JP2005033570A - Method and system for providing mobile body image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a mobile body imaging system not making students and general audience dizzy when watching a terminal, allowing a teacher side to grasp the face of a questioner on a screen, and allowing a person in a remote place to have a feeling of attendance at a lecture. <P>SOLUTION: The system includes a first imaging means 3 with a platform side as a visual field, and a second imaging means 4 with a student side from the platform side as a visual field. The first imaging means 3 comprises a first wide angle camera 1 and a first tracking camera 2. The second imaging means 4 comprises a second wide angle camera 5 and a second tracking camera 6. The first imaging means 3 detects a target (student or teacher) on the platform, and tracks the new target by zoom-up even when a plurality of targets exist. The second imaging means 4 displays the image of the whole attending students or the zoom-up image of the questioner on the monitor 10 of a host computer. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a moving body image providing system that appropriately captures a moving object as a moving object using a plurality of imaging devices and can track an image at an appropriate speed according to the situation of the moving object. is there.

  In recent years, large stores such as supermarkets, convenience stores, and pachinko machines are increasing.

  Such stores tend to have long hours such as business hours in the middle of the night with changes in the way of civic life. For this reason, there is a tendency to use a surveillance system that monitors the behavior of customers and store clerk day and night with a television camera (TV). It is well known that such a monitoring system is considered so as to grasp the degree of congestion of customers and prevent crimes.

  As a result of the recent development of multimedia-related technologies and the widespread use of network technologies, broadband capable of transmitting large amounts of information over networks has been adopted.

  Such broadband is capable of transmitting large-capacity information such as 1.5M, 8M to 20M, for example, and is being adopted for transmission of high-definition information such as a remote medical system using a TV camera and remote education. There is the present situation.

  That is, it is possible to acquire information in a form that exceeds time and space by a remote viewing system using a TV camera.

  For example, videos of university lectures, weddings, concerts, etc. can be transmitted in real time to remote locations so that users at remote locations can view them.

An imaging system that can also be used for education or monitoring, such as a remote viewing system, is disclosed in JP-A-5-334572. In this document 1, a wide-angle TV camera for tracking and a TV camera for tracking are shared, and the entire space is photographed on the one hand. The video is being acquired.
Patent Document 1: JP-A-5-334572

  However, since the above-mentioned system is a system that distributes video viewed from one direction, for example, in a school lecture, video on the teacher's side can be distributed, but video that captures the face of the questioner cannot be distributed. There was a problem.

  For example, when a lecture is given in a large classroom, there is a distance between the instructor and the student, and the teacher cannot grasp the questioner's face.

  Furthermore, when performing remote viewing education using the imaging system as described above, for example, if an automatic tracking system employed in a normal monitoring system is used as it is, an attempt is made to capture the behavior of a teacher on the platform, for example. At the time, the tracking performance is reflected as it is, and the viewer (student) of the terminal cannot respond to the dizzying change of the tracking video that can be transmitted, causing not only dizziness but also focusing on lectures etc. Cannot watch.

  In order to solve this, manual operation of the camera not only reduces productivity but also increases labor costs, and the quality of the images taken varies depending on the sensitivity of each photographer. This is also not an effective means.

  Therefore, students, general auditors, etc. will not be dizzy when they see the terminal, and the teacher can grasp the questioner's face on the screen, and the teacher's video and the questioner's face during the lecture will be distributed. Therefore, it is desirable to obtain a system that can give a feeling of participating in a lecture even at a remote location.

  The present invention provides a lecturer side video to be lectured by the lecturer at the auditorium to the student terminal via the network, while the student side video viewed from the lecturer is designated as a lecturer terminal. It is the moving body imaging method provided to the terminal.

(A). Taking a lecturer-side image comprising the first wide-area video and the first zoom video on the lecturer side with a first imaging means; and (b). Capturing a student-side image composed of the second wide-area video and the second zoom video on the student side by a second imaging means.

  The main controller is (c). Providing the instructor-side image of the first imaging means to the instructor-side terminal or a designated terminal via the network; and (d). A first target is detected and tracked from the first wide-area video of the lecturer-side image, and the first zoom video of the first target is connected to the teacher-side terminal and designated terminal via the network. Providing (e). Providing a student-side image of the second imaging means to the instructor-side terminal or a designated terminal via the network; (f). A second target is detected from the student-side image, and the instructor's terminal and the designated terminal are caused to take a second zoom image of the second target and the instructor's terminal is designated via the network. And a step of providing to the terminal.

