JPH0498991A - Automatic visual field adjustment device - Google Patents

Abstract

PURPOSE:To display a picture of an object while catching it in the center of a screen at all times by generating a video signal resulting from tracing automatically a moving part of a television signal. CONSTITUTION:A television signal and another television signal sent succeedingly are compared by a color element comparator circuit 14, from which difference information is generated and 1st position information of a part having a change on a television screen is generated based on a time difference between the time when the difference information is detected and the time of a synchronizing signal of a relevant television signal. Then 2nd position information 16 to segment a new magnified television picture from an original television signal is generated based on the 1st position information and specific horizontal scanning line information is extracted by the 2nd position information. As soon as the horizontal scanning line information is generated, scanning line information for a prescribed period is extracted from one horizontal scanning line information and the information is used to expand the time with respect to one horizontal scanning line information of a prescribed form of the television signal.

Description

Detailed Description of the Invention (Industrial Field of Application) The present invention is a method for automatically displaying images of moving subjects on the screen in video devices that mainly use people as subjects, such as video door intercoms, video telephone devices, and surveillance devices. This relates to a device that automatically adjusts the display field of view by processing video signals so that the field of view is centered.

(Prior art) In conventional video door intercoms, videophone devices, surveillance devices, and other video devices that photograph people as subjects, the imaging range is fixed, so in order to ensure that the human image of the subject is in the center of the screen, it is necessary to In each case, the direction of the imaging camera was manually adjusted, the person being photographed corrected his or her position while looking at a monitor screen, or the direction of the camera was adjusted using a remote control unit.

In addition, when such adjustment cannot be carried out, the imaging viewing angle of the camera is set to a wide angle so that even if the subject moves a little, it still falls within the imaging range.

Additionally, a method has been proposed that captures the characteristics of a subject as an image and determines the shape and position of the head.

(Problems to be Solved by the Invention) It is quite cumbersome to operate so that the human image of the subject is placed in the center of the screen, and even when using a wide-angle lens, the image of the subject becomes small. The disadvantage is that it is difficult to see. In addition, in order to capture the characteristics of the subject image and determine the shape and position of the head, a large amount of information processing and complex algorithms are required to avoid misjudging the position, making the device more compact and economical. It was difficult.

(Object of the Invention) The present invention eliminates the drawbacks of the above-mentioned prior art, and automatically captures the subject even if the subject moves in a video device that takes a person as a subject, such as a video door intercom, a video telephone device, or a surveillance device. To provide a small and economical automatic field of view adjustment device that can track an object and display it at the center of the screen.

(Means for Solving the Problems) In order to solve the above problems and achieve the objects, the present invention provides means for storing television signals, and a means for storing television signals stored in the television signal storage means and transmitting the television signals next. difference information detecting means for generating difference information by comparing the incoming television signal; and detecting changes on the television screen due to the time difference between the time when the difference information is detected by the difference information detecting means and the synchronization signal of the television signal. means for generating first positional information of a portion where the original television signal was located; means for generating positional information, and information from the second positional information generating means, extracts specific horizontal scanning line information from among the plurality of horizontal scanning line information of the television signal, and uses this information to generate a specified At the same time, the horizontal scanning line information of the television signal is generated, and at the same time, scanning line information for a predetermined period is extracted from one horizontal scanning line information of the television signal, and this is converted into one horizontal scanning line information of the television signal in a specified format. and a means for time expansion, and is characterized in that it generates a video signal that automatically tracks a moving portion of the television signal.

(Function) The present invention compares a television signal with the next transmitted television signal to generate difference information, and uses the time difference between the time when the difference information is detected and the synchronization signal of the television signal to control the television signal. First position information of a portion of the screen that has changed is generated.

This first position information is used to generate second position information for cutting out a new enlarged television image from the original television signal, and this second position information is used to generate information on a plurality of horizontal scanning lines of the television signal. Extract specific horizontal scanning line information from inside. Then, at the same time as generating horizontal scanning line information of a specified television signal, scanning line information for a specified period is extracted from one horizontal scanning line information of the television signal and converted into a specified format. The time is expanded to one horizontal scanning line information of the television signal.

In this way, a video signal that automatically tracks the moving parts of the television is generated. The video signal is processed so that a moving subject such as a person is automatically placed in the center of the screen, and the display field of view is automatically adjusted.

