People Detection System

Abstract

To detect a stray child appropriately while reducing labor.SOLUTION: A person detection system according to the present invention includes a person extraction unit that extracts person information representing a person from image information taken by a monitoring camera, a determination unit that extracts behavior information representing a behavior of the person from the image information, determines a group formed by a plurality of persons based on the behavior information and registers group information including the person information, a first person identification unit that identifies a first person from the image information taken by the monitoring camera based on the extracted person information and the extracted behavior information, a second person identification unit that identifies a second person belonging to the same group as the first person from the image information taken by the monitoring camera based on the person information of the person identified as the first person and the group information, and a position identification unit that identifies a position of the first person and a position of the second person based on position information of the monitoring camera that takes the image information in which the first person and the second person are respectively identified.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to a person detection system, and more particularly to a person detection system that detects a person from within an image. [Background technology]

[0002] In large entertainment and commercial facilities, when a child gets separated from his / her guardian and gets lost, the child is discovered only when a staff member or another customer spots the crying child and guides him / her to the lost child center. However, since finding a lost child relies on human and visual inspection, there are problems with low accuracy and time.

[0003] Additionally, in order to reunite lost children with their guardians after they have been rescued, the current practice is to broadcast the characteristics of the lost child or rely on the guardians' inquiries to the lost children center. However, it is difficult to obtain information from lost children (especially small children). In addition, some entertainment facilities do not broadcast information, and there are cases where the means of alerting guardians are limited. This results in the problem that it takes even more time to reunite lost children with their guardians.

[0004] On the other hand, some methods of detecting lost children include having children carry wireless communication devices that detect their location or notify them by sound, but these devices must be worn by children and their guardians at all times, which creates a burden for them. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2007-60528 A Summary of the Invention [Problem to be solved by the invention]

[0006] Here, Patent Document 1 discloses a system for detecting lost children. In the technology disclosed in Patent Document 1, a registration camera is prepared in advance, and facial images of the chaperone and the child are taken with the registration camera, and each facial image is associated with an email address and registered in a database. Then, a facial image of a person is detected from each image acquired by a surveillance camera, and it is determined whether or not there is a matching facial image. If a matching facial image is found, the position of the child is identified from the position of the surveillance camera, and further, the chaperone is identified from the database.

[0007] However, the technology of the above-mentioned Patent Document 1 requires that the leader and the child be photographed and registered in advance using a registration camera, which is time-consuming. On the other hand, if the photograph is neglected, the lost child cannot be detected.

[0008] Therefore, an object of the present invention is to provide a person detection system that can appropriately detect a lost child while reducing the amount of work required, which is the problem described above. [Means for solving the problem]

[0009] A human detection system according to one embodiment of the present invention comprises: a person extraction unit that extracts person information representing a person from image information captured by a surveillance camera; a group determination unit that extracts behavioral information representing the behavior of the person from the image information, determines a group formed by a plurality of people based on the behavioral information, and registers group information including the person information; a first person identification unit that identifies a first person based on the extracted person information and the extracted behavior information from the image information captured by the surveillance camera; a second person identification unit that identifies a second person belonging to the same group as the first person based on the person information of the person identified as the first person and the group information from the image information captured by the surveillance camera; a position identification unit that identifies the positions of the first person and the second person based on position information of the surveillance camera that captured the image information in which the first person and the second person are respectively identified; Equipped with The structure is as follows.

[0010] Further, a human detection method according to one aspect of the present invention includes: Extracting personal information representing a person from image information captured by a surveillance camera, extracting behavioral information representing the behavior of the person from the image information, determining a group formed by a plurality of people based on the behavioral information, and registering group information including the person information; Then, a first person is identified based on the extracted person information and behavior information from the image information captured by the surveillance camera, Identifying a second person who belongs to the same group as the first person based on the person information of the person identified as the first person and the group information from the image information captured by the surveillance camera; identifying the positions of the first person and the second person based on position information of the surveillance camera that captured the image information in which the first person and the second person are respectively identified; The structure is as follows.

