JPH0472991A - Separation system and separation circuit for television signal - Google Patents

Abstract

PURPOSE:To reduce crosstalk by grouping a consecutive field of a television signal into fields each comprising n (n is an integer being 2 or over) fields and applying processing to the divided fields and applying completed separation processing in each field group. CONSTITUTION:Two frames of a film correspond to five fields of a television signal. In this case, suppose that a television signal of 1st and 2nd fields is generated from a 1st film and a television signal of 3rd, 4th and 5th fields is generated from a 2nd film, then an inputted NTSC (composite signal) X525 is inputted to delay circuits 1-4 and respective output signals X263, X0, X-263,X-525 are obtained. The signals are switched by using a changeover device 5. In this case, a changeover device 4 is controlled for each field by a control circuit 6 so that X263, X-263, X525, X-525 are outputted when a scanning line X0 is resident on fields No 1, 2, 3, 4, 5.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to a separation method and separation circuit for television signals, and in particular to a separation method and separation circuit suitable for use with television signals converted from movie films, etc. Regarding circuits.

[Conventional technology]

In the current television system (NTSC system), images are transmitted by dividing one frame into two fields using interlace scanning, and transmitting images at a rate of 30 frames (60 frames per second).
field).

In contrast, standard films such as 16+nm and 35mm have a rate of 24 frames per second. If a standard screen projector is used in conjunction with a television camera to transmit film, the relationship between the shutter of the projector and the scanning of the television will cause distinct horizontal stripes that move up and down or severe flickering to appear in the image, making it impractical for telecine. A special projector called a converts the film image into a television signal.

Telecine has 24 frames per second for film and 6 frames per second for television.
In order to correspond to the 0 field, 2 frames of film are converted into 5 fields with 1712 seconds as one cycle. In telecines using the 23 pulldown method (or the fast pulldown method, which is an improved version of it) that is commonly used today, odd-numbered films are processed in two fields (2/6
By scraping off every 3 fields (3/60 seconds) for even-numbered films, 2 frames of film and 5 fields are made to correspond. This type of technology is available, for example, from Yu Iwase - 2"Video Glossary," Shashin Kogyo Shuppansha.

It is described in "Television Engineering Handbook," edited by the Television Society of Japan, August 1989, and published by 7 Oichimusha, December 1969.

By the way, in the NTSC television system, luminance signals and color signals (color signals modulated with subcarriers) are frequency-multiplexed in a three-dimensional frequency domain expressed by horizontal, vertical, and temporal frequencies, and transmitted as a composite signal. ing. Conventionally, in conventional television receivers, luminance signals and chrominance signals are processed using a BPF (band pass filter) with a center frequency of 3.58 MHz or a two-dimensional comb filter that incorporates calculations between scanning lines within the field. Separation (YC total separation is performed.YC separation method using a three-dimensional comb filter that improves separation accuracy by introducing interframe calculations has also been proposed. For example, Achiba et al.

"Motion Adaptive Signal Processing for IDTV Receivers," Journal of the Society of Television Engineers, 41, 7, 1987.

The conventional three-dimensional comb yc separation method will be explained using FIGS. 5 to 7. In FIG. 5, the phase of the color subcarrier modulating the color signal is reversed between the upper and lower scanning lines and between the front and rear scanning lines, as shown by + and - symbols in the figure.

Here, if we pay attention to, for example, the scanning line G among the transmitted scanning lines A-M, if the image is still, G'l'=0.25B+0.5G+0. 25LGC=-
The calculation 0.25B+0.5G-0.25L is performed to separate the luminance signal GV and color signal GC. Also, if the image is moving, the above separation characteristics deteriorate, so from the upper and lower scanning lines in the field, GY = 0.25F + 0.5G + 0.25HGC = -0,
The calculation 25F+0.5G-0,25H is performed to separate the luminance signal GY and the color signal GC. These two modes
Continuously (
(multi-valued) switching.