(Embodiment) FIG. 1 shows a configuration diagram of an embodiment of the present invention, in which 1 is a television camera unit composed of an image sensor;
2 is a process amplifier circuit that generates a composite video signal including a synchronization signal, 3 is a television monitor that divides the video signal, 4 is an A/D converter, 5 is a synchronization signal separation circuit, 6 is a timing generation circuit, and 7 is a Clock signal generation circuit, 8 horizontal address counter, 9 vertical address counter, 1o address generation circuit, 11 image signal memory, 12 D/A
Converter, 13 is a process amplifier circuit that generates a composite video signal, 14 is a pixel comparison circuit, 15 is a pixel comparison circuit 14
16 is a display position determining circuit;
Reference numeral 18 indicates a screen cutting circuit composed of a video memory such as a RAM, and 18 an enlargement control circuit.

In addition, G is a gate for screen switching in frame units,
G,, G, is the screen extraction gate in frame units,
G3 is a gate for the image cutting circuit 17.

Next, the operation will be described with reference to the explanatory diagram of detecting the moving position of the subject in Fig. 2, the diagram showing the timervJ relationship between the synchronization signal of the image signal and the detecting signal of the moving position of the subject in Fig. 3, and the original diagram of Fig. 4. This will be explained with reference to a diagram showing the relationship between the signal and the image signal obtained by automatic field of view enlargement processing.

A process amplifier circuit 2 performs necessary waveform processing such as a synchronization signal on the video signal photographed by the television camera unit 1, and transmits it to the video signal processing section of the present invention.

The A/D converter 4 samples the video signal at regular intervals, converts it into a digital signal, and stores it in the image signal memory 11 via the gate G. Further, the synchronization signal separation circuit 5 separates only the synchronization signal from the video signal and supplies it to the horizontal address counter 8 , the vertical address counter 9 , and the timing generation circuit 6 .

The timing generation circuit 6 generates conversion timing for the A/D converter 4 and timing signals for updating the horizontal address counter 8 and vertical address counter 9 from the master clock and synchronization signal of the clock signal generation circuit 7. The horizontal address counter 8 updates its count with the horizontal synchronization signal and is reset every vertical synchronization signal, that is, every field.

The vertical address counter 9 updates its count at the conversion timing and is reset every horizontal synchronization signal. The values of these horizontal and vertical address counters 8.9 become logical addresses indicating positions on the screen.

In the address generation circuit 1O, a physical address signal on the memory is constructed from the values of the horizontal and vertical address counters 8 and 9, respectively. There are various methods of configuring the image signal memory J1 as a means for storing television signals, but for example, as shown in FIG. The area and read area are used by switching alternately.

In other words, since writing and reading cannot be performed on a memory medium such as a RAM at the same time, while the digital video signal output from the A/D converter 4 is being written to one area via the gate G, the digital video signal is being written from the other area. A digital video signal stored one field before is read out via gate G8 by a method described later. By repeating this operation alternately, it is possible to simultaneously generate a video signal delayed by one frame for generating a difference between a video signal from the television camera unit 1 and a moving image.

That is, the pixel comparison circuit 14 serving as the difference information detection means receives the video signal S1 shown in FIG. 2 from the television camera unit 1 and the video signal S delayed by one frame.

Which difference is taken (subtraction processing), and if it is not zero [20-1.2O-2 in Figure 2 (b)], it is determined that there is an image of a moving subject in the incoming video signal. can.

The output 5out of this pixel comparison circuit I4 is input to a motion position address generation circuit 15, which is a first position information generation means. That is, as shown in FIG. 3(a), the time difference (
t v, t h) or the moment when the difference occurs [20 in Figure 3 (a)], the horizontal and vertical address information can be regarded as the movement position information of the subject on the screen [M in Figure 3 (b)] I can do it. This moving position address information of the subject can be obtained for all moving positions of the subject on the screen, so among these moving position information, from the horizontal and vertical synchronization signals,
The address information obtained first is regarded as the address information of the top of the head of the subject.

