JP6593742B2 - In-facility person search support apparatus, in-facility person search support system, and in-facility person search support method - Google Patents

Description

  The present invention relates to an in-facility person search support device, an in-facility person search support system, and an in-facility person search support that supports searching for a person who is separated from a companion in the facility based on images taken by a plurality of cameras in the facility It is about the method.

  Facilities such as department stores and shopping centers have lost child centers that protect children who have lost their parents, but children who have lost their children or lost children It is desirable to search for parents and bring them together quickly.

  As a technique related to searching for a person in such a facility, conventionally, a technique for searching for a search target person based on a video of a camera that captures the inside of the facility is known (see Patent Documents 1 to 3). . In particular, in the techniques disclosed in Patent Documents 1 and 2, a captured image of a search requester, that is, a companion of a search target person is acquired, and a search that exists in the vicinity of the search requester is based on the captured image. The target person is discovered from the camera image, and the search target person is tracked to grasp the latest location of the search target person. In the technique disclosed in Patent Literature 3, a photographed image of a search target person is acquired, and the search target person is searched by person matching using a person's face image.

JP 2004-206584 A JP 2004-151820 A JP 2004-126711 A

  In the conventional technique, a person to be searched is searched using a person detection technique for detecting a person from a camera image or a person recognition technique for identifying a person detected from a picture. The person to be searched can be found by person tracking using detection technology or person recognition technology. However, human detection technology and human recognition technology have limitations in accuracy, and do not necessarily give correct results, and human detection and human recognition are not performed properly, which reduces the accuracy of tracking information. There was a problem that the latest location of the search subject could not be presented reliably.

  The present invention has been devised to solve such problems of the prior art, and the main purpose of the present invention is to allow the user to perform selection and correction operations appropriately even when the accuracy of the tracking information is low. Accordingly, an object of the present invention is to provide an in-facility person search support device, an in-facility person search support system, and an in-facility person search support method configured to be able to reliably present the latest location of a search target person.

  An in-facility person search support device of the present invention is an in-facility person search support device that supports searching for a person who is separated from a companion in a facility, based on video images of the facility taken by a plurality of cameras. Based on the search condition setting unit that sets the search condition for specifying the search target person according to the input operation, the tracking information acquired by the tracking process for the video taken inside the facility, and the search condition, A search target search unit that searches for a person who is likely to be a search target person from the staying person, and a video of a person who is likely to be a search target based on the search result of the search target search unit A search target candidate presentation unit that extracts a predetermined number and presents those videos as search target candidate videos, and the selection according to the user's operation of selecting the search target candidate videos in which the search target person is captured A search target setting unit for setting a person corresponding to the searched candidate video as a search target, and acquiring the latest tracking position of the person set as the search target based on the tracking information, And a latest position presenting section that presents the tracking position as the latest location of the search target person.

  In addition, the in-facility person search support system of the present invention is an in-facility person search support system that supports searching for a person who is separated from a companion in the facility based on images taken by a plurality of cameras in the facility. The camera that captures the inside of the facility, a display device that displays video for each camera, and a plurality of information processing devices, any of the plurality of information processing devices depending on a user's input operation A search condition setting unit for setting a search condition for specifying a search target person, tracking information acquired by tracking processing for a video shot in the facility, and a person staying in the facility based on the search condition From the search target search unit that searches for a person who is likely to be a search target, and a predetermined number of images of the person who is likely to be a search target based on the search result of the search target search unit A search target candidate presenting section that presents those videos as search target candidate videos, and a user's operation to select the search target candidate videos in which the search target person is photographed. Based on the tracking information, a search target setting unit that sets a person corresponding to the candidate video as a search target, acquires the latest tracking position of the person set as the search target, and searches for the latest tracking position. And a latest position presenting unit that presents the latest location of the target person.

  Further, the in-facility person search support method according to the present invention is an in-facility facility that causes an information processing apparatus to perform a process for supporting a search for a person who is separated from a companion in a facility, based on images taken by the plurality of cameras in the facility. A method for supporting a person search, the step of setting a search condition for specifying a search target person according to a user's input operation, the tracking information acquired by a tracking process for a video taken inside the facility, and the search Based on the conditions, a step of searching for a person who is likely to be a search target person from among persons staying in the facility, and a person who is highly likely to be a search target person based on the search result in this step Extracting a predetermined number of videos, presenting those videos as search target candidate videos, and according to the user's operation to select the search target candidate videos in which the search target person is captured, Setting the person corresponding to the selected candidate video of the search object to be searched, and acquiring the latest tracking position of the person set as the search object based on the tracking information, And a step of presenting the tracking position as the latest location of the search target person.

  According to the present invention, since a plurality of videos of a person who is likely to be a search target person are presented as search target candidate videos and selected by the user, the correct person can be searched even when the accuracy of tracking information is low. It is set as a target, and the latest location of the search target person can be surely presented.

Overall configuration diagram of in-facility person search support system according to the present embodiment Plan view showing the installation status of the camera 1 in the facility Functional block diagram showing schematic configuration of PC3 Explanatory drawing which shows the transition state of the screen displayed on the monitor 7 A flow chart showing a procedure of processing performed in each part of the PC 3 according to the operation of the supervisor performed on each screen Explanatory drawing which shows the search condition input screen displayed on the monitor 7 Explanatory drawing which shows the search object selection screen displayed on the monitor 7 Explanatory drawing which shows the whereabouts display screen displayed on the monitor 7 Explanatory drawing which shows the candidate selection screen displayed on the monitor 7 Explanatory drawing which shows the tracking process confirmation screen displayed on the monitor 7 Explanatory drawing which shows the map tracking screen displayed on the monitor 7 Explanatory drawing showing the map tracking screen when tracking fails Explanatory drawing showing the map tracking screen when tracking fails

  A first invention made to solve the above-described problem is an in-facility person search support device that supports searching for a person who is separated from a companion in a facility based on images taken by a plurality of cameras in the facility. Based on a search condition setting unit that sets a search condition for specifying a search target person according to a user's input operation, tracking information acquired by a tracking process for a video taken inside the facility, and the search condition A search target search unit that searches for a person who is likely to be a search target person from among people staying in the facility, and a search target search unit based on the search result of this search target search unit is highly likely A search target candidate presentation unit that extracts a predetermined number of human videos and presents those videos as search target candidate videos, and a user operation that selects the search target candidate videos in which the search target is shown A search target setting unit for setting a person corresponding to the selected candidate video of the search target as a search target, and acquiring the latest tracking position of the person set as the search target based on the tracking information The latest tracking position is configured to include a latest position presenting unit that presents the latest tracking position as the latest location of the search target person.