[Problem to be solved by the invention]

FIG. 6 shows the yc separation characteristics in the still image mode in the vertical-temporal frequency domain. In the case of a still image, both the luminance signal Y and the color signal C have a small temporal frequency spread, so complete YC separation can be achieved by frequency separation in the time direction, and there is no crosstalk between the two. Also, at this time, the luminance signal Y is up to 52
It can reproduce up to a vertical resolution of 5/2 c.p.h.

FIG. 7 shows the yc separation characteristic in the moving image mode in the vertical-temporal frequency domain. In the case of a moving image, the temporal frequency spread of both the luminance signal Y and the color signal C becomes large, and it becomes impossible to separate the two even if calculations are performed between frames. For this reason, calculations are performed between scanning lines within a field to perform separation based on the difference in vertical frequency. However, at this time, the luminance signal is 525/
Since it is possible to reproduce only up to a vertical resolution of 8 c p h, and a luminance signal exceeding this is reproduced as a color signal, a cross 1 hake occurs.

In addition, since the still image mode and video mode are switched depending on the size of movement for each pixel, when the movement is slow or at the beginning (end of movement), the playback image may appear unnatural due to the difference in resolution caused by switching between the two modes. In some cases,

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a YC separation method and a separation circuit that improve the vertical resolution that can be reproduced in a moving image mode for television signals converted from film, and that have less crosstalk.

[Means to solve the problem]

The above object is achieved by performing YC separation on a television signal converted from a film for each field group created from the same film.

[Effect] The operating principle of the present invention will be explained using FIG.

As mentioned above, two frames of film correspond to five fields of television signals. At this time, for example, as shown in the figure, television signals of the first and second fields are created from the first film, and television signals of the 3rd, 4th, and 5th fields are created from the second film. shall be carried out. Even in the case of a video, the first field and the second field together can be considered a still image.
YC separation is possible using all of the information in this field group. For example, focusing on scan lines B and E, BY= 0.5 B + 0.5 E Bc=0.5B-0,5E EY= 0.5 B + 0.5 E EC=-0,5B + 0 , 5 E to obtain the luminance signals BY and EY and the color signal BC.
and find EC.

FIG. 2 shows the YC separation characteristics at this time in the vertical-time frequency domain. From the figure, it can be seen that even a moving image can be reproduced with a vertical resolution of up to 525/4 cph.

Similarly, since the third to fifth fields are signals created from blank film, they can be considered still images even if they are moving images, and YC separation can be performed by performing calculations between fields. becomes. In particular, by taking the sum signal and difference signal of the scanning lines at the same position in the third and fifth fields, it is possible to reproduce the luminance signal and the color signal, and it is possible to reproduce the complete YC signal without deterioration of vertical resolution or crosstalk.
Separation can be achieved.

Therefore, by performing YC separation processing for each field group created from the same film, it is possible to achieve YC separation with less deterioration in vertical resolution and less crosstalk between luminance signals and color signals even in video parts, achieving the above objectives. can do.

Furthermore, by performing static and motion determination for each pixel, performing the above-mentioned YC separation between field groups for the moving image portion, and performing conventional inter-frame YC separation for the still image portion, complete YC, separation is performed for the still image portion. can be realized.

Also, for the third and fifth fields, complete Y
Since C separation is possible, the luminance signal and chrominance signal reproduced from this field group can be used as is, or filtered in the vertical-temporal frequency domain, and the nearby fields (for example, 1st, 2nd, 4th, 6th .7 field, etc.) may be replaced with the luminance signal and color signal.

Furthermore, for fields where complete YC separation is not possible (for example, the 1st, 2, and 4th fields), crosstalk components may be removed in advance on the transmitting side. At this time, 2
Dimensional or three-dimensional brifilter (bricombing)
can be used.

〔Example〕

Hereinafter, the present invention will be explained in detail using the drawings.