Next, in order to obtain screen cropping position information from the horizontal and vertical address information of the apex of the head of the subject, the address of the apex of the head of the subject is sent to the display position determination circuit 16, which is the first position information generating means. The information is transferred, and a predetermined amount of address is subtracted (or added) to generate screen cutout position address information so that the moving subject is in the center of the screen. This screen cutout position address information is transferred to the enlargement control circuit 18 and stored there.

The screen cutout circuit 17 is composed of a video memory such as a RAM, and cuts out video information to be enlarged and displayed from the image information transferred from the image signal memory 1. The cropping range at this time is determined by the screen cropping position address information generated by the display position determining circuit 16.

FIG. 4 shows an example of means for cutting out video information to be enlarged and displayed from the cut-off position address information of the screen.

In FIG. 4, (a) shows one frame of the video signal transferred from the image signal memory, hl-h
8 indicates the horizontal scanning line number. An actual television signal consists of 525 scanning lines, but in order to illustrate the principle of operation, fewer scanning lines are used here.

The image information to be cut out and enlarged from the original image information is 4 in kl~ included in the horizontal scanning line numbers h4 to h7, and based on this information, the television signal after being cut out and enlarged is shown in FIG. 4(b). generate.

There are various means to achieve this function, but for example, as shown in FIG. Of the received video signal, hl, h2, and h3 are discarded, and information on h4 to h7 is stored in the first frame a. At this time, the frame memory of the screen cutting circuit 17 only has the storage capacity after cutting out, and the information of h8 is automatically discarded.

At the same time, since the gate of the frame memory is closed except during the period when 1 is being transferred to the information to be enlarged and displayed among the information in h4, the information other than 1 is discarded to the information to be enlarged and displayed.

By performing such information processing, the enlarged image information kl is stored in the first frame memory a of the screen cutting circuit 17.
, to 2. 3. 4 is stored in .

The next frame information transferred from the image signal memory 11 undergoes similar information processing and is stored in the second frame b of the screen cutout circuit 17.

The reason for preparing two frame memories here is that image information is continuously transmitted from the image signal memory 11, so while image information is being read from the first frame, image information of the next frame is being read out. Since the second frame memory cannot be received, a second frame memory is prepared and the image information of the next frame is stored therein. If it is a memory that can be read and written at the same time, it can be realized with one frame memory.

While the image information of the next frame is being stored in the second frame memory b, the enlarged image information is read from the first frame memory a. By repeating this operation alternately, the information continuously sent from the image signal memory 11 is enlarged. At this time, the image signal memory 11
Which horizontal scanning line information is selected from among the image information transmitted from the frame memories a and b of the screen cutting circuit 17
It is determined by the vertical address of the screen cutout position address information stored in the enlargement control circuit 18.

That is, the vertical address information of the screen cropping position stored in the enlargement control circuit 18 and the screen cropping circuit I7
The image information transmitted from the image signal memory 11 is discarded until the values of the vertical address counter 9 match, and the image information transmitted from the image signal memory 11 is discarded from the point when the vertical address information and the value of the vertical address counter 9 match. The incoming image information is taken into the screen cutting circuit 17.

In addition, until the horizontal address information of the screen cropping position stored in the enlargement control circuit 18 and the value of the horizontal address counter 9 of the screen cropping circuit 17 match,
The image information transmitted from the image signal memory 11 is discarded, and from the time when the horizontal address information and the value of the horizontal address counter 8 match, the image information transmitted from the image signal memory 11 is taken into the screen cutting circuit 17. . The amount of information stored in the screen cutting circuit 17 is determined by the storage capacity of the screen cutting circuit 17. In the example of FIG. 4, it becomes Σkn.

Next, an example of means for generating the normal television video signal shown in FIG. 4(b) from the information in the frame memory of the screen cutting circuit 17 will be described. Screen extraction circuit 1
kl in the frame memory of 7.

k2. 3. 4 information is stored in . In order to generate the video signal corresponding to FIG. 4(b) from this information, it is necessary to specify the address information and the period of the reading clock using a different method from that used in the case of the image signal memory 11. . That is, to generate a horizontal scanning line corresponding to the period of Hl, first specify the vertical address of the line memory where the kl information is stored, and at the same time, specify the vertical address of the line memory where the kl information is stored, and at the same time, specify the vertical address of the line memory where the kl information is stored. A lock signal (lower speed than the image information memory write/read clock signal) is supplied. As a result, information 1 is read out as image information H1 corresponding to the period of one horizontal scanning line.