[0011] A human detection system according to one embodiment of the present invention comprises: In the information processing device, a person extraction unit that extracts person information representing a person from image information captured by a surveillance camera; a group determination unit that extracts behavioral information representing the behavior of the person from the image information, determines a group formed by a plurality of people based on the behavioral information, and registers group information including the person information; a first person identification unit that identifies a first person based on the extracted person information and the extracted behavior information from the image information captured by the surveillance camera; a second person identification unit that identifies a second person belonging to the same group as the first person based on the person information of the person identified as the first person and the group information from the image information captured by the surveillance camera; a position identification unit that identifies the positions of the first person and the second person based on position information of the surveillance camera that captured the image information in which the first person and the second person are respectively identified; To achieve this, The structure is as follows. Effect of the Invention

[0012] With the above-described configuration, the present invention can appropriately detect a person while reducing the effort required by the user. [Brief description of the drawings]

[0013] [Figure 1] 1 is a block diagram showing a configuration of an information processing system according to a first embodiment of the present invention. [Diagram 2] 2 is a flowchart showing the operation of the information processing system disclosed in FIG. 1. [Diagram 3] 2 is a diagram illustrating an operation of the information processing system disclosed in FIG. 1. [Figure 4] 2 is a flowchart showing the operation of the information processing system disclosed in FIG. 1. [Diagram 5] 2 is a diagram illustrating an operation of the information processing system disclosed in FIG. 1. [Figure 6] 2 is a flowchart showing the operation of the information processing system disclosed in FIG. 1. [Figure 7] 2 is a diagram illustrating an operation of the information processing system disclosed in FIG. 1. [Figure 8] FIG. 11 is a block diagram showing a configuration of an information processing system according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] <Embodiment 1> A first embodiment of the present invention will be described with reference to Fig. 1 to Fig. 7. Fig. 1 is a diagram for explaining the configuration of an information processing system in the first embodiment. Fig. 2 to Fig. 7 are diagrams for explaining the operation of the information processing system.

[0015] The information processing system (person detection system) of the present invention can be used mainly in large entertainment facilities and commercial facilities to detect lost children and reunite them with their guardians. However, the information processing system of the present invention is not limited to detecting lost children, and may be used to detect any person in any place.

[0016] [composition] The information processing system 10 in this embodiment is composed of one or more information processing devices each having a calculation device and a storage device. As shown in Fig. 1, the information processing system 10 includes a feature extraction unit 11, a person extraction unit 12, an individual feature analysis unit 13, a behavior analysis unit 14, a group determination unit 15, a lost child identification unit 16, a guardian identification unit, and a person position identification unit 18, which are constructed by the calculation device executing a program. The information processing system 10 also includes a location information database (DB) 21 and a group database (DB) 22, which are formed in the storage device.

[0017] Furthermore, a surveillance camera 31 and a camera 32 can be connected to the information processing system 10. The surveillance cameras 31 are installed at multiple locations within the facility, capture images of the locations where the cameras are installed, and input the image information to the information processing system 10. At this time, the surveillance cameras 31 and location information corresponding to the locations where the surveillance cameras 31 are installed are associated and registered within the information processing system 10. In other words, the information processing system 10 can recognize the shooting location of the image information input from the surveillance camera 31. For example, the surveillance cameras 31 are installed at the entrance to the facility or at landmark locations within the facility.

[0018] The camera 32 is a different camera from the surveillance camera 31, and is provided at a predetermined location within the facility. The camera 32 may be a digital camera or a camera equipped on a mobile phone, and is connected to the information processing system wirelessly or by wire, so that captured image information can be input to the information processing system 10.

[0019] The feature extraction unit 11, the person extraction unit 12, and the individual feature analysis unit 13 cooperate to function as a person extraction unit that extracts person information representing a person from image information captured by the surveillance camera 31. Specifically, the feature extraction unit 11 extracts feature information representing external features such as the outline of a person from the image information. The person extraction unit 12 automatically identifies a person from the extracted feature information. The individual feature analysis unit 13 performs an individual feature analysis of the identified person's age, clothes, physique, etc., and stores the individual information for each person in the location information database 21 in association with the location information and the shooting time of the surveillance camera 31. As a result, the location information database 21 stores the features of each person and the place and time when the person appeared, and is updated as the person moves.

