JPS58711B2 - Kounouritsufugōkahoushiki - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an improvement in a high-efficiency encoding system that makes encoding a composite color television signal in which a luminance signal and a carrier chrominance signal are frequency-multiplexed more efficiently than before.

In order to pulse encode the standard color television signal as it is and transmit it digitally, the sampling frequency for converting the analog color television signal into a digital signal is selected to be 10 to 11 MHz, and the number of bits per pixel is set to 8. Since the bit rate of the durable color television signal is extremely large, on the order of 80 to 90 megabits/second, an extremely wide transmission band is required.

Therefore, for high-efficiency encoding to reduce the huge bit rate as mentioned above and perform high-efficiency transmission of composite color television signals, it has conventionally been necessary to combine the luminance signal Y and the carrier chrominance signal C. is spectrally separated using a comb filter, the carrier color signal C is further demodulated into two color signals C1 and C2, and each of these signals Y and C0 and C2 are differentially pulse encoded, etc. Research and development is underway in various places regarding encoding methods.

However, these encoding systems use a comb filter to separate the high-frequency components of the luminance signal and the carrier color signal, whose respective signal spectra are arranged alternately in the same frequency band, and the luminance signal is separated from the high-frequency components. Since signal levels in the entire frequency domain, including the luminance signal, are sampled in the same quantization step and pulse encoded, the circuit configuration for encoding and decoding on the receiving side becomes complex, and the sample level for the luminance signal is It is difficult to increase the efficiency of pulse encoding because it is necessary to increase the encoding frequency and therefore the bit rate for pulse encoding, and a wide transmission frequency band is required to transmit such encoded signals. There were drawbacks such as the need for

The object of the present invention is to eliminate the above-mentioned drawbacks, to simplify the circuit configuration for encoding and decoding on the receiving side, to compress the quantization step of the signal level of the luminance signal and to reduce the sampling frequency, thus reducing the sampling frequency. It is an object of the present invention to provide a highly efficient encoding system for color television signals, which performs highly efficient pulse encoding by lowering the encoding bit rate and narrows the transmission frequency band of the encoded signal.

That is, in encoding a color television signal configured by frequency multiplexing a luminance signal and a carrier chrominance signal, the high-efficiency encoding method of the present invention shares a frequency band with the carrier chrominance signal of the luminance signal. The two signal components of the carrier color signal, each including a low-midband signal component of a portion of the luminance signal that does not share a frequency band with the carrier color signal, and a high-frequency signal component of a portion of the luminance signal that shares a frequency band with the carrier color signal, are This feature is characterized in that it is encoded as follows.

The present invention will be described in detail below with reference to the drawings.

FIG. 1 shows an aspect of signal spectrum separation between a luminance signal Y and a carrier color signal C in the conventional color television signal encoding system described above.

In a composite color television signal based on frequency multiplexing of a luminance signal and a carrier color signal, as shown in Figure 1a,
For example, an odd multiple of 1/2 of the horizontal scanning frequency fh/2 (2n+
1) Since the signal spectrum sequence of the carrier color signal centered on the component (n is a positive integer) is frequency interpolated, the carrier color signal C and the high frequency component of the luminance signal Y share the frequency band. There is.

Therefore, conventionally, prior to respectively encoding the luminance signal Y and the carrier color signal C, a delay time of one or two periods of horizontal scanning of the preview signal, that is, 1/fh or 2/fh is provided. A comb filter configured to use a delay line and utilize the fact that the polarity of the carrier color signal is reversed in adjacent horizontal scanning periods to obtain the sum and difference of composite signals in adjacent horizontal scanning periods. By adding the composite color television signal of FIG.
The luminance signal Y and the carrier color signal C as shown in FIGS. As a result, the two color signals C1 and C2 as shown in FIG. 1d are demodulated.

In this case, for example, in Japan's standard television system, the upper limit fb of the frequency band of the brightness signal Y is 4.2 MH.
z, and the upper limit fc of the frequency band of the color signals C1 and C2 is, for example, 1.5 MHz.

In the conventional YC separation encoding method, when each of the luminance signal Y and chrominance signal C1°C2 separated as described above is encoded, the luminance signal Y is encoded as shown in FIG. 3a, which will be described later. As shown in FIG. The sampled luminance signal components are sequentially encoded by reducing the bit rate by, for example, differential pulse encoding.

- In the above encoding method, the sampling frequency fs for the luminance signal Y is set as fs=2fb-α, and the luminance signal Y
As shown in the luminance signal spectral distribution in Figure 3a, the minimum Nyquist frequency 2fb is lower than the spectrum of the original signal with fb as the upper limit of the frequency domain, and the frequency centered at the sampling frequency fs. This is to ensure that the fundamental wave component spectrum of the sampled signal whose upper and lower limits of the band are also fb is alternately interpolated in frequency as in the frequency multiplexing of the luminance signal Y and the carrier color signal C described above. , hence the sampling frequency f.

is set to a frequency at which the spectral interpolation described above is a 1/2 line offset performed by a frequency difference of 4fh.

