JPS6085695A - Television system - Google Patents

Abstract

PURPOSE:To make a time base multiple signal having less-deteriorated picture quality without spreading a frequency zone by utilizing a little of high frequency components in the oblique direction, by zone-compressing a luminance signal, making a space in term of time due to the time base compression, and multiplying the time base-compressed chrominance components in the space. CONSTITUTION:The first input terminal 1 inputs a wide zone luminance signal, which is quantized into a digital signal. The first adder 15 sums up these signals, which is divided by a half. The difference of these signal is calculated by a subtractor 16. The second and third input terminals 7 and 8 input (R-Y) and (B-Y), respectively, which are stored in the second and third buffer memory 22 and 23, respectively. Among a horizontal scanning period when signals are not written in each memory, a differential component of the luminance signal is read out through the fourth switch 24 in a half of an image signal period during said period, (R-Y) is read out in its one fourth period and (B-Y) is read out in the remaining one fourth period.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a television system, and particularly to a television system suitable for realizing higher quality images than conventional color television devices.

[Background of the invention]

The television whitening system currently being broadcast includes NTSO,
There are PAL and SEOAM, both of which transmit a color difference signal by frequency multiplexing it with the Ushiro component of a luminance signal. Therefore, a separate band is not required for the color difference signal, and narrowband transmission is possible.

However, since the color difference signals are frequency multiplexed, image quality is significantly degraded due to mutual interference between cross color signals, luminance signals, and color difference signals.

Furthermore, FM modulation is planned to be used in satellite broadcasting, which is expected to become more popular in the future, due to transmission power. There are problems such as the problem of triangular noise and difficulty in ensuring the S/N of color difference signals at high frequencies.Therefore, it is necessary to separate the luminance signal and color difference signal on the time axis and multiplex them on the time axis. Multiplex signal formats began to be considered. For example, MAO method (Multiple
ue Component: Literature uDirect T
e1evision B r oadcast by
Satellite Desirability of
a New 'rranmission-8tand
ar'ds//, I B A, B/D Report
tIL6/.

In 81), the luminance signal is temporally compressed to E, the two color difference signals are temporally compressed to M, and this is time-axis multiplexed with the compressed luminance signal alternately every horizontal scanning period. Transmit. When the # anonymous calculation is compressed on the time axis in this way, in addition to its reciprocal, ψ1'', for example, a 1.5 times wider frequency band is required in the MAC method. As a result, the required transmission band becomes wider and the number of channels becomes smaller. In addition, since transmission power and antenna gain are required to ensure S/N, there are many practical problems.

Similar time axis multiplexed signal formats include T, CI ('I
'ime Compressed Integrati
on: Literature - Development of high-definition television and its future〃
, Television Television 7 Academic Journal Va 1.36 NO10,19
82). This differs the frequency band of the luminance signal for each horizontal scan, and increases the time axis compression ratio on the narrow band side.
During the horizontal scanning period during which this narrowband luminance signal is transmitted, the broadband signal among the color difference signals is time-axis multiplexed, and during the horizontal scanning period during which the broadband luminance signal is transmitted, the narrowband color-difference signal is time-axis multiplexed. By doing so, the increase in the required bandwidth of multiplexed signals is reduced. However, in this method, the S/N differs depending on the consumption period. Furthermore, every horizontal scanning period, the frequency band of the signal is different, so if a signal with a small area moves slowly up and down, the signal with this small area will be split between the wide-band scanning line and the narrow-band scanning line. There is a phenomenon in which the image appears clearly and sometimes cannot be seen depending on the period of movement, resulting in a very unnatural pattern.

Furthermore, in both of these two signal systems, the color difference signals are performed line-sequentially, so at the vertical color change, one color difference signal is the signal for the upper part of the screen, and the other is the signal for the lower part. Therefore, during that horizontal scanning period, the upper and lower colors will be reproduced as completely different colors.
This results in significant image quality deterioration.

[Object of the Invention] An object of the present invention is to eliminate the drawbacks of the conventional time axis multiplexing method, to make the highest frequency of the transmission signal the same as the layer high frequency of the luminance signal, and to prevent image quality deterioration, as in the case of frequency multiplexing. [Summary of the Invention] In order to achieve the above object, the present invention adds luminance signals of horizontal scanning periods that match the tiles, and adds DJ$1. Then, a sum component (additional signal) and a difference component (subtraction signal) are created, and the difference component is temporally compressed to produce the compressed signal and two time-axis compressed color difference signals.

