JP2010067008A - Imaging management system, imaging management method, authentication system, and authentication method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To acquire a target image with high resolution by recognizing one or more objects from a video image of a first imaging means for overall imaging, and controlling a plurality of second imaging means according thereto. <P>SOLUTION: The imaging system includes: a first imaging means; and a plurality of second imaging means, and configured to perform the imaging management of one or more objects included in the video image of the first imaging means. The imaging management system includes: an object specification means for specifying one or more objects included in the video image of the first imaging means; an imaging control means for assigning each of the second imaging means to one or more specified objects, and for controlling each of the second imaging means so as to capture a video image with resolution higher than that of the video image of the first imaging means for the one or more objects; and a high resolution image management means for simultaneously acquiring and managing the images with high resolution of the one or more objects captured by each of the second imaging means. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an imaging management system, an imaging management method, an authentication system, and an authentication method, for example, by recognizing a person from an image of a parent camera that captures the entire image and controlling a plurality of child cameras accordingly, thereby achieving high resolution. It can be applied to a system that acquires and authenticates human images.

  For example, there is an authentication technique for authenticating feature points and feature information such as faces and belongings. As a system using this authentication technology, for example, as described in Patent Document 1 and Patent Document 2, a person who enters or leaves a building is monitored and managed to enter a room or facility that requires security. There are entrance / exit management systems and monitoring systems that manage exits.

  For example, when using authentication technology that authenticates the feature points of the entrance / exit's face in the entrance / exit management system, the entrance / exit person stops in front of the camera, and the camera captures the facial features from the images of each person It is common to acquire and authenticate points, and it is difficult for a person who moves like a pedestrian to recognize the feature points and feature information.

  In order to deal with such a problem, the technology described in Patent Document 1 controls the entry / exit of the entrance / exit to the facility of the entry / exit according to the result of collation of the biometric information of the entry / exit by the biometric collation. In other words, the imaging means captures the face image of the entrance / exit person, and manages the biometric matching result of the entrance / exit person and the face image as the entrance / exit history.

  In the technique described in Patent Document 2, when the face detection unit detects a person's face area from the captured image of the first imaging unit that captures an image including the face of the moving person, and there are a plurality of face areas, However, the registration target is selected from the captured image captured by the second imaging unit. A feature point is a technique in which feature points are extracted based on the face area selected by the clerk and registered in the dictionary as face feature information.

JP 2004-78687 A JP 2007-272811 A

  In the system using the authentication technology as described above, there are the following problems in order to make the authentication technology function effectively.

  There may be a problem that it is difficult to recognize a plurality of persons simultaneously by simply controlling a single camera. For example, the technique described in Patent Document 2 extracts facial feature information from the face area captured by the first imaging means by selecting an attendant, and when a plurality of persons are simultaneously shown in the captured image of the camera, A human face area may not be sufficiently large, and face recognition may not be performed effectively.

  In addition, a high-resolution image is required to acquire human features such as face authentication technology as data and to reliably identify a person based on these feature data.

  As a system using the authentication technology, there are not only the entrance / exit management system as described above but also a monitoring system at a station, an airport, etc., and such a monitoring system has the same problem as above. .

  Therefore, in the present invention, first imaging means for photographing the whole so as to recognize a plurality of objects (for example, a person, an object, etc.) is prepared, and each object is assigned to a plurality of second imaging means. An object of the present invention is to provide an imaging management system, an imaging management method, an authentication system, and an authentication method that acquire a high-resolution target image by control and require a high-definition image such as a face authentication technique.

  In order to solve such a problem, an imaging management system according to a first aspect of the present invention includes a first imaging unit and a plurality of second imaging units that capture a range narrower than the image of the first imaging unit. An imaging management system that performs imaging management of one or a plurality of objects that are reflected in a video of a first imaging means, and (1) a feature of an object to be identified based on a video acquired from the first imaging means A target specifying unit that searches for a feature region including information and specifies one or more targets appearing in the video of the first imaging unit; and (2) one or more targets specified by the target specifying unit. An imaging control unit that assigns each second imaging unit to each of the plurality of objects and controls each second imaging unit so as to capture a higher resolution image than the image of the first imaging unit for one or a plurality of objects; (3) One or more images captured by each second imaging means Characterized in that it comprises a high-resolution image managing means for simultaneously managing to obtain a high resolution image of the subject.

  An imaging management method according to a second aspect of the present invention includes a first imaging unit and a plurality of second imaging units that capture a range narrower than an image of the first imaging unit, and an image of the first imaging unit. An imaging management method for managing imaging of one or a plurality of targets shown in (1), wherein (1) the target specifying means includes feature information of the target to be specified based on the video acquired from the first imaging means A target specifying step of searching for a feature region and specifying one or more objects shown in the image of the first imaging means; and (2) one or more of the imaging control means specified by the target specifying means. An imaging control step of assigning each second imaging means to a target and controlling each second imaging means so as to capture a higher resolution video than the video of the first imaging means for one or a plurality of targets (3) The high-resolution image management means There characterized by having a high resolution image management step of managing simultaneously obtain high resolution images of one or more target captured.

