JP2006013832A - Video photographing apparatus and video photographing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a video photographing apparatus and a video photographing program whereby a plurality of users can utilize video images of different azimuths, the need for expensive specific software for distortion correction can be eliminated, and only video images in particular azimuths can continuously be extracted. <P>SOLUTION: The video photographing apparatus 1 extracts a frame image by each particular azimuth from an omniazimuth video image resulting from photographing an omniazimuth of a center point for a photographing object, outputs the frame image, and is provided with a photographing means 3, a rotation means 5, and a frame image extract means 7. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a video photographing apparatus that photographs a photographing target, and more particularly, to a video photographing apparatus and a video photographing program that output a video in an arbitrary direction from an omnidirectional video obtained by photographing a photographing target in all directions.

  Conventionally, a fixed-type robot camera (conventional camera) that installs a video camera on a pan head fixed on the rooftop of a building and controls the video camera to shoot a video in a desired direction from the installation point (installation point) Is known. This fixed robot camera captures a video in a desired direction and transmits the weather, traffic jam situation, and vibration situation during an earthquake around the installation point to a TV station (broadcast station) or the like in real time.

  In addition, a conical mirror with its apex facing downward so that incident light from all directions (surroundings) is emitted as reflected light, and reflected light emitted from the conical mirror installed below this conical mirror are used as an input for photographing. There is known an omnidirectional imaging apparatus comprising a camera body that shoots a circular omnidirectional video by a video processing unit and a video processing unit that demodulates the omnidirectional video captured by the camera body into a square omnidirectional video. (For example, refer to Patent Document 1).

  That is, in the omnidirectional imaging device described in Patent Document 1, incident light from all directions (around) is reflected by the conical mirror to the surface of the photographic lens of the camera body. The camera body captures the reflected light from the conical mirror as an input, and stores the captured image as each pixel in each memory element of a memory provided in the image processing unit.

  Note that the captured image captured by the camera body is a circular image because the camera body captures the reflected light from the conical mirror as an input. In addition, the captured image is distorted due to reflection by the conical mirror, and this distortion becomes smaller as the center of the captured image (the apex of the conical mirror) is approached. For this reason, each pixel stored in each memory element of the memory provided in the video processing unit is demodulated into a regular video by processing of a distorted circular shot video by software provided in the video processing unit. As a result, it was obtained.

As a result, in the omnidirectional imaging apparatus, it is possible to shoot in all directions (surroundings) with one camera body, and the video processing unit can demodulate the captured video into undistorted video (each pixel). Become.
Japanese Unexamined Patent Publication No. 2000-261705 (paragraphs 0011 to 0012, FIG. 1)

  However, the fixed type robot camera (conventional video shooting device) is exclusively used by a specific user such as a technician of the TV station that owns the robot camera, and uses the same fixed type robot camera. Thus, there is a problem that a third party cannot use videos in different directions (directions).

  Further, the omnidirectional photographing apparatus has a problem that a distorted circular photographed image photographed by the camera body must be demodulated into a regular image by processing of software provided in the image processing unit. .

  Further, in this omnidirectional photographing apparatus, there is a problem that software for distortion correction, which is provided in the video processing unit, must be developed, and the entire apparatus becomes expensive (high cost).

  Furthermore, if you want to obtain local information about the surrounding area where the camera body is installed, based on all or part of the video (omnidirectional video) taken with this omnidirectional imaging device, specify There is a problem that it is difficult to use for acquiring the regional information because there is no means for designating the video in the direction (specific direction). As an example of this problem, for example, the omnidirectional imaging device can be remotely operated to visually recognize in a specific direction, and cannot be used for a survey such as measuring the number of pedestrians and traffic vehicles on the road.

  Alternatively, in this omnidirectional imaging apparatus, the captured video (omnidirectional video) is drawn in one frame, so that the image quality deteriorates. For example, the role of a security camera for monitoring the inside of a building Cannot be achieved, and there is a problem that versatility is low.

  In order to solve the above problems, the present invention allows a plurality of users to use videos of different orientations, eliminates the need for expensive dedicated software for distortion correction, and degrades image quality. An object of the present invention is to provide a video shooting apparatus and a video shooting program that can continuously extract only videos in a specific direction.