The present invention provides a television system in which signals are temporally multiplexed during a horizontal scanning period, and the sum component and the temporally multiplexed signal are switched and transmitted every horizontal scanning period.

Furthermore, the present invention provides a television system in which the time-domain compressed signal is transmitted in advance among the sum component and the time-domain compressed signal that are purchased with the luminance signal of the same horizontal scanning period.

The present invention utilizes the fact that there is a strong correlation between image signals of adjacent horizontal scanning lines.

That is, as is well known, since the beam of the image pickup tube has a spread, the high frequency components in the diagonal direction have less energy than the high frequency components in the vertical direction due to the integration effect within the beam. Visually, there is also a large tolerance for frequency characteristic deterioration in diagonal directions.

Taking advantage of the fact that the energy of the still frequency component in the needle direction is extremely small and the visual tolerance is large, the current NTSO system uses the color signal as the diagonal component of the sieve frequency of the 1st luminance. They succeeded in achieving compatibility with black and white by multiplexing (that is, frequency interleaving).

In the present invention, taking advantage of the fact that there are few high-frequency components in the diagonal direction, the luminance signal is band-compressed and the time axis is compressed to create a gap in time, and the time axis compressed color signal is applied to this part. By multiplexing, a time-domain multiplexed signal is created without widening the frequency band and with less deterioration in image quality. This shows that there are very few high frequency components in the long component of the signal. Therefore, even if the band is limited for the difference component, the effect on image quality is extremely small. In the 1-order scanning method, which is different from the current interlaced scanning such as NTSO, there is no problem as long as the two 14Ds match horizontally, and even if it is compressed to about 1, the 11! Quality II is obtained.

Therefore, the difference between the two horizontal eclipsing line signals is determined, the band of the difference component is limited, and the time axis is compressed to the extent that the maximum frequency of this limited signal does not exceed the maximum frequency of the luminance signal. Therefore, the time-axis compressed color difference signal is transmitted during the free time.Furthermore, in the present invention, the transmission order of the sum component, the difference component, and the color difference signal of the multiplexed luminance signal is determined by the image receiving imitation press. The order is such that the configuration of j is the simplest. That is,
In general broadcasting, the radio waves sent from a small number of broadcasting stations are received by the general broadcasting station, and the number of transmitters, including local broadcasting stations, is in the hundreds, while the number of receivers is in the thousands. Ashi, 1
There are about 0 cracks. Therefore, it is extremely important to make the configuration of the receiving side as simple as possible.

By the way, in the present invention, since the sum component is not compressed on the time axis, it can be used as it is, whereas the difference component is not compressed, so it is necessary to juxtapose it at the original time on the receiver side. .

For this purpose, it is necessary to provide a line memory and read out the contents of the line memory by delaying the bladder output clock with respect to the input clock. Therefore, if the configuration is such that the difference component is transmitted first and then the sum component, and the difference component is recorded for one line, and then the difference component 1 is read out in synchronization with the timing of the sum component, this output Since the times of the daily component and the difference component are aligned, and the sum and difference of these two components are easily calculated, the signal of the original horizontal scanning period can be restored.

[Embodiments of the invention]

An embodiment of the circuit configuration on the transmission side of the present invention is shown in FIG.

In FIG. 1, i, 2.3.4 is an input terminal, 5 is an output terminal, and 6. Z8 is A/D converter, 910, 11.2
4 is a switch, 12.13.14.48.49 is 1H, however, H is a delay line indicating the horizontal scanning period, 15.17° 18
16 is an adder, 16 is a subtracter, 19 is a low-pass filter, 20 is a resampling circuit, 21, 22, 23 is a buffer memory, 25 is a D/A converter, and 26 is an adder.

A broadband luminance signal is input from the tenth input terminal 1. The first A/D converter 6 quantizes it into an 8-bit digital signal. The clock frequency (fc) of the first A/I converter 6 needs to be at least twice the highest frequency (fmαX) of the luminance signal input from the input terminal 1.

For example, when f7'rLax w 16MHz, fc is 3
The value is 2 MHz or more, for example, 40 MHz. Here, the quantized waveform is transmitted to the first switch 1, and the IH
Each time, the first delay line 12 delays the signal by 1H. This 1H delay 12 delays the digital signal of 0.1, so for example, it is used in a shift register or a normal DRA.
This can be easily configured by using a memory such as M and reading it out using IHL after convolution. In terms of capacity, if we consider a sequential set meal with a frame frequency of 60 MHz (z s) and 525 scanning lines, the A/D sampling frequency is 40 MHz.
It is z. Even if quantization is set to 8 bits, the memory capacity may be approximately 10 Kbits.