  According to this, in order to present a plurality of images of a person who is likely to be a search target candidate video to be searched and allow the user to select, the correct person is selected as the search target even when the tracking information accuracy is low. It is set, and the latest location of the search target person can be surely presented.

  Further, in the second invention, the latest position presentation unit presents position information regarding the latest location of the search target person, the latest tracking video corresponding to the latest tracking position, and the search target person is the latest The time information related to the time at which the user stays is presented.

  According to this, since the location information regarding the whereabouts is presented, the user can grasp the whereabouts of the search target person, and furthermore, the latest tracking video of the search target person is presented, so the latest search target person's latest information is presented. The user can grasp the situation and the time information about the time when the search target person stays in the latest location is presented. The user can grasp whether time has passed.

  According to a third aspect of the present invention, the latest position presentation unit presents a live video of a camera that captures the latest tracking video together with the latest tracking video.

  According to this, since the shooting time of the latest tracking video is relatively old, that is, even if a long time has elapsed since the tracking of the search target person was completed, the live video is presented, so the latest location If the user can grasp the current situation and the search target person is not moving, the search target person can be found in the live video.

  According to a fourth aspect of the present invention, the latest position presentation unit displays the live video on a display device, and displays an operation unit for switching the live video camera on the display screen of the live video. .

  According to this, when the search subject moves to the camera shooting area near the camera that shoots the latest tracking video and the search target is not reflected in the live video, by switching the live video camera, The search target person can be found out.

  Further, the fifth invention is the same as the person set as the search target based on the tracking information when there is an error in the latest location of the search target person presented by the latest position presentation unit. A tracking correction candidate presenting unit that extracts a predetermined number of videos of the next highest person after the person whose latest location was presented by the latest position presenting unit and presents those videos as candidate videos for tracking correction The latest position presenting unit includes the latest tracking position of the person corresponding to the selected candidate video for tracking correction in response to a user operation to select the candidate video for tracking correction in which the search target is captured. And the latest tracking position is presented as the latest location of the search target person.

  According to this, if the latest location of the search target presented by the latest position presenter is incorrect, that is, if the tracking information after the search target is set as the search target is incorrect, the tracking correction is performed. Therefore, the tracking information is appropriately corrected, and the latest location of the search target person can be surely presented.

Further, the sixth invention is further provided when the search target setting unit sets a person as a search target when there is an error in all of the tracking correction candidate videos presented by the tracking correction candidate presenting unit. There is a possibility that it is the same as the person set as the search target based on the tracking information during the period from the search candidate video to the latest tracking video presented by the latest position presentation unit. A confirmation video presentation unit that presents a video for each camera in which a person is captured as a confirmation video,
Based on the tracking information, the tracking correction candidate presentation unit may select the possibility that the tracking correction candidate presentation unit is the same as the person set as the search target based on the tracking information. In addition, a predetermined number of videos of the next person higher than the person corresponding to the confirmation video are extracted, and these videos are presented as the tracking correction candidate videos.

  According to this, the video for each camera from the candidate video when set as the search target to the latest tracking video is presented as a confirmation video, and there is a confirmation video with an error, that is, the confirmation that the search target is not shown By allowing the user to select the video, the location where there is an error in the tracking information is clarified, the candidate video corresponding to the location where there is an error in the tracking information is presented, and the user is allowed to select an appropriate candidate video As a result, it is possible to reliably correct a portion having an error in the tracking information.

  In addition, according to a seventh aspect of the invention, the confirmation video presentation unit displays a time series tracking screen on which a plurality of the confirmation videos are arranged in order of shooting time on a display device.

  According to this, since the confirmation video having the error can be found by sequentially checking the confirmation video arranged in the order of the photographing times, the confirmation video can be confirmed quickly.

  Further, according to an eighth aspect of the present invention, the confirmation video presenting unit displays a map tracking screen in which a plurality of the confirmation videos are arranged on the map image indicating the state of the facility according to the actual arrangement state of the camera. The display device is configured to display.

  According to this, the user can easily grasp where the tracking of the search target person has failed by sequentially confirming the confirmation video while looking at the actual arrangement state of the camera.

  In addition, the ninth invention is configured such that the search condition setting unit sets, as the search condition, an image obtained by photographing a companion and information on the characteristics of the search target person as the search condition.

  According to this, since the search target person is searched based on the image obtained by photographing the accompanying person and the information on the characteristics of the search target person, it is possible to improve the search accuracy of the search target person.

  A tenth aspect of the invention is an in-facility person search support system that supports searching for a person who is separated from a companion in a facility, based on images taken in the facility by a plurality of cameras. The camera, a display device that displays video for each camera, and a plurality of information processing devices, and any one of the plurality of information processing devices selects a search target person according to a user input operation. Based on the search condition setting unit for setting the search condition for specifying, the tracking information acquired by the tracking process for the video captured in the facility, and the person who stays in the facility based on the search condition, the search target person A search target search unit that searches for a person who is likely to be a search target, and a predetermined number of videos of a person who is highly likely to be a search target person based on the search result of the search target search unit, and those videos Looking for A search target candidate presenting unit to be presented as a candidate video of the target, and a person corresponding to the selected candidate video of the search target in response to a user operation to select the candidate video of the search target in which the search target person is captured Based on the search target setting unit to be set as the search target and the tracking information, the latest tracking position of the person set as the search target is acquired, and the latest tracking position is set as the latest location of the search target person. And a latest position presenting unit to be presented.

  According to this, similarly to the first invention, even when an error occurs in the tracking information, the user can surely present the latest location of the search target person by selecting an appropriate search target candidate video. Can do.

  An eleventh aspect of the invention is an in-facility person search support method for causing an information processing apparatus to perform a process of supporting a search for a person who is separated from a companion in a facility based on images taken by a plurality of cameras in the facility. In accordance with the input operation of the user, based on the step of setting the search condition for specifying the search target person, the tracking information acquired by the tracking process for the video taken inside the facility, and the search condition, A step of searching for a person who is likely to be a search target person from among persons staying in the room, and a predetermined number of images of a person who is highly likely to be a search target person based on the search result in this step Extracting and presenting those videos as search target candidate videos, and in response to a user operation to select the search target candidate videos in which the search target person is captured, the selected A step of setting a person corresponding to a candidate video to be searched as a search target, and acquiring a latest tracking position of the person set as the search target based on the tracking information, and searching for the latest tracking position And a step of presenting as the latest whereabouts of the target person.

  According to this, similarly to the first invention, even when an error occurs in the tracking information, the user can surely present the latest location of the search target person by selecting an appropriate search target candidate video. Can do.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the description of the present embodiment, terms having similar meanings of “tracking” and “tracking” are used. This is merely used for convenience of explanation, but “tracking” is mainly used. This is used when there is a strong connection with the action of the observer, and “tracking” is used when the connection with the processing mainly performed by the apparatus is strong.

