Information processing device

Abstract

To provide an information processing device, an information processing method, and a program for detecting a desired person present in a target location.SOLUTION: In an information processing system, a monitoring device 10 includes: a reference information storage unit 15 that acquires environment information related to time corresponding to a target location, acquires, on the basis of the environment information, reference attribute information related to behavior of a person corresponding to the target location and stores the acquired reference attribute information; a person extraction unit 11 that extracts behavior information of the person appearing in an image photographed at the target location; and a detection unit 13 that detects, on the basis of the reference attribute information and the behavior information, a person who performs predetermined behavior in the target location.SELECTED DRAWING: Figure 2A

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program for detecting a person present in a target location. [Background technology]

[0002] In places where large numbers of people gather, such as airports, train stations, stores, and event venues, there may be people who may be committing crimes or causing trouble. Such people may be behaving suspiciously or may have abnormal biological conditions, and they may be identified as suspicious just by appearance, allowing for proactive measures to be taken.

[0003] On the other hand, there are cases where there are few or no security personnel to detect suspicious individuals, or where it is not possible to identify suspicious individuals by appearance alone. Taking such situations into consideration, technology that can automatically detect suspicious individuals is desired. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2018-37075 Summary of the Invention [Problem to be solved by the invention]

[0005] For example, a technology for automatically detecting suspicious individuals at airports and the like is disclosed, as described in Patent Document 1. In Patent Document 1, as an example, suspicious individuals are detected by collecting biometric data of individuals in the immigration inspection area of ​​an airport.

[0006] However, with the method of detecting suspicious individuals from biometric data as described in Patent Document 1, it is not easy to detect biometric data itself from an unspecified number of people. For this reason, it is not possible to detect suspicious individuals, and it is also not possible to detect not only suspicious individuals but also desired individuals in a certain location.

[0007] Therefore, an object of the present invention is to provide an information processing device that can solve the above-mentioned problem of not being able to easily detect a desired person in a target location. [Means for solving the problem]

[0008] An information processing device according to one aspect of the present invention includes: a storage means for storing reference attribute information representing attributes of a person according to a target location; extraction means for extracting person attribute information representing attributes of people in a photographed image of the target location; a detection means for detecting a predetermined person in the photographed image based on the reference attribute information and the person attribute information; Equipped with The structure is as follows.

[0009] Furthermore, a program according to one aspect of the present invention includes: In the information processing device, extraction means for extracting person attribute information representing attributes of a person in a photographed image of a target location; a detection means for detecting a predetermined person in the captured image based on reference attribute information representing attributes of a person according to the target location stored in a storage means and the person attribute information; To realize The structure is as follows.

[0010] Furthermore, an information processing method according to one aspect of the present invention includes: extracting person attribute information representing the attributes of people in a photographed image of the target location; detecting a predetermined person in the photographed image based on the person attribute information and standard attribute information representing attributes of people according to the target location stored in a storage means; The structure is as follows. [Effects of the Invention]

[0011] With the above-described configuration, the present invention can easily detect a desired person in a target location. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram showing the overall configuration of an information processing system according to a first embodiment of the present invention. [Figure 2A] 1 is a block diagram showing a configuration of an information processing system according to a first embodiment of the present invention. [Figure 2B] 2B is a diagram showing an example of information stored in a reference information storage unit disclosed in FIG. 2A. FIG. [Figure 3] 2 is a diagram illustrating an example of information stored in the monitoring device disclosed in FIG. 1. FIG. [Figure 4] 2 is a diagram showing an example of an image output by the monitoring device disclosed in FIG. 1. FIG. [Figure 5] 2 is a diagram showing an example of an image output by the monitoring device disclosed in FIG. 1. FIG. [Figure 6] 2 is a diagram showing an example of an image output by the monitoring device disclosed in FIG. 1. FIG. [Figure 7] 2 is a flowchart showing a processing operation of the monitoring device disclosed in FIG. 1. [Figure 8] 2 is a flowchart showing a processing operation of the monitoring device disclosed in FIG. 1. [Figure 9] FIG. 10 is a diagram showing an example of an image output by a monitoring device according to a second embodiment of the present invention. [Figure 10] FIG. 10 is a block diagram showing the configuration of an information processing device according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] <Embodiment 1> A first embodiment of the present invention will be described with reference to Figures 1 to 8. Figures 1 to 6 are diagrams for explaining the configuration of an information processing system, and Figures 7 and 8 are diagrams for explaining the processing operation of the information processing system. Below, the configuration and operation of the information processing system will be described together.