[0020] The behavior analysis unit 14 and the group determination unit 15 cooperate to determine a group formed by a plurality of people, and function as a group determination unit that registers group information including information on a plurality of people determined to be a group in the group database 22. Specifically, the behavior analysis unit 14 analyzes the behavior of people from a moving image to extract behavior information. Based on the behavior information, the group determination unit 15 performs behavior analysis of the behavior of a plurality of people who are close to each other, such as the direction and timing of movement, whether or not they are talking, and automatically determines whether they are a plurality of people or a group. If it is determined to be a group, the information on the plurality of people is registered in the group database 22 as one group information.

[0021] The group registration process by the group determination unit 15 is performed, for example, from video captured by a surveillance camera 31 installed near the entrance of the facility. However, the group registration process may be performed from video captured by any surveillance camera 31, and the group information may be updated each time a group is determined. For example, when it is determined that a group originally consisting of four people has been split into groups of two, the group information for each pair may be registered, and these group information may also be registered in association with each other.

[0022] The lost child identification unit 16 (first person identification unit) identifies a lost child (first person) based on the extracted person information and behavior information from the image information captured by the surveillance camera. For example, the lost child identification unit 16 extracts features such as contours from an image captured by the surveillance camera 31 installed at a certain location using the above-mentioned feature extraction unit 11, person extraction unit 12, and individual characteristic analysis unit 13, automatically grasps the person, and extracts only people of an age that may be particularly likely to get lost based on the person information resulting from the individual characteristic analysis of each person's age, clothing, physique, etc. Furthermore, if the lost child identification unit 16 determines that a person may be lost based on information such as anxious facial expressions and behavior, whether the person is acting alone (alone), etc., which is the result of the behavior analysis of the person performed in parallel by the above-mentioned behavior analysis unit 14, the lost child identification unit 16 identifies the person as lost.

[0023] The lost child identification unit 16 may also be provided with a case where it does not determine that a child is lost depending on the installation location of the surveillance camera 31 or the situation. For example, in a place where a child is considered to be exceptionally separated from his / her parents, such as a daycare center, the child may not be identified as lost even if he / she is acting alone.

[0024] The person position identifying unit 18 identifies and outputs the position of the lost child identified by the lost child identifying unit 16 as described above, based on the position information of the security camera 31 that captured the image of the lost child. The position information of the lost child is output, for example, to notify the facility staff so that the staff can be dispatched to the position of the lost child.

[0025] The guardian identification unit 17 (second person identification unit) identifies a guardian (second person) who belongs to the same group as the lost child, based on the personal information of the lost child identified by the lost child identification unit 16 and the group information registered in the group database 22. For example, the guardian identification unit 17 queries the group database 22 for characteristic information of the lost child, identifies the group to which the lost child belongs, and identifies a guardian who is thought to be an adult from the personal information of the identified group.

[0026] The person position identifying unit 18 identifies the position of the guardian identified by the guardian identifying unit 17 as described above by inquiring about the guardian's personal information from the position information database 21. The guardian's position information is output to notify the facility staff, for example, so that the staff can be dispatched to the guardian's position.

[0027] The above-mentioned feature extraction unit 11, person extraction unit 12, and individual feature analysis unit 13 work together to function as a person extraction unit that extracts person information representing a person from image information captured by the camera 32. Specifically, the camera 32 captures a guardian searching for a lost child, and from the image information, similar to the above, feature information such as the outline of the person is extracted, the person is automatically identified, and individual feature analysis is performed. The extracted person information is then passed to the guardian identification unit 17.

[0028] The guardian identification unit 17 (third person identification unit) searches the group database 22 based on the personal information extracted from the image captured by the camera 32. Then, among the personal information included in the group information registered in the group database 22, a person who matches the personal information extracted from the image is identified as a guardian (third person). At this time, the group to which the identified guardian belongs is also identified.

[0029] The lost child identifying unit 16 (fourth person identifying unit) identifies a lost child (fourth person) from within the group of guardians identified as described above based on the person information included in the group information, and extracts the person information of the lost child. At this time, for example, the person information within the group is checked, and a person determined to be a child from the characteristic information is identified as a lost child, and the person information is extracted. Furthermore, the lost child identifying unit 16 identifies the extracted lost child from an image captured by the surveillance camera 31. For example, the location information database 21 is inquired about the extracted person information of the lost child, and a person with matching person information is identified as a lost child.