  FIG. 1 is an overall configuration diagram of an in-facility person search support system according to the present embodiment.

  This in-facility person search support system is constructed for large-scale commercial facilities such as department stores and shopping centers, and includes a plurality of cameras 1, a recorder (video storage means) 2, a PC (facility). Internal person search support device) 3 and in-camera tracking processing device 4.

  The camera 1 is installed at an appropriate place in the facility, the inside of the facility (monitoring area) is photographed by the camera 1, and the image of the facility photographed by the camera 1 is recorded on the recorder 2.

  Connected to the PC 3 are an input device 6 such as a mouse for a monitor to perform various input operations, and a monitor (display device) 7 for displaying a monitoring screen. This PC 3 is installed in the security room of the facility, etc., and the supervisor (guard) can view the video in the facility captured by the camera 1 in real time on the monitoring screen displayed on the monitor 7, In addition, it is possible to browse the video in the past facility recorded in the recorder 2.

  Note that a monitor (not shown) is also connected to the PC 11 provided in the headquarters, and the video in the facility photographed by the camera 1 is viewed in real time, and the video in the past facility recorded in the recorder 2 is browsed. The situation in the facility can be confirmed at the headquarters.

  In the in-camera tracking processing device 4, a person (moving body) detected from the video of the camera 1 is tracked, and processing for generating in-camera tracking information for each person is performed. At this time, in addition to acquiring the position information for each person, information on the characteristics of the person, that is, the attributes of the person (such as age and gender) and the characteristics on the appearance of the person (such as the height and the color of the upper and lower clothes) Also get information. For this in-camera tracking processing, a known image recognition technique (such as a person detection technique and a person recognition technique) may be used.

  In the present embodiment, the in-camera tracking processing device 4 always performs the in-camera tracking processing independently of the PC 3. However, the in-camera tracking processing device 4 may execute the tracking processing in response to an instruction from the PC 3. Good. Further, in the in-camera tracking processing device 4, it is desirable to perform tracking processing for all persons detected from the video. However, tracking is limited to a person designated as a tracking target and a person highly related to the tracking target. Processing may be performed.

  Next, the installation situation in the facility of the camera 1 shown in FIG. 1 will be described. FIG. 2 is a plan view showing the installation status of the camera 1 in the facility.

  In the facility (monitoring area), a passage is provided between the product display areas, and a plurality of cameras 1 are installed so as to mainly photograph the passage.

  When the person moves through the passage in the facility, the person is photographed by one or a plurality of cameras 1, and the photographing of the person is taken over by the next camera 1 as the person moves. At this time, the camera 1 that takes over the photographing of the person is limited by the form of the passage in the facility and the photographing area of the camera 1, and in this embodiment, the camera that is in a relation that the photographing of the person is taken over has a cooperative relationship. It is called a camera. Information relating to the cooperation relationship between the cameras is set in advance and held in the PC 3 as camera cooperation information. In addition, the information regarding the cooperation relationship between the cameras is obtained in preparation for changing the number of cameras 1 and the installation location, and the installation information of each camera 1 is individually acquired by the PC 3 when the system is started up. Information may be updated.

  Next, a schematic configuration of the PC 3 shown in FIG. 1 will be described. FIG. 3 is a functional block diagram showing a schematic configuration of the PC 3.

  The PC 3 includes a tracking information storage unit 21, an input information acquisition unit 22, a search condition setting unit 23, a search target search unit 24, a search target candidate presentation unit 25, a search target setting unit 26, and an inter-camera tracking process. Unit 27, latest position presentation unit 28, tracking correction candidate presentation unit 29, confirmation video presentation unit 30, and screen generation unit 31.

  In the tracking information storage unit 21, the in-camera tracking information generated by the in-camera tracking processing device 4 is stored. In the tracking information storage unit 21, the inter-camera tracking information generated by the inter-camera tracking processing unit 27 is stored.

  In the input information acquisition unit 22, processing for acquiring input information based on the input operation is performed in accordance with the input operation of the supervisor using the input device 6 such as a mouse.

  In the search condition setting unit 23, a process for setting a search condition for specifying a search target person is performed in accordance with a user input operation. In the present embodiment, a search condition input screen (see FIG. 6) is displayed on the monitor 7, and an image obtained by photographing a companion (search requester) as a search condition on this search condition input screen, the characteristics of the search target person Information, specifically age, gender, and appearance characteristics (height, top and bottom clothing colors, etc.), and information about stray time and place.

  In the search target search unit 24, based on the tracking information (in-camera tracking information) stored in the tracking information storage unit 21 and the search conditions set by the search condition setting unit 23, the search target search unit 24 A process of searching for a person who is highly likely to be a search target person is performed. The tracking information storage unit 21 stores information about the characteristics of each person staying in the facility. The information about the characteristics of each person is compared with the search condition set by the search condition setting unit 23. Thus, the high possibility of being a search target person is determined.

  In this process, first, a face image is cut out from the captured image of the accompanying person acquired as the search condition, and the same based on the similarity between the accompanying person's face image and the face image of each person staying in the facility. Person verification is performed to determine whether the person is a person, and person verification based on the color information (clothing color) of the accompanying person is performed. By this person verification, a person who is highly likely to be an accompanying person is extracted. . Here, priority is given to the similarity of the companion's face image, and the companion's video is extracted, but the present invention is not limited to this, and other feature information such as clothes color information is prioritized. The video of the accompanying person may be extracted.

  Next, information on the characteristics of the search target person (age, gender, height, and the color of the upper and lower clothes) acquired as search conditions for a person who is in the vicinity of a person who is likely to be a companion A person who has a high possibility of being a search target person is extracted by person verification based on the above. At this time, the search target is narrowed down based on the separation location and separation time acquired as the search conditions. That is, the search is performed with priority on the person detected from the video of the camera 1 corresponding to the separation place and the camera 1 in the vicinity thereof, and the search is performed with priority on the person detected near the separation time.

  The search target candidate presenting unit 25 extracts a predetermined number of videos of a person who is highly likely to be a search target person from the top based on the search result of the search target search unit 24, and these videos are searched for candidate videos. Is presented. In the present embodiment, a search target selection screen (see FIG. 7) for displaying a search target candidate video is displayed on the monitor 7, and a video in which a search target person is captured from the candidate video displayed on the search target selection screen. Let the user select

  The candidate video presented by the search target candidate presentation unit 25 is the latest tracking video for each person who is likely to be a search target person, that is, a person who is highly likely to be a search target person, and a companion It is recommended to use the latest video that has been detected at the same time as a person who is likely to be a (search requester). In this case, the correct candidate video showing the search target is accompanied by the search target. It shows the situation just before the break.