[0014] The information processing system of the present invention is used to detect a desired person, such as a person who is determined to be suspicious based on preset criteria, among people P present in a predetermined target location R, such as a store or facility. In the following description, the target location is assumed to be a "cosmetics counter in a department store," and a person who can be identified as a "suspicious person" in such a location is used as a detection target. However, in the present invention, the target location may be any location, such as a jewelry store, a game center, or an amusement park. Furthermore, the detection target of the present invention is not limited to suspicious people, and any person may be detected, such as a person with attributes desired by a store. As a result, the present invention can be used for security purposes by monitoring suspicious people in the target location, or for marketing activities, such as actively serving customers with target attributes. The person to be detected may also be a lost child, a sick person, an elderly person, a person receiving care, or the like.

[0015] As shown in FIG. 1, the information processing system of this embodiment includes a camera C for capturing an image of a space that is a target location R, a monitoring device 10 for monitoring a person P in the captured image, and an output device 20 for outputting the monitoring results. The monitoring device 10 is configured with one or more information processing devices that include a calculation device and a storage device. The output device 20 is configured with one or more information processing devices that include a calculation device and a storage device, and further includes a display device. As will be described later, the display device is used to display and output the detected person together with the captured image captured by the monitoring device 10. The configuration of the monitoring device 10 will be mainly described in detail below.

[0016] 2A, the monitoring device 10 includes a person extraction unit 11, a learning unit 12, and a detection unit 13, which are constructed by a computing device executing a program. The monitoring device 10 also includes a person information storage unit 14 and a reference information storage unit 15, which are formed in a storage device. Each component will be described in detail below.

[0017] The person extraction unit 11 (extraction means) first receives images of a target location R captured by a camera C at regular time intervals. For example, as shown in FIG. 4, images of the target location R in which multiple people P are present are received and temporarily stored. Note that, although only one camera C is connected to the monitoring device 10 in this embodiment, multiple cameras C may be connected, and extraction processing, learning processing, and suspicious person detection processing, as described below, may be performed on the images captured by each camera C.

[0018] Then, the person extraction unit 11 extracts person P from the photographed image based on the shape, color, movement, etc. of objects appearing in the photographed image (step S1 in FIG. 7). Furthermore, the person extraction unit 11 extracts person attribute information representing the attributes of person P based on the image portion of person P extracted from the photographed image (step S2 in FIG. 7). The person attribute information is information representing, for example, person P's gender, age (generation), and personal belongings such as clothing and belongings, and is extracted from person P's facial image, body image, etc. by image analysis.

[0019] The person extraction unit 11 also extracts behavioral information that represents the behavior of person P in the captured image (step S3 in FIG. 7). For example, the person extraction unit 11 extracts, as behavioral information, facial orientation, facial expression, line of sight, moving path, and whether the person is a single person or a group, from the facial image, body image, and distance from other people of the person. In this embodiment, the person extraction unit 11 acquires captured images consisting of moving images from the camera C and extracts the attributes and behavior of the person as described above. However, the person extraction unit 11 may also extract the attributes and behavior of the person from captured images consisting of still images from the camera C. In this case, attributes such as gender of the person and behavior such as facial orientation may be extracted from a single still image, or the attributes and behavior of the person may be extracted from multiple still images that are consecutive in time series.

