JPS59208983A - Automatic tracing camera - Google Patents

Abstract

PURPOSE:To realize a simplified considerably in constitution, small sized and inexpensive automatic tracing camera by detecting the difference between pictures at prescribed time intervals, calculating the center of gravity between the said pictures, detecting the moving amount and the direction of the center of gravity to an image picked-up screen of an object picture to control an image pickup device. CONSTITUTION:A picture signal from a camera circuit 3 is applied from an input terminal 6 to a separating circuit 7, where the signal is separated into a luminance signal and a chrominance signal. The picture signal from the separating circuit 7 is applied to an analog/digital converting circuit 8, where the picture signal is digitized by setting M1XM2 sampling points per one frame sampling the picture signal. A subtraction circuit 10 processes subtraction for the stored picture signal for one frame's share and the picture signal of other frame having a prescribed time interval, generates a picture difference signal and stores the result to a frame memory 11. An operating device 12 uses a supplied picture difference signal, calculates a center of gravity of the picture represented by the picture difference signal by using a reference point of the picked-up screen as the origin, generates a control signal and applies the signal to a control section 5.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to an automatic tracking camera capable of automatically tracking and imaging a target object.

[Background of the invention]

Conventionally, in order to constantly observe a moving object (hereinafter referred to as a subject), the orientation of the imaging device as well as the movement of the subject,
In other words, it automatically changes the camera direction, tracks the subject,
It is known that it is an automatic tracking camera that uses shooting rules. In such an automatic tracking camera, it is first necessary to recognize the subject, and for this purpose, conventionally, a processing technique using the art industry method (to perform turn recognition of an image obtained by capturing an image of the subject (hereinafter referred to as a subject image)) is used. In other words, the shape, color, etc. of the subject were extracted to show the characteristics of the subject, and the pattern was compared with the preset standard turns to determine the subject. I try to recognize it.

However, in order to recognize these kinds of turns, a large Amashi calculation machine is used, and a huge number of programs must be used to perform the processing, making conventional automatic tracking cameras unavoidable. It became very popular in Japan, and was also very biographical.

-In recent years, video cameras have become increasingly popular, along with video table recorders, for industrial use.

It has become widely used for household use.

Moreover, such video cameras are not only used to capture images of desired subjects by direct operation by the user, but also used as sensors for monitoring devices. When used as
There is some meaning in fixing a video camera and monitoring within a certain field of view, but if the subject to be monitored moves and the video camera's angle of view is restricted, It is necessary for an unmanned monitoring device or a video camera to have an automatic tracking function to track and image a moving subject.

However, as mentioned earlier. As conventional automatic tracking cameras are large-scale and extremely expensive, it is practically impossible to provide an automatic tracking function to a video camera for industrial or home use.

[Purpose of the invention]

The object of the present invention is to eliminate the drawbacks of the above-mentioned prior art.

To provide a compact and inexpensive automatic tracking camera with a greatly simplified configuration.

[Structure of the invention]

In order to achieve this object, the present invention detects the difference between images at regular time intervals and calculates the center of gravity of the difference between the images,
The amount of movement of the center of gravity, the amount of movement wJ relative to the imaging screen of the subject image from the direction of movement, and the direction of movement are detected, the imaging device is controlled, and the subject is set in a reference area of the imaging screen. The feature is that it is possible to do this.

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing one embodiment of an automatic tracking camera according to the present invention, where 1 indicates a subject.

2 is a photographing device, 6 is a camera circuit, and 4 is a signal processing MS-
5 is a control section.

In the figure, an imaging device 2 images a moving subject 1 and generates an image signal. This image signal is supplied to the camera circuit 3, subjected to predetermined processing such as amplification, and then supplied to the signal processing section 4.

The signal processing circuit 4 processes the supplied image signal to create an image of the subject 1 on the imaging screen (viewing screen projected onto the imaging surface) of the imaging device 2, that is, a movement of the subject image.
1 generates a control key signal representing the quantity and direction of movement. This control signal is supplied to the control unit 5, and the control unit 5 controls the imaging device 2 to change its camera direction so that the subject image is within a predetermined head position (for example, the center) of the imaging screen. Track as shown in .

