JPH1032746A - Method and device for automatically following up camera - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for automatically following up camera by which a moving object can automatically be followed up, a request for saving equipment and the labor of staffs can be dealt with, the object can be caught large and sharp and follow-up operation can be executed similarly to the followup operation by a skilled cameraman. SOLUTION: The motion of a target object in an image pickup range is detected by the video signals of two cameras 1 and 2 with different magnifications, parameters for controlling universal heads 5 and 18 of two cameras 1 and 2 are calculated by two parameters corresponding to these different magnifications, one camera 1 has the field of view covering the monitored range, the other camera 2 monitors the concerned object, the optimum moving amount of cameras is estimated from the correlative relation between the motion of the entire target object extracted by one camera 1 and the motion of the target object extracted by the other camera 2 and based on this estimated value, the respective cameras are moved, so that the image over the optimum image pickup range can be picked up.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic camera tracking method and apparatus capable of capturing an image of a subject while automatically following the movement of the subject unattended.
It is applicable to unmanned television cameras, unmanned surveillance systems, and the like.

[0002]

2. Description of the Related Art In recent years, with the development of digital video technology, for example, digital satellite television broadcasting of multi-channels, such as 150 channels, and various and more information are efficiently transmitted by packetizing and multiplexing. Digital lines (such as ISDN) that can be used are becoming widespread. As the number of channels and multimedia increases, it is required to produce high-quality programs efficiently and at low cost.

[0003] Focusing on television broadcast programs, drama,
It can be classified into various genres such as music, sports, news, commercial messages (CM), etc., and the recording form and editing technique differ according to each genre.
On the other hand, in the multi-channel era described above, there are many regional programs that can be edited in a short period of time, have low production costs, can be edited in a short time, and have many documentary and timely programs. Is expected. Among them, the number of event programs and sports broadcasting is expected to increase in the future, but how to reduce the number of relay staff is a key prerequisite. This is because, in the multi-channel era, it is difficult to physically secure the relay staff for all the channels, both costly and physically. Therefore, labor saving of equipment and staff will be increasingly required in the future.

[0004] Therefore, it has been considered that the camera is moved following the movement of the specific subject by focusing on a specific subject in the field of view of the camera. However, according to this method, in order to reliably capture a specific subject, it is necessary to set a wide screen and pay attention to some of the subjects, and the image of the target subject is small and unclear. There is a difficulty that becomes.

FIG. 9 shows an example of a conventional television relay. One television camera 31 is provided with one cameraman 33, and the cameraman 33 has the television camera 3 on site.
The subject 32 is photographed while operating the camera 1. Subject 3
In many cases, there is a moving thing in 2
The cameraman 33 performs camera work such as panning, tilting, and zooming while constantly observing a moving object in the field of view of the television camera 31 so as to obtain an optimal image according to the situation. ing.

[0006] Camera work differs depending on the program. For example, even in sports broadcasting, baseball, soccer, tennis,
There are athletes' movements, stadiums, etc., which are specific to the sport, such as marathons, etc., and the installation position of the camera, the camera work, etc. greatly differ depending on these various conditions. The camera work for obtaining a more realistic image has been cultivated through the long experience of photographers, and it is difficult to quantitatively analyze the photographing method.

FIG. 8 shows an example of a relationship between eye movements of a veteran photographer and camera work (here, panning), in which time is plotted on the horizontal axis and speed is plotted on the vertical axis. is there. As can be seen from FIG. 8, the cameraman follows the moving object in the visual field and moves the camera following the moving object. The relationship between the panning angle and the maximum speed of the eye movement is shown. For, it can be seen that rather than increasing the panning angle in proportion to the maximum speed of eye movement, there is an upper limit on the speed, and photographers tend to empirically try to achieve a panning image that is easy for the viewer to see . In addition, the panning speed curve has an asymmetric shape in which acceleration is fast and slow is slow. Generally, the deceleration time tends to be about 60% longer than the acceleration time.