  In addition, the present invention provides a video image of the lecturer side lectured by the lecturer at the attendance hall to the student terminal via the network, while the video image of the student side viewed from the lecturer is shown by the lecturer. This is a mobile image providing system provided to a terminal on the side and a designated terminal. A first imaging device including a first wide-angle camera that captures a wide-area image of the instructor side and a first tracking camera that captures a zoom image by panning and tilting operations, and a wide-area image of the student side A second imaging device comprising a second wide-angle camera and a second tracking camera that captures a zoom image by panning and tilting operations; and detecting the instructor-side target from the wide-area video of the first wide-angle camera. A first main controller for tracking and transmitting the first zoom image in place of the wide-area image of the first wide-angle camera, and a second main unit from the second photographing device. When there is movement in the wide-area video, the second tracking camera is controlled at the position where the target on the student side is recognized based on the motion detection image, and the second zoom video of the target on the student side is obtained. Let me shoot And a second main controller to transmit a second zoom image.

  The host computer is a combination of the first imaging device and the second main control device or a combination of the second imaging device and the second main control device based on a mode designated by the instructor's terminal. Means for operating in combination; an image of a first wide-angle camera from the first main control device; an image of the first zoom image; an image of a second wide-angle camera from the second main control device; And a means for transmitting the zoom image of No. 2 to the instructor side terminal and the remote student terminal.

  As described above, according to the present invention, during the lecturer's lecture, the lecturer's zoom image is displayed on the remote viewer terminal while tracking the movement of the lecturer. In addition, when there is a question in the classroom, the instructor's monitor and remote viewer terminal can get a zoomed-in image of the questioner, so the instructor and remote viewer can understand who asked the question. There is.

  In addition, when the instructor moves within the sense area on the viewer's terminal, the instructor's zoom image (zoom-in) is displayed, and when leaving the sense area, the zoom-out image is immediately replaced. It can be seen naturally without learning dizziness.

  For this reason, since a person like a specific professional photographer is not used, the cost can be reduced.

  Also, the video of the lecture (instructor side, questioner in the classroom) is provided to the external viewer's terminal, and when there is a question from the viewer, the face image is displayed on the instructor's terminal. There is an effect that the lecturer can attend the lecture as if the lecturer is in the same classroom as the viewer at the location and the viewer in the classroom.

  Furthermore, when a large number of students go up to the platform, the remote viewer terminal is provided with images with a zoom ratio according to the number of students. be able to.

<Embodiment 1>
FIG. 1 is a schematic configuration diagram of a mobile object image providing system according to the first embodiment. As shown in FIG. 1, a first imaging unit 3 having a visual field on the side of the platform and a second imaging unit 4 having a visual field on the side of the student from the platform side are provided. Wide-angle camera 1 and first tracking camera 2. Further, the second imaging means 4 has a second wide-angle camera 7 and a second tracking camera 6.

  Then, the first imaging means 3 detects targets on the platform (students and teachers). When there are a large number of student targets, tracking shooting is performed at a fixed zoom rate according to the number of student targets, and the tracking video is recorded. The host computer 10 (arranged on the teacher side) transmits to the remote viewer terminal 31. If there is only one student target, the student target is tracked and zoomed in when stopped.

  Further, the second imaging means 4 takes images of all the students in the classroom (referred to as student images or viewer images) or zoomed-in images of the questioners, and these images are provided on the host computer 10 (the teacher's side). ) Can be displayed on the monitor 8.

(Configuration of each part)
The first tracking camera 2 (dome camera) and the second tracking camera 6 described above have a dome-like outer shape and have a camera part (not shown) inside, and the base part has a panning camera part. Each drive unit for tilting is built in and fixed to the ceiling or the like.

  The first wide-angle camera 1 is located behind the first tracking camera 2 and the second wide-angle camera 7 is located behind the second tracking camera 4 at predetermined intervals. Attached to the ceiling of the classroom. The first wide-angle camera 1 has a view of the lower dome of the tracking camera 2 and the platform. The second wide-angle camera 7 has a field of view below the second tracking camera 6 and the student side (listener side).

  The first tracking camera 2 and the first wide-angle camera 1 are connected to a first automatic tracking device 12 and are remotely controlled by the first automatic tracking device 12. The first automatic tracking device 12 is connected to the first image output device 13. These are collectively referred to as the main control device 14 for monitoring the teaching side.

    On the other hand, as shown in FIG. 1, the second imaging unit 4 connects a second wide-angle camera 7 and a second tracking camera 6 that monitor students from the teaching side to a second automatic tracking device 16. The second automatic tracking device 16 is connected to the second image output device 17. These are also collectively referred to as a student-side monitoring main controller 19.