  An authentication system according to a third aspect of the present invention includes a first imaging unit and a plurality of second imaging units that capture a range narrower than the image of the first imaging unit. An authentication system for authenticating one or a plurality of objects shown, wherein (1) a feature area including feature information of an object to be identified is searched based on a video acquired from a first imaging unit; (1) assigning each second imaging means to one or more objects specified by the object specifying means; (2) specifying the one or more objects reflected in the image of one imaging means; Imaging control means for controlling each of the second imaging means so as to pick up an image having a higher resolution than the image of the first imaging means for one or a plurality of objects; and (3) each second imaging means takes an image. Simultaneously acquire high-resolution images of one or more objects Resolution image acquisition means, (4) registration information storage means for storing in advance one or more registration information to be authenticated, and (5) one or more target high resolution images acquired by the high resolution image acquisition means, And an authentication unit that performs a predetermined authentication process based on each registration information of the registration information storage unit.

  An authentication method according to a fourth aspect of the present invention includes a first imaging unit and a plurality of second imaging units that capture a range narrower than the image of the first imaging unit, and the image of the first imaging unit An authentication system for authenticating one or a plurality of objects shown, comprising (1) a registration information storage unit that stores in advance one or a plurality of registration information to be authenticated; A target specifying step of searching for a feature region including feature information of a target to be specified based on a video acquired from one imaging means and specifying one or a plurality of targets shown in the video of the first imaging means; (3) The imaging control means assigns each second imaging means to one or a plurality of objects specified by the object specifying means, and the one or more objects are higher than the video of the first imaging means. Each second imaging means is controlled so as to capture a resolution image. (4) a high-resolution image acquisition step in which the high-resolution image acquisition unit simultaneously acquires a high-resolution image of one or a plurality of objects captured by each second imaging unit; and (5) authentication. The means includes an authentication step of performing a predetermined authentication process based on the high-resolution image of one or a plurality of objects acquired by the high-resolution image acquisition means and each registration information in the registration information storage means. .

  According to the present invention, an image of a high-resolution target is obtained by recognizing one or a plurality of objects from the video of the first imaging unit that captures the entire image and controlling the plurality of second imaging units accordingly. Can be acquired.

(A) First Embodiment Hereinafter, a first embodiment of an imaging management system, an imaging management method, an authentication system, and an authentication method of the present invention will be described with reference to the drawings.

  The first embodiment implements an entrance / exit management system using the present invention, and will be described by exemplifying an embodiment particularly applied to an attendance / exit management system in a company.

(A-1) Configuration of First Embodiment FIG. 2 is an explanatory diagram for explaining an image of camera arrangement in the entrance / exit management system of the first embodiment. In FIG. 2, the camera arrangement at the entrance gate G is shown, but the same configuration can be adopted for the exit gate.

  As shown in FIG. 2, in the entrance / exit management system of the first embodiment, the whole camera 101 that captures the person entering from the entrance gate G and a plurality of (four in FIG. 2) children around the whole camera 101. Cameras 106-1 to 106-4 are arranged. In FIG. 2, an entrance gate G such as a fence is provided so that visitors can enter as much as possible in order to improve shooting accuracy and efficiency.

  In the entrance / exit management system according to the first embodiment, face recognition of one or a plurality of subjects (persons) is performed from images captured by the entire camera 101, and each person whose face is recognized is assigned to the child cameras 106-1 to 106-4. Face authentication of each subject (person) is performed from captured images of one or a plurality of child cameras 106-1 to 106-4, and entrance / exit is controlled and managed.

  FIG. 1 is a configuration diagram showing the configuration of the entrance / exit management system of the first embodiment.

  In FIG. 1, an entrance / exit management system 100 according to the first embodiment includes an overall camera 101, a moving object detection device 102, a person recognition device 103, a camera position calculation device 104, a child camera control device 105, a plurality of child cameras 106, a face. It has at least a recognition device 107, an employee database 108, and an attendance management device 109.

  The whole camera 101 is an image pickup means for picking up an image of the entire entrance / exit gate, and gives a picked-up image to the moving object detection apparatus 102. The whole camera 101 has an angle of view that can capture one or a plurality of persons passing through an entrance / exit gate, and captures an image with a resolution that can recognize each person's face from the captured image.

  The moving object detection apparatus 102 receives a captured image in real time from the entire camera 101 and detects a moving object from the image. In addition, when a moving object is detected, the moving object detection device 102 gives position information of the moving object to the person recognition device 103.