In this way, the output of one side I of the first switch 9 is obtained every 1H as shown in FIG. 2(h),
On the other hand, as shown in FIG. 2(C), an output is obtained at the opposite position. This will be delayed by 1H in the first delay/delay line 12, so as shown in Figure 2 (d),
Outputs are obtained every 11-1 degrees at the same time as in FIG. 2(b).

Therefore, if the sum of both is divided by 1 in the first adder 15, the average value of FIG. 2(h) and (d) (for example, FIG. 2(
(e) Yl+Y2, etc.) is obtained. When the difference is taken by the subtractor 16, Kei (for example, the second
o.

Y+-Y2 in figure (f), etc.) are obtained.

The low-pass filter 9 is composed of a digital filter.

It will be done. As is well known, for signals with a frequency of 7 or less, where the clock frequency is 1 fc, a filter similar to a normal analog filter can be constructed by adding yA of digital signals with a time difference of a.・This digital filter reduces the band of the difference component of the luminance signal to 1
That is, the highest frequency of the difference component is limited to -2 fmax (for example, 8 MHz). like this.

Then, even if the data is thinned out every clock by the resampling circuit and the clock frequency is set to -Lfc, the difference component information will be resampled correctly.

I will be able to do it.

Therefore, the first buffer memory 21 contains the first I
It becomes possible to store the difference component with the memory capacity (approximately 5 Kbit) of the H delay line.

On the other hand, the performance of color difference signals does not deteriorate compared to luminance signals even in a narrow band (Currently, NTSO operates in a band of about 8.0cm, but there is no problem at all), so for example, 4 (a. fmax (
For example, it can be limited to 4MFJo) ). Therefore, from the second input terminal 7 (R, -Y), the input terminal 8 of m5
When (B-Y) are input, the second A/D
The signals are quantized by the converter 7 and the third A/D converter 8, respectively. This 20th A/D converter 6th A/D converter
The sampling frequency of the D converter is one! -fc (for example, 10 MHz) is sufficient.

When the second switch 10 and the third switch 11 are operated in synchronization with the first switch 9, the outputs of the second adder 17 and the third adder 18 are respectively shown in FIG. and(
As shown in e), information on the average value of two scanning lines (for example, Ul+, IJzVl+V2, etc.) is obtained in 1H@. These color difference signals are stored in the buffer memory 22 of Mouse 2 and the third buffer memory 23, respectively. The storage capacity of this buffer memory is 7t for each first delay line 12.
No Yu no Yo 4.

(Therefore, for example, 2sKbjt) is sufficient. Here, since the difference component of the luminance signal is in the negative band of the original luminance signal 1, the time axis is compressed to i. It will be in the same band as the signal.

Similarly, two color difference signals (几-Y) and (B-Y・).

Since it is in the band of 1, 4 which forces the time axis to 1 becomes the same band as the luminance signal. Therefore, the first buffer memory 21.42 buffer memory 1822, the third buffer memory 21.
All the read clocks of the buffer memory 26 of
c, the horizontal scanning period during which signals are not spread to each memory.

The components are read out through the fourth switch 24. (R, -Y) are read out during the period L and 7, and (B-Y) is read out during the remaining period.
By reading out the difference component, (R, -Y) and (B-'Y) can be multiplexed within the highest frequency of the luminance signal.

If this is done, all the contents of these six buffer memories will be read out, so that the convolution information will be input in the next horizontal scanning period after reading out, and new next information can be written.

The sum component of the luminance signal is generated by the ninth 1H delay line 48 and the tenth 1H delay line 49 (memory capacity is 10 bits each)
The difference component of the luminance signal and the color difference signal are delayed by 2H and output as they are through the switch 24 during the 1H period after the color difference signal.

As is clear in the above embodiments, the order of the output signals is the temporally multiplexed signal followed by the sum component of the luminance signal. As mentioned above, this output order has the effect of configuring the receiver side in the simplest manner. if,
If this effect l is not required, the sum component may be output first. In this case, the above-mentioned 1H delay lines 48 and 49 become unnecessary on the transmitting side and are installed on the receiving side. .

By switching the fourth switch 24 as described above.

Output and convert to analog quantity with D/A converter 25,
It is mixed with the composite synchronization signal and audio signal input from the fourth input terminal 4 in the fourth adder 26.

Together, the desired transmission signal is obtained. .

Here, the audio signal is an extremely narrow band signal.