[0020] Furthermore, the person extraction unit 11 acquires scene information of the target location R when the personal attribute information and behavior information for each person P are extracted as described above. Specifically, the person extraction unit 11 acquires, as the scene information, location information representing the attributes of the target location R and a specific location, and environmental information representing the surrounding environment of the target location R. For example, the person extraction unit 11 acquires, as the location information, camera identification information and camera position information assigned to the camera C capturing the image of the target location R, and acquires location information associated with the camera identification information and camera position information. As an example, the location information may be information representing the attributes of a location, such as "a cosmetics counter in a department store," or information representing a specific location, such as "XX Department Store Ginza Branch." Furthermore, the person extraction unit 11 acquires, as the environmental information, information such as the date, time, season, and weather from another information processing device connected via a network. Note that the location information and environmental information described above are not limited to the above information and may be acquired by any method. The person extraction unit 11 then associates the extracted person attribute information and behavior information with scene information consisting of location information and environmental information, and stores them in the person information storage unit 14.

[0021] The learning unit 12 (reference attribute information generating means) reads out the person attribute information and behavior information extracted from the captured image and stored in the person information storage unit 14 as described above. Then, by learning the person attribute information and behavior information, reference attribute information representing the attributes and behavior of person P corresponding to the target location R, i.e., the scene information, is generated (step S4 in FIG. 7). For example, the learning unit 12 generates reference attribute information by learning the age distribution by gender of person P visiting the "cosmetics counter" which is the target location R, from the extracted attributes such as gender and age of person P, as shown in FIG. 3. In the example of FIG. 3, it can be seen that the attributes of person P visiting the "cosmetics counter" which is the target location R are predominantly female (see symbol W) between the ages of 20 and 60, and extremely rare male (see symbol M). In this way, the reference attribute information represents the attributes of multiple people. The learning unit 12 then associates the generated reference attribute information with scene information and stores it in the reference information storage unit 15 (storage means) (step S5 in FIG. 7).

[0022] The learning unit 12 may also learn environmental information and behavioral information associated with the person attribute information to generate the reference attribute information. For example, the learning unit 12 may generate, from the behavioral information, an age distribution by gender for only person P who is acting alone, as the reference attribute information, or an age distribution by gender for person P who is acting in a group of multiple people, as the reference attribute information. Furthermore, the learning unit 12 may generate, from the behavioral information, statistics on the facial orientation and movement path of person P by gender, as the reference attribute information. As another example, the learning unit 12 may generate, from the behavioral information, specific behaviors by person P, such as taking a product from a shelf / returning a product to the shelf, wandering, staying in a specific location for a certain period of time, gazing at a specific product for a certain period of time, or looking around the area other than the shelf, as the reference attribute information. Furthermore, the learning unit 12 may generate, from the environmental information, an age distribution by gender for person P for each season or time period.

[0023] Here, the learning unit 12 generates reference attribute information using person attribute information, etc. for each piece of scene information stored when the target location R is determined to be in a normal state by a prior setting or an external input. In other words, the reference attribute information for each piece of scene information stored in the reference information storage unit 15 is information for when the location information and environmental information of the target location R are in a normal state. Furthermore, the learning unit 12 may continue learning using person attribute information, etc. extracted from new captured images by the person extraction unit 11 described above, and update the reference attribute information in the reference information storage unit 15. Note that the reference attribute information stored in the reference information storage unit 15 is not necessarily limited to information learned by the learning unit 12 as described above, and may instead be information prepared in advance.

[0024] FIG. 2B shows an example of reference attribute information generated by the learning unit 12 and stored in the reference information storage unit 15. For example, for the situation information "cosmetics counter" and the environment information "holiday," person attributes such as "women in their 20s or older" and "male-female couples in their 30s or older" and actions such as "looking at products" and "picking up products" are stored as the reference attribute information. As another example, for the situation information "XX Department Store Ginza Branch" and the environment information "summer," a person attribute such as "light clothing (clothes)" is stored as the reference attribute information. However, the reference attribute information shown in FIG. 2B is merely an example, and the distribution of ages by gender of person P by person season or time period as shown in FIG. 3 may be stored as the reference attribute information, or any other information may be used.

[0025] Note that in the example of FIG. 2B, the attributes and behavior of person P in a normal state are used as the reference attribute information for the location information and environmental information. However, the attributes and behavior of person P that match an abnormal state, i.e., a suspicious person, may also be used as the reference attribute information. For example, for the situation information "cosmetics counter" and the environmental information "weekday," a person attribute such as "male in a group" and a behavior such as "looking around other than at the products" may be stored as the reference attribute information. Furthermore, the attributes and behavior of a person who is a potential customer who wants to visit the store in the location information and environmental information may also be stored as the reference attribute information.