  Here, various methods can be applied as the moving object detection method. For example, the moving object detection device 102 has a storage device for storing video frames, and a comparison between a past frame and a current frame is performed. A method of detecting a moving object by taking a difference and detecting a motion vector can be applied.

  In addition, the moving object detection device 102 has a moving object sorting function for specifying and managing each moving object when a plurality of moving objects are detected. As this moving object classification function, for example, each moving object can be specified and managed by labeling the area pixels of each moving object with a labeling process.

  When receiving the position information of the moving object from the moving object detection apparatus 102, the person recognition apparatus 103 recognizes the person's part from the position information of the moving object. Thereby, the position information of the subject can be acquired. Here, the parts recognized by the person recognition device 103 include, for example, a face, an upper body, a lower body, an arm, and presence / absence of belongings.

  The first embodiment exemplifies a case where the person recognition apparatus 103 performs face recognition. Here, various face recognition methods can be widely applied as the face recognition method by the person recognition device 103. For example, a face region including feature information of a person's face is detected from an image region based on moving body position information. By doing so, the face of the person can be recognized. Further, even when the recognized part is other than the face, a method of detecting the recognition target part by detecting a specific region including the feature information of each part can be applied.

  In addition, the person recognition device 103 may ignore a person's part when the person's part cannot be detected from the region based on the position information of the moving object and the person cannot be recognized. At this time, the person recognizing device 103 may notify the moving object detecting device 102 that it is not a recognition target, and prompt the user to cancel the labeling.

  Furthermore, the person recognition apparatus 103 gives each face position information to the camera position calculation apparatus 104, if one or more face recognition is performed.

  Upon receiving one or more pieces of face position information from the person recognition apparatus 103, the camera position calculation apparatus 104 obtains position information of each subject on the captured image of the whole camera 101. The position information obtained by the camera position calculation device 104 is contributed as position information for controlling the child camera 106. Note that by setting the coordinate axis of the captured image in advance using the whole camera 101 as a fixed point camera, it is possible to obtain subject position information based on the face position information of each face area.

  When receiving the control information from the camera position calculation device 104, the child camera control device 105 assigns each subject to the child camera 106 based on the control information. The child camera control device 105 is connected to the child camera 106 and performs imaging control of each child camera 106. As a method for controlling the child camera 106, for example, H.264. 224 and H.C. A method performed using a control protocol such as 281 is applied.

  FIG. 3 is an internal configuration diagram showing an internal configuration of the child camera control device 105. In FIG. 3, the child camera control device 105 has at least a subject position analysis unit 31, a child camera assignment unit 32, a child camera assignment management unit 33, and a child camera imaging control unit 34 as its main functions.

  The subject position analysis means 31 determines the assignment of the child cameras 106 that capture the respective subjects based on the position information and feature areas of each subject on the image of the whole camera 101 from the child camera position calculation device 104. This is to analyze the position of the subject.

  The child camera assigning unit 32 assigns a subject to the child camera 106 based on the analysis result of the subject position analyzing unit 31.

  Here, as a method for assigning the child cameras 106, various methods can be applied.

  For example, when a plurality of subjects are shown in the image of the entire camera 101, the subject position analysis unit 31 determines whether the image of the entire camera 101 is based on the position information and the feature area of each subject acquired from the child camera position calculation device 104. Find the distance of the subject in the depth direction. The child camera assigning unit 32 can apply a method in which the assignment order of the child cameras 106 is set in advance, and the child cameras 106 are assigned according to the assignment order set in advance from the distance of the subject in the depth direction. As a method for obtaining the distance in the depth direction of the subject, a method of obtaining using the size of the feature region of the person on the image of the whole camera 101 can be applied. As a result, the child cameras 106 can be assigned to subjects that pass through the entrance gate G in order in a preset assignment order.

  Further, for example, the subject position analysis unit 31 also acquires motion vector information of the moving object via the camera position calculation device 104, analyzes the direction of movement of each subject passing through the entrance gate G, and the child camera assignment unit 32 A method of assigning the child cameras 106 according to the movement direction of each subject can be applied. For example, when the motion vector direction of the moving object is leftward, the child camera assignment unit 32 selects and assigns the child camera 106 arranged on the right side of the whole camera 101. Thus, when a person coming from the left in the image of the whole camera 101 passes through the entrance gate G, the child camera arranged according to the moving direction of the person is assigned to the child camera 106 on the right side of the whole camera 101. 106 can accurately capture the face of a person.

  Further, the child camera assigning unit 32 may change the subject assigned to the child camera 106.

  The child camera assignment management unit 33 manages the assignment status of the child cameras 106 assigned to the subject by the child camera assignment unit 32. For example, the child camera allocation management unit 33 associates the identification information of the subject with the identification information of the child camera 106, and manages which subject is captured by which child camera 106.