[0008]

SUMMARY OF THE INVENTION The present invention has been made in view of the above points, and has been made so as to automatically follow a moving object and to provide the above-described equipment and staff. A camera automatic tracking method capable of responding to a demand for labor saving, capturing a target subject in a large and clear manner, and further, making the following operation comparable to the following operation by a veteran photographer, and a method thereof. It is intended to provide a device.

[0009]

The invention according to claim 1 relates to a camera automatic tracking method, wherein the movement of a target object within an imaging range is detected from video signals of two cameras having different magnifications. The parameters for controlling the camera platform of the two cameras are calculated using the two parameters corresponding to different magnifications. One camera has a field of view covering the monitoring range, and the other camera monitors the object of interest. And 1
The optimum amount of movement of the camera is estimated from the correlation between the movement of the entire target object extracted by one camera and the movement of the object of interest extracted by another camera, and each camera is determined based on the estimated value. It is characterized in that an optimal imaging range is imaged by moving the image.

[0010] The invention according to claim 2 relates to an automatic camera follow-up device, and has two cameras having different magnifications.
One camera has a field of view covering the monitoring range, and the other camera monitors an object of interest within the monitoring range, and monitors the movement of the target object within the imaging range of the two cameras. A parameter for controlling the camera platform of the two cameras is calculated based on the motion detection unit to be detected and the two parameters corresponding to different magnifications of the two cameras, and the motion of the entire target object detected by the one camera A calculation unit for estimating an optimal movement amount of each camera from a correlation between the movement of the object of interest detected by the other camera and
The camera further includes a camera platform control unit that controls the camera platform of each camera based on the estimated value to capture an image of an optimal image capturing range.

According to a third aspect of the present invention, in the second aspect of the present invention, there is provided a synthesizing means for synthesizing an image picked up by two cameras on one screen and displaying it.

[0012]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of an automatic camera follow-up method according to the present invention. FIG. 2 shows an outline of the camera automatic tracking method and the apparatus according to the present invention. It has a main camera 1 with a relatively wide viewing angle θ and a sub camera 2 with a viewing angle α smaller than the viewing angle θ, and focuses on a part of the subject within the field of view 21 of the main camera 1. This is set as the field of view 21a of the sub camera 2, the movement of the target object in the field of view 21a of the sub camera 2 is tracked by the sub camera 2, and the movement of the target object during the entire imaging of the main camera 1 is detected. The optimum motion of the main camera 1 and the sub camera 2 is estimated and quantified from the correlation between the motion of the entire target object extracted by the sub-camera 1 and the motion of the target object extracted by the sub-camera 2, and based on the estimated value, Then, the respective cameras 1 and 2 are moved to capture an image of the optimal imaging range. That is, the sub camera 2 corresponds to a sensor camera that detects the movement of the subject, and corresponds to the eyes of the cameraman. The main camera 1 and the sub camera 2 can perform panning, tilting, and zooming movements.

Next, a specific example of the camera automatic tracking method and its apparatus will be described. In FIG. 1, a main camera 1 and a sub camera 2 are mounted on pan heads 5 and 18 that can perform panning and tilting, respectively, and have imaging lenses that can perform zooming. The main camera 1 and the sub camera 2 have different magnifications, the main camera 1 has a wide field of view covering the range to be monitored, and the sub camera 2 has a narrower field of view than the main camera 1 and is within the field of view of the main camera 1. Monitor the object of interest. The video signal from the main camera 1 is input to the motion detector 10. The motion detection unit 10 includes a video sampling control unit having a filtering function, a key processing unit to which an output of the video sampling control unit is input, a center of gravity calculation unit to which an output of the key processing unit is input, and an output of the center of gravity calculation unit. It has an input geometric operation processing unit and the like, detects the motion of the entire target object extracted by the main camera 1, and outputs this as a vector amount.