[0030] The person position identifying unit 18 identifies the position of the lost child from the person information of the lost child identified as described above, using the position information database 21. The position information of the lost child is output to notify the facility staff, for example, so that the staff can be dispatched to the position of the lost child.

[0031] [Operation] Next, the operation of the above-mentioned information processing system will be described in detail below: (1) Group registration operation, (2) First lost child finding operation, and (3) Second lost child finding operation.

[0032] (1) Group registration operation 2 and 3, the operation of determining customer location information and group information using the surveillance camera 31 in the facility and updating the database in real time will be described.

[0033] First, visitors are photographed by the surveillance camera 31 at each entrance of the facility (step S1 in FIG. 2). Then, features such as contours are extracted from the image (step S11 in FIG. 3) and the person is automatically identified (step S12 in FIG. 3). Furthermore, personal features such as age, clothing, and build of each person are analyzed to extract personal information about each person, and this personal information is linked to the position information of the surveillance camera 31 and stored in the position information database 21 (step S13 in FIG. 3).

[0034] In addition, behavioral analysis is performed between people who are close to each other, such as the direction and timing of movement and whether they are talking (step S14 in FIG. 3), and it is automatically determined whether there are multiple people or a group (step S2 in FIG. 2, step S15 in FIG. 3). For example, height and age recognition is used to determine a combination of a child and his / her guardian as a group. When multiple people are determined to be a group, group information including personal information about each person is registered in group database 22 (step S3 in FIG. 2). As a result, as will be described later, when a person in a certain group is the subject of a search, information about other people determined to be in the same group can be output.

[0035] Any surveillance camera 31 in the facility always captures images in real time, extracts and analyzes people, and updates and registers group information including the person information along with the position information of the surveillance camera 31. This allows the location of individuals and groups to be updated appropriately. Also, when it is determined that all members of a certain group have left the facility based on the capture and analysis of the images by the surveillance cameras 31 near each exit, the group information is deleted from the group database 22.

[0036] At the same time, every time an image is captured by each surveillance camera 31, personal information is extracted from the image as described above, and the position information of the personal information is registered in the position information database 21. The position information database 21 is updated in accordance with the movement of the person.

[0037] (2) First operation to find a lost child 4 and 5, the operation up to contacting the guardian will be described in the case where a lost child is discovered before being contacted by the guardian.

[0038] First, images are taken by the surveillance cameras 31 installed at various locations in the facility (step S21 in FIG. 4). Then, features such as contours are extracted from the images (step S31 in FIG. 5), and the person is automatically identified (step S32 in FIG. 5). Furthermore, personal features such as age, clothing, and physique of each person are analyzed (step S33 in FIG. 5), and only people of an age where they may be lost are extracted. In parallel, the behavior of the person is analyzed (step S34 in FIG. 5), and if it is determined that there is a possibility of a person being lost based on information such as anxious facial expressions and behavior, whether the person is acting alone, etc., the person is identified as a lost person (step S22 in FIG. 4, step S35 in FIG. 5). At this time, the facility staff is notified so that staff can be dispatched to the site immediately (step S36 in FIG. 5). If the staff determines that the person is lost at the site, they will protect the lost person.

[0039] In addition to the above-mentioned processing, the group database 22 is queried for personal information of the identified lost child, and personal information of the guardian is identified (step S23 in FIG. 4, step S37 in FIG. 5). For example, from personal information in the group to which the identified lost child belongs, a person who appears to be an adult is identified as the guardian. Then, the location information database 21 is queried for personal information of the guardian, and the guardian's current location is identified (step S24 in FIG. 4, step S38 in FIG. 5). At this time, by dispatching a staff member to the location, the guardian is informed that the lost child is being looked after (step S39 in FIG. 5), and it becomes possible to quickly reunite the guardian and the lost child.

[0040] (3) Second operation to find lost child 6 and 7, the operation up to the protection of a lost child when a parent inquires about the lost child will be described.