  The child camera imaging control means 34 controls the imaging processing of the child camera 106 by instructing the child camera 106 about an imaging position, magnification, brightness, and the like. The child camera imaging control means 34 can also cause the child camera 106 to track the subject.

  The child camera 106 (106-1 to 106-4) captures a specific area with high resolution under the control of the child camera control device 105. That is, when the child camera 106 receives the position information of the subject from the child camera control device 105, the child camera 106 gives the face authentication device 107 a video image of the subject with the position information. In addition, the child camera 106 has a resolution that can authenticate the face of the subject, and it is desirable to narrow the angle of view of the entire camera 101 of the child camera control device. Furthermore, the child camera 106 is provided with a tracking function, and can track a moving subject. For this tracking function, various technologies can be widely applied. For example, the tracking function is based on the motion vector of the moving object acquired by the moving object detection device 102 in real time according to the size of the motion vector of the subject. This can be realized by moving the shooting direction in the direction of the motion vector at a high speed.

  When the face authentication apparatus 107 receives a captured image from the child camera 106, the face authentication apparatus 107 refers to the employee database 108 based on each image and performs face authentication. As a face authentication method, a face feature area is extracted from the image of the child camera 106, the face feature area is compared with face registration information registered in the employee database 108, and authentication is performed when the similarity exceeds a threshold value. Success is assumed and authentication failure is assumed when the similarity is equal to or less than a threshold value.

  In addition, when the face authentication is successful, the face authentication device 107 gives authentication information, which is a result of the face authentication, to the time and attendance management device 109. Further, when face authentication fails, the face authentication device 107 feeds back information that is insufficient for the captured image of the child camera 106 to the child camera control device 105. Information includes position correction, magnification, brightness, and the like.

  FIG. 4 is an internal configuration diagram showing an internal configuration of the face authentication device 107. In FIG. 4, the face authentication device 107 includes at least a face authentication unit 41, a failure analysis unit 42, and a notification unit 43.

  The face authentication means 41 extracts face feature information from the image captured by the child camera 106, compares the extracted face feature information with the face registration information registered in the employee database 108, and calculates the similarity. The face authentication is performed based on the similarity.

  The failure analysis unit 42 analyzes the cause of the authentication failure when the authentication process by the face authentication unit 41 fails, and gives the analysis result to the notification unit 43. The cause of this authentication failure is that, when comparing the face feature information in the image of the child camera 106 and the registered face information, if the size of the comparison area is different, the magnification is corrected or the image on the child camera 106 is corrected. If the face feature area cannot be imaged, the imaging position is corrected.

  The notification means 43 gives authentication information as an authentication result to the attendance management device 109 when the face authentication is successful. In addition, the notification unit 43 gives the analysis result from the failure analysis unit 42 to the child camera control device 105 when face authentication fails.

  The employee database 108 is a database that holds employee face registration information used for face authentication of the face authentication apparatus 107. Further, the employee database 108 holds attendance / leaving management information of each employee by processing of the attendance / exit management device 109.

  The attendance and attendance management device 109 receives authentication information that is a face authentication result acquired from the face authentication device 107, and performs predetermined attendance and attendance management.

(A-2) Operation of the First Embodiment Next, the operation of processing in the entrance / exit management system 100 of the first embodiment will be described with reference to the drawings.

  FIG. 5 is a flowchart showing processing in the entrance / exit management system 100 of the first embodiment.

  First, the whole camera 101 always images the entrance / exit gate G (step S101). The video imaged by the whole camera 101 is given to the moving body detection device 102, and the moving body detection device 102 monitors the video in real time and detects one or more moving bodies moving in the video from the whole camera 101 (step). S102).

  The moving object detection apparatus 102 recognizes a moving object by moving object detection using real-time video from the whole camera 101, identifies the position of the moving object, and updates the position of the moving object in real time. In addition, when a plurality of motions are detected, the moving body detection device 102 assigns numbers to the pixels of each moving body, performs a labeling process, and identifies and manages the position of each moving body.

  When one or more moving objects are detected by the moving object detection device 102, position information of each moving object is given to the person recognition device 103. The person recognition apparatus 103 recognizes a person by detecting a face area including facial feature information from the position information area of each moving object (step S103).

  If it is determined that the moving object detected by the person recognition device 103 is not a person, the person recognition device 103 notifies the moving object detection device 102 to cancel the labeling process (step S104).

  When the person is recognized by the person recognition device 103, each face position information is given to the camera position calculation device 104, and the camera position calculation device 104 obtains face position information on the image of the whole camera 101 (step S105). .