The output of the motion detector 10 is input to the CPU 9. The CPU 9 controls these while exchanging signals with the key processing unit and the arithmetic processing unit based on a command signal input from the operation panel 17,
Further, it controls the camera platform control unit 8 and the display position setting unit 30.
The pan head controller 8 controls the pan head 5 of the main camera 1 based on the two vector amounts calculated by the CPU 16 to perform panning and tilting control. Head control unit 8
Includes a smoothing process for smoothly performing panning and tilting of the main camera 1 without discomfort.

A video signal from the sub-camera 2 is input to an object extracting unit 20. The object extracting unit 20 includes a key processing unit, a mask processing unit that receives an output of the key processing unit as an input, and converts a chroma key signal, that is, a specific color component from a video signal from the sub camera 2 into a key. And the mask processing is performed on the key signal by the mask processing unit. The masked key signal is input to the motion detection unit 14 and the CPU 16 as an output of the object extraction unit 20. The motion detecting unit 14 includes a center of gravity calculating unit that receives an output of the mask processing unit as an input, an arithmetic processing unit that receives an output of the center of gravity calculating unit, and the like. The output of the motion detector 14 is input to the CPU 16. The CPU 16 controls these based on a command signal input from the operation panel 17 while exchanging signals with the key processing unit, the arithmetic processing unit, and the pan head control unit 15. Head control unit 1
5 controls the pan head 18 of the sub camera 2 based on the vector amount calculated by the CPU 16 to perform panning and tilting control. The vector amount calculated by the CPU 16 is input from the CPU 16 of the sub camera 2 to the CPU 9 of the main camera 1.

The video signal A of the main camera 1 and the video signal B of the sub camera 2 are input to a synthesizing unit 31 and synthesized.
The display position signal from the display position setting unit 30 is input to the synthesizing unit 31, and the display position of the video signal B of the sub camera 2 is set. The video signal synthesized by the synthesizing unit 31 is input to the monitor 32 and displayed on the monitor 32.

Next, the operation of the above camera automatic tracking method and apparatus will be described. The viewing angle θ of the main camera 1 is relatively wide and covers a predetermined monitoring range. An example of an image of the visual field 21 is shown in the upper part of FIG. In addition, a part of the target object within the field of view 21 of the main camera 1, for example, a range in which one person is focused on by focusing on one person is set as the field of view 21 of the sub camera 2.
a. The CPU 21 determines which range of the target object within the field of view 21 of the main camera 1 should be included in the field of view 21a of the sub camera 2 by designating from the operation panel 17.
It is set by 16 commands. The target object extraction unit 20 extracts the target object from the video 23 captured in the field of view 21a of the sub camera 2. This object extraction unit 20
Then, key processing is performed to extract a specified specific target object portion from the field of view 21a of the sub camera 2 as a key signal. When recognizing a specific target part by color, the operator designates the number of colors from the operation panel 17 and extracts the specified specific color component from the RGB signal from the sub camera 2. If the specific object is a human,
For example, one method is to specify a human skin color as a specific color component.

Mask processing is performed on the extracted key signal. This extracts the object of interest from the video captured with the chroma component corresponding to the specified object,
In order to delete objects having the same chroma component in the captured image, a mask is applied to objects other than the object of interest. In the masked image, the center of gravity of the specific target object portion is obtained by the center of gravity calculation unit. A well-known dedicated IC for calculating the center of gravity
Can be used. If the specific object of interest moves, its center of gravity also moves, and this movement is detected by the motion detection unit 14. The motion detection unit 14 extracts the motion of the specific target object as a vector amount, and inputs this to the CPU 16. CP
U16 controls the camera platform 18 of the sub camera 2 via the camera platform control unit 15 with the detection signal from the motion detection unit 14 during the blanking period, and terminates the above series of processing in one frame period.