  The first automatic tracking device 12 and the second automatic tracking device 16 described above each include a motion detection unit, a target tracking unit, a multiple target range setting unit, a CPU, a DSP, a RAM, a ROM, and the like (not shown). A tracking camera control unit (including a zoom function) is realized.

  That is, a moving object (student, teacher, etc .: target) in a motion detection area of a wide-area image captured by a wide-angle camera is detected by a motion detector, and the moving object is a student, and the number of students is constant. For example, the target range setting unit determines a range with these students as the field of view and sets the center as a tracking point. Then, the target tracking unit tracks this tracking point.

  That is, when there are a large number of targets, tracking shooting can be performed at a constant zoom rate.

  The first image output device 13 and the second image output device 17 are inputted with the supervision side view image from the automatic tracking device 12 and the student side monitoring image from the automatic tracking device 16 and amplify these images. To the main computer 21 of the host computer 10.

(Configuration of automatic tracking device)
The first automatic tracking device 12 and the second automatic tracking device 16 have the configuration shown in FIG. In the present embodiment, the first automatic tracking device 12 will be described as a representative.

  FIG. 2 is a schematic configuration diagram of the first automatic tracking device 12. As shown in FIG. 2, the first automatic tracking device 12 includes a motion detection unit 120, a target tracking unit 121, a tracking camera control unit 122, a multiple target range setting unit 123, and the like.

  The motion detection unit 12 converts the photoelectrically converted output signal of the first wide-angle camera 1 into a digital image signal, saves it in the first frame memory, and saves it in the second frame memory (for the previous image). The pixel value (for example, luminance) of the image is determined for each pixel, and a pixel having a large difference (absolute value) is “1” and a small pixel is “0” (this is referred to as a motion detection image). Referred to in this embodiment). For the above-described motion detection, the motion detection unit 120 detects a target based on a motion detection area, a mask area, or the like preset in the memory 124.

  The target tracking unit 121 cuts out an “island (connected region)” of a pixel of value “1” from the motion detection image, and recognizes this as a new target when the cut out target exists in a preset sense area. . Then, a symbol code for the cut target (symbol code including coordinates estimated to be present: tracking position information) is generated (tracking). However, just before the system is started, only one target is detected. When a large number of targets are detected, a symbol code (also referred to as tracking position information) is output to the multiple target range tracking setting unit.

  Further, when one target is detected and the target stops moving, a target tracking stop signal (including a symbol code) is output.

  As shown in FIG. 3, the multiple target range tracking setting unit 123 includes, in the sense area, for example, a symbol code (2) indicating a second target (student) and a third target (student) in addition to the first target (teacher). Are input from the target tracking unit 121, the second target and the third target corresponding to this symbol code are allocated. That is, the first target (teacher) is excluded.

  Then, the center of gravity of the second target and the third target is determined, the distances (L1, L2) from both ends of the sense area to the center of the second target and the third target are obtained, and the sense area length L to L1, A horizontal range is determined by subtracting L2 and adding a predetermined length m. Then, a range Gi of a predetermined height h is obtained, and the center of this range is set as the tracking position. That is, the tracking point of the range Gi is determined on a straight line connecting the centroids of the second and third targets.

  The target tracking unit 121 outputs tracking information including the range Gi and the tracking point (coordinate position) to the tracking camera control unit 122.

  The tracking camera control unit 122 stores the tilt angle, pan angle, zoom ratio, and the like of the first tracking camera 2, and every time the tracking information is input, the tilt angle and pan angle based on the tracking position of the tracking information. Is controlled so that the target falls within the viewing angle of the first tracking camera 2 and the zoom ratio of the range Gi is controlled. In this control, each angle and speed are controlled so as to achieve a target angle. That is, the speed control with the tracking position as the target value is performed to switch to the angle control.

  The video superimposing unit 31 superimposes the wide-angle camera video and / or tracking camera video from the wide-angle camera 1 and sends them to the image output device 13.

  However, the automatic tracking device 16 of the student-side monitoring main control device 19 is connected to the second transmission / reception device 19, and the wide-angle camera image (student-side wide-angle camera image) of the second wide-angle camera 7 when the system is turned on. Is sent to the host computer 8 via the image output device 17.

  Further, when the first target (teacher) is obtained after the system is turned on and the student stands up, the second automatic tracking device 16 detects the movement of the student and recognizes it as the questioner's target). Then, the zoom is increased at a predetermined zoom ratio. Then, this zoom-up image is sent to the host computer 8.

  The sense area and detection area at this time are set in a rectangular shape above the seating range as shown in FIG.

  On the other hand, the host computer 10 is connected to a transmission / reception device 25 that is an access point.

  In addition, a student image from the student monitoring control device 19 or a remote person image taken by the remote viewer terminal 31 is displayed on the monitor 8.

