaging area to the second imaging area, or the time it takes for a person to move from the first imaging area to the second imaging area. The information processing apparatus according to claim 3.

5. The identifying unit further determines the degree of relevance based on the arrangement of pathways for people to walk around the first imaging area and the second imaging area. The information processing apparatus according to claim 1.

6. The identifying unit determines the degree of relevance based on the positional relationship between the first imaging area and the second imaging area, and at least one of the orientation of the person and the imaging position of the person in the first or second imaging image. The information processing apparatus according to claim 1.

7. The identifying unit identifies the degree of relevance for each possible state of at least one of the elements of weather and temperature, time of day, date and time, and date category in an area including at least one of the first imaging area and the second imaging area. The estimation unit estimates the probability that a person having the first and second attributes is included in the first or second image, based on the correlation between the weather, temperature, and time of day of an area including at least one of the first and second imaging areas, for each possible state of the area. The information processing apparatus according to claim 1.

8. The setting unit sets attribute tendencies of people appearing in the captured image corresponding to the imaging area, based on attributes corresponding to the imaging area which is the area captured by the first imaging device or the second imaging device, or to one or more facilities in the vicinity of the imaging area. The information processing apparatus according to claim 1.

9. The setting unit sets the attribute tendencies of people appearing in the captured images corresponding to the imaging area, for each weather and temperature, time of day, and date category in the imaging area which is the area captured by the first imaging device or the second imaging device. The information processing apparatus according to claim 1.

10. A computer executes A step of identifying a degree of correlation indicating the strength of the relationship between the increase or decrease in the number of people appearing in a first image captured by a first imaging device and the increase or decrease in the number of people appearing in a second image captured by a second imaging device that captures a second imaging area within a predetermined range from the first imaging area, which is the imaging area of ​​the first image, wherein the degree of correlation is identified based on the distance from the first imaging area to the second imaging area, or the time it takes for people to move from the first imaging area to the second imaging area, and the similarity between the increase or decrease in the number of people appearing in a plurality of first images captured in a first time period and the increase or decrease in the number of people appearing in a plurality of second images captured in a second time period. The steps include setting a first attribute probability, which is the probability that a person appearing in the first captured image has a first attribute corresponding to one or more facilities in the first imaging area or the surrounding area, as the first attribute trend, which is the tendency of the attributes of the people appearing in the first captured image, and setting a second attribute probability, which is the probability that a person appearing in the second captured image has a second attribute corresponding to one or more facilities in the second imaging area or the surrounding area, as the second attribute trend, which is the tendency of the attributes of the people appearing in the second captured image, The steps include: estimating the probability that a person possessing the first attribute and the second attribute is included among the multiple persons captured in the first or second captured image by multiplying the identified correlation by the first attribute probability as the set first attribute tendency and the second attribute probability as the second attribute tendency; An information processing method having

Images