  When each face position information is obtained by the camera position calculation device 104, each face position information on the image of the whole camera 101 is given to the child camera control device 105, and the child camera control device 105 applies to each person's face. Each child camera 106-1 to 106-4 is assigned, and the imaging of each child camera 106-1 to 106-4 is controlled (step S106).

  Under the control of the child camera control device 105, each of the child cameras 106-1 to 106-4 images the subject at the position instructed by the child camera control device 105 (step S107), and performs face authentication on the captured image. To device 107.

  The face authentication device 107 performs face authentication by referring to the face registration information registered in the employee database 108 based on the video acquired from each of the child cameras 106-1 to 106-4 (step S108). Note that the employee database 109 may be accessible via a network or the like during face authentication by the face authentication apparatus 107.

  If the face authentication device 107 can identify the face authentication, the authentication information is given to the attendance / attendance management device 109 (step S109), and the attendance / attendance management device 109 records the attendance / attendance management information of the employee in the employee database 108 (step S110). . Further, after the authentication, the face authentication device 107 notifies the moving object detection device 102 to release the labeling of the person (step S111).

  If face authentication cannot be specified by the face authentication apparatus 107, the face authentication apparatus 107 analyzes information that is insufficient for successful face authentication based on the image information from the child cameras 106-1 to 106-4 (step). S112).

  If information that is insufficient for successful face authentication is obtained, the face authentication device 107 feeds back the information to the child camera control device 105 (step S113). Accordingly, the child camera control device 105 can correct the imaging position, magnification, brightness, and the like instructed to each child camera 106-1 to 106-4 in real time based on the fed back information.

(A-3) Effects of the First Embodiment As described above, according to the first embodiment, the following effects can be expected.

  According to the first embodiment, by linking the whole camera and one or a plurality of child cameras, face authentication of a plurality of persons shown in the captured video of the whole camera can be performed at a time.

  In addition, according to the first embodiment, a high-resolution video image of a person shown in a video image captured by the overall camera can be acquired by linking the overall camera and one or more child cameras.

  Furthermore, according to the first embodiment, the whole camera and one or a plurality of child cameras are linked, and the moving object detection device updates the moving object position in real time based on the video from the whole camera, so that the subject moves. But tracking can be done.

  Furthermore, according to the first embodiment, even when face authentication fails, the face authentication device feeds back control information to the child camera control device based on the failed information, so that the face can be immediately re-applied. Authentication is possible.

(B) Second Embodiment Next, a second embodiment of the imaging management system, imaging management method, authentication system, and authentication method of the present invention will be described with reference to the drawings.

  In the second embodiment, an embodiment for realizing a passer-by monitoring system that manages persons passing through a security gate such as an airport using the present invention will be described.

  In the first embodiment, only the face information used for face authentication is used among the parts of the person specified by the person recognition device 103 (for example, the presence of the face, upper body, lower body, arms, possession, etc.) In the second embodiment, a plurality of child cameras capture and authenticate a person's face, upper body, and possessed items.

(B-1) Configuration of Second Embodiment FIG. 6 is a configuration diagram illustrating a configuration of a passer-by monitoring system according to the second embodiment. In FIG. 6, a passer-by monitoring system 200 according to the second embodiment includes an overall camera 101, a moving object detection device 102, a person recognition device 103, a camera position calculation device 104, a child camera control device 105, and child cameras 106-1 to 106. -3, the face authentication device 201, the upper body analysis device 202, the belongings analysis device 203, the criminal database 204, the notification device 205, and the passerby database 206.

  In the second embodiment, the configuration of the whole camera 101, the moving object detection device 102, the person recognition device 103, the camera position calculation device 104, the child camera control device 105, and the child camera 106 already described in the first embodiment. I'll omit the explanation.

  As a method of controlling the child camera 106 by the child camera control device 105, for example, the area captured by each child camera 106 is determined as, for example, a face, an upper body, or a belonging, and when the person recognition device 103 recognizes a person, A method in which the child camera control device 105 instructs each child camera 106 to image a specific area of the person can be applied.

  Further, for example, the imaging region of the child camera 106 may not be determined in advance, and the child camera control device 105 may arbitrarily determine the imaging region of the child camera 106 and control the imaging process when recognizing a person.

  The face authentication device 201 acquires the video of the child camera 106-1 from the child camera 106-1, and performs face authentication while referring to the criminal database 204. The face authentication device 201 records information used for face authentication, still images, and the like in the passerby database 206 as passerby information when performing face authentication of a criminal. At this time, the passer-by information includes, for example, a video acquired from the child camera 106-1, face information matched from the criminal database 204, reference information of the criminal, and travel time information. Further, as a still image, for example, compression processing may be performed using a still image encoder such as JPEG.

  The face authentication apparatus 201 extracts a face feature area from the video of the child camera 106-1 as pre-processing for face authentication, and if the face feature area can be extracted, the extraction is successful, and the face feature area cannot be extracted. The extraction fails.