  In order to achieve the above object, the video photographing apparatus according to claim 1, wherein the video photographing apparatus extracts and outputs a frame image for each specific azimuth from an omnidirectional video photographed by photographing all azimuths at a certain central point. The apparatus is configured to include a photographing unit, a rotating unit, and a frame image extracting unit.

  According to such a configuration, the video image capturing apparatus captures an image to be imaged by the image capturing unit, and rotates the image capturing unit around the center point by the rotating unit. In other words, the imaging object is continuously imaged in all directions (omnidirectional) on the circumference of a certain center point while the imaging means is rotated by the rotation means. The shooting target includes a subject and a background. Then, the video imaging apparatus extracts and outputs a frame image for each specific direction set in advance by the frame image extraction means. The specific orientation refers to a direction at a certain angle (for example, an angle in increments of 12 °) from a reference line passing through a certain center point. In this frame image extraction means, from a omnidirectional image at a certain angle. Frame images are extracted sequentially.

  According to a second aspect of the present invention, in the video photographing apparatus according to the first aspect, the rotating means stops the photographing means at every predetermined angle and circulates around the center point at a constant time. And

  According to such a configuration, the video photographing apparatus is configured such that when the rotating unit rotates the photographing unit around the center point for a predetermined time (for example, one rotation / one second), the photographing unit captures each frame ( For example, the rotation unit stops at a certain time (for example, 1/30 second) so as to match 1/30 second per frame), so that the imaging unit does not rotate (stops) when shooting each frame. Image blurring can be prevented.

  According to a third aspect of the present invention, there is provided the video photographing apparatus according to the first or second aspect, further comprising an omnidirectional video recording means for recording the omnidirectional video photographed by the photographing means. .

  According to such a configuration, the video photographing apparatus records the omnidirectional video by the omnidirectional video recording means. Therefore, when extracting the frame image by the frame image extracting means, the video imaging apparatus refers to the omnidirectional video recording means. It can be processed even if the number of frame images to be extracted is increased.

  According to a fourth aspect of the present invention, there is provided a video shooting program in which a frame image for each specific direction is obtained from an omnidirectional video imaged by the imaging unit by rotating the imaging unit with the omnidirectional image at a central point. The apparatus for extracting and outputting is configured to function as rotation control means, photographing control means, and frame image extraction means.

  According to such a configuration, the video shooting program causes the rotation control unit to stop the shooting unit at a predetermined angle set in advance, and causes the shooting unit to take a shooting target when stopped by the shooting control unit. . Then, the video shooting program extracts and outputs a frame image for each predetermined specific direction from the omnidirectional video by the frame image extraction means.

  According to the first and fourth aspects of the present invention, since frame images with different orientations are extracted from the omnidirectional video, a plurality of users can use the video with different orientations, and a conical mirror or the like is used. Because there is no distortion in the captured omnidirectional video, it is possible to eliminate the need for expensive dedicated software for distortion correction, by rotating the imaging means multiple times with the rotation means, The frame image extraction means can continuously extract frame images in the same direction (specific direction).

  According to the second aspect of the present invention, the photographing unit is stopped at every certain angle, and the photographing unit photographs the photographing object and circulates in a certain time, so that the photographing unit is stopped at every certain angle and each Since the frame is photographed, blurring of the photographed image can be prevented.

  According to the third aspect of the present invention, when the omnidirectional video is recorded in the omnidirectional video recording means, the frame image can be extracted and the omnidirectional video recording means can be referred to. It is possible to process even if it is increased.

Next, embodiments of the present invention will be described in detail with reference to the drawings as appropriate.
(Configuration of video shooting device)
FIG. 1 is a block diagram of a video photographing apparatus. The video photographing device 1 extracts a frame image for each specific azimuth (angle) from an omnidirectional video obtained by photographing all directions around a certain point (center point). As shown in FIG. A rotation means 5, a frame image extraction means 7, a control means 9, and an omnidirectional video recording means 11.