  In the search target setting unit 26, the selected search target is selected in response to a user's operation for selecting a candidate video in which the search target person is shown from the search target candidate videos presented by the search target candidate presentation unit 25. A process of setting a person corresponding to the candidate video as a search target is performed.

  The inter-camera tracking processing unit 27 includes a link score calculation unit 41 and a tracking information generation unit 42.

  The link score calculation unit 41 acquires in-camera tracking information related to each camera 1 from the tracking information accumulation unit 21, and the persons detected and tracked in the in-camera tracking processing related to each camera 1 may be the same person. A link score (evaluation value) representing the height is calculated. In this process, the link score is calculated based on the person detection time (frame shooting time), the person detection position, the person moving speed, the color information of the person image, and the like.

  The tracking information generating unit 42 has the highest link score among the people tracked by the in-camera tracking related to the camera 1 that is in the cooperative relationship, starting from the person set as the search target by the search target setting unit 26. That is, a process is performed in which a person who is most likely to be the same person is sequentially selected for each camera 1 and information that associates those persons as the same person is generated.

  Specifically, first, the person with the highest link score is selected from the persons tracked by the in-camera tracking of the camera 1 that is linked to the camera 1 that has captured the candidate video for the search target selected by the user. Next, the person with the highest link score is selected from the persons tracked by the in-camera tracking of the camera 1 that is linked to the camera 1 that photographed the selected person, and such a person is selected. Repeat for each linked camera 1.

  Information on the link score calculated by the link score calculation unit 41 of the inter-camera tracking processing unit 27 and information on the correspondence between the persons generated by the tracking information generation unit 42 are stored in the tracking information storage unit 21 as inter-camera tracking information. Accumulated.

  The latest position presentation unit 28 acquires the latest tracking position of the person set as the search target by the search target setting unit 26 based on the tracking information (inter-camera tracking information) stored in the tracking information storage unit 21. A process of presenting the latest tracking position as the latest location of the search target person is performed. In particular, the latest position presentation unit 28 presents position information regarding the latest location of the search target person, and the latest tracking video corresponding to the latest tracking position, that is, the latest video in which the person has been detected in the tracking process. And time information related to the time when the search target person stays at the latest location (tracking video shooting time). Furthermore, the latest position presentation unit 28 presents a live video (current video) of a camera that captures the latest tracking video at the same time as the latest tracking video. In the present embodiment, a whereabouts display screen (see FIG. 8) on which the latest tracking video and live video are displayed side by side is displayed on the monitor 7.

  In the tracking correction candidate presenting unit 29, when there is an error in the latest tracking video presented by the latest position presenting unit 28, that is, when the search target person is not reflected in the latest tracking video, the tracking information storage unit 21 A person who has the next highest possibility of being the same as the person set as the search target based on the accumulated tracking information (inter-camera tracking information) next to the person corresponding to the latest tracking video presented by the latest position presentation unit 28 A predetermined number of these videos are extracted from the top, and processing for presenting these videos as tracking correction candidate videos is performed. In the present embodiment, a candidate selection screen (see FIG. 9) for displaying a tracking correction candidate video is displayed on the monitor 7, and a video showing the search target person is displayed among the candidate videos displayed on the candidate selection screen. To select.

  Here, the latest position presentation unit 28 presents the video of the person most likely to be the same as the person set as the search target based on the link score, and the tracking correction candidate presentation unit 29 presents the latest position. A predetermined number of persons with a higher link score are selected from the top among persons whose link score is lower than that of the person corresponding to the image presented by the unit 28, and images of those persons are presented.

  In this way, when one of the tracking correction candidate videos presented by the tracking correction candidate presenting unit 29 is selected by the user, the latest position presenting unit 28 displays the latest of the person corresponding to the candidate video selected by the user. Is acquired, and the latest tracking position is presented as the latest location of the search target person, that is, the whereabouts display screen on which the latest tracking video corresponding to the latest tracking position is displayed (see FIG. 8). ) Is displayed on the monitor 7.

  The tracking correction candidate video presented by the tracking correction candidate presentation unit 29 may be the latest tracking video of each person. In this case, the candidate video selected by the user is updated as the latest tracking video. Presented by the position presentation unit 28.

  In the confirmation video presenting unit 30, the tracking correction candidate presenting unit 29 presents an error in all of the tracking correction candidate videos, that is, if the search target person is not reflected in all of the tracking correction candidate videos. Based on the tracking information stored in the information storage unit 21, the camera from the candidate video when the search target setting unit 26 sets the person as the search target to the latest tracking video presented by the latest position presentation unit 28 A process of presenting each video as a confirmation video is performed. In the present embodiment, a tracking process confirmation screen (see FIG. 10) in which confirmation videos are arranged and displayed in order of shooting time is displayed on the monitor 7, and the user is allowed to select a confirmation video that does not include the search target person. When there are a plurality of confirmation videos with errors, the user is allowed to select the confirmation video with the oldest shooting time.

  In this way, when one of the confirmation videos presented by the confirmation video presentation unit 30 is selected by the user, the tracking correction candidate presentation unit 29 searches based on the tracking information stored in the tracking information storage unit 21. A predetermined number of images of the next highest person after the person corresponding to the confirmation image selected by the user (link score) are extracted from the top, and the images are tracked. Present as a candidate video for correction.

  The screen generation unit 31 generates a screen to be displayed on the monitor 7. Specifically, the screen generation unit 31 generates screen information of a search condition input screen (see FIG. 6) in response to an instruction from the search condition setting unit 23. The screen information of the search target selection screen (see FIG. 7) is generated according to the instruction from the search target candidate presentation unit 25, and the whereabouts display screen (see FIG. 8) according to the instruction from the latest position presentation unit 28. Information is generated, screen information of a candidate selection screen (see FIG. 9) is generated according to an instruction from the tracking correction candidate presentation unit 29, and a tracking process confirmation screen (FIG. 10) is generated according to an instruction from the confirmation video presentation unit 30. (See) screen information is generated.

  Each unit of the PC 3 shown in FIG. 3 is realized by causing the processor (CPU) of the PC 3 to execute an in-facility person search support program (instruction) stored in a memory such as an HDD. These programs are pre-installed in the PC 3 as an information processing device and configured as a dedicated device, or recorded in an appropriate program recording medium as an application program that operates on a predetermined OS, and via a network, It may be provided to the user.

  Next, each screen displayed on the monitor 7 shown in FIG. 1 and processing performed in each part of the PC 3 in accordance with the operation of the supervisor performed on each screen will be described. FIG. 4 is an explanatory diagram showing the transition state of the screen displayed on the monitor 7. FIG. 5 is a flowchart showing a procedure of processing performed in each part of the PC 3 in accordance with the operation of the supervisor performed on each screen.