[0026] Here, the reference information storage unit 15 in which the reference attribute information is stored as described above is not necessarily limited to being provided within the monitoring device 10, but may be provided in another information processing device. In this case, the monitoring device 10 may connect to the other information processing device via a network to store the generated reference attribute information or read out the stored reference attribute information as described below.

[0027] The detection unit 13 (detection means) detects a suspicious person among persons P in a newly captured image. To this end, for a newly captured image, the person extraction unit 11 first extracts person P from the captured image as described above (step S11 in FIG. 8), extracts person attribute information, behavior information, and environmental information of the person P (steps S12 and S13 in FIG. 8), and passes the information to the detection unit 13. At this time, the detection unit 13 acquires camera identification information and camera position information assigned to the camera C that captured the captured image, and extracts location information associated with these pieces of information. Then, the detection unit 13 reads out reference attribute information associated with the extracted location information and environmental information and stored in the reference information storage unit 15, and compares the reference attribute information with the person attribute information, etc. extracted from the newly captured image (step S14 in FIG. 8).

[0028] Then, when the extracted person attribute information and behavior information do not match the reference attribute information based on preset criteria, the detection unit 13 detects the person P as a suspicious person (Yes in step S15 of FIG. 8). For example, the reference attribute information may be the age distribution by gender of a person P acting alone as shown in FIG. 3, or the content of "women in their 20s or older" corresponding to the scene information "cosmetics counter" and the environmental information "weekday" in FIG. 2B. In this case, if the newly extracted person attribute information is "a man in his 60s" or "a group (of men)," the person P or the person group is detected as a suspicious person as not matching the reference attribute information. In this way, the detection unit 13 can detect multiple people P (groups) as suspicious by comparing the person attribute information, etc. of multiple people P with the reference attribute information. The determination that the extracted person attribute information and behavior information do not match the reference attribute information is made, for example, when the probability (e.g., likelihood) that the extracted person attribute information or behavior information matches the reference attribute information is equal to or less than a threshold (e.g., 20% or less). However, such a determination may be made by any method.

[0029] Then, the detection unit 13 outputs and displays the captured image including information about the suspicious person on the output device 20 (step S16 in FIG. 8). Here, FIG. 4 shows the state when the captured image itself is output to the output device 20, and a person P1 detected as a suspicious person is displayed on such a captured image so that he or she can be distinguished. For example, as shown in FIGS. 5 and 6, the suspicious person P1 may be displayed by filling it with a predetermined color, or the suspicious person P1 may be displayed in a way that makes it stand out, such as by displaying a border around the suspicious person P1 or by attaching a mark to the suspicious person P1. Furthermore, the detection unit 13 may track the person P in the captured image as the person P moves, and may also track the suspicious person P1 and display the suspicious person.

[0030] Here, another example of the case where the detection unit 13 detects a suspicious person will be described. For example, the reference attribute information is a person attribute such as "female in her twenties or older" corresponding to the scene information "cosmetics counter" and the environment information "weekday" in Fig. 2B, and the content of the behavior such as "looking at a product" and "picking up a product." In this case, even if the person attribute information extracted from person P is "female in her twenties," if the behavior information is "face looking around (not facing the product shelves)," this does not match the reference attribute information, and therefore person P is detected as a suspicious person.

[0031] As another example, if the reference attribute information represents the size of the bag carried by person P and the information is "small size," and the size of the bag carried as the personal attribute information extracted from person P is "large size," this does not match the reference attribute information, and therefore person P is detected as a suspicious person. As another example, if the reference attribute information represents person P's clothing and the information is "lightly dressed," and the clothing as the personal attribute information extracted from person P is "heavily dressed," this does not match the reference attribute information, and therefore person P is detected as a suspicious person.

[0032] In the above, a case where a person whose personal attribute information extracted from person P does not match the reference attribute information is exemplified, but person P whose personal attribute information matches the reference attribute information may be detected. In other words, if the reference attribute information itself is information that matches a suspicious person, a suspicious person can be detected, and if the reference attribute information itself is content that indicates a potential customer that the store wants to visit, a desired potential customer can be detected.