  The photographing means 3 is a photographing camera (TV camera, video camera, etc.), for example, taking a subject, a background (landscape), etc. as a subject to be photographed while being rotated by the rotating means 5. The photographing means 3 is capable of photographing a photographing target in real time and outputting a video (omnidirectional video [details will be described later)) obtained by photographing the photographing target. The photographing means 3 is controlled to start and stop photographing in accordance with control signals (shooting start signal and photographing stop signal) output from the control means 9.

  In this embodiment, the video imaged by the imaging unit 3 is output to the frame image extraction unit 7 and the omnidirectional video recording unit 11. For example, other devices (such as the Internet and an intranet) may be used. Server, display device, etc.). In this case, means for processing the video imaged by the imaging means 3 (corresponding to the frame image extraction means 7) except for the case where the video imaged by the imaging means 3 is simply displayed on the other device side. ).

  Incidentally, the installation location of this photographing means 3 is, for example, the rooftop of a building or an intersection. By appropriately installing in these places, it is possible to obtain a desired video that changes every moment.

  The rotating means 5 rotates (circulates) the photographing means 3 around a certain point in accordance with the control signals (rotation start signal, rotation stop signal, rotation number per unit time) output from the control means 9. A pan head (not shown), a shaft (not shown), and a drive source (not shown) are included. While the photographing unit 3 is rotated by the rotating unit 5, the photographing unit 3 shoots images in all directions (omnidirectional) on the circumference of the center point of the rotating unit 5. The video to be played is an omnidirectional video.

  The camera platform (not shown) mounts and fixes the photographing means 3 and includes a control line connection interface connected from the control means 9 to the photographing means 3. With this connection interface, the control signal output from the control means 9 is input to the photographing means 3. When the photographing means 3 is fixed to the pan head (not shown), the photographing means 3 can be placed horizontally and fixed, or the photographing means 3 is inclined upward or downward. It can also be fixed.

  The shaft (not shown) is connected (axially supported) to a pan head (not shown) so that the photographing means 3 can be adjusted to an arbitrary height according to the installation location of the photographing means 3. .

  The drive source (not shown) rotates a shaft (not shown) clockwise or counterclockwise a predetermined number of times in a fixed time. Further, this drive source (not shown) can be stopped at every arbitrary angle while rotating the photographing means 3. For example, it is configured by a stepping motor or the like.

  Further, when a stepping motor or the like is used for this drive source (not shown), the rotation can be stopped accurately, so that the image photographed by the photographing means 3 is prevented from being shaken in the rotation direction. be able to.

  For example, when the frame rate at the time of shooting by the shooting unit 3 is 30 frames / second and the shaft (not shown) is rotated once per second, the image shot by the shooting unit 3 is rotated in the rotation direction. In order to prevent shaking, a shaft (not shown) may be stopped at every rotation angle of 12 °.

Note that the number of rotations by which the photographing means 3 is rotated by this drive source (not shown) can be arbitrarily set by a control signal from the control means 9.
Since the photographing unit 3 continuously photographs while being rotated by the rotating unit 5, the photographed image is obtained by photographing a different direction for each frame image. By extracting a frame image obtained by photographing the same direction from the output video and arranging the extracted frame images in time series, a specific direction video can be obtained.

  Here, if the rotation speed of the photographing means 3 by the rotation means 5 is k (rotation / second) and the frame rate of the image photographed by the photographing means 3 by the rotation means 5 is k (rotation / second), the image is output from the photographing apparatus. The frame rate of the specific direction video is k (frame / second), and the extraction angle to be set (unit angle that can be set (selected) by the user) is (360 × k) / j unit angle. In order to acquire a video in the same direction (specific direction video), the value of k needs to be divisible by the value of j.

  The frame image extraction means 7 extracts the omnidirectional video imaged by the imaging means 3 as a frame image for each direction (each direction), and outputs it as a specific direction video in which the frame images in the same direction are continued in time series. Is. Here, referring to FIG. 2, the operation of the frame image extracting means 7 is carried out when the photographing means 3 is rotated at a rotational speed of 1 rotation / second and the photographing means 3 captures an image at a frame rate of 30 frames / second. A form is demonstrated as an example. In this embodiment, since the photographing means 3 is photographing at a frame rate of 30 frames / second, one frame is photographed every 1/30 seconds. Since the photographing means 3 is rotating at 1 rotation / second, one frame is photographed every time the camera rotates 12 °, and 30 frames are photographed after one second, and the photographing means 3 is rotated 360 ° to change the original direction. It will turn.