Next, to generate a horizontal scanning line corresponding to the period H2,
The address of the line memory where the same <kl information is stored is specified again, and at the same time, the same clock signal necessary to read the 1 information again during the Hl period is supplied. As a result, information 1 is read out as image information for period H2.

In the periods H3 and H4, information 2 is read out in the same way.Hereafter, by repeating the same operation, k
Information of 1 to 4 can be converted into a television signal of 1 frame of H1 to H8.

This H1 to H8, l-frame television signal is
The A converter 12 converts the signal into an analog signal, the process amplifier circuit 13 performs necessary waveform processing such as a synchronization signal, and when the signal is displayed on the television monitor 3, a moving subject can be automatically followed and enlarged. That is, as the subject moves moment by moment, the movement position address generation circuit 1
5 changes every moment, the display position address information of the display position determining circuit 16 changes correspondingly, and the screen cut out by the screen cutting circuit 17 changes every moment as it follows the subject.

The above explanation describes a circuit that uses two frames each for the frame memory to obtain differential information to detect the movement of the subject and the frame memory to store the enlarged image after the movement position is detected, but the memory capacity can be reduced. There are various circuit configurations.

Conceivable. For example, in the image signal memory 11,
Instead of taking the difference for every frame, take one difference for every two frames.
If you want to take the difference from the image two frames before,
The capacity of the image signal memory 11 is only one frame, and the memory capacity can be halved.

Further, the memory control function described above can also be realized by software using a storage program method.

(Effects of the Invention) As explained above, the present invention provides an image frame memory (
Since it can be configured from RAM), A/D, D/A converter, counter, etc., it can be easily integrated into LSI, and a compact and economical automatic visual field adjustment device can be realized.

In addition, in video door intercoms, videophones, surveillance equipment, and other video devices that take pictures of people as subjects, it is possible to automatically track the subject using only electronic circuits, without requiring manual operation or camera swinging functions as in conventional technology. This is convenient because it allows you to capture the image of the subject in the center of the screen and display it enlarged.

Further, the automatic field of view enlargement function of the present invention can be installed anywhere in a television transmission system such as a video intercom or video telephone device, such as on the camera side, on the monitor side, or in the middle of the transmission system.

[Brief explanation of the drawing]

Fig. 1 is a configuration diagram of an embodiment of the present invention, Fig. 2 is an explanatory diagram for detecting the moving position of the subject in Fig. 1, and Fig. 3 is a synchronization signal of the image signal and the moving position of the subject in Fig. 1. FIG. 4 is a diagram showing the time relationship with the detection signal, and is a diagram showing the relationship between the original signal and the image signal obtained by the automatic field of view enlargement function processing in FIG. 1. l...TV camera unit, 2,13...process amplifier circuit, 3...TV monitor, 4...
A/D converter, 5... synchronous signal separation circuit,
6...Timing generation circuit, 7...Clock signal generation circuit, 8...Horizontal address counter, 9.
...Vertical address counter, 10... Address generation circuit, 11... Image signal memory, 12... D/A
Converter, I4... Pixel comparison circuit, 15...
Movement position address generation circuit, 16...Display position determination circuit, 17...Screen cutting circuit, 18...Enlargement control circuit. Patent applicant Nippon Telegraph and Telephone Corporation

Claims (1)

[Claims] means for storing a television signal; a difference information detecting means for generating difference information by comparing the television signal stored in the television signal storage means with a television signal to be transmitted next; and the difference information. means for generating first positional information of a portion of a television screen that has changed based on the time difference between the time at which the difference information is detected by the detection means and the synchronization signal of the television signal; and the first positional information generating means. means for generating second position information for cutting out a new enlarged television image from the original television signal based on the first position information; and information from the second position information generating means. Extracts specific horizontal scanning line information from multiple horizontal scanning line information of the signal,
At the same time, the horizontal scanning line information of the specified television signal is generated, and at the same time, the scanning line information of a specified period is cut out from one horizontal scanning line information of the television signal, and this is used to generate the horizontal scanning line information of the specified format of the television signal. 1. An automatic field of view adjustment device comprising: means for time-expanding one horizontal scanning line information, and generating a video signal that automatically tracks a moving portion of a television signal.