  First, when an operation for starting the in-facility person search support process is performed on the PC 3, a search condition input screen for inputting a search condition for specifying a search target person is displayed on the monitor 7 (ST101). When the user inputs a search condition on this search condition input screen (Yes in ST102), the search condition setting unit 23 performs a process of setting the search condition based on the input information. Then, the search target search unit 24 performs a process of searching for the search target person from the persons staying in the facility based on the search condition (ST103). Next, in the search target candidate presentation unit 25, a process for displaying a search target selection screen for displaying the search target candidate video on the monitor 7 is performed (ST104).

  Next, on the search target selection screen, when the user performs an operation for selecting a candidate video in which the search target person is captured (Yes in ST105), the search target setting unit 26 selects a person corresponding to the candidate video selected by the user. Is set as a search target (ST106). Then, in the latest position presentation unit 28, a process for causing the monitor 7 to display a location display screen for displaying the latest tracking video of the person set as the search target is performed (ST107).

  When there is an error in the latest tracking video displayed on the whereabouts display screen, that is, when the search target person is not shown in the latest tracking video, an operation for displaying the candidate selection screen is performed by the user (in ST108). Yes), the tracking correction candidate presenting unit 29 performs a process of displaying a candidate selection screen for displaying a candidate video for tracking correction on the monitor 7 (ST109).

  When a video showing the search target person is found in the candidate video displayed on the candidate selection screen, an operation for selecting the candidate video is performed by the user (Yes in ST110). A process of displaying on the monitor 7 a whereabouts display screen for displaying the latest tracking image of the person corresponding to the selected candidate image is performed (ST107).

  On the other hand, when a video showing the search target person is not found in the candidate video displayed on the candidate selection screen, an operation for displaying a tracking process confirmation screen is performed by the user (Yes in ST111), and a confirmation video presentation unit At 30, a process for displaying a tracking process confirmation screen on the monitor 7 for displaying the confirmation videos in order of photographing time is performed (ST112).

  Then, the user performs an operation of selecting an erroneous confirmation video from the confirmation video displayed on the tracking process confirmation screen, that is, a confirmation video that does not include the search target person (Yes in ST113). In the correction candidate presenting unit 29, a process of displaying a candidate selection screen for displaying the tracking correction candidate video on the monitor 7 is performed (ST109).

  The subsequent processing (ST107 to ST113) from the display of the whereabouts display screen can be repeated until the latest tracking video showing the search target person is displayed on the whereabouts display screen. The user can grasp the latest location of the search target person by displaying the video on the location display screen.

  Hereinafter, each screen shown in FIG. 4 will be described in detail.

  First, the search condition input screen shown in FIG. 4 will be described. FIG. 6 is an explanatory diagram showing a search condition input screen displayed on the monitor 7.

  In this search condition input screen, the user inputs search conditions for specifying the search target person. This search condition input screen includes information regarding the date and time, the location, the characteristics of the search target person, and the search requester. Each input part 51-54 of this picked-up image is provided.

  In the date and time input unit 51, the user inputs the time (separation time) when the search requester is separated from the search target person (accompanied person). In the place input unit 52, the user inputs a place (separation place) where the search requester is separated from the search target person. The location input section 52 is provided with a “map search” button. When this button is operated, a map search screen (not shown) is displayed, and a separation location is displayed on the map displayed on the map search screen. Can be specified.

  In the information input unit 53 relating to the characteristics of the search target person, the user inputs information related to the attributes of the person (age and gender) and the characteristics on the appearance of the person (height and upper and lower clothing colors). The upper and lower clothing colors can be selected from a pull-down menu. The input unit 53 is also provided with a “color palette designation” button. When this button is operated, a color palette (not shown) is displayed, and a clothing color can be selected in detail from a pull-down menu. .

  What is necessary is just to listen to a search requester and input the information input into these input parts 51-53. Regarding clothing colors, it is difficult to express the clothing color by color name, especially for children, but the color name is inaccurate, but by selecting the color directly from the pull-down menu or color palette, the clothing color is accurate. You can type well.

  The search client's photographed image input unit 54 is provided with “image capture”, “image capture”, and “redo” buttons for inputting an image of the search client. The “image capture” button is used when a captured image captured by the camera is captured to the PC 3, and the “image capture” button is used when the PC 3 itself has a camera like a tablet terminal. The “Redo” button is used when a photographed image is input again. When the photographed image of the search requester is input by operating these buttons, the photographed image of the search requester is displayed on the input unit 54.

  In this search condition input screen, when the user inputs necessary information to each of the input sections 51 to 54 and operates the “search” button, the search condition setting section 23 sets the search conditions based on the input information. Processing is performed.

  Next, the search target selection screen shown in FIG. 4 will be described. FIG. 7 is an explanatory diagram showing a search target selection screen displayed on the monitor 7.

  This search target selection screen displays a plurality of images of a person who is likely to be a search target person as search target candidate videos, and allows the user to select a video image of the search target person. On the screen, candidate videos are displayed side by side in descending order of the possibility of being a search target person. The candidate video displayed on the search target selection screen is the video immediately before the search requester (accompanied person) is separated from the search target person. The client is shown.

  In this search target selection screen, when the user selects a candidate video in which the search target is shown and operates the “decision” button, the search target setting unit 26 searches for a person corresponding to the candidate video selected by the user. Processing to set the target is performed. In this example, the thumbnail image of the candidate video is selected and determined, but a person frame is displayed for each person shown in the thumbnail image, and after selecting the person frame corresponding to the search target person, the determination button is pressed. You may make it operate. In this case, by performing highlighting such as changing the display color or line type of the selected person frame, the user can check whether the search target person is correctly selected. In addition to the search target selection screen, if the person frame for each person and the person frame corresponding to the search target person are highlighted on various display screens, the selection status of the search target person is always displayed. Can be identified.

  Next, the whereabouts display screen shown in FIG. 4 will be described. FIG. 8 is an explanatory diagram showing a location display screen displayed on the monitor 7.

  This location display screen displays the latest tracking video corresponding to the latest tracking position of the person set as the search target and presents the latest location of the search target to the user. 1 includes a tracking video display unit 61, a position / time information display unit 62, a playback operation unit 63, a live video display unit 64, and a camera switching unit 65.

  The tracking video display unit 61 displays the latest tracking video corresponding to the latest tracking position of the person set as the search target, that is, the latest video in which the person is detected in the tracking process. When the tracking of the person is continued, the current video is displayed, and when the tracking of the person is completed, the video corresponding to the position where the tracking is completed is displayed. In addition, if tracking of the search target person fails in the middle and a change with another person occurs, an image of a person different from the search target person is displayed. The tracking video is displayed as a moving image. When the search target person moves to the imaging area of another camera 1 and is detected from the video of the camera 1, the tracking video is switched to that of the other camera 1.