FIG. 2 is a block diagram showing a specific example of the signal processing section of the Ren1st, in which 6 is an input terminal, 7 is a separation circuit, 8 is an analog-to-digital conversion circuit, 9 is a frame memory, and 10 is a subtraction circuit. , 11 is a frame memory, 12 is a performance N device, 1
3 is an output terminal.

In the figure, a camera 1 circuit 3 (helmet 1) image signal is supplied from an input terminal 6 to a separation circuit 7, where it is separated into a luminance signal and a color signal. The separation circuit 7 is provided because the image signal obtained from the imaging device 2 (FIG. 1) is generally a color video signal, but the luminance signal and the color m number are subjected to the same processing, and these Since it is not particularly important for the present invention that the signals be separated and processed, they will be described below as image signals.

Picture fiJ from separation circuit 7! The signal is supplied to an analog-to-digital conversion circuit 8. The analog-to-digital conversion circuit 8 samples and digitizes the image signal by setting M, XM sample points per frame.

That is, as shown in FIG. 3(,4), the frame I°
A number of sample points are set for each of M2 scanning lines in the scanning axis 14 of
Ru. This can be done by sampling and digitizing the image signal for every 01/M□ tansashi line in one frame, and every 1/M0 period of the tansashi period.

Now, let the position of the sample point 15 be ('', y) (where x, y are positive integers, 1≦X≦M1.1≦y≦”2), and the position of the sample point in the upper left corner be ( 1,1).

Let the position of the sample point in the lower right corner be (M+, #2).

The digitized squadron signal can be expressed as YCx,y).

The picture number from the analog-to-digital conversion circuit 8 is supplied to a subtraction circuit 10 and a frame memory 9. The frame memory 9 stores an image signal for one frame, and the subtraction circuit 10 calculates a fixed time interval from the image signal f for one frame stored in the frame memory 9 and this 7L/-m. A subtraction process is performed on images A and H of other frames to generate a single image difference signal and store it in the frame memory 11.

By the way, since the subject 1 (Fig. 1) is moving, the position of the subject image with respect to the imaging screen changes from time to time, and for this reason, the frame image 1M signal stored in the frame memory 9 and the What is an image signal of frames at fixed time intervals?

The °C waveform changes depending on the movement of the subject 1, and as a result, the image difference signal obtained from the subtraction circuit 10 is a signal corresponding to the difference between the images FW of the two frames as the subject 1 moves.

That is, in FIG. 4, an 11th frame is obtained by the imaging scan of the imaging device 2, but now, to the frame,
..., assuming N frames of frame FN-1+frame FM-I+frame FN, and.

A is a subject image, B is an image of a stationary object, and frame FH
When the image of Frame-kun is subtracted from the image of , the image of stationary object B is canceled out, and an image ΔF consisting only of the difference image A' between the subject image A in frame FN and the subject image A in frame F1 is obtained. It will be done. Image difference No. 43 stored in the frame memory 11 is 4 corresponding to the image ΔF.
It is M number. The image signal of frame p′1 is Y+ (”
, y), the image signal of frame F〃 as Yn(''-y)
Then, Yn (x, y) - Y, ('', y) - ΔY (x
, y) is the self-IW difference signal. In addition, −” + 3’ is
This is the direct position defined in the explanation of Figure 5 (/J).

In this case, the image difference number 4g from the subtraction circuit 10 is stored in the frame memory 11 as a signal representing 1, E to even 4.

Image difference No. 48 stored in the frame memory 11 is as follows.

The data is read out and supplied to a performance device 12 consisting of a microcomputer or the like.

The Shinson device 12 first uses the supplied image difference signal,
Imaging image iM+ service! ■A point (for example, Nakagami, point) is a hot point, ゛'C, Sogen running number 1 is wearing 1 stroke 1 y, and the point is a1 cho-phrase.

Therefore, 1z, l = Hm 1M No. ΔYCx, y)k
Qr bar threshold 1 is greater than 1. It was made into 21ut as a small Kerepo O, and it was made into 2 pieces.
y), the on-vehicle position (-ta, '/a) is given by the following formula, f.

Calculated by

By the way, the amount of movement and movement of the subject image on the screen, as well as the amount of movement of the hairs and points in the image expressed by the surface difference signal as the subject moves.

The relationship with the direction of movement is as follows.

1) The amount of movement of the seven base points is 4, which is the amount of movement of one subject image.

2) The moving direction of the center of gravity matches the moving direction of the subject image.