  In addition, the host computer 10 transmits the teacher-side image from the teacher-monitoring main control device 14 to the remote viewer terminal 31 via the network (broadband).

  Further, a speaker (not shown) and a liquid crystal TV are connected to the host computer 10, and voices of questioners (remote viewers, students, and teachers) and teachers are generated from the speakers and based on instructions from the portable controller 28. The video of the remote viewer is displayed on a liquid crystal TV (or a projector).

  Further, this embodiment may be used in a large classroom. In such a case, a large number of wide-angle cameras and tracking cameras may be provided as shown in FIG. In addition, as shown in FIG. 5B, a wide-angle camera may be provided so that the center of the optical axis is parallel to the floor of the staircase classroom, and the tracking camera may be provided on the wide-angle camera.

  Furthermore, the host computer 10 stores in advance lecture information in which a classroom number is associated with a date, a lecture start time, an end time, a lecture name, a lecturer name in charge (see FIG. 6)). This lecture information is preferably managed on a monthly or weekly basis.

  Further, the host computer 10 stores a mode table shown in FIG. 7 in advance. This mode table includes an audiovisual classroom mode, a study mode, a monitoring mode, and the like, and determines whether to perform mask, pen, tilt, speed, sense color, and the like.

  For example, when the displayed audiovisual classroom mode, attendance mode, and monitoring mode shown in FIG. 8 are selected, the reference control information is determined by a matrix corresponding to the mode table of the host computer. The selected matrix information is set in the automatic tracking device.

  Furthermore, it is connected to a control box of each facility (not shown), and this control box is connected to a liquid crystal TV, a viewer, a projector, a lighting control box, and the like.

(Description of operation)
FIG. 9 is a sequence diagram for explaining the operation of the system according to the first embodiment. In this embodiment, the questioner in the classroom stands up. Further, the remote viewer has decided to select the question button displayed on the terminal 31 when there is a question.

  Under this arrangement, when the teacher enters the classroom, the teacher operates the portable controller 28 to transmit a system-on (learning mode) to the host computer 10 (d1). When the host computer 10 receives the system-on signal via the transmission / reception device 25, the host computer 10 immediately activates the student-side monitoring main control device 19 and the textbook monitoring main control device 14 (d2). This system-on signal includes the ID of the portable controller 28, the ID of the transmission / reception device 25, and the lecturer code (lecture name, lecturer in charge, date, time).

  Further, when receiving the system-on signal, the host computer 10 extracts an initial classroom image (see FIG. 10) stored in advance for the corresponding classroom and transmits it to the monitor 8 (d3).

  In addition, the main control device 14 for monitoring the platform side sends the platform side image captured by the wide-angle camera 1 to the host computer 10 when the system is turned on (d4), and the host computer 10 remotely transmits the platform side image via the network. To the user terminal 31 (d5) and to the monitor 8 (d6). The transition of the text side image at this time will be described later.

  Further, the student-side monitoring main control device 19 sends the student-side image captured by the wide-angle camera 7 to the host computer 10 when the system is turned on (d7). In response to the input of the student side image, the host computer 10 outputs the student side image to the monitor 8 instead of the platform side image (d8).

  When the student starts up, the student-side monitoring main control device 19 sends the student-side image including the student to the host computer 10 (d9), and the host computer 10 transmits it to the remote viewer terminal 31 and the monitor 8. (D10, d11).

  As shown in FIG. 4, the student-side monitoring main control unit 19 has a motion detection area (the same as the sense area). Therefore, when the student starts up, the motion detection unit Capture motion detection images.

  Next, when the target tracking unit 121 exists in the sense area shown in FIG. 4, the target tracking unit 121 recognizes it as a new target, determines the center of gravity (center portion) of this target (questioner target), and sends it to the tracking camera control unit 122. In addition to tracking, the interrogator target is zoomed in at a predetermined zoom ratio (see FIG. 11).

  In other words, the student-side monitoring main control device 19 sends the interrogator's zoomed-in image to the host computer 10 (d12).

  The host computer 10 transmits the zoomed-in image of the questioner to the remote viewer terminal 31 and the monitor 8 (d13, d14). That is, the remote viewer terminal 31 and the monitor 8 display a zoom-up image of the questioner who has stood up and indicated the intention of the question.

  Then, as shown in FIG. 9, when the questioner sits, the target tracking unit 121 cancels the tracking of the target of the questioner because the sense area and the motion detection area shown in FIG. 4 are left. With this tracking cancellation, the tracking camera control unit 122 zooms out at a zoom ratio that captures all students.