On the other hand, the video signal from the main camera 1 is input to the motion detection unit 10 and the accuracy of the pixels of the main camera 1 is determined by using the limited image range information from the mask processing unit on the sub camera 2 side. Is reduced to the key size of the sub camera 2. FIG. 4A shows the concept of the filtering process. In the field of view 21 of the main camera 1, an image range equivalent to the field of view 21a masked by the sub camera 2 is set.
As shown in (b), the field of view 21 of the main camera 1 is partitioned into an image range equivalent to the masked field of view 21a.

The filtered image information is input to the key processing unit in the motion detection unit 10. When recognizing a specified specific subject portion by color as described above, the key processing unit extracts a specified color component as a key signal. By the above filtering process, the sub camera 2
The key signal is extracted with the same parameters as in. This key signal is input to the center-of-gravity calculation unit in the motion detection unit 10,
The center of gravity is calculated similarly to the sub camera 2 side. FIG. 4 (b)
Shows an example in which the center of gravity of the entire image shown in FIG. 4A is obtained, and four locations having designated color components are extracted as key signals. Since the position of the center of gravity moves with the movement of the subject, the movement of the subject is detected while comparing the position of the center of gravity between the fields or frames in the geometric operation processing unit.

Since the sub-camera 2 follows the movement of a human in the above example, when a specific person in the field of view 21 of the main camera 1 moves as shown in FIG. The center of gravity P of the specific color component of the image of the main camera 1 also moves to P ′ (see FIG. 5B). This movement amount takes the center of gravity G (Xt, Yt) with respect to the extracted video. Here, t indicates an arbitrary field. Also, the center of gravity taken in the next field is G (Xt + 1, Yt +
1) and the moving amount (vector amount) of the object of interest is Rb
Assuming (X, Y), Rb (X, Y) = G (Xt + 1, Yt + 1) -G (Xt, Yt) (1), and the movement amount Rb can be calculated from the equation (1). . If the main camera is moved according to the calculated moving amount and in the same moving direction, an optimal image can always be obtained according to the movement of the subject.

In the main camera 1, the filtering process is performed on the entire area of the captured image, and the center of gravity is set on the entire area. The moving amount Ra of the center of gravity in the entire image is given in the same manner as in the above equation (1). Movement amount R of the center of gravity in the entire image
“a” is a sensory vector amount of a captured image, that is, an amount indicating a direction in which a motion is observed when viewing the image as a whole. In this manner, in the main camera 1, as shown in FIG. 6, the center of gravity P also moves to P-2 due to a motion other than the object of interest in the field of view 21a of the sub camera 2. The reason is that the key processing unit 4 extracts the specified chroma component as a key signal as described above, so that the field of view 2 of the sub camera 2 is within the field of view 21 of the main camera 1.
This is because, if there is a target object having the specified chroma component other than 1a and moves, the center of gravity moves from P to P-2 as shown in FIG. 6B. As a result, it can be considered that this is equivalent to a state where the photographer is silently looking at another moving object in the field of view 21 of the main camera 1 through the viewfinder.

The scalar amounts Rb 'and Ra' in the movement amount Rb obtained by the sub camera 2 and the movement amount Ra obtained by the main camera 1 are converted by the CPU 9 according to the imaging magnification ratio of the two cameras 1 and 2. And processed by the same scale. That is, Ra '= αRa (α: imaging magnification ratio) (2) Rb ′ = βRb (β = 1 / α) (3) Condition the movement of the object of interest in the movement of
The camera platform control amounts (camera movement amounts) of the cameras 1 and 2 according to the conditions are quantified. As described above, the CPU 9 calculates the parameters for controlling the camera platforms 5 and 18 of the two cameras 1 and 2 based on the two parameters corresponding to different magnifications of the two cameras 1 and 2, and Has a function as a calculation unit for estimating the optimal movement amount of each camera from the correlation between the movement of the entire target object detected by the above and the movement of the target object detected by the other camera 2. Thus, a stable operation of the automatic camera follow-up device can be realized.