FIG. 3 shows the positional relationship of the scanning lines of the transmitted television signal. In the figure, the horizontal axis is time (field)
, the vertical axis indicates the vertical positional relationship. A scanning line that is temporally in the future by 525H (LH is one horizontal scanning period) with respect to the XO scanning line as the center is expressed as X525, and a scanning line that is temporally in the past by 525H is expressed as X-525. Other scanning lines (X263.X262.Xi, X-1, X-26
The same applies to 2°X-263). Also shown in the figure are field groups created from the same film. The field number in the same figure is the field number shown in FIG.
, corresponds to For example, the detection of field NO.
The method described in Japanese Patent Application No. 2-45408 may be used. The positional relationship of these scanning lines will be described with reference to below.

FIG. 4 shows a specific configuration diagram of an embodiment of the present invention.

In the figure, the scanning line groups shown in FIG. 1 are used when performing YC separation. In the same figure, the input NTS
The C signal (composite signal) is set to X525, this is input to delay circuits 1 to 4, and the respective output signals are
．． XO, X-263°X-525. These signals are switched using a switch 5. At this time, the control circuit 6 causes the switch 4 to change the scanning line XO to the field Nα.
When it is on 1°2.3, 4, 5, respectively X 26
3.

X-263, X525. X263. Control each field so that X-525 is output.

This is repeated every 5 fields. A subtracter 7 is used to take the difference between the output of the switch 5 and xO, and a multiplier 8 is used to multiply the difference by 1/2 to obtain a color signal C for the moving image mode. On the other hand, adder 99 multiplier 10. Subtractor 11 and multiplier 1
(-X-525/4+XO/2-X525/
4) (7) Create a signal and use it as color signal C in still image mode. Separately, a motion detection circuit 13 is used to detect the motion amount k (0≦to≦1) for each pixel from the input signal.

The amount of motion includes the low frequency of the frame difference of the input signal and the
Obtained from frame difference, etc. Multiplier 14 is used to multiply the motion mode color signal C by k, multiplier 15 is used to multiply the still image mode color signal C by (1-k), adder 16 is used to add both, and then 3.58MHz as required
A bandpass filter (BPF) 17 with a center frequency of
After passing through the color signal C, the color signal C is obtained. At the same time, the color signal C is subtracted from xO using the subtracter 18 to obtain the luminance signal Y. Note that the multiplier in the figure may be realized by bit shifting of the signal.

FIG. 8 shows a specific configuration diagram of another embodiment of the present invention. The configuration shown in the same figure is based on the configuration shown in FIG.
In addition to separation, it also allows switching to conventional motion-adaptive yc separation. In the figure, the input NTSC signal (composite signal) is designated as X525, which is input to delay circuits 19 to 26, and the respective output signals are transmitted to X263.
X262゜Xi, XO, X-1, X-262, X-26
3°X-525. Multipliers 27 to 3 are applied to these signals.
5 is used to multiply the coefficients of h1 to h9, respectively, and the adder 3
7 is added to form the color signal C in the moving image mode. The coefficients h1 to h9 are switched by the control circuit 36 for each field. Details of this coefficient will be described later.

On the other hand, adder 99 multiplier 10. Using subtracter 11 and multiplier 12, (-X-525/4+XO/2-X525
/4) is created and used as the color signal C in the still image mode.

Separately, a motion detection circuit 13 is used to detect motion 1i-k (0≦to≦1) for each pixel from the input signal.

The amount of motion includes the low frequency of the frame difference of the input signal and the
Obtained from frame difference, etc. Multiplier 14 is used to multiply the motion mode color signal C by k, multiplier 15 is used to multiply the still image mode color signal C by (1-k), adder 16 is used to add both, and 3,58M as required
A bandpass filter (BPF) with a center frequency of Hz
) 17 and then becomes color signal C. At the same time, subtractor 1
8 to subtract the color signal C from xO to obtain the luminance signal Y. Note that the multiplier in the figure may be realized by bit shifting of the signal.