  That is, the above-mentioned text image is sent to the host computer 10 (d15) and transmitted to the monitor 8 and the remote viewer terminal 31 (d16, d17).

  On the other hand, the remote viewer may ask a question by operating a question button (not shown) of the remote viewer terminal 31. By selecting this question button, the remote viewer terminal 31 transmits the viewer ID and the question code to the host computer 10 (d18).

  The host computer 10 stores in advance lecture information in which a classroom number is associated with a date, a lecture start time, an end time, a lecture name, a lecturer name, a viewer registration permission ID, and the like. This lecture information is preferably managed on a monthly or weekly basis.

  The host computer 10 compares the viewer ID of the viewer question information from the remote viewer terminal 31 with the lecture information, and if there is a matching ID, the host computer 10 receives the question from the viewer and receives the viewer ID. The area name and the name (remote questioner information) stored in correspondence with are transmitted to the monitor 8 (d19). In other words, someone is trying to understand.

  When the teacher operates the portable controller 28 and selects the remote questioner information displayed on the monitor 8 (d20), the host computer 10 immediately transmits a video request command to the remote viewer terminal 31 (d21). An image of the viewer at the point is taken and transmitted (d22).

  The host computer 10 transmits the image of the viewer at this remote location to the monitor 8 (d23). In other words, the teacher can understand the face of the viewer at the remote location, so that he can understand who is asking the question.

(Operation of the main control device for teaching side monitoring)
Here, the description of the operation of the main control device for monitoring the teaching side is supplemented. First, when the teacher operates the host computer 10 to turn on the system, operation icons for operating various devices shown in FIG. 10 and an image of the classroom are displayed. This image is displayed according to the classroom.

  For example, when the teacher enters the platform, the video captured by the wide-angle camera 1 becomes a platform-side image captured of the entire image shown in FIG.

  When the teacher further proceeds, the motion detection unit 120 of the automatic tracking device 12 of the main control device 14 for monitoring the platform detects the teacher's motion image by the processing described below, and the target tracking unit 121, the tracking camera control unit 122 obtains the zoom image of FIG.

  Specifically, the motion detection unit 120 compares the motion detection area (defined by coordinates) stored in the memory (ROM, RAM) with the platform side image captured by the wide-angle camera 1, and the platform side image detects motion. When it exists in the area, it is detected as a motion detection image (teacher).

  Next, the target tracking unit 121 cuts out the teacher's motion detection image and generates tracking position information of the target. The target tracking unit 121 sends tracking position information to the tracking camera control unit 122 when a target exists (when a target is detected) in the sense area stored in the memory (see c in FIG. 12). Output and zoom in.

  The sense area is a single horizontal character as a simple method for suppressing the movement of the tracking camera in the tilt direction.

  The tracking camera control unit 122 controls pan / tilt, zoom ratio, and the like of the tracking camera based on the tracking position information. As a result, the zoom image shown in FIG. 11B is obtained.

  That is, the video of the wide-angle camera is taken out as a predetermined video by a video cutting means (not shown), and camera control information based on the video is sent to a driving unit (not shown) of the tracking camera. Performs pan / tilt operation.

  This zoom image (teacher) is taken into the image output device, outputted to the host computer, and stored in the HDD (compression).

  If the student goes up to the platform with some question (see a in FIG. 13), the motion detection unit 120 compares the previous wide-angle camera image with the current wide-angle camera image, and determines the student in the sense area. Recognize as a new target. Due to this recognition, the remote viewer terminal 31 is transmitted with the entire image shown in FIG. That is, the images are switched. Then, the target tracking unit 121 calculates tracking position information of a new target and outputs it to the tracking camera control unit 122. The tracking camera control unit 122 obtains a tilt, pan, and zoom ratio for a new target and zooms in (see FIG. 13B).

  Furthermore, when a new student enters the platform and there are two students (see Fig. 14 (a)), the tracking of the teacher and the previous student is stopped, and a large number of tracking information of new targets are targeted. The data is output to the range tracking setting unit 123.

  Further, since the multiple target range tracking setting unit 123 has a plurality of targets, tracking position information is determined at the center of two students.

  That is, the centroids of the second target and the third target are determined, the distances (L1, L2) from both ends of the sense area to the centroid of the second target and the third target are obtained, and the sense area length L to L1 , L2 is subtracted and a predetermined length m is added to define the horizontal range. Then, a range Gi of a predetermined height h is obtained, and the center of this range is set as the tracking position. That is, the tracking point of the range Gi is determined on a straight line connecting the centroids of the second and third targets.

  As a result, the tracking camera control unit obtains the tilt, pan angle, and zoom ratio so as to accommodate two students, and zooms in to track (see FIG. 14B).