From this, the amount and direction of movement of the subject image can be detected based on the amount and direction of movement of the seven points.

Now, let us assume that the position of the center point of the imaging screen is ("o = yo), and add the coordinates with this center point ("o = 3'o) as the origin, so that the subject image will be located at the center of the imaging screen that includes this origin. The image signal of the frame stone (Fig. 4) when set to the chain acid (hereinafter referred to as the reference area) is stored in the frame memory 9 (second
Assuming that 1 frame is stored in Figure), the position of the center of gravity of the difference between the image of frame FsCm4 (Figure) and the image of frame F1 after a certain period of time is (Jta, 3'G). The center of gravity has moved horizontally by <3:G-xo) and vertically by (3'G3',). However, as for the subject image.

This means that it has moved 2X (xG-0) in the horizontal direction and 2X (3'G'10) in the vertical direction. It should be noted that the horizontal direction is the direction of the scanning wave, and the vertical direction is the direction perpendicular to the short bet win.

Based on the take obtained as described above, the arithmetic unit 12 generates a control signal, and this control signal is supplied to the control section 5 (FIG. 1) through the output terminal 13.

For this purpose, in FIG. 1, the imaging device 2 is controlled to change the camera direction so that the subject image is set in the reference area of the image plane.

The processing unit 4 for each issue has stopped its processing operation.

When the camera direction is directed toward the subject and the subject image is set in the reference area of the imaging screen, the No. 1g processing unit 4 resumes the above processing operation.

In this way, measurement of the movement of the subject and control of the imaging device 1 are alternately repeated to perform automatic tracking.

According to this embodiment, automatic tracking can be performed without using pattern recognition technology, and it can be constructed using a circuit for processing No. 4G and a relatively small arithmetic device such as a microcomputer. Further, since the processing target is the change in the image due to the movement of the subject, the amount of information to be processed is extremely small, and calculation processing can be performed quickly.

The frame memory used can also have a relatively small capacity, and the response speed is the same as above.

It is small, lightweight, and inexpensive.

Note that the origin of the photographic image area is not necessarily the point I and 1 within it.
- It is possible to set it at any point, and therefore, the subject image set by automatic tracking can be set at a desired position.

FIG. 5 is an explanatory diagram of another embodiment of the automatic tracking camera according to the present invention, where 16 is an imaging screen and 17° and 18 are angles of view.

In this first embodiment, the subject is moved using one angle of view.
The basic configuration is the same as that in FIGS. 1 and 2.

In FIG. 5, the imaging screen 16 of the imaging device 2 (FIG. 1)
Assume that there are two angles of view 17 and 18 within.

The center points of the respective angles of view 17 and 18 are set as the captured image [1111(
The small angle of view 17 is set to match the center point of S, and the angle of view 18 is set to a desired size. Therefore, the subject image always has an angle of view of 17
Automatic tracking is performed as shown in the image below. In addition, the angle of view 18 is used to determine whether the subject has moved or not, and the area between the angle of view 17 and the angle of view 18 (hereinafter referred to as
If image difference No. 4g occurs in the movement determination area), it is determined that the subject has moved.

Figure 6 is a flowchart showing the operation of this embodiment, and is based on Figures 2, 5, f, and 6.

Let me explain the operation of this embodiment.

As explained in the previous embodiment, first, the frame memory 9
Process 19) to store the image signal of one frame in memory, and store the image difference signal between each image of one frame after a certain period of time in the frame memory 11 using the subtraction circuit 10 (process 2υ
). Next, the processing unit 1 case 12 reads out the addresses corresponding to the movement determination area between the viewing angles 17 and 18 of the frame memory 11, and when the image difference signal is written into these addresses, the j-line is determined. In other words, when there is a take of 11', a flag is set (process 21).

Next, it is determined whether or not the flag is set (determination 22), and if the 7 rung is set or rL, it is assumed that there is no movement M of the subject and the process returns to process 20. In this case, memory? stores the previously stored image signal of one frame, and furthermore obtains an image difference No. 18 with the image signal of one frame after the time jbJ.