  In the position / time information display unit 62, the latest tracking video shooting time is displayed as time information related to the time when the search target person stays in the latest location, and the position information regarding the latest search target location is as follows: The name (camera ID) of the camera that captures the latest tracking video is displayed.

  The playback operation unit 63 performs operations related to playback of the video displayed on the tracking video display unit 61. The playback operation unit 63 includes buttons for playback, reverse playback, stop, fast forward, and rewind. The video can be efficiently browsed by operating these buttons.

  The live video display unit 64 displays the live video (current video) of the same camera 1 as the latest tracking video displayed on the tracking video display unit 61. When the tracking of the person continues, the same video is displayed on the tracking video display unit 61 and the live video display unit 64, and when the tracking of the person fails, the tracking video display unit 61 and Different images are displayed on the live image display unit 64.

  In the live video display unit 64, up / down / left / right direction indication marks (operation unit) 66 for switching the camera 1 are displayed on the live video, and by operating the direction indication mark 66, the direction is displayed. It is possible to switch to the image of the camera 1 located in the direction corresponding to the instruction mark 66.

  The camera switching unit (operation unit) 65 can switch the camera 1 that displays the video on the live video display unit 64 using a pull-down menu.

  Next, the candidate selection screen shown in FIG. 4 will be described. FIG. 9 is an explanatory diagram showing a candidate selection screen displayed on the monitor 7.

  This candidate selection screen is set as a search target when there is an error in the latest tracking video displayed on the whereabouts display screen (see FIG. 8), that is, when the search target person is not shown in the latest tracking video. Displaying multiple images of the person who has the highest possibility of being the same as the person next to the person in the latest tracking video as candidate videos for tracking correction, and allowing the user to select the video in which the search target is shown In this candidate selection screen, candidate videos for tracking correction are displayed side by side in order of shooting time.

  In this candidate selection screen, when the user selects a candidate video in which the search target person is photographed and operates the “decision” button, the latest position presentation unit 28 displays the latest of the person corresponding to the candidate video selected by the user. Processing for acquiring the tracking position and displaying the whereabouts display screen (see FIG. 8) on which the latest tracking video corresponding to the latest tracking position is displayed on the monitor 7 is performed.

  In addition, in the candidate selection screen, when a candidate video showing the search target person is not found, the user operates the “route confirmation” button to change to the tracking process confirmation screen (see FIG. 10).

  Next, the tracking process confirmation screen (time series tracking screen) shown in FIG. 4 will be described. FIG. 10 is an explanatory diagram showing a tracking process confirmation screen displayed on the monitor 7.

  This tracking process confirmation screen is displayed when a search target person is not shown in the candidate video displayed on the candidate selection screen (see FIG. 9) and a person is selected as a search target on the search target selection screen (see FIG. 7). The video for each camera 1 from the video to the latest tracking video displayed on the whereabouts display screen (see Fig. 8) is displayed as a confirmation video, and the confirmation video that does not show the search target is selected by the user On this tracking process confirmation screen, confirmation images for each camera 1 are displayed side by side in order of photographing time. Each confirmation video displays a video shooting time and a camera name (camera ID). This tracking process confirmation screen is displayed by operating the “route confirmation” button on the candidate selection screen (see FIG. 9).

  If the latest tracking video displayed on the whereabouts display screen (see FIG. 8) is from another person, that is, if there is an error in the tracking information, this tracking process confirmation screen displays the confirmation video from a certain time onward. All the confirmation videos are replaced with those of another person, and the user is allowed to select the first confirmation video that has been replaced with that of another person. Thereby, it is possible to identify a video in which an error has occurred during the tracking process.

  In this tracking process confirmation screen, when the user selects a confirmation video with an error and operates the “decision” button, the tracking correction candidate presenting unit 29 may be the same as the person set as the search target. A candidate selection screen (see FIG. 9) for extracting a predetermined number of videos of the next highest person from the person corresponding to the confirmation video selected by the user from the top and displaying those videos as tracking correction candidate videos. Processing to be displayed on the monitor 7 is performed.

  Next, a map tracking screen which is another example of the tracking process confirmation screen shown in FIG. 10 will be described. FIG. 11 is an explanatory diagram showing a map tracking screen displayed on the monitor 7.

  On the time-series tracking screen shown in FIG. 10, the confirmation videos for each camera 1 are displayed side by side in order of shooting time. On the map tracking screen shown in FIG. 11, a map image showing the layout (facility status) in the facility A plurality of video display windows (video display frames) 72 for displaying the confirmation video for each camera 1 are arranged on 71 corresponding to the actual arrangement status of the camera 1. In addition, since the video display window 72 is connected with arrows in the order of shooting times, confirmation videos for each camera 1 can be checked in the order of shooting times.

  The map tracking screen may be displayed instead of the time series tracking screen shown in FIG. 10, but the user can select the map tracking screen and the time series tracking screen as appropriate. Also good.

  12 and 13 are explanatory diagrams showing a map tracking screen when the tracking process has failed. The map tracking screen shown in FIG. 11 is a case where the tracking process is successful, and the search target person is continuously captured from the confirmation video of the camera 1 that started tracking to the confirmation video of the camera 1 at the latest tracking position. ing. On the other hand, as shown in FIG. 12 and FIG. 13, when the tracking process has failed, the search target person is replaced by another person in the confirmation video of the camera 1 in the middle of tracking, and where the search target person has failed to be tracked I can understand. In the example shown in FIG. 12 and FIG. 13, a confirmation video is displayed on the map image 71 of each floor of the facility, and tracking of the search target person has failed on the third floor shown in FIG. .

  As described above, in the present embodiment, the search condition setting unit 23 sets the search condition for specifying the search target person in accordance with the user's input operation, and the search target search unit 24 images the inside of the facility. Based on the tracking information acquired by the tracking processing for the video and the search condition, a person who is highly likely to be a search target person is searched from the persons staying in the facility, and the search target candidate presentation unit 25 searches for this search. Based on the search result of the target search unit 24, a predetermined number of videos of a person who is highly likely to be a search target are extracted from the top, and these videos are presented as search target candidate videos, and the search target setting unit 26 , The person corresponding to the selected candidate video for the search target is set as the search target in accordance with the operation of the user who selects the candidate video for the search target. In section 28, based on the tracking information, obtain the latest tracking position of the person is set to search the object, the latest tracking position, and shall be presented as the latest location of the search subject.

  According to this, in order to present a plurality of images of a person who is likely to be a search target candidate video to be searched and allow the user to select, the correct person is selected as the search target even when the tracking information accuracy is low. It is set, and the latest location of the search target person can be surely presented.