  Accordingly, the automatic tracking device 7 outputs an image in which two students are zoomed in together.

    Further, it is preferable that a mask area can be set as shown in FIG. 15 so that, for example, an image projected by a projector is not erroneously recognized as a target. This mask region is preferably colored.

  Next, the description of the motion detection and tracking process described above will be supplemented using the flowchart of FIG. In this embodiment, the automatic tracking device 12 will be described as a representative.

  When the system is turned on, the motion detection unit 120 internally sets control values based on the motion detection area of the memory, the sense area, the mask area, and the mode table from the host computer 10 (S1601).

  Next, motion detection / cutout processing is performed (S1602). The motion detection unit 120 performs motion detection in the motion detection area, and the target tracking unit 122 cuts out the motion detection video to determine the symbol code.

  Since the tracking camera 2 is zoomed out since the system is turned on in the present embodiment, an image of the entire classroom of the wide-angle camera 1 is acquired.

Next, it is determined whether or not the system is immediately on (S1603). If it is determined in step S1603 that the system has just been turned on, it is determined whether there is one target (S1604). If it is determined in step S1604 that the number of targets is one, the tracking processing unit performs a tracking process for zooming in on a certain range around the first target (S1605). That is, when the teacher enters the motion detection area and exists in the sense area, it is recognized as a new target, and zoom-in processing is performed. When it is determined in step S1604 that there is not one target (student ), It is determined whether there are two or more targets (S1606).

  If it is determined in step S1606 that there are two or more targets, the number of targets is determined (S1607). If it is determined in step S1607 that the number of targets is two or less, the target tracking unit generates a signal for zooming out the first target (teacher) (S1611a), and the first small target (student) A control signal for zooming in on a predetermined range is generated with reference to (S1611b), and the process proceeds to step S1610.

  If it is determined in step S1607 that the number of targets is three or more, the first target is zoomed out (S1608), the distance between the small targets (students) is obtained, and a zoom-in range based on this distance is determined. A decision is made to zoom in (S1609), and a tracking process is performed (S1610). In other words, when a student leaves the sense area, tracking is handed over to the teacher's target.

  Therefore, when the target moves out of the sense area, it zooms out, but then the same wide-angle camera image that the camera angle is pulled will be displayed on the student terminal, so it will be in a natural state, which is preferable for distance education etc. It becomes a form. For this reason, auditors can acquire natural images without camera sickness.

  Then, when the lecture is started and multiple students approach the teacher and enter the sense area (such as report submission and role playing), the system is multiple (for example, three or more). The image is zoomed out by recognizing and transmitting a zoom-out signal to the tracking camera 2). This zoom-out operation is executed after the imaging area is returned to the center position. In addition, the zoom-out operation is also effective because it can be zoomed out first and then returned to the center position, returning to the preset center position while copying the target, so that the object will not be missed. .

<Embodiment 2>
FIG. 17 is a detailed configuration diagram of the automatic tracking apparatus according to the second embodiment. In FIG. 17, the same components as those in FIG.

  In the system according to the second embodiment, the motion detection unit 120a of the motion detection unit 120 inputs a wide-area video image of the wide-angle camera 2, and the motion detection unit 120 detects only the image information (cell set) that has changed. Then, it is determined whether or not this detected image exists in the motion detection area of the memory 120b. At this time, the detection is performed by avoiding the mask area of the memory 120c so as not to detect an unrelated part.

  The memory 121b stores a sense color for detecting a specific color and a sense shape for detecting whether there is a specific shape.

  When the clipping unit 121a of the target tracking unit 121 detects a moving image, the clipping unit 121a passes through the sense area, the presence / absence of a moving object, the size, position, moving direction, moving speed, and sensing area of the clipped motion detection image. Whether the motion detection image has a specific color or not is stored in the buffer 50.

  In addition, the cutout unit 121 a stores in the buffer 51 whether or not a specific shape based on the memory 121 b is detected in the buffer 51, the size and type of the specific shape.

  These are connected to the system bus and decoded by the computer 41 (CPU 41) so that control data of the tracking camera 2 can be obtained.

  The analog video signal from the tracking camera 2 is digitally converted by the A / D conversion means 53 and stored in the image storage device 60 (HDD, MPEG, JPEG compression) by the CPU 41.

  The audio signal from the sound collection microphone 43 is digitally converted by the A / D conversion means 54 and stored in the image storage device 16 by the CPU 41.

  On the other hand, the CPU 41 sequentially reads the data stored in the buffer memories 50 and 51 and stores the data in the buffer 56, and sends tracking control signals for controlling pan, tilt, and zoom ratio based on the position information in the buffer 56. It is generated from the camera state and output to the tracking camera control unit 122.