The flag is set and 'l'lJ foot 22 says 'YE'.
When it is determined that there is a condition to terminate the automatic tracking operation (Part 3), if this naming is 71 Itogi, it is determined that there is a problem that will cause the automatic tracking operation to end. Similar to the previous embodiment, processing is performed to calculate the point b of the nail of the colleague, to detect the amount and direction of movement of the subject image, and to set the camera direction by controlling the imaging device 2 (FIG. 1). 24)
When the setting of the camera direction is completed, the process returns to process 19 and performs the same operation again.1-0 These series of operations are repeated, and if there is an instruction to end the automatic tracking operation, a 1-decision 26 returns "YES". It is determined that the automatic tracking operation stops.

In this embodiment, since the image difference number 4g within one field of view 18 is processed, the amount of information processed is .

This is much smaller than when processing image difference signals over the entire imaging screen, and the automatic tracking operation is performed more quickly. You can use it. Since the processing range of the shuffled one-image difference No. 4M is limited to a small range within the angle of view 18, the influence of noise on the arithmetic processing for detecting the subject image is significantly reduced.

The accuracy of automatic tracking is greatly improved.

Furthermore, in this embodiment, a movement determination area between the viewing angles 17 and 18 is provided to detect movement of the subject image, and after the movement of the subject is detected, arithmetic processing for setting the camera direction is performed. .

As soon as it is determined that there is no movement of the subject, the next frame is detected as the image difference between the frame stored in the frame memory 9 (Fig. i2) and the number of strokes of the subject. When it moves (i.e., when the subject image hangs out from the angle of view 17), it is detected in one section, and the response characteristic of self-tracking becomes the same as above.

7. The angles of view 17 and 18 can be set at any position on the imaging screen 16, as long as the angles of view 17 and 18 coincide with each other. Therefore, the range in which the subject is captured can be set at will. You can do it.

FIG. 7 is a block diagram showing the main parts of still another embodiment of the self-TfJ@tail camera according to the present invention.

Reference numeral 25 denotes an automatic focus adjustment device, and parts corresponding to those in FIGS. 1 and 2 are given the same reference numerals and some parts are omitted.

This lol? When the moving direction of the moving object 1 (Fig. 1) is in the direction approaching the imaging device 2 (Fig. 1), l+ changes the zoom magnification to take wide-angle shots or increase the angle of view. This allows the subject image to be limited to a predetermined size, and similarly to the embodiment described above with reference to FIG.
Assume 1B.

FIG. 8 is a flowchart showing the main part of the operation of this embodiment. The operation of this embodiment is also represented by -y in the flowchart of FIG. 6, except that process 24 performs the processing shown in the flowchart of FIG.

In FIGS. 5, 7, and 8, as explained in the previous embodiment, when it is detected that the subject image is moving by the movement determination area between the viewing angles 17 and 18, the arithmetic unit 12, a detection process is performed to determine in what part of the movement determination area the image difference signal exists (process 2
6). That is, the movement WtJJ judgment area is divided into several partial movement judgment areas, and in process 26, the presence or absence of an image difference signal is detected for each partial movement judgment area.

Next, calculation processing for calculating the center of gravity of the next image 1 #difference A] at the 5th angle of view 17 and detecting the G shift amount and moving direction of the subject image is performed as explained in the previous embodiment. , a control signal based on the calculation result is supplied to the control section 5.

Next, based on the detection result in the process 26, the processing unit 12 determines whether or not an image difference signal exists in the partial movement determination area in a plurality of directions opposite to each other with respect to the angle of view 11 (determination). 28). This determination 28 is to detect whether the moving direction of the subject 1 (Fig. 1) is the direction of the photographer's device 2 (Fig. 1). 1 is not close to the imaging device 2,
Returning to process 19 in FIG.

On the other hand, if the decision 28 is f-YESJ,
A calculation process is performed to calculate the zoom magnification (process 29). This calculation process is to find the distance from the center point of the angle of view 17 after changing the camera direction in process 27 to the farthest sample point (FIG. 3) on the object plane.

Now, the movement vector of one object plane is (ΔX.

Δy) 1 The coordinates of the upper middle point of the angle of view 17 are (X・, 3',
).

The coordinates of each sample point of the subject image before processing 27 are (x.

y), the subject image becomes (-ΔX
, -Δy), the distance d from the center point of the angle of view 17 of each sample point of the subject image after processing 27 is.

It is. Once the distance d of each sample point is determined from this equation, the distance d□ which is the maximum value among these is selected. The sample point for this direction 61 is the sample point farthest from the center point of the angle of view 17 of the subject image after the camera direction is set.