(The band is 20 KHz, which is about 10-3 of the image signal.) Therefore, for example, POM (Pulse Coded Mo
(duration) signal and multiplexing it during the horizontal blanking period. Furthermore, the composite synchronization signal may be a conventional synchronization signal, or it may be the same as a video signal by encoding synchronization information.

Of course, it is also possible to use so-called digital synchronization (for example, the above-mentioned MAO signal) that is multiplexed within a level range.

Still, there is no problem or inconvenience in the synchronization system even if the synchronization signal is transmitted once every 2H like the TOI signal mentioned above.

Therefore, once every 2H, there is a horizontal pull.

The ranking period can also be used to transmit video signals.

is also possible. In this case, this spread time? .

Assign it to the difference component or color difference signal, lower the compression ratio,
It is best to use it for wideband ps/N improvement. Since the horizontal ranking period is about 20% of one horizontal scanning period,
Even if you assign it to the difference component and assign it to the raw two color difference components,
As can be seen in Figure 2 (0 or Q), the allocation time is increased by about 50%.

In other words, when assigned to the difference component, the time axis is compressed.

3 Can be from 1 to 7. In this case, since the difference component can be made into a wide band, the slope of the low-pass filter 19 can be made gentler, making it easier to create a circuit that is less likely to cause ringing, etc., and since the filter can be cut off, it is easier to configure a band-limiting filter. Also, it becomes easier to obtain a waveform with good response.

Also, depending on how the synchronization information is entered, Fig. 2 (r) and (S)
A format such as this is also possible. Furthermore, it is obvious that there is no problem in changing the order in which the two color difference signals are input, or the order in which the difference component of the luminance signal and the color difference signal are input.

Furthermore, it is obvious that the time compression ratio is not essentially limited to the above numerical value.

Next, FIG. 6 shows a circuit that demodulates a luminance signal and a color difference signal from this transmitted signal. This 9° circuit is included in the middle of the transmission path or in the final receiver. 27 is an input terminal, 28, 29, 30 is an output terminal 31 is an A/D converter, 52, 43, 44° 45 is a switch, 40.
41.42. is 1H delay axis, 34° 35, 361d, backup memory, 38 is adder, 39 is subtracter, 46 is synchronous separation circuit, 47 is clock regeneration circuit, 50.51.5
2 is a D/A converter.

For a sequential tanshi with a frame frequency of 60 Hz and 525 scanning lines, the horizontal repetition frequency (fu) is approximately 31.5
It becomes KHz.

Therefore, when the synchronization signal separated by the synchronization separation circuit 46 is injected into the clock recovery circuit 1147, which is composed of the VQO oscillating at 40 MI-1z, a frequency division circuit that divides the frequency by 1280, and a phase detection circuit, the vCO becomes 40.32 M)(
z - Oscillates in phase synchronization with the Eiju synchronization signal. Therefore, using this clock, the fourth A/D converter 3
1, and further, by counting this clock with reference to the horizontal synchronizing signal, a control signal for separating each section of the multiplexed signal shown in FIG. 2 (rL) can be easily obtained. Using this control signal, the fifth switch 62
By controlling the multiplexed signal, the sum component of the luminance signal, the difference component of the luminance signal, the color difference signal (R-Y): the color difference signal (B
-Y, ) can be separated from each other. .

Here, the v# degree-folded signal component (Fig. 4 (υ)) color difference signal (RY) (Fig. 4 (to)), the color difference signal (
Bs-Y) (FIG. 4 ()) are compressed on the time axis, so they are stored in the fourth buffer memory 64, the fifth buffer memory 35, and the buffer memory 36 of M6,
In the next 1H, the clock frequency is set to 640 f.
Read with H1320fH and 320fH.

As a result, a signal extended to 1H period is obtained.

The timing at which the difference component of the luminance signal is read from the buffer memory is exactly the timing at which the sum component of the luminance signal is read out from the A/D (see Figure 4 (u)), as shown in Figure 4 (W). Because we have everything.

When the sum component and the difference component are added, the signal of the preceding horizontal scanning period of the two horizontal scanning periods is obtained, as shown in Fig. 4 (→), and when the difference component is subtracted from the sum component, the signal of the preceding horizontal scanning period is obtained as shown in Fig. 4 (, The signal for the subsequent horizontal scanning period is obtained as shown in VI.Now, using the sixth 1H delay line 40,
By delaying the output of the subtracter 39' by 1H and outputting it by switching every 1H with the sixth switch 43, a luminance signal in the same order as the original luminance signal can be reproduced.