On the monitor for displaying the object, only the image taken by the main camera 1 may be displayed. In the example shown in FIG. 1, the image A taken by the main camera 1 and the image taken by the sub camera 2 are displayed. The captured image B is synthesized by the synthesizing unit 31,
The image B is enlarged and displayed on the screen of one monitor 32 with the image A as a background. Where the image B is displayed in the image A is set by the display position setting unit 30 according to a command from the CPU 9. CPU 9
Is constantly recognizing where the image B is in the image A, and while observing the movement of the entire target object, an unobstructed area in the image A, for example, the image B in the image A is hidden by the enlarged image. A position that will not be displayed is specified, and an enlarged image of the image B is displayed at the specified position through the display position setting unit 30. Therefore, if the object of interest moves and the image B moves in the image A and overlaps with the enlarged image display portion of the image B, the enlarged image display portion of the image B is moved to a region in the image A where there is no problem. Let it.

By the way, if the main camera 1 sensitively follows a slight change in the position of the subject, the image becomes restless.
It is desirable that the dead area can be set so that the camera head does not move unless the pan head control parameter quantified from the movement amount extracted from the cameras is equal to or more than a certain value. Here, assuming that the pan amount of the quantified pan head control parameter is P, the tilt amount is C, the actual pan amount is P ′, and the actual tilt amount is C ′, P ′ = iP (4) C '= JC (5)

In the above equations (4) and (5), the insensitive area can be set by changing the parameters i and j by software. By setting the dead area, it is possible to realize a smooth movement of the imaging screen by the main camera 1. This processing is called smoothing processing, and each of the camera platform controllers 8 and 10 has this smoothing processing function. The dead area can be arbitrarily changed according to various imaging conditions and the like. Further, the above-described smoothing process and the cumulative addition of the actual pan amount P ′ and the actual tilt amount C ′ are performed at a designated time T (the designated time T is variable by software). By performing the operation of moving the table in parallel, a smoother movement can be obtained.

In the embodiment described above, even if the target object monitored by the sub camera 2 disappears outside the field of view 21 of the main camera 1 as shown in FIG. Is constantly tracked, and the main camera 1 is moved following the movement of the object of interest, which is captured by the sub camera 2 and masked, so that the field of view 21 of the main camera 1 is again shown in FIG.
The monitored object can be captured within.
Therefore, even if the viewing angle of the main camera 1 is narrowed to a specific target object, the target target object is not missed.

According to the embodiment of the present invention described above, two cameras having different magnifications are used, one camera 1 is provided with a field of view covering the monitoring range, and the other camera 2 is used. The object of interest is monitored, and the optimal movement amount of the camera is estimated from the correlation between the motion of the entire target object extracted by one camera 1 and the motion of the object of interest extracted by another camera 2. By moving each of the cameras 1 and 2 based on the estimated value, the optimum imaging range is imaged, so that a moving target object can be automatically followed, and the demand for labor saving of equipment and staff is required. In addition to focusing on the target object, which can be caught in a large and clear image.Furthermore, the following operation is not inferior to the following operation by a veteran photographer. It is possible.

In addition, in the field of view of the main camera 1, there are many cases where a moving object exists in addition to the object of interest. For example, it is similar to a tree whose branches sway due to the wind, a wavy water surface, and the like. In such a case, if the camera moves in response to a tree whose branches are swaying due to the wind, a wavy water surface, or the like, a malfunction occurs. However, if the object to be followed is specified by a color such as only a skin color portion as in the above-described embodiment of the present invention, malfunction can be prevented. Alternatively, a tree whose branches are swaying due to the wind as described above, a wavy water surface, and the like are moving with regularity, and those moving with regularity may be ignored.