[0033] Furthermore, the detection unit 13 may identify a person P to be tracked from behavioral information of the person P in the captured image, and detect a suspicious person P1 from the behavioral information of the identified person P. For example, a person P performing a preset behavior such as repeatedly stopping by the same place or wandering around may be tracked, and if the person P performs a behavior stored as reference attribute information, the person P may be detected as a suspicious person P1. Furthermore, the detection unit 13 may store the number of times the person P is detected as a suspicious person P1, and only when the number of times is equal to or greater than a threshold, may the detection unit 13 finally determine and output the person P as a suspicious person P1.

[0034] Furthermore, for example, when standard attribute information representing the age distribution of person P by gender as shown in FIG. 3 is stored, the detection unit 13 may detect "men in the age group" who account for less than a few percent of the total as suspicious persons. As another example, when attributes of person P who caused an abnormal situation such as a crime are stored as standard attribute information based on a captured image when an abnormal situation such as a crime occurs at target location R, person P who has the potential to cause an abnormal situation can be detected as suspicious. As an example, when a person's behavior such as "picking up an item other than merchandise" is stored as standard attribute information for scene information "cosmetics counter," person P who has performed such behavior is detected as suspicious.

[0035] As described above, in the present invention, by preparing reference attribute information that represents the attributes of people according to a target location and comparing it with the attributes of people in a photographed image of the target location, it is possible to detect a desired person in the target location. This can be used, for example, to detect suspicious people who are considered inappropriate for the target location and use it for security purposes, or, in another example, to detect potential customers who are considered appropriate for the target location and use it for customer service.

[0036] Although the above example illustrates a case where both a person's attributes and a person's behavior are used as the reference attribute information, the present invention may use only one of these pieces of information as the reference attribute information. That is, the present invention may detect a suspicious person only from the person's attributes extracted from a new captured image based on the person's attributes stored as the reference attribute information, or may detect a suspicious person only from the person's behavior extracted from a new captured image based on the person's behavior stored as the reference attribute information.

[0037] <Embodiment 2> Next, a second embodiment of the present invention will be described with reference to Fig. 9. Fig. 9 is a diagram showing an example of an image output by the monitoring device 10 in the second embodiment.

[0038] The information processing system of this embodiment has a configuration similar to that of the above-described embodiment 1. In the embodiment 1, the monitoring device 10 monitors almost the entire image captured by one camera C as the target location R, but in this embodiment, the monitoring device 10 divides the image captured by one camera C into divided areas, and each of the divided areas is used as the target location R to be monitored.

[0039] 9 shows an image captured by a single camera C, with two different locations (for example, two sales areas) shown on the left and right sides of the horizontal direction of the captured image. In this case, the monitoring device 10 divides the captured image into left and right halves, setting the left side as a first divided area R1 and the right side as a second divided area R2. The monitoring device 10 then extracts a person P from each of the first divided area R1 and the second divided area R2, generates reference attribute information through learning, and detects a suspicious individual from the reference attribute information and the newly extracted attributes and behavior of the person P.

[0040] Specifically, the monitoring device 10 acquires scene information such as location information and environmental information corresponding to the first divided area R1 from the camera identification information of the camera C and the area information specifying the divided area. The monitoring device 10 then extracts attribute information and behavior information of the person P appearing only in the first divided area R1, learns this information, and generates and stores reference attribute information corresponding to the location information and environmental information of the first divided area R1. The monitoring device 10 then extracts the person P within the first divided area R1, and detects the person P as a suspicious person if the attribute information and behavior information of the person P match / does not match the reference attribute information stored corresponding to the first divided area R1.

[0041] The monitoring device 10 also performs the same processing on the second divided region R2 of the same captured image as that performed on the first divided region R1. First, the monitoring device 10 acquires scene information, such as location information and environmental information, corresponding to the second divided region R2 from the camera identification information of the camera C and the region information specifying the divided region. The monitoring device 10 then extracts attribute information and behavior information of the person P appearing only in the second divided region R2, learns this information, and generates and stores reference attribute information corresponding to the location information and environmental information of the second divided region R1. The monitoring device 10 then extracts the person P from the second divided region R2, and detects the person P as a suspicious person if the attribute information and behavior information of the person P match / does not match the reference attribute information stored corresponding to the second divided region R2.