[0041] When a guardian inquires at the lost child center about a lost child, the guardian is photographed by the camera 32 (step S41 in FIG. 6). Then, from the photographed image of the guardian, personal information of the guardian is extracted through feature extraction (step S51 in FIG. 7), person extraction (step S52 in FIG. 7), and individual feature extraction (step S53 in FIG. 7). Then, based on the guardian's personal information, an inquiry is made to the group database 22 to identify lost children in the same group as the guardian (step S42 in FIG. 6, step S54 in FIG. 7), and personal information describing the characteristics of the lost children is obtained.

[0042] Next, the location information database 21 is queried for information about the lost child, and location information about the lost child is obtained (step S43 in FIG. 6, step S55 in FIG. 7). By informing staff of this, the staff can quickly protect the lost child on site and introduce him / her to his / her guardian.

[0043] In this embodiment, an image of the guardian taken by the camera 32 is used to query the group database 22, which has the effect of preventing kidnapping by a false guardian who poses as a child and eliminating human error.

[0044] As described above, according to the present invention, the process from the occurrence of a lost child to detection and the subsequent reunion with the guardian can be carried out in a shorter time, more accurately, and with less human intervention. In this case, the system can be constructed using camera equipment installed within the facility without requiring a mobile device such as a mobile phone, thereby reducing the burden on service providers and users.

[0045] In the above-mentioned information processing system, if it is difficult to detect lost children or groups due to issues with camera accuracy or behavior recognition accuracy, multiple candidates may be given as an answer rather than just one. In that case, one correct answer may be selected visually from the multiple selected targets and given hints (information such as what percentage probability of match, similar faces but different clothing, etc.). Even when multiple candidates are given as an answer, if the probability of a match can be calculated during the matching process, sorting and outputting them in order of the probability of a match can help match lost children with their guardians more quickly.

[0046] <Embodiment 2> Next, a second embodiment of the present invention will be described with reference to Fig. 8. Fig. 8 is a block diagram showing the configuration of a human detection system in the second embodiment. Note that the human detection system in this embodiment shows an outline of the configuration of the information processing system described in the first embodiment.

[0047] As shown in FIG. 8, the human detection system 100 in this embodiment includes a human extraction unit 101 that extracts human information representing a human from image information captured by a surveillance camera 110; a group determination unit 102 that extracts behavioral information representing the behavior of a person from the image information, determines groups formed by a plurality of people based on the behavioral information, and registers group information including the person information; a first person identification unit (103) that identifies a first person based on extracted person information and behavior information from image information captured by a surveillance camera (110); a second person identification unit that identifies a second person belonging to the same group as the first person based on the person information and group information of the person identified as the first person from the image information captured by the surveillance camera; a position identification unit that identifies the positions of the first person and the second person based on position information of a security camera that captured image information in which the first person and the second person are identified; It is equipped with:

[0048] Each of the above-mentioned components 101 to 105 may be constructed by executing a program using a calculation device installed in an information processing device.

[0049] According to the person detection system having the above configuration, first, person information and behavior information are extracted from image information captured by a surveillance camera, and group information consisting of a plurality of people is automatically registered from the person information and behavior information. Therefore, the group is always registered without any effort on the part of the user. After that, when a first person (e.g., a lost child) is identified based on the person information and behavior information extracted from image information captured by surveillance cameras in various locations, a second person (e.g., a guardian) of the group to which the first person belongs is identified from the person information and group information of the first person. Then, the positions of the first person and the second person can be identified by using the position information of the surveillance cameras in which the first person and the second person were captured. As a result, there is no need for preparations such as taking pictures with a registration camera in advance, and people can be detected appropriately while reducing the user's efforts.

[0050] In addition to the above configuration, the person extraction unit extracts person information of the person from image information of the person captured by a camera other than the surveillance camera. The person detection system further includes a third person identification unit that identifies a third person from group information based on the person information of the person extracted from image information captured by a camera other than the surveillance camera, and a fourth person identification unit that identifies a fourth person belonging to the same group as the identified third person based on the person information of the person identified as the third person and the group information, and identifies the fourth person based on the person information extracted from the image information captured by the surveillance camera. The position identification unit further identifies the position of the fourth person based on the position information of the surveillance camera that captured the image information in which the fourth person is identified.