The camera automatic tracking method and apparatus according to the present invention can be used even in a place where humans cannot enter.
An image can be taken while tracking a specific target object.
In particular, if the sub-camera is an infrared camera or a high-sensitivity camera, it can be used as an unmanned surveillance system or an abnormal alarm system. If a video tape recorder or a frame memory (in units of several minutes) is combined with the apparatus of the present invention, it is possible to record information at the time of disaster.

In the illustrated embodiment, the main camera and the sub camera are separated from each other.
One camera has a wide angle of view, the other lens has a narrower angle of view, the main lens is a wide angle of view,
If a lens having a narrow angle of view is used as a sub in the same manner as in the above-described embodiment, the same operation and effect as in the illustrated embodiment can be obtained. Also, a large number of cameras such as three or four cameras may be used. In this case, a plurality of sub cameras or a plurality of main cameras may be used. Also, the image screen of the sub camera may be displayed simultaneously with the image screen of the main camera. In this case, the designated area is appropriately processed, and the imaging screen of the sub camera is displayed in an area where there is no problem while observing the entire movement. By doing so, it is possible to simultaneously see the processing of the designated portion while watching the entire movement.

[0032]

According to the first aspect of the present invention, two cameras having different magnifications are used, one camera has a field of view that covers the monitoring range, and the other camera focuses on an object of interest. Monitoring and estimating the optimal movement amount of the camera from the correlation between the movement of the entire target object extracted by one camera and the movement of the target object extracted by the other camera, and based on this estimated value By moving each camera to capture the optimal imaging range, it is possible to automatically follow a moving target object, respond to the demand for labor saving of equipment and staff, and The object can be narrowed down to be large and clear, and the following operation can be made comparable to the following operation by a veteran photographer.

According to the second aspect of the present invention, two cameras having different magnifications are provided, one camera has a field of view covering the monitoring range, and the other camera is within the monitoring range. It monitors an object of interest, and detects a movement of a target object within the imaging range of the two cameras, and two parameters corresponding to different magnifications of the two cameras. The parameters for controlling the camera platform are calculated, and the optimum of each camera is determined from the correlation between the motion of the entire target object detected by one camera and the motion of the target object detected by another camera. The invention according to claim 1, comprising a calculation unit for estimating the movement amount, and a camera platform control unit for controlling the camera platform of each camera based on the estimated value to capture an optimal imaging range. Similarly, dynamic Can automatically follow a target object that has an object, responding to the demand for labor-saving equipment and staff, and narrowing down to the target object to capture it in a large and clear manner. The operation can be made comparable to the following operation by a veteran photographer.

According to the third aspect of the present invention, in the second aspect of the present invention, since the synthesizing means for synthesizing the images captured by the two cameras on one screen and displaying the same is provided, a predetermined monitoring range is provided. Can be superimposed and displayed in a magnified form on the screen of the entire object covering the object.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a camera automatic tracking method and apparatus according to the present invention.

FIG. 2 is a conceptual diagram showing a state of imaging according to the embodiment.

FIG. 3 is a conceptual diagram showing an operation of detecting a movement of a target object according to the embodiment.

FIG. 4 is a conceptual diagram showing a state of a filtering process according to the embodiment.

FIG. 5 is a conceptual diagram showing a state of movement of the center of gravity due to the movement of the object of interest in the main camera according to the embodiment.

FIG. 6 is a conceptual diagram showing a state of movement of the center of gravity due to the movement of a target object other than the target object in the main camera according to the embodiment.

FIG. 7 is a conceptual diagram showing an example of another movement of the object of interest according to the embodiment.

FIG. 8 is a diagram showing a relationship between a cameraman's eye movement and camera work.

FIG. 9 is a conceptual diagram showing a state of imaging by a conventional general camera.