It should be noted that processing 29 is performed only after an image difference signal is detected in the movement between the viewing angles 17 and 18, so the image difference signal in the movement A-size foot trap area is the subject of processing 27ii7. The image of the image can be regarded as the image number IH, and therefore -x+, V are the coordinates 111 of the sample point in the moving leg area. ΔX, Δy are movement pairs of the subject plane image detected in the process 27.

Obtained from the direction of movement.

In this way, the zoom magnification is calculated so that the sample points corresponding to d and na are within the angle of view 17 using the calculated ctma and the zoom magnification from the 7-point eye movement focus adjustment device 25, As a result, a control signal is supplied to the automatic focus adjustment device 25, and the zoom lens is controlled so that the entire subject image is included in the angle of view 17.When the control of the zoom lens is completed, the sixth Return to 8!19 in the figure.

In addition, in this embodiment, the frame written in the frame memory 9 is updated after the calculation process including the process for increasing the movement of the subject plane image in the calculation unit 12 (J -rj.
If it is determined that this is the case, the process returns to process 19). This is to prevent errors in zoom magnification from occurring.

In this embodiment, the distance between the subject and the imaging device is determined.
(- It is possible to capture the image within the angle of view of 17, and it is possible to accurately check the movement of the subject image, and the amount and direction of movement of the subject image are also accurate. Automatic tracking is performed accurately regardless of the direction of movement of the subject.

Note that in one or more embodiments, frame memories are used as the nodes 1Ji1, 12, but field memories may also be used.

〔Effect of the invention〕

As explained above, according to the invention, there is no need for complicated processing, there is no error in the circuit's X configuration, and there is no need for continuous speed. It is possible to provide an eye movement tracking camera that can perform 1I71 tracking, is small, inexpensive, and has advanced functions not found in the prior art.

[Brief explanation of the drawing]

Figure 1 shows self-invention due to non-invention! J-Block diagram showing an example of an embodiment of the tracking camera, 2nd V, 1 shows a specific example of the processing section of the 1st diagram, and FIG. (Continued diagram showing the sample points, 3rd one is i CB)
Figure 6(,4) is a rectangular and medium-sized self-portrait (g). 5 is an explanatory diagram of another eye tracking camera according to the invention, and FIG.
FIG. 7 is a Gronk diagram showing the main parts of yet another embodiment of the automatic tracking camera according to the invention; FIG. 8 is a main part of the operation of the embodiment shown in FIG. 7. This is a simple flowchart. 2... Imaging device 6... Camera circuit 4...
Signal processing unit 5...Control unit 9...Frame memory 10...Subtraction l! 2 circuits 11・-Frame memory
12...Performance device 16...Repentance ball picture ii 1
7.18... Angle of view 25... Eye movement focus 7 Adjustment device direct 11th ? Field No. 5 Figure 5 (8) 5 Figure 4 Figure 5 Field No. 1 "1)nJ λ 7" 3, Continued from page 1 ■Applicant Hitachi Video Engineering Co., Ltd. 292 Yoshida-cho, Totsuka-ku, Yokohama City =516

Claims (1)

[Claims] (1) Measuring the amount and direction of movement of the subject to be tracked;
A first means for obtaining an image difference signal representing a difference between images at a fixed time interval by supplying an image signal obtained from the imaging device to an automatic tracking camera controlled to be able to image the subject according to the measurement result; a second means for calculating the position of the center of gravity of the difference between the images IW on the imaging screen based on the image difference signal and detecting the amount and direction of movement of the subject image on the imaging screen; and a detection signal from the second means and a fifth means for controlling the imaging device based on the above, and the eye movement tracking camera is configured such that the subject image can be set within a reference area of a front camera screen. . . (2) Claim No. 11 (wherein the reference area is a first angle of view, and the second means is outside the first angle of view, and the first angle of view is An automatic tracking camera characterized in that the automatic tracking camera detects that there is a difference between the images within a second angle of view including the second angle of view, and generates the detection signal to the third means. (In Clause 11 or Clause (2), the automatic tracking camera is characterized in that the second means generates the detection (g) to set the camera direction of the 1m photographing device. (4) In claim 21, the second means sets the camera direction for the imaging @ position, and
An automatic tracking camera whose % sign is to generate the detection signal to reduce the size of the object image to fit within the first angle of view.