[0051] According to the above configuration, first, a third person (e.g., a guardian) is photographed by a camera other than the surveillance camera, and the third person in the group information is identified from the person information extracted from the image information. Then, a fourth person (e.g., a lost child) who belongs to the same group as the third person is identified from the group information, and the person information of the fourth person is obtained. The fourth person is identified from the image information photographed by the surveillance camera, and the position of the fourth person can be determined. In this way, even if the third person is searching for the fourth person, the person can be appropriately detected.

[0052] <Additional Notes> A part or all of the above-described embodiments can be described as follows: The following describes an outline of the configurations of the human detection system, the program, and the human detection method according to the present invention. However, the present invention is not limited to the following configurations.

[0053] (Appendix 1) a person extraction unit that extracts person information representing a person from image information captured by a surveillance camera; a group determination unit that extracts behavioral information representing the behavior of the person from the image information, determines a group formed by a plurality of people based on the behavioral information, and registers group information including the person information; a first person identification unit that identifies a first person based on the extracted person information and the extracted behavior information from the image information captured by the surveillance camera; a second person identification unit that identifies a second person belonging to the same group as the first person based on the person information of the person identified as the first person and the group information from the image information captured by the surveillance camera; a position identification unit that identifies the positions of the first person and the second person based on position information of the surveillance camera that captured the image information in which the first person and the second person are respectively identified; Equipped with People detection system.

[0054] (Appendix 2) 2. The person detection system of claim 1, the person extraction unit extracts, as the person information, feature information that represents an external feature of a person; The first person identification unit identifies a lost child as the first person based on the characteristic information. People detection system.

[0055] (Appendix 3) 3. The person detection system according to claim 2, The first person identification unit identifies a child who is acting alone as the first person, that is, a lost child, based on the behavior information and the characteristic information. People detection system.

[0056] (Appendix 4) 4. The person detection system according to claim 2, the second person identification unit identifies the second person as a guardian of the lost child identified as the first person based on the characteristic information. People detection system.

[0057] (Appendix 5) 2. The person detection system of claim 1, The person extraction unit extracts personal information of a person from image information obtained by photographing the person with a camera other than the surveillance camera, moreover, a third person identification unit that identifies a third person from the group information based on the person information of a person extracted from image information captured by a camera different from the surveillance camera; a fourth person identification unit that identifies a fourth person belonging to the same group as the identified third person based on the personal information of the person identified as the third person and the group information, and identifies the fourth person based on the personal information extracted from the image information captured by the surveillance camera; Equipped with The position identification unit identifies a position of the fourth person based on position information of the surveillance camera that captured the image information in which the fourth person is identified. People detection system.

[0058] (Appendix 6) 6. The person detection system according to claim 5, The third person identification unit identifies the third person as a guardian, the fourth person identification unit identifies the fourth person as a child of the guardian identified as the third person; People detection system.

[0059] (Appendix 7) A person detection system according to any one of claims 1 to 6, the group determination unit updates and registers the group information based on the behavior information extracted from the new image information. People detection system.

[0060] (Appendix 8) Extracting personal information representing a person from image information captured by a surveillance camera, extracting behavioral information representing the behavior of the person from the image information, determining a group formed by a plurality of people based on the behavioral information, and registering group information including the person information; Then, a first person is identified based on the extracted person information and behavior information from the image information captured by the surveillance camera, Identifying a second person who belongs to the same group as the first person based on the person information of the person identified as the first person and the group information from the image information captured by the surveillance camera; identifying the positions of the first person and the second person based on position information of the surveillance camera that captured the image information in which the first person and the second person are respectively identified; Person detection methods.

[0061] (Appendix 9) 9. The person detection method according to claim 8, further comprising: Identifying a third person from the group information based on the person information of the person extracted from image information obtained by photographing a person with a camera other than the surveillance camera; Identifying a fourth person who belongs to the same group as the identified third person based on the personal information of the person identified as the third person and the group information, and identifying the fourth person based on the personal information extracted from the image information captured by the surveillance camera; identifying a position of the fourth person based on position information of the surveillance camera that captured the image information in which the fourth person is identified; Person detection methods.