  In addition, when the face feature region extraction fails, the face authentication device 201 feeds back information lacking in the captured image of the child camera 106-1 to the child camera control device 105. Information includes position correction, magnification, brightness, and the like.

  The upper body analysis apparatus 202 is an apparatus that acquires an image of the child camera 106-2 from the child camera 106-2 and analyzes the upper body of the person. For example, when the upper body analyzing apparatus 202 receives an image captured by the child camera 106-2, the upper body analyzing apparatus 202 analyzes characteristics of the person's clothing (for example, color, type of clothing estimated) from the image. The upper body analysis apparatus 202 records the analysis result in the passerby database 206 as passerby information.

  The upper body analysis apparatus 202 analyzed the characteristics of the person's clothing from the image of the child camera 106-2, and when the characteristics of the person's clothes were analyzed, the analysis was successful, and the characteristics of the person's clothing could not be analyzed. Sometimes analysis fails.

  Further, the upper body analysis device 202 feeds back to the child camera control device 105 information that is insufficient in the captured image of the child camera 106-2 when the analysis of the characteristics of the person's clothing fails. Information includes position correction, magnification, brightness, and the like.

  The personal belonging analysis device 203 is a device that acquires the video of the child camera 106-3 from the child camera 106-3 and analyzes the personal belongings of the person. The belongings analysis device 203 analyzes, for example, the characteristics of the belongings (for example, the color, shape, length of the belongings, the kind of the belongings to be inferred, etc.). Also, the belongings analysis device 203 records the analysis result in the passerby database 106 as passerby information.

  The belonging analysis device 203 analyzes the characteristics of the belongings from the image of the child camera 106-3, and when the characteristics of the belongings can be analyzed, the analysis is successful, and when the characteristics of the belongings cannot be analyzed. Analysis failure.

  In addition, when the analysis of the characteristics of the possessed item fails, the possessed item analyzing device 203 feeds back the information lacking in the captured image of the child camera 106-3 to the child camera control device 105. Information includes position correction, magnification, brightness, and the like.

  The criminal database 204 is a database that holds criminal face information, and is used when the face authentication apparatus 201 performs face authentication.

  When the face authentication apparatus 201 authenticates a criminal, the notification apparatus 205 receives a report from the face authentication apparatus 201 and notifies an alarm signal to, for example, a security guard, a security company, or the police.

  The passerby database 206 is a database that records the authentication result in the face authentication device 201 or the analysis information in the upper body analysis device 202 and the belongings analysis device 203 as passerby information.

(B-2) Operation | movement of 2nd Embodiment Next, operation | movement of the process in the passer-by monitoring system 200 of 2nd Embodiment is demonstrated.

  FIG. 7 is an explanatory diagram illustrating an image of camera arrangement in the passer-by monitoring system 200 according to the second embodiment.

  In FIG. 7, in the passer-by monitoring system 200 according to the second embodiment, the entire camera 101 that images passers-by passing through the security gate SG, and a plurality of (three in FIG. 7) child cameras around the entire camera 101. 106-1 to 106-3 are arranged.

  In FIG. 7, in the first embodiment, the child camera 106 is assigned to each of a plurality of persons reflected on the whole camera 101. However, in the second embodiment, the child camera 106 is assigned to each feature region of one person. assign.

  FIG. 8 is an explanatory diagram showing an example of assignment of each of the child cameras 106-1 to 106-3 in the person feature area. In FIG. 8, for example, the child camera 106-1 images a human face part, the child camera 106-2 images a person's upper body part, and the child camera 106-3 images the presence / absence of the person's belongings. To do.

  As in the first embodiment, the moving object detection device 102 detects a moving object based on the video of the whole camera 101 that captures a passerby passing through the security gate SG, and the person recognition device 103 detects the person from the position information of the moving object. Recognize

  Then, based on the position information on the image of the entire camera 101 by the camera position calculation device 104, the child camera control device 105 assigns one or more feature regions of the person to the child cameras 106-1 to 106-3. .

  In the following, it is assumed that the face area of the person is assigned to the child camera 106-1, the upper body area of the person is assigned to the child camera 106-2, and the belonging area of the person is assigned to the child camera 106-3.

  The child camera 106-1 assigned to the face of the person by the child camera control device 105 images the face area of the person based on the position information instructed from the child camera control device 105.

  When the image captured by the child camera 106-1 is given to the face authentication device 201, the face authentication device 201 refers to the criminal database 204 to perform criminal face authentication.

  When the face authentication device 201 obtains the criminal face authentication, the face authentication device 201 records the passerby information in the passerby database 206. Further, the face authentication device 201 notifies the notification device 205, and the notification device 205 notifies the security guard, the police, and the like. Further, when the criminal face authentication is obtained, a gate control device (not shown) may close the security gate SG.