  FIG. 2 shows in which direction the photographing means 3 is photographing after the start of photographing (after the start of rotation) and after n seconds (n = 0, 1, 2, 3,...). The frame image extraction means 7 extracts 30 frame images photographed per second by the photographing means 3 × n = 30 n frame images by extracting frame images photographed in the same direction in an arbitrary direction. A moving image corresponding to an arbitrary direction is generated as a specific direction video. In this embodiment, since the photographing unit 3 faces the same direction every time one second elapses, the same direction is photographed by extracting frame images at intervals of 30 frames from consecutive frame images photographed by the photographing unit 3. The extracted image can be generated.

  The extraction unit angle of the frame image extracted by the frame image extraction means 7 can be arbitrarily set within a range that satisfies the condition based on the relationship between the frame rate and the rotational speed. In the embodiment of FIG. 2, extracted images in 30 directions can be obtained at intervals of 12 ° from the relationship between the shooting frame rate and the number of rotations, and the extraction unit angle is 12 °.

  The control means 9 is responsible for controlling the video photographing apparatus 1, and the photographing means 3 and the rotating means 5 according to an instruction (operation signal) input by an operator of the video photographing apparatus 1 through an input means (not shown). And a control signal to the frame image extraction means 7 and the omnidirectional video recording means 11. The control signals output from the control means 9 to the photographing means 3 are a photographing start signal and a photographing stop signal, and the control signals output to the rotating means 5 are a rotation number / angle setting signal, a rotation start signal, and a rotation stop signal. The control signal output to the frame image extraction unit 7 is a signal for specifying the extracted video direction, and the control signal output to the omnidirectional video recording unit 11 is an omnidirectional video output signal.

  The omnidirectional video recording means 11 records the omnidirectional video imaged by the imaging means 3 and is constituted by a general recording medium such as a hard disk. The omnidirectional video recorded in the omnidirectional video recording unit 11 is appropriately selected from the frame image extraction unit 7 or the video photographing apparatus 1 according to the control signal (omnidirectional video output signal) output from the control unit 9. Output to the outside. When an omnidirectional video is output to the frame image extraction unit 7, a frame image in a specific direction is extracted by the frame image extraction unit 7, and is output from the video photographing apparatus 1 as a specific direction video. That is, when output to the outside of the video shooting device 1, it is handled as an omnidirectional video (no processing), or a frame image in a specific direction is extracted from this omnidirectional video (extraction processing). Processing will be done.

(Examples of omnidirectional images shot by shooting means)
Next, with reference to FIG. 3, an example of an omnidirectional video imaged by the imaging means 3 of the video imaging device 1 will be described (see FIG. 1 as appropriate). FIG. 3 shows an example of an omnidirectional video photographed by the photographing means 3.

  As shown in FIG. 3, an omnidirectional image for the first rotation (n = 0), an omnidirectional image for the second rotation (n = 1), an omnidirectional image for the third rotation (n = 2), The fourth rotation (n = 3) omnidirectional images are listed from the top to the bottom of FIG. Each frame image constituting the omnidirectional video has an angle at which the frame image is captured and a frame serial number (frame #) of 0 ° (30n + 1), 12 ° (30n + 2), 24 ° (30n + 3), 36. ° (30n + 4), 48 ° (30n + 5), 60 ° (30n + 6), ... 348 ° (30n + 30) are given.

  Then, the frame image extraction means 7 generates a specific direction video based on the frame serial number. In the present embodiment, the photographing means 3 takes pictures at the frame rate of 30 frames / second while being rotated by the rotating means 5 at a rotation speed of 1 revolution / second, so the same direction is taken every 30 frames. Therefore, in order to extract a video in a specific direction, it is only necessary to extract frame images at intervals of 30 frames and to continue the extracted frame images. For example, the specific direction video in the 0 ° direction is a series of frame images of frame serial numbers, frame # = 1, frame # = 31, frame # = 61, frame # = 91,. The specific direction video in the 36 ° direction is obtained by continuing frame images of frame serial numbers, frame # = 4, frame # = 34, frame # = 64, frame # = 94,.