  In the present embodiment, the latest position presentation unit 28 presents position information related to the latest location of the search target person, the latest tracking video corresponding to the latest tracking position, and the search target person to the latest location. Time information on the time of stay was presented. According to this, since the location information regarding the whereabouts is presented, the user can grasp the whereabouts of the search target person, and furthermore, the latest tracking video of the search target person is presented, so the latest search target person's latest information is presented. The user can grasp the situation and the time information about the time when the search target person stays in the latest location is presented. The user can grasp whether time has passed.

  In the present embodiment, the latest position presentation unit 28 presents a live video of a camera that captures the latest tracking video along with the latest tracking video. According to this, since the shooting time of the latest tracking video is relatively old, that is, even if a long time has elapsed since the tracking of the search target person was completed, the live video is presented, so the latest location If the user can grasp the current situation and the search target person is not moving, the search target person can be found in the live video.

  In the present embodiment, the latest position presentation unit 28 displays a live video on the monitor 7 and switches the live video camera on the live video display screen and a direction indication mark (operation unit). Is to be displayed. As a result, the search target person moves to the shooting area of the camera 1 near the camera 1 that captures the latest tracking video, and the live video camera 1 is switched when the search target is not shown in the live video. With this, you can find out who to search for.

  In the present embodiment, in the tracking correction candidate presenting unit 29, when there is an error in the latest location of the search target person presented by the latest position presenting unit 28, the person set as the search target based on the tracking information A predetermined number of images of the next highest person after the person who presented the latest whereabouts in the latest position presenting unit 28 are extracted from the top and presented as candidate images for tracking correction. In the latest position presentation unit 28, the latest tracking position of the person corresponding to the selected candidate video for tracking correction is acquired according to the operation of the user who selects the candidate video for tracking correction in which the search target person is captured. The latest tracking position is presented as the latest location of the search target person.

  According to this, when there is an error in the latest location of the search target person presented by the latest position presenting unit 28, that is, when there is an error in the tracking information after setting the search target person as the search target, In order to present a plurality of candidate videos for correction and allow the user to select them, the tracking information is appropriately corrected, and the latest location of the search target person can be reliably presented.

  Further, in the present embodiment, in the confirmation video presentation unit 30, when all of the tracking correction candidate videos presented by the tracking correction candidate presentation unit 29 are erroneous, the search target setting unit 26 sets a person as a search target. The video for each camera from the search target candidate video to the latest tracking video presented by the latest position presentation unit 28 is presented as a confirmation video, and the tracking correction candidate presentation unit 29 confirms that there is an error. Depending on the operation of the user who selects the video, based on the tracking information, the video of the person who has the next highest possibility of being the same as the person set as the search target is next to the person corresponding to the confirmation video selected by the user. Then, a predetermined number is extracted from the upper level, and these videos are presented as candidate videos for tracking correction.

  According to this, the video for each camera 1 from the candidate video when set as the search target to the latest tracking video is presented as the confirmation video, and there is a confirmation video with an error, that is, the search target person is not shown. By allowing the user to select the confirmation video, the location where there is an error in the tracking information becomes clear, the candidate video corresponding to the location where there is an error in the tracking information is presented, and the appropriate candidate video is selected by the user By doing so, it is possible to surely correct a portion having an error in the tracking information.

  In the present embodiment, the confirmation video presentation unit 30 displays a time series tracking screen in which a plurality of confirmation videos are arranged in order of shooting time on the monitor 7. Accordingly, since the user can sequentially confirm the confirmation videos arranged in the order of the shooting times, it is possible to find out the confirmation video having an error, so that the confirmation video can be confirmed quickly.

  In the present embodiment, the confirmation video presenting unit 30 monitors a map tracking screen in which a plurality of confirmation videos are arranged on the map image indicating the facility state in accordance with the actual arrangement state of the camera 1. 7 to be displayed. According to this, the user can easily grasp where the tracking of the search target person has failed by sequentially confirming the confirmation video while viewing the actual arrangement state of the camera 1.

  Further, in the present embodiment, the search condition setting unit 23 sets, as the search condition, an image obtained by photographing a companion and information on the characteristics of the search target person as the search condition. Thus, the search target person is searched based on the image of the companion and the information related to the characteristics of the search target person, so that the search accuracy of the search target person can be improved.

  As mentioned above, although this invention was demonstrated based on specific embodiment, these embodiment is an illustration to the last, Comprising: This invention is not limited by these embodiment. The constituent elements of the in-facility person search support device, the in-facility person search support system, and the in-facility person search support method according to the present invention described in the above embodiment are not necessarily all required, and at least the scope of the present invention As long as they do not deviate from the above, it is possible to make appropriate selections.

  For example, in the above-described embodiment, an example of a commercial facility such as a department store or a shopping center has been described, but the present invention can also be applied to a facility other than a commercial facility such as a public facility.

  In the present embodiment, the latest position presentation unit 28 uses the latest tracking video of the search target person, the live video of the camera 1 that captures the latest tracking video, and the time when the search target person stays in the latest location. Although the information and the position information regarding the latest location of the search target person are presented, only the position information (for example, which sales floor on which floor) may be presented. In addition to displaying the whereabouts with characters, the whereabouts may be displayed on a map.

  In the above embodiment, as shown in FIGS. 1 and 3, the in-camera tracking processing is performed by the in-camera tracking processing device 4, and the inter-camera tracking processing and the in-facility person search support processing are performed by the PC 3. Although an example has been described, a configuration in which in-camera tracking processing is also performed by the PC 3 may be employed. A configuration in which the camera tracking processing unit is provided in the camera 1 is also possible. Further, all or part of the inter-camera tracking processing unit 27 can be configured by a tracking processing device different from the PC 3.

  In the above embodiment, as shown in FIG. 2, the camera 1 is a box-type camera with a limited viewing angle. However, the present invention is not limited to this, and an omnidirectional camera capable of shooting a wide range is used. You can also.

  In this embodiment, the tracking process and the in-facility person search support process are performed by an apparatus installed in the facility. However, these necessary processes are performed in the headquarters as shown in FIG. You may make it make the provided computer 11 and the cloud computer 12 which comprises a cloud computing system perform. In addition, the necessary processing is shared by a plurality of information processing apparatuses, and information is transferred between the plurality of information processing apparatuses via a communication medium such as an IP network or a LAN, or a storage medium such as a hard disk or a memory card. It may be. In this case, the tracking support system is configured by a plurality of information processing apparatuses that share necessary processing.

  In particular, in the system configuration including the cloud computer 12, in addition to the PCs 3 and 11 provided in the facility and the headquarters, necessary information is obtained by a portable terminal such as the smartphone 13 and the tablet terminal 14 connected to the cloud computer 12 via a network. It is preferable that the information can be displayed, so that necessary information can be confirmed at any place such as a place to go besides the facility and the headquarters.