[Explanation of symbols]

 Reference Signs List 1 main camera 2 sub camera 5 pan head 8 pan head control section 9 CPU as operation section 10 motion detection section 14 motion detection section 15 pan head control section 16 CPU as operation section 18 pan head 31 synthesis section

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] April 8, 1997

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0008

[Correction method] Change

[Correction contents]

[0008]

SUMMARY OF THE INVENTION The present invention has been made in view of the above points, and has been made so as to automatically follow a moving object and to provide the above-described equipment and staff. it is possible to meet the demand for labor saving, and an object thereof is to provide an automatic camera tracking method and apparatus subject as possible out to capture the large and sharp objects.

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0028

[Correction method] Change

[Correction contents]

According to the embodiment of the present invention described above, two cameras having different magnifications are used, one camera 1 is provided with a field of view covering the monitoring range, and the other camera 2 is used. The object of interest is monitored, and the optimal movement amount of the camera is estimated from the correlation between the motion of the entire target object extracted by one camera 1 and the motion of the object of interest extracted by another camera 2. By moving each of the cameras 1 and 2 based on the estimated value, the optimum imaging range is imaged, so that a moving target object can be automatically followed, and the demand for labor saving of equipment and staff is required. it is possible to meet, as possible out to capture this large and sharp narrow the object of interest.

[Procedure amendment 3]

[Document name to be amended] Statement

[Correction target item name] 0032

[Correction method] Change

[Correction contents]

[0032]

According to the first aspect of the present invention, two cameras having different magnifications are used, one camera has a field of view that covers the monitoring range, and the other camera focuses on an object of interest. Monitoring and estimating the optimal movement amount of the camera from the correlation between the movement of the entire target object extracted by one camera and the movement of the target object extracted by the other camera, and based on this estimated value By moving each camera to capture the optimal imaging range, it is possible to automatically follow a moving target object, respond to the demand for labor saving of equipment and staff, and narrow the object as possible out to capture this large and sharp.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0033

[Correction method] Change

[Correction contents]

According to the second aspect of the present invention, two cameras having different magnifications are provided, one camera has a field of view covering the monitoring range, and the other camera is within the monitoring range. It monitors an object of interest, and detects a movement of a target object within the imaging range of the two cameras, and two parameters corresponding to different magnifications of the two cameras. The parameters for controlling the camera platform are calculated, and the optimum of each camera is determined from the correlation between the motion of the entire target object detected by one camera and the motion of the target object detected by another camera. The invention according to claim 1, comprising a calculation unit for estimating the movement amount, and a camera platform control unit for controlling the camera platform of each camera based on the estimated value to capture an optimal imaging range. Similarly, dynamic Can automatically follow the target object with, it is possible to meet the demand for labor saving equipment and staff as possible out to capture this large and sharp narrow the object of interest.

Claims (3)

[Claims] 1. A movement of a target object within an imaging range is detected from video signals of two cameras having different magnifications, and a parameter for controlling a camera platform of the two cameras is determined by two parameters corresponding to the different magnifications. Calculate, give one camera a field of view that covers the monitoring range, monitor the object of interest with another camera, and calculate the movement of the entire target object extracted by one camera and the other one A camera automatic tracking method for estimating an optimal movement amount of a camera from a correlation with a movement of an object of interest extracted by the camera, and moving each camera based on the estimated value to capture an optimal imaging range. 2. A camera having two cameras having different magnifications, one camera having a field of view covering the monitoring range, and the other camera monitoring an object of interest in the monitoring range. A motion detection unit that detects a motion of a target object within an imaging range of the two cameras; and a parameter that controls a camera platform of the two cameras by two parameters corresponding to different magnifications of the two cameras. A calculation unit for calculating the optimal movement amount of each camera from the correlation between the movement of the entire target object detected by one camera and the movement of the target object detected by the other camera An automatic camera tracking device comprising: a camera platform control unit that controls the camera platform of each camera based on the estimated value of the arithmetic unit to capture an optimal imaging range. 3. A synthesizing means for synthesizing an image picked up by two cameras onto one screen and displaying the synthesized image.
The automatic camera follow-up device described in the above.