(Example of frame image [specific direction video] extracted by frame image extraction means)
Next, an example of a frame image (specific direction video) extracted by the frame image extraction means 7 of the video photographing device 1 will be described with reference to FIG. FIG. 4 shows an example of the frame image (specific direction video) extracted by the frame image extraction means 7.

  The specific direction image (specific direction image in the 0 ° direction) shown in FIG. 4A has a frame serial number, frame # = 1, frame # = 31, frame # = 61, frame # = 91,. A series of frame images. It is a moving image with a frame rate of 1 frame / second taken in the same direction (0 ° direction).

  Also, the specific direction image (specific direction image in the direction of 48 °) shown in FIG. 4B has a frame serial number, frame # = 5, frame # = 35, frame # = 65, frame # = 95,. These frame images are continuous. Similar to the specific direction video in the 0 ° direction, the video is a moving image with a frame rate of 1 frame / second taken in the same direction (48 ° direction).

(Internal structure of frame image extraction means)
Next, the internal configuration of the frame image extracting means 7 and the operation of each configuration will be described with reference to FIG. 5 (see FIG. 1 as appropriate). FIG. 5 shows an example of the configuration of the frame image extraction means.

  In FIG. 5, reference numeral 71 denotes an extraction processing control means that controls the frame image extraction means. The extraction process control means 71 receives the control signal from the control means 9 and controls each component of the frame image extraction means 7 according to the content of the received control signal.

  The omnidirectional video input to the frame image extraction means 7 is first temporarily stored in the input video buffer 72. From the omnidirectional video stored in the input video buffer 72, the video frame filter means 73 extracts only a frame image in a specific direction. The frame image extracted by the video frame filter means 73 is output to the output video buffer 74, and the output video buffer 74 temporarily stores it. In the case of the present embodiment, since the photographing means 3 photographs all directions while rotating and faces the same direction every second, one frame image is intermittently input to the output video buffer 74 at intervals of one second. Come.

  The output video creation means 75 complements the frame images between intermittent frame images (in this example, 1 frame / second) stored in the output video buffer 74 and can display a frame rate (for example, 30). Frame / sec) video is created and output as a video in a specific direction.

  The extraction processing control means 71 is responsive to the content of the control signal from the control means 9 so that the specific direction video in the designated direction is output in an appropriate format that can be displayed on a display or the like. The means 73, the output video buffer 74 and the output video creation means 75 are controlled. The control signal output from the extraction processing control means 71 to the input video buffer 72 is the number of frame images to be temporarily stored, the control signal output to the video frame filter means 73 is the frame number to be extracted, and output The control signal output to the video buffer 74 is the retention time of the input extracted frame image, and the control signal output to the output video creation means 75 is a control signal for controlling the frame image complementing method.

(Operation of video camera)
Next, the operation of the video photographing apparatus 1 will be described with reference to the flowchart shown in FIG. 6 (see FIG. 1 as appropriate).
First, the video imaging apparatus 1 selects (sets) the extraction angle (extraction angle) of the frame image extracted by the frame image extraction means 7 and the rotation number by which the imaging means 3 is rotated by the rotation means 5 (S1).

  Then, the video photographing apparatus 1 rotates the photographing unit 3 with the rotating unit 5 (S2), and photographs the photographing target with the rotating photographing unit 3 (S3).

  Then, the video imaging apparatus 1 determines whether or not the omnidirectional video captured by the imaging unit 3 is output as it is as an omnidirectional video (S4), and when it is determined to output (S4, Yes), the operation of the video imaging apparatus 1 Ends.

  If it is not determined in S4 to output as an omnidirectional video (No in S4), the frame image extraction means 7 extracts each frame image in the direction set from the omnidirectional video (S5), and this extracted frame The images are continued in time series and output as a specific direction video (S6).

(Example of system construction using video camera)
Next, with reference to FIG. 7 and FIG. 8, a construction example of a video photographing system using the video photographing device 1 will be described. 7 and 8 are diagrams illustrating a video shooting system in which each configuration of the video shooting device 1 is divided into two (three).