  When face authentication cannot be specified by the face authentication apparatus 201, the face authentication apparatus 201 analyzes information that is insufficient for successful face authentication based on image information from the child camera 106-1. When information that is insufficient for successful face authentication is obtained, the face authentication device 201 feeds back the information to the child camera control device 105. Accordingly, the child camera control device 105 can correct the imaging position, magnification, brightness, and the like instructed to the child camera 106-1 on the basis of the fed back information in real time.

  Further, the child camera 106-2 assigned to the upper body of the person by the child camera control device 105 images the upper body of the person based on the position information instructed from the child camera control device 105. Then, when the image captured by the child camera 106-2 is given to the upper body analysis apparatus 202, the upper body analysis apparatus 202 analyzes the upper body feature information based on the image from the child camera 106-2, and the analysis result is obtained. It is recorded in the passerby database 206 as passerby information. Here, the upper body analysis apparatus 202 includes clothes features (color, type of analogy) and the size of the upper body of the person as the upper body feature information.

  If the upper body analysis device 202 cannot identify the upper body feature information, the upper body analysis device 202 analyzes information that is insufficient to identify the upper body feature based on the image information from the child camera 106-2. . When the information that is insufficient to identify the upper body feature is obtained, the upper body analysis apparatus 202 feeds back the information to the child camera control apparatus 105. Accordingly, the child camera control device 105 can correct the imaging position, magnification, brightness, and the like instructed to the child camera 106-2 in real time based on the fed back information.

  Furthermore, the child camera 106-3 to which the feature region of presence / absence of a person's belongings is assigned by the child camera control device 105 performs imaging based on position information instructed from the child camera control device 105. Based on the image of the child camera 106-3, the possession analysis device 203 determines whether the possession of the person is present, and when possessing the possession, the color, shape, length, Analyze the kind of analogy. Also, the belonging analysis device 203 records the analysis information in the passerby database 206 as passerby information.

  Also, if the possession analysis device 203 cannot identify the color, shape, length, type of analogy, etc. of the possession, the possession analysis device 203 uses the possession based on the image information from the child camera 106-3. Analyze the missing information to identify features such as the color, shape, length, and analogy of the object. Then, when information that is insufficient to specify the characteristics such as the color, shape, length, and analogy type of the possession is obtained, the possession analysis device 203 feeds back the information to the child camera control device 105. . Accordingly, the child camera control device 105 can correct the imaging position, magnification, brightness, and the like instructed to the child camera 106-2 in real time based on the fed back information.

  The processes of the face authentication device 201, the upper body analysis device 202, and the belongings analysis device 203 may be performed in parallel processing. The information recorded in the passerby database 206 can also be used as information when a crime or the like occurs later.

(B-3) Effects of Second Embodiment As described above, according to the second embodiment, the following effects can be expected.

  According to the second embodiment, a plurality of feature areas of a person can be photographed at a time by cooperation of the whole camera and the child camera.

  In addition, according to the second embodiment, it is possible to capture a high-resolution video in a person's feature area by cooperation of the whole camera and the child camera.

  Furthermore, according to the second embodiment, it is possible to store information on the characteristic area (passerby information) based on the video acquired from the child camera in the passerby database.

  Furthermore, according to the second embodiment, even if the analysis of the specific part fails, the control information is fed back based on the failed information, so that the analysis can be immediately performed again.

(C) Other Embodiments In the first embodiment, the entrance / exit management system is shown. However, if the employee database 108 in the first embodiment is replaced with a customer database and the attendance management device is replaced with a customer management device, store customer management is performed. It is also applicable to the system. The present invention can also be applied to a system that provides information useful to management of registered customers and registered customers.

  Although the passer-by monitoring system is shown in the second embodiment, the present invention can also be applied to a statistical information management system that focuses on a specific part of a passing person.

It is a block diagram which shows the structure of the entrance / exit management system of 1st Embodiment. It is explanatory drawing explaining the image of camera arrangement | positioning in the entrance / exit management system of 1st Embodiment. It is a block diagram which shows the function structure of the sub camera control apparatus of 1st Embodiment. It is a block diagram which shows the function structure of the face authentication apparatus of 1st Embodiment. It is a flowchart which shows the process in the entrance / exit management system of 1st Embodiment. It is a block diagram which shows the structure of the passer-by monitoring system of 2nd Embodiment. It is explanatory drawing explaining the image of camera arrangement | positioning in the passer-by monitoring system of 2nd Embodiment. It is explanatory drawing explaining allocation of the imaging | photography area | region of the child camera of 2nd Embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 101 ... Whole camera, 102 ... Moving body detection apparatus, 103 ... Person recognition apparatus, 104 ... Camera position calculation apparatus, 105 ... Child camera control apparatus, 106 (106-1 to 106-4) ... Child camera, 107 ... Face authentication apparatus , 108 ... employee database, 109 ... attendance and attendance management device, 201 ... face authentication device, 202 ... upper body analysis device, 203 ... belongings analysis device, 204 ... criminal database, 205 ... reporting device, 206 ... passerby database.