  The tracking camera control unit 122 reads the camera parameters in the memory 40 and applies a control signal based on the tracking control signal to the tracking camera 20.

  The memory 55 stores information for each classroom (mask range, motion detection area, sense area, specific sense color, etc.) by the operation means, and these classroom information is stored in the motion detection unit 120 when the system is turned on. The target tracking unit 121 is set.

  Furthermore, the memory 55 stores video gestures and voice gestures so that zoom-in is performed when a specific gesture or voice is generated.

  Then, by calling a menu for switching the camera use mode along with the operation of various devices, it is possible to appropriately change to, for example, the audiovisual education mode, the monitoring mode, and the attendance mode. In this embodiment, the audiovisual education mode is set.

<Embodiment 4>
The fourth embodiment is a system in which each student has a student terminal, and a platform image or a zoomed-in image of a questioner is displayed on the screen of the student terminal.

FIG. 18 is a schematic configuration diagram of a moving body imaging system according to the fourth embodiment. In the present embodiment, the platform-side monitoring main control device 26 includes an automatic tracking device 12, an image output device 13, and a transmission / reception device 27 as an access point. The student-side monitoring main control device 28 includes an automatic tracking device 16, an image output device 17, and a transmission / reception device 29 as an access point.

  The table in the classroom is equipped with student terminals 9i (9a, 9b, 9c,..., Preferably with a microphone).

  Then, a tracking image (an image on the side of the teacher) on the platform (student, teacher) photographed by the first imaging means 3 is sent via the host computer 40 to the student terminal 9i (9a, 9b, 9c,...) It transmits to the viewer terminal 31.

  In addition, the student-side image captured by the second imaging unit 4 is transmitted to the student terminal 9 i, the monitor 8, and the remote viewer terminal 31.

(Configuration of each part)
When the student terminal 9i (9a, 9b,...) Receives the video of this system, the terminal request including the student number, the lecture number, the date, the student ID, the content type (whether or not the tracking zoom video is present), etc. Enter information and receive a lecture-side image (including wide-angle, tracking, zoom) or student-side image. It has a function that can be performed by the wireless PC card for transmitting and receiving the above-mentioned terminal request information and video.

  Each of the aforementioned transmission / reception devices (access points) is provided on the classroom wall and ceiling. These transmission / reception devices are internally provided with a MAC circuit, a baseband processor, a frequency conversion circuit, an antenna (diversity method), a RAM, a ROM, a CPU, and the like. When terminal request information is received from the student terminal 9i, this terminal request When the information is sent to the image output device and the permission signal is obtained, the wide-angle camera image and the tracking zoom image of the automatic tracking device are transmitted to the student terminal 9i.

  On the other hand, the host computer 40 has a function of determining the content type from the student terminal and performing a billing process when the questioner requests zooming. In other words, the questioner's zoom screen charges for this system.

<Embodiment 5>
The fifth embodiment is a case where the teacher is tracking, and when a plurality (1 to 5 students) of students come up to the platform, only the teacher is tracked first.

  The automatic tracking device 16 changes the target when the student enters the sense area with one, two, and so on. When the student enters more than one, all the students who have entered ( The tracking camera 2 is gradually or stepwise zoomed so that the limit can be photographed by, for example, five people.

At this time, if a student (moving) crosses the teacher, the teacher is also regarded as a student and is recognized as a zoom target. For example, when a sixth student comes in, the zooming is performed at the center position. ,
In addition, you may comprise each automatic tracking apparatus of the said embodiment as a camera apparatus integrated with the camera.

It is a schematic block diagram of the mobile body image provision system of this Embodiment 1. It is a schematic block diagram of the automatic tracking apparatus of this Embodiment 1. It is explanatory drawing explaining zoom-in of many targets. It is explanatory drawing explaining the detection area for detecting the questioner in a classroom, and a sense area. It is explanatory drawing at the time of providing many imaging devices. It is explanatory drawing of lecture information. It is explanatory drawing of a setting condition mode table | surface. It is explanatory drawing explaining the setting screen of a camera mode. It is a sequence diagram explaining operation | movement of this Embodiment. It is explanatory drawing explaining an initial screen. It is explanatory drawing explaining a questioner zoom display. It is explanatory drawing explaining a wide angle camera, pan tilt, and tracking. It is explanatory drawing of a detection area (sense area, motion detection area). It is explanatory drawing explaining multiple target tracking. It is explanatory drawing of the mask area of this Embodiment. 3 is a flowchart for explaining the operation of the automatic tracking device according to the first embodiment. It is a schematic block diagram of the automatic tracking apparatus of this Embodiment 2. It is a schematic block diagram of the mobile body imaging system of this Embodiment 4.