Further, the two color difference signals are transferred to a fifth buffer memory 55.
If the outputs of the and sixth buffer memory 36 are delayed by 1H in the seventh 1H delay line 41 and the eighth 1H delay/delay line 42, and the same information is read over a 2H period, as shown in FIG. ) and (F), continuous color difference signals can be reproduced. Here, these two color difference signals are generated by Ul + T while the luminance signals are being read out, for example, Yl and Y2.
J2 and ■1±.

Since V2 is read out, the vertical centers of gravity of the luminance signal and the color difference signal coincide. Of course, since the same signal is read out over 2H for the color difference signal, the vertical resolution is i compared to the luminance signal, but what was said above regarding the horizontal band also holds true for the vertical direction, and this is sufficient for the color difference signal band. is obtained.

There is also a line-sequential method for transmitting color difference signals, such as the MAO signal and TCZ signal mentioned above, but in this method, the ratio of (R-Y) and (B-Y) components is determined at the vertical color change line. The image will differ from the actual image, and false colors will appear at the transition points, resulting in significant image quality deterioration. On the other hand, the color difference signal transmission described in this embodiment does not have this problem, and extremely good color reproduction is possible.In addition, in this embodiment, on the transmitting side, the following scanning line is Since the explanation was made in terms of subtracting the signal of If you make it into a subtracting form,
On the playback side, the signal of the subsequent scanning line is obtained from the adder output, so a 1H delay line is required for this output.
Of course, there is no essential difference.

In this embodiment, since the difference component of the luminance signal is configured to precede the sum component, the 1H delay line that adjusts the timing of the read signal on the playback side is not required for the sum component. . Therefore, on the playback side, it is of great significance because it is incorporated into television receivers that are produced in extremely large quantities. Of course, as mentioned in the explanation of FIG. 1, it is also possible to precede the oro component, but
In this case, 21 1H delay arms are required between the switch 32, adder 38, and subtracter 390 in FIG.

Furthermore, if the present invention is made into a sequential method with the same field frequency as the current NTSO and twice the number of scanning lines (i.e., a sequential method with 525 Tz scanning lines at a frame frequency of 60), the signal of this sequential method will be Converting to NTSO format is very easy. In other words, if in a certain field the sum of the Nth and N+1th lines is 0, then in the next field the sum is found as the sum of the (N-1)th and Nth lines, and the sum of the (N+1)th and (N+2)th lines. It turns out. Therefore, the sum signal of one field and the sum signal of the next field have a slightly interlaced relationship. Therefore, the sum component of the luminance signal and the sum component of the two color difference signals are stored as they are in the buffer memory, and the luminance signal is twice,
By time-expanding the color difference signal by eight times and performing the necessary band limiting, a 4 degree signal of N T 80 and two color difference signals can be obtained.

In this way, signal conversion to the conventional system is extremely easy.

〔Effect of the invention〕

As explained above, the time axis multiplexing method according to the present invention can multiplex color difference signals within the highest frequency band of the luminance signal in the same way as conventional frequency multiplexing, and eliminates the drawback of conventional time axis multiplexing, which is the broadband nature. Moreover, it has become possible to completely eliminate problems such as cross color, which were problems with frequency multiplexing systems. Furthermore, in the present invention, by making the difference component of the luminance signal temporally precede the sum component, it has become possible to perform black with a simple configuration on the reproduction side.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an embodiment of the signal converter of the television device according to the present invention, FIG. 2 is a diagram showing waveform examples of each part of FIG. 1, and FIG. FIG. 4 is a diagram showing an example of a signal demodulator, and FIG. 4 is a diagram showing an example of waveforms of each part in FIG. 3. 9, 10.11.24.32.45.44.45...
...Switches 12, 13.14.40.41.42.
48.49...1H delay line 21, 22.23.
34.35.56...Buffer memory 15, 1
7.18.58... Assister 16.39...
... Subtractor Patent Attorney Akio Takahashi Continued from page 1 ■ Inventor Village 1) Toshinori ■ Inventor Michitaka Osawa 29 Akiji, Yoshida-cho, Totsuka-ku, Yokohama Hitachi, Ltd. Home Appliance Research Laboratory

Claims (1)

[Claims] 1. Adding luminance signals of adjacent horizontal scanning periods. The subtracted signal is compressed in the time axis, and the compressed 5 signals and the signal obtained by compressing the time axis of the color difference signal are temporally multiplexed in the horizontal scanning period. A television system characterized by switching between and transmitting the above-mentioned temporally multiplexed signal. 2. The television system according to claim 1, wherein of the addition signal 0 and the temporally multiplexed signal, which are configured in the same horizontal scanning period, the multiplexed signal is transmitted first. .