[0062] (Appendix 10) In the information processing device, a person extraction unit that extracts person information representing a person from image information captured by a surveillance camera; a group determination unit that extracts behavioral information representing the behavior of the person from the image information, determines a group formed by a plurality of people based on the behavioral information, and registers group information including the person information; a first person identification unit that identifies a first person based on the extracted person information and the extracted behavior information from the image information captured by the surveillance camera; a second person identification unit that identifies a second person belonging to the same group as the first person based on the person information of the person identified as the first person and the group information from the image information captured by the surveillance camera; a position identification unit that identifies the positions of the first person and the second person based on position information of the surveillance camera that captured the image information in which the first person and the second person are respectively identified; A program to achieve this.

[0063] (Appendix 11) 11. The program of claim 10, The person extraction unit extracts personal information of a person from image information obtained by photographing the person with a camera other than the surveillance camera, Furthermore, the information processing device a third person identification unit that identifies a third person from the group information based on the person information of a person extracted from image information captured by a camera different from the surveillance camera; a fourth person identification unit that identifies a fourth person belonging to the same group as the identified third person based on the personal information of the person identified as the third person and the group information, and identifies the fourth person based on the personal information extracted from the image information captured by the surveillance camera; Realize this, The position identification unit identifies a position of the fourth person based on position information of the surveillance camera that captured the image information in which the fourth person is identified. program.

[0064] The above-mentioned program is stored in a storage device or a computer-readable recording medium. For example, the recording medium is a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, or a semiconductor memory.

[0065] Although the present invention has been described with reference to the above-mentioned embodiment, the present invention is not limited to the above-mentioned embodiment. Various modifications that can be understood by a person 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. Information Processing Systems 11 Feature Extraction Unit 12 Person extraction part 13 Personal Characteristics Analysis Unit 14 Behavior Analysis Department 15 Group Judgment Department 16. Lost Child Identification Department 17 Parent Identification Department 18 Person location identification section 21 Location Database 22 Group Database 31 Surveillance Camera 32 Camera 100 Person Detection System 101 Person extraction part 102 Group Judgment Section 103 First Person Identification Department 104 Second Person Identification Department 105 Location identification part 110 Surveillance Camera

Claims

1. Acquire an image from an imaging device; acquiring a first feature and a second feature related to the person based on the image; acquiring a candidate that satisfies a predetermined condition based on the first characteristic and the second characteristic related to the person; In response to presenting the candidates that satisfy the predetermined condition, presenting, together with the candidates that satisfy the predetermined condition, an indicator that indicates to what extent the candidates that satisfy the predetermined condition are related to the first feature and an indicator that indicates to what extent the candidates that satisfy the predetermined condition are related to the second feature, expressed as a percentage. A program that causes a computer to execute a process.

2. The program according to claim 1, The first characteristic includes at least one of the person's age, clothing, and physical build. program.

3. 3. The program according to claim 1 or 2, storing the first feature and the second feature related to the person acquired based on the image in a database; A program that causes a computer to execute a process.

4. 4. The program according to claim 1, presenting, for each candidate that satisfies the predetermined condition, an index indicating how the candidate that satisfies the predetermined condition is related to the first characteristic and an index indicating how the candidate that satisfies the predetermined condition is related to the second characteristic; A program that causes a computer to execute a process.

5. Acquire an image from an imaging device; acquiring a first feature and a second feature related to the person based on the image; acquiring a candidate that satisfies a predetermined condition based on the first characteristic and the second characteristic related to the person; In response to presenting the candidates that satisfy the predetermined condition, presenting, together with the candidates that satisfy the predetermined condition, an indicator that indicates to what extent the candidates that satisfy the predetermined condition are related to the first feature and an indicator that indicates to what extent the candidates that satisfy the predetermined condition are related to the second feature, expressed as a percentage. Information processing methods.

6. a first means for acquiring an image from an imaging device; a second means for acquiring a first feature and a second feature related to the person based on the image; a third means for acquiring a candidate that satisfies a predetermined condition based on the first characteristic and the second characteristic related to the person; a fourth means for presenting, in response to presenting the candidates that satisfy the predetermined condition, an indicator that indicates to what extent the candidates that satisfy the predetermined condition are related to the first characteristic and an indicator that indicates to what extent the candidates that satisfy the predetermined condition are related to the second characteristic, in percentages, together with the candidates that satisfy the predetermined condition; An information processing device comprising:

Images