Explanation of symbols

3 First image pickup means 4 Second image pickup means 9 Student terminal 12 in the classroom First automatic tracking device 14 Main controller for teaching platform side 16 Second automatic tracking device 19 Main control device for student side monitoring

Claims (9)

The lecturer's video for the lecturer at the auditorium is provided to the student terminal via the network, while the student's video as viewed from the lecturer is provided to the instructor's terminal and the designated terminal. A moving body imaging method,
(A). A step of photographing a lecturer-side image composed of a first wide-area video on the lecturer side and a first zoom video by a first imaging means;
(B). Capturing a student-side image composed of the second wide-area video and the second zoom video on the student side with a second imaging means,
The main control unit
(C). Providing the instructor-side image of the first imaging means to the instructor-side terminal or a designated terminal via the network;
(D). A first target is detected and tracked from the first wide-area video of the lecturer-side image, and the first zoom video of the first target is connected to the teacher-side terminal and designated terminal via the network. The process of providing
(E). Providing a student-side image of the second imaging means to the instructor-side terminal or a designated terminal via the network;
(F). A second target is detected from the student-side image, and the instructor's terminal and the designated terminal are caused to take a second zoom image of the second target and the instructor's terminal is designated via the network. And a step of providing the terminal to a terminal. (G). Based on a preset mode, a combination including the steps (a), (c), and (d) or a combination including the steps (b), (e), and (f) is selected. Process,
The moving body image providing method according to claim 1, further comprising: The mobile image providing method according to claim 1, wherein the designated terminal is a remote viewer terminal at a remote location or a student terminal in the attendance hall. While the host computer provides the lecturer's video that the lecturer at the auditorium lectures to the student terminal via the network, the student's video as seen from the lecturer by the lecturer's terminal A mobile image providing system provided to a terminal,
A first imaging device comprising a first wide-angle camera that captures a wide-area image on the instructor side and a first tracking camera that captures a zoom image by pan and tilt operations;
A second imaging device comprising a second wide-angle camera that captures a wide-area image of the student side and a second tracking camera that captures a zoom image by pan and tilt operations;
The instructor side target is detected and tracked from the wide-area image of the first wide-angle camera, and the first zoom image is transmitted instead of the wide-area image of the first wide-angle camera. A first main controller;
When there is movement in the second wide-area video from the second imaging device, the second tracking camera is controlled at a position where the target on the student side based on the motion detection image is recognized, and the attendance is A second main control device for capturing a second zoom image of the target on the user side and transmitting the second zoom image;
The host computer
Means for operating in a combination of the first imaging device and the second main control device or a combination of the second imaging device and the second main control device based on a mode designated by the instructor's terminal When,
The lecturer displays the first wide-angle camera image from the first main controller, the first zoom image, the second wide-angle camera image from the second main controller, and the second zoom image. And a means for transmitting to a remote student terminal. One of the first imaging device and the second imaging device is a tracking imaging unit that performs pan and tilt operations, and the other is a monitoring imaging unit that images a certain direction at a certain viewing angle. 5. The moving body image providing system according to claim 4, wherein a zoomed-in or zoomed-out image is obtained. The second main control device and the first main control device are:
Means for determining whether the moving object is present in a preset sense area for each detection when the moving object is detected in a preset detection region;
When the target is present in the sense area in a predetermined number or less, means for zooming out in stages until a predetermined zoom ratio including these targets is obtained;
6. The moving body image providing system according to claim 4, wherein the moving body image providing system is provided. The target recognition is
A motion detection area; and a sense area, wherein a target of the wide-area video is newly present in the motion detection area, and when the target is present in the sense area, the target position is zoomed in. The moving body image providing system according to claim 4. The first main control unit and the second main control unit are connected to a host computer,
The host computer
At least a learning mode and an audiovisual mode, and when the learning mode is selected, the second imaging unit is operated via the second main control unit to capture a zoom image of the student When,
When the audiovisual mode is selected, means for operating the first imaging means via the first main control unit to photograph the lecturer side zoom image;
When both the attendance mode and the audiovisual mode are selected, the first main control unit and the second main control unit are operated to take a zoom video on the student side and a zoom video on the lecturer side. The moving body image providing system according to claim 4, wherein the moving body image providing system is provided. The host computer has a web server function, and when an external student terminal is selected from the instructor side terminal, sends a shooting command to the external student terminal, and The moving body image according to any one of claims 4, 5, 6, 7 and 8, further comprising means for transmitting an external student video from the student terminal and providing the video to the instructor terminal. Offer system.

Images