  The video shooting system A shown in FIG. 7 includes a shooting unit 3 and a rotation unit 5 of the video shooting apparatus 1, a frame image extraction unit 7, a control unit 9, and an omnidirectional video recording unit 11 (these are stored in the user servers a and b). The internal and omnidirectional video recording means 11 are connected to each other by a distributor 15 and a monitor 17 for displaying the output from the frame image extracting means 7 is connected.

  This video shooting system A enables a video shot by one shooting means to be used simultaneously by a plurality of users. The omnidirectional video captured by the imaging unit 3 rotated by the rotation unit 5 is distributed, and a specific frame is extracted by the frame image extraction unit 7 of each of the user servers a and b. Video). In this video shooting system A, the omnidirectional video shot by the shooting means 3 is simply distributed by the distributor 15 so that a plurality of users can use it simultaneously. With such a configuration, each user can freely select a video in a favorite direction (specific direction video).

  The video imaging system B shown in FIG. 8 connects the imaging means 3 and the rotation means 5 of the video imaging apparatus 1 to the control means 9 and the omnidirectional video recording means 11 (which are included in the server c) via a data cable or the like. Then, the server c is connected to the frame image extraction means 7 and the control means 9 (which are respectively included in the user servers d and e) via the network 19 to display the output from the frame image extraction means 7 17 is connected.

  This video imaging system B enables a plurality of users to simultaneously use videos taken by one imaging means via a network. A plurality of users at different locations acquire an omnidirectional video of a remote location photographed by the imaging means 3 via the network, and are included in the user server from the omnidirectional video of the remote location obtained via the network. The frame image extraction means 7 can acquire videos in any direction (specific direction videos).

  As mentioned above, although embodiment of this invention was described, this invention is not limited to the said embodiment. For example, although the present embodiment has been described as the video photographing device 1, for example, the processing of each component of the video photographing device 1 can be regarded as a video photographing program described in a general-purpose computer language. In this case, the same effect as that of the video photographing apparatus 1 can be obtained.

It is a block diagram of the video imaging device concerning an embodiment. It is a figure explaining the timing at which each frame image which comprises an omnidirectional video is image | photographed when an imaging | photography means image | photographs an omnidirectional video, rotating by the rotation means. It is a figure explaining the example of the omnidirectional video image | photographed with the imaging | photography means. It is a figure explaining the example of the specific direction image | video extracted by the frame image extraction means. It is a figure explaining the detail of a structure of the frame image extraction means. 2 is a flowchart for explaining the operation of the video photographing apparatus shown in FIG. 1. It is a figure explaining that this system can be simultaneously used by a plurality of users by separating and configuring each configuration of the video photographing apparatus of the present invention. It is a figure explaining that this system can be used simultaneously by a plurality of users in different places by separating each configuration of the video photographing apparatus of the present invention and using a network.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image | photographing device 3 Imaging | photography means 5 Rotating means 7 Frame image extraction means 9 Control means 11 Omnidirectional video recording means

Claims (4)

A video shooting device that extracts and outputs a frame image for each specific direction from an omnidirectional video that has been shot with the omnidirectional image of a certain central point as a shooting target,
Photographing means for photographing the photographing object;
Rotating means for rotating the photographing means around the center point;
A frame image extraction means for extracting and outputting a frame image for each specific orientation set in advance from an omnidirectional video imaged by the imaging means;
A video photographing apparatus comprising:   The video photographing apparatus according to claim 1, wherein the rotating unit stops the photographing unit at a predetermined angle and circulates the center point for a predetermined time.   The video imaging apparatus according to claim 1, further comprising an omnidirectional video recording unit that records an omnidirectional video captured by the imaging unit. An apparatus for extracting and outputting a frame image for each specific direction from an omnidirectional video imaged by the imaging means by rotating the imaging means with the rotating means, with the omnidirectional image of a certain center point as an imaging target,
A rotation control unit for controlling the rotation unit to stop the photographing unit at a predetermined angle;
A photographing control means for controlling the photographing means whose rotation is controlled by the rotation controlling means to photograph the photographing object;
A frame image extraction means for extracting and outputting a frame image for each specific orientation set in advance from an omnidirectional video imaged by the imaging means;
A video shooting program characterized by functioning as

Images