Claims (8)

One or a plurality of objects including a first imaging means and a plurality of second imaging means for imaging a narrower range than the image of the first imaging means, and appearing in the image of the first imaging means An imaging management system for performing imaging management of
A target for identifying one or a plurality of targets appearing in the video of the first imaging means by searching for a feature region including feature information of the target to be specified based on the video acquired from the first imaging means. Specific means,
The second imaging means is assigned to the one or more objects specified by the object specifying means, and an image having a higher resolution than the image of the first imaging means is assigned to the one or more objects. Imaging control means for controlling each of the second imaging means so as to image,
An imaging management system comprising: a high-resolution image management unit that simultaneously acquires and manages the high-resolution images of the one or more objects captured by each of the second imaging units. The target identification means is
A moving object detection unit for detecting one or more moving objects based on the video acquired from the first imaging means;
Feature information for recognizing each target is stored in advance, and it is searched whether or not there is a feature region including the feature information in the region of the one or more moving objects detected by the moving object detection means. An object recognition unit for recognizing each object,
The imaging management system according to claim 1, further comprising: a position calculation unit that obtains position information on the image of the first imaging unit for each of the objects recognized by the object recognition unit.   The imaging control means assigns the second imaging means based on the position information of the objects on the image of the first imaging means acquired from the position calculation unit of the target specifying means. The imaging management system according to claim 2, wherein the imaging management system is one.   The imaging management system according to any one of claims 1 to 3, wherein the high-resolution image management means feeds back video information of the second imaging means to the imaging control means. One or a plurality of objects including a first imaging means and a plurality of second imaging means for imaging a narrower range than the image of the first imaging means, and appearing in the image of the first imaging means An imaging management method for performing imaging management of
The target specifying unit searches for a feature region including the feature information of the target to be specified based on the video acquired from the first imaging unit, and the one or more reflected in the video of the first imaging unit An object identification process for identifying an object;
The imaging control means assigns each of the second imaging means to the one or more objects specified by the object specifying means, and the one or more objects are more than the images of the first imaging means. An imaging control step of controlling each of the second imaging means so as to capture a high-resolution image;
A high-resolution image management means comprising: a high-resolution image management step of simultaneously acquiring and managing a high-resolution image of the one or more objects captured by each of the second imaging means. Method. One or a plurality of objects including a first imaging means and a plurality of second imaging means for imaging a narrower range than the image of the first imaging means, and appearing in the image of the first imaging means An authentication system that performs authentication of
A target for identifying one or a plurality of targets appearing in the video of the first imaging means by searching for a feature region including feature information of the target to be specified based on the video acquired from the first imaging means. Specific means,
The second imaging means is assigned to the one or more objects specified by the object specifying means, and an image having a higher resolution than the image of the first imaging means is assigned to the one or more objects. Imaging control means for controlling each of the second imaging means so as to image,
High-resolution image acquisition means for simultaneously acquiring high-resolution images of the one or more objects captured by each of the second imaging means;
Registration information storage means for storing one or more pieces of registration information to be authenticated in advance;
An authentication unit that performs a predetermined authentication process based on the high-resolution image of the one or more objects acquired by the high-resolution image acquisition unit and the registration information of the registration information storage unit. Authentication system.   When the authentication unit fails to authenticate based on the high-resolution image of each target, cause analysis relating to the authentication failure is performed, and the cause analysis information is fed back to the imaging control unit. The authentication system according to claim 6. One or a plurality of objects including a first imaging means and a plurality of second imaging means for imaging a narrower range than the image of the first imaging means, and appearing in the image of the first imaging means An authentication system that performs authentication of
A registration information storage means for storing one or more registration information to be authenticated in advance;
The target specifying unit searches for a feature region including the feature information of the target to be specified based on the video acquired from the first imaging unit, and the one or more reflected in the video of the first imaging unit An object identification process for identifying an object;
The imaging control means assigns each of the second imaging means to the one or more objects specified by the object specifying means, and the one or more objects are more than the images of the first imaging means. An imaging control step of controlling each of the second imaging means so as to capture a high-resolution image;
A high-resolution image acquisition unit that simultaneously acquires high-resolution images of the one or more objects captured by each of the second imaging units;
An authentication unit includes an authentication step of performing a predetermined authentication process based on the one or more target high-resolution images acquired by the high-resolution image acquisition unit and the registration information of the registration information storage unit. An authentication method characterized by that.

Images