  In the above-described embodiment, the recorder 2 that accumulates the video of the camera 1 is installed in the facility. However, the processing necessary for the person search support in the facility is performed on the PC 11 and the cloud computer 12 installed in the headquarters. In such a case, the video from the camera 1 may be transmitted to the headquarters or the operating facility of the cloud computing system, and the video from the camera 1 may be stored in a device installed there.

  The in-facility person search support device, the in-facility person search support system, and the in-facility person search support method according to the present invention allow the user to appropriately perform selection and correction operations even when the tracking information accuracy is low. In-facility person search that has the effect of reliably presenting the latest location of the target person and supports the search for persons who have been separated from their companions in the facility based on images taken by multiple cameras in the facility. It is useful as a support device, a facility person search support system, a facility person search support method, and the like.

1 Camera 2 Recorder 3 PC (In-facility person search support device)
4 In-camera tracking processing device 6 Input device 7 Monitor 11 PC
DESCRIPTION OF SYMBOLS 12 Cloud computer 13 Smartphone 14 Tablet terminal 21 Tracking information storage part 22 Input information acquisition part 23 Search condition setting part 24 Search target search part 25 Search target candidate presentation part 26 Search target setting part 27 Inter-camera tracking process part 28 Latest position presentation part 29 Tracking correction candidate presentation unit 30 Confirmation video presentation unit 31 Screen generation unit

Claims (11)

An in-facility person search support device that supports searching for a person who is separated from a companion in a facility based on images taken by a plurality of cameras in the facility,
A search condition setting unit for setting a search condition for specifying a search target person according to a user input operation;
A search target search unit for searching for a person who is highly likely to be a search target person among the persons staying in the facility based on the tracking information acquired by the tracking process for the video imaged in the facility and the search condition; ,
Based on the search result of the search target search unit, a predetermined number of videos of a person who is likely to be a search target are extracted, and a search target candidate presenting unit that presents those videos as search target candidate videos,
A search target setting unit that sets a person corresponding to the selected candidate video for the search target as a search target in response to a user operation for selecting the candidate video for the search target in which the search target person is captured,
Based on the tracking information, obtain the latest tracking position of the person set as the search target, and present the latest tracking position as the latest location of the search target person,
An in-facility person search support device characterized by comprising:   The latest position presentation unit presents position information related to the latest location of the search target person, and includes the latest tracking video corresponding to the latest tracking position and time information related to the time when the search target person stays in the latest location. The in-facility person search support device according to claim 1, wherein:   The in-facility person search support device according to claim 2, wherein the latest position presentation unit presents a live video of a camera that captures the latest tracking video together with the latest tracking video.   The said latest position presentation part displays the said live image | video on a display apparatus, and displays the operation part which switches the camera of the said live image | video on the display screen of the said live image | video. In-facility person search support device. Furthermore, when there is an error in the latest location of the search target person presented by the latest position presentation unit, the possibility that the search target person is the same as the person set as the search target is based on the tracking information. A tracking correction candidate presentation unit that extracts a predetermined number of videos of the next highest person next to the person who presented the latest location in the presentation unit and presents those videos as tracking correction candidate videos,
The latest position presenting unit obtains the latest tracking position of the person corresponding to the selected candidate video for tracking correction in response to a user operation for selecting the candidate video for tracking correction in which the search target person is captured. The in-facility person search support device according to any one of claims 1 to 4, wherein the latest tracking position is presented as the latest location of the search target person. Further, when there is an error in all of the tracking correction candidate videos presented by the tracking correction candidate presentation unit, from the search target candidate video when the person is set as a search target in the search target setting unit, From the latest tracking video presented by the latest position presentation unit , based on the tracking information, a video for each camera in which a person having the same possibility as the person set as the search target is captured. It has a confirmation video presentation part to present as a confirmation video,
Based on the tracking information, the tracking correction candidate presentation unit may select the possibility that the tracking correction candidate presentation unit is the same as the person set as the search target based on the tracking information. 6. The in-facility person according to claim 5, wherein a predetermined number of videos of a person next to the person corresponding to the confirmation video are extracted and presented as candidate videos for the tracking correction. Search support device.   The in-facility person search support device according to claim 6, wherein the confirmation video presenting unit causes a display device to display a time-series tracking screen in which a plurality of the confirmation videos are arranged in order of shooting time.   The confirmation video presenting unit causes the display device to display a map tracking screen in which a plurality of the confirmation videos are arranged on the map image indicating the state of the facility according to the actual arrangement state of the camera. The in-facility person search support device according to claim 6, wherein the device is a person search support device.   9. The search condition setting unit according to claim 1, wherein the search condition setting unit sets, as the search condition, an image obtained by photographing a companion and information on characteristics of a search target person as a search condition. In-house person search support device. An in-house person search support system that supports searching for a person who is separated from a companion in a facility based on images taken by a plurality of cameras in the facility,
The camera for photographing the inside of the facility;
A display device for displaying video for each camera;
A plurality of information processing devices;
Have
Any of the plurality of information processing devices
A search condition setting unit for setting a search condition for specifying a search target person according to a user input operation;
A search target search unit for searching for a person who is highly likely to be a search target person among the persons staying in the facility based on the tracking information acquired by the tracking process for the video imaged in the facility and the search condition; ,
Based on the search result of the search target search unit, a predetermined number of videos of a person who is likely to be a search target are extracted, and a search target candidate presenting unit that presents those videos as search target candidate videos,
A search target setting unit that sets a person corresponding to the selected candidate video for the search target as a search target in response to a user operation for selecting the candidate video for the search target in which the search target person is captured,
Based on the tracking information, obtain the latest tracking position of the person set as the search target, and present the latest tracking position as the latest location of the search target person,
In-facility person search support system characterized by comprising An in-facility person search support method for causing an information processing device to perform processing for supporting a search for a person who is separated from a companion in a facility based on images taken by a plurality of cameras in the facility,
Setting a search condition for specifying a search target person in accordance with a user input operation;
Based on the tracking information acquired by the tracking process for the video imaged in the facility and the search condition, searching for a person who is highly likely to be a search target person from the persons staying in the facility;
Based on the search result in this step, extracting a predetermined number of videos of persons who are likely to be search subjects, and presenting those videos as search target candidate videos; and
In response to a user's operation to select the search target candidate video in which the search target person is captured, setting a person corresponding to the selected search target candidate video as a search target;
Obtaining the latest tracking position of the person set as the search target based on the tracking information, and presenting the latest tracking position as the latest location of the search target person;
An in-facility person search support method characterized by comprising:

Images