[0042] By doing the above, it is possible to set reference attribute information for each divided area R1, R2 in the captured image. For example, if the attributes (gender, age, etc.) and behaviors (e.g., just browsing products or wanting to try them out) of customers vary for each sales floor corresponding to divided areas R1, R2, it is possible to set appropriate reference attribute information that represents the attributes and behaviors of people in normal and abnormal states for each sales floor. As a result, it is possible to appropriately detect suspicious individuals for each divided area R1, R2.

[0043] <Embodiment 3> Next, a third embodiment of the present invention will be described with reference to Fig. 10. Fig. 10 is a block diagram showing the configuration of an information processing device in the third embodiment. Note that this embodiment shows an outline of the configuration of the monitoring device described in the first and second embodiments.

[0044] As shown in FIG. 10, the information processing device 100 according to this embodiment includes: a storage means 130 for storing reference attribute information representing attributes of people according to a target location; extraction means 110 for extracting person attribute information representing the attributes of people in a photographed image of a target location; a detection means 120 for detecting a person in a photographed image based on the reference attribute information and the person attribute information; Equipped with.

[0045] The extraction means 110 and detection means 120 described above may be constructed by a calculation device included in the information processing device 100 executing a program, or may be constructed using electronic circuits.

[0046] The information processing device 100 having the above configuration: extracting person attribute information representing the attributes of people in a photographed image of the target location; Detecting a person in the photographed image based on the reference attribute information representing the attributes of the person according to the target location stored in the storage means and the person attribute information; It operates to execute the process.

[0047] According to the above invention, a desired person in a target location can be detected by comparing standard attribute information representing the attributes of a person corresponding to a pre-prepared target location with the attributes of a person in a captured image of the target location.

[0048] <Additional Notes> A part or all of the above-described embodiments can be described as follows: The following provides an overview of the configurations of the information processing system, information processing method, and program according to the present invention. However, the present invention is not limited to the following configurations.

[0049] (Appendix 1) a storage means for storing reference attribute information representing attributes of a person according to a target location; extraction means for extracting person attribute information representing attributes of people in a photographed image of the target location; a detection means for detecting a predetermined person in the photographed image based on the reference attribute information and the person attribute information; An information processing device comprising:

[0050] (Appendix 2) 10. The information processing device according to claim 1, the detection means detects a predetermined person in the captured image based on the reference attribute information representing attributes of a plurality of people and the person attribute information representing attributes of one or a plurality of people; Information processing device.

[0051] (Appendix 3) 3. The information processing device according to claim 1, the detection means detects a predetermined person in the captured image corresponding to the person attribute information when the person attribute information matches the reference attribute information based on a predetermined criterion, or when the person attribute information does not match the reference attribute information based on a predetermined criterion; Information processing device.

[0052] (Appendix 4) An information processing device according to any one of Supplementary Notes 1 to 3, the reference attribute information is set for each surrounding environment, the detection means detects a predetermined person in the photographed image based on the reference attribute information set in accordance with the surrounding environment of the photographed image and the person attribute information; Information processing device.

[0053] (Appendix 5) An information processing device according to any one of Supplementary Notes 1 to 4, the extraction means extracts behavior information representing the behavior of a person in the captured image; the detection means detects a predetermined person in the captured image corresponding to the person attribute information based on the reference attribute information, the person attribute information, and the behavior information; Information processing device.

[0054] (Appendix 6) 6. An information processing device according to any one of Supplementary Notes 1 to 5, a reference attribute information generating means for extracting attributes of people in the photographed image of the target location, and generating and storing the reference attribute information corresponding to the target location based on the extracted attributes of the people; Information processing device.

[0055] (Appendix 7) An information processing device according to any one of Supplementary Notes 1 to 6, The attribute is information representing the person's age, information representing the person's gender, or information representing the person's personal belongings. Information processing device.