FIG. 9 shows coefficients h1 to h9 used in the configuration shown in FIG.
An example is shown below. In the figure, when the scanning line xO is on the field Nα1, the coefficients are selected from among the coefficient groups shown in item numbers 1 to 3. For example, as in item number 1, h2=-1/
2. If h5=1/2, the result of YC separation will be the same as that obtained by the configuration shown in FIG. Similarly, items 4 to 6 are examples of coefficients when the scanning line XO is on the field Nα2,
Item number 7 represents the case where the scanning line XO is on field Nα3, item numbers 8 to 16 represent the case where the scanning line XO is on field No.
5. These steps are repeated every 5 fields. Item No. 18 is the same as YC separation in the conventional moving image mode, and is used for general images (images not converted from film).

Further, even for signals converted from film, conventional YC separation may be performed manually according to the viewer's preference. Item number 19 is the same as YC separation in the conventional still image mode, and the coefficients in item numbers 1 to 18 above are multiplied by the motion amount k in advance, and the coefficient in item number 19 is multiplied by (1-k) in advance to separate both. By adding , the adder 99 and the multiplier 10 . of the configuration shown in FIG. The subtracter 112, the multiplier 122, the multiplier 149, the multiplier 15°, the adder 16, etc. can be eliminated, the amount of hardware can be reduced, and it is economical. Also, the coefficients are not limited to the above, and it is also possible to combine the sums or products of the coefficients of each item number, or to make the vertical frequency characteristics steeper by increasing the number of taps in the vertical direction. .

Note that the present invention can be used not only for transmitted television signals but also for signals recorded on VTRs, video discs, and the like.

〔Effect of the invention〕

By applying the present invention, it is possible to perform accurate separation of luminance and color signals even for moving images by signal processing only on the receiver side for television signals converted from film. Since it is possible to improve the vertical resolution of the separated signals and realize a YC separation circuit with less crosstalk, the effect of improving image quality is extremely large.

[Brief explanation of the drawing]

1 and 2 are diagrams explaining the operating principle of the present invention,
FIG. 3 is a diagram explaining the positional relationship of scanning lines, FIG. 4 is a configuration diagram of an embodiment of the present invention, FIGS. 5 to 7 are diagrams explaining the operating principle of conventional YC separation, and FIG. This figure is a block diagram of another embodiment of the present invention, and FIG. 9 is a diagram showing an example of coefficients used in another embodiment of the present invention.

Claims (1)

[Claims] 1. A system and circuit for separating each signal from a television signal in which a plurality of signals are frequency multiplexed,
A television characterized in that continuous fields of the television signal are processed by dividing them into field groups consisting of n fields (where n is an integer of 2 or more), and complete separation processing is performed within each field group. John signal separation method. 2. A claim characterized in that the motion of each pixel of the television signal is detected, and depending on the detection result, adaptive switching is made between separation that is completed within the field group and separation that is not completed within the field group. The television signal separation method according to item 1. 3. The separation of television signals according to claim 1 or 2, wherein the integer n is 2 or 3, and processing of field groups of n=2 and processing of field groups of n=3 are performed alternately. method. 4. A method and circuit for separating each signal from a television signal in which a plurality of signals are frequency multiplexed,
A television characterized in that continuous fields of the television signal are processed by dividing them into field groups consisting of n fields (where n is an integer of 2 or more), and complete separation processing is performed within each field group. John signal separation circuit. 5. A claim characterized in that the motion of each pixel of the television signal is detected, and depending on the detection result, the separation is adaptively switched between the separation that is completed within the field group and the separation that is not completed within the field group. Item 4. The television signal separation circuit according to item 4. 6. Separation of television signals according to claim 4 or 5, wherein the integer n is 2 or 3, and processing of field groups of n=2 and processing of field groups of n=3 are performed alternately. circuit.