[0056] (Appendix 8) In the information processing device, extraction means for extracting person attribute information representing attributes of a person in a photographed image of a target location; a detection means for detecting a predetermined person in the captured image based on reference attribute information representing attributes of a person according to the target location stored in a storage means and the person attribute information; A program to achieve this.

[0057] (Appendix 8.1) 9. The program of claim 8, the reference attribute information is set for each surrounding environment, the detection means detects a predetermined person in the photographed image based on the reference attribute information set in accordance with the surrounding environment of the photographed image and the person attribute information; program.

[0058] (Appendix 8.2) 1. A program according to any one of Appendix 8 or 8.1, comprising: the extraction means extracts behavior information representing the behavior of a person in the captured image; the detection means detects a predetermined person in the captured image corresponding to the person attribute information based on the reference attribute information, the person attribute information, and the behavior information; program.

[0059] (Appendix 8.3) A program according to any one of appendices 8 to 8.2, The information processing device includes: a reference attribute information generating means for extracting attributes of people in the photographed image of the target location, and generating and storing the reference attribute information corresponding to the target location based on the extracted attributes of the people; A program to further realize this.

[0060] (Appendix 9) extracting person attribute information representing the attributes of people in a photographed image of the target location; detecting a predetermined person in the photographed image based on the person attribute information and standard attribute information representing attributes of people according to the target location stored in a storage means; Information processing methods.

[0061] (Appendix 10) 10. The information processing method according to claim 9, the reference attribute information is set for each surrounding environment, detecting a predetermined person in the photographed image based on the reference attribute information and the person attribute information set in accordance with the surrounding environment of the photographed image; Information processing methods.

[0062] (Appendix 11) 11. The information processing method according to claim 9 or 10, extracting behavior information representing the behavior of a person in the captured image; detecting a predetermined person in the captured image corresponding to the person attribute information based on the reference attribute information, the person attribute information, and the behavior information; Information processing methods.

[0063] (Appendix 12) 12. An information processing method according to any one of Supplementary Notes 9 to 11, extracting attributes of people in the photographed image of the target location, and generating and storing the reference attribute information corresponding to the target location based on the extracted attributes of the people; Information processing methods.

[0064] The above-described program can be stored and supplied to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may also be supplied to a computer by various types of transitory computer-readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transitory computer-readable media can supply the program to a computer via a wired communication path such as an electric wire or optical fiber, or via a wireless communication path.

[0065] Although the present invention has been described above with reference to the above-mentioned embodiments, the present invention is not limited to the above-mentioned embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. [Explanation of symbols]

[0066] 10 Monitoring equipment 11 Person extraction part 12 Learning Department 13 Detector 14 Person information storage section 15 Standard information storage section 20 Output Devices 100 Information processing device 110 Extraction means 120 Detection means 130 Memory means C Camera P person P1 Suspicious person R Target location

Claims

1. an acquisition unit that acquires time information corresponding to a target location and acquires standard behavior information corresponding to the target location based on the time information; an extraction unit that extracts behavior information of a person appearing in an image taken at the target location; a detection unit that detects a person performing a predetermined behavior in the target location based on the reference behavior information and the behavior information; An information processing device comprising:

2. 2. The information processing device according to claim 1, The reference behavior information and the behavior information include a travel route. Information processing device.

3. 2. The information processing device according to claim 1, the acquisition unit acquires the reference behavior information based on the time information and identification information of a camera that captures the captured image. Information processing device.

4. 2. The information processing device according to claim 1, the target location is a plurality of divided regions in the captured image, the acquisition unit acquires the reference behavior information corresponding to the divided region based on the time information. Information processing device.

5. an acquisition unit that acquires time information corresponding to a target location and acquires standard behavior information corresponding to the target location based on the time information; an extraction unit that extracts behavior information of a person appearing in an image taken at the target location; a detection unit that detects a person performing a predetermined behavior in the target location based on the reference behavior information and the behavior information; An information processing system comprising:

6. The information processing device acquiring time information corresponding to a target location, and acquiring standard behavior information corresponding to the target location based on the time information; extracting behavioral information of people appearing in images taken at the target location; detecting a person performing a predetermined behavior in the target location based on the reference behavior information and the behavior information; Information processing methods.

Images