In the conventional YC separate encoding system, two color signals C
1 and C2 are also sampled in the same way as for the luminance signal Y described above to form a sampling signal similar to that shown in FIG. Sampling frequency f for °C2
, C respectively fs(Y)=8~6.02MHz, f8
(C) is set appropriately as described above within the range of 3 to 1.77 MHz.

In contrast to the conventional method described above, in the coding method of the present invention, when a composite color television signal is frequency-separated and then coded as in the conventional method, each signal spectrum string is alternately interpolated in frequency. In order to enable highly efficient encoding without frequency-separating the high frequency component of the luminance signal and the carrier color signal C using a comb filter, the color television signal is Frequency separation is performed in an extremely simple manner as shown in the figure.

That is, the composite color television signal shown in FIG. 2a is converted into a luminance signal Y that does not share a frequency band with the carrier color signal C by frequency interpolation of each signal spectrum sequence, as shown in FIG. is separated by a simple circuit configuration such as adding the composite color television signal to a low-pass filter whose cutoff frequency is the upper limit frequency fL of the frequency region of the component.

On the other hand, as for the carrier color signal C, as shown in FIG. Separation is achieved by a simple circuit configuration such as adding a composite color television signal to a high-pass filter of frequency fL.

The frequency separated signals in FIG. 2C are further synchronously detected using color subcarrier signals having a phase difference of 900, and as shown in FIG. Take out signals C1 and C2.

In the color signals C1 and C2 shown in FIG. 2d, the high-frequency component spectrum of the luminance signal Y is converted into a spectral train of two low-frequency components having mutually orthogonal phases, resulting in a spectral train of the color signal. The frequency is interpolated inside.

In the coding method of the present invention, the Y signal and the C1 and C2 signals formed as described above are independently and highly efficiently coded by differential pulse coding etc. as in the conventional method. The manner of sampling in the process of conversion is shown in FIG. 3 for the above-mentioned low-mid range luminance signal Y.

The sampling frequency f of the luminance signal Y in the conventional method.

As described above, for the upper limit frequency fb of the luminance signal Y, as shown in FIG. 3a, fs = 2fb - α. Since only the frequency components are separated, the upper limit frequency fL is a value lower than the above-mentioned upper limit frequency fb.
As shown in Figure 3, the sampling frequency fs, which is set slightly lower than the Nyquist frequency which is twice
L, which is a value lower than the sampling frequency in the conventional method.

Another aspect of sampling the luminance signal Y in the encoding system of the present invention is shown in FIG. 3C.

That is, the upper limit of the low-mid luminance signal band and the lower limit of the fundamental wave component band of the sampled signal for this original signal are defined as the bandwidth f.

Frequency multiplexing is performed by frequency interpolation as shown in Figure 3d, and the F localized wave number of the fundamental wave component of the sampled signal is
Then, the original signal band and the sampled signal band are superimposed as shown in FIG. 3C, and the sampling frequency fs is set to f8-2fI, -f.

, and the sampling frequency is set to a value even lower than fs≈2fL shown in FIG.

Therefore, the frequency band that is visually important as a horizontal spatial frequency characteristic in standard TV signals in Japan is about 1.5 MHz, so the upper limit frequency fL of the original signal mentioned above is 30 MH7, and the fundamental wave component of the sampling signal is When the lower limit frequency f'L is set to 1.5 MHz, the band f0 in which these signals are superimposed is fo-=fLf'I, -1. ．． 5 MIIz, and the sampling frequency f can be set to a low value of about 4.5 MHz.

Note that in order to superimpose the frequency bands as described above, the sampling frequency f is set to f, = 4 - fh (2n + 1), as in the case of frequency interpolation between the carrier color signal and the luminance signal. do.

As described above, the sampling frequency for the luminance signal Y is significantly reduced compared to the conventional method, and for example, the sampling frequency for the luminance signal Y is reduced compared to the conventional method.
It can be selected to be around 6 MHz or 4.5 MHz.

Furthermore, regarding the sampling frequency for the color signals C1 and C2 that include high-frequency luminance signal components, even if the quantization step of the signal level for the high-frequency components of the luminance signal is reduced to the same degree as that for the color signal, the image quality will still be poor. Therefore, the upper limit of the color signal band of 1.5 MHz can be set to about 3 MHz or less.

In this way, the sampling frequency when encoding a composite color preview signal by YC separation can be significantly reduced compared to the conventional method, and the bit rate during pulse encoding can be reduced accordingly. This, combined with the bit rate reduction by differential pulse encoding, allows extremely highly efficient encoding.

Furthermore, in the encoding system of the present invention, as mentioned above, since YC separation is performed extremely simply using only a low-pass or high-pass filter, for example, the configuration of the encoding circuit and the configuration of the decoding circuit on the receiving side can also be changed. , it can be much simpler than the conventional method.

FIGS. 4a and 4l) respectively show configuration examples of a high-efficiency encoding circuit and its decoding circuit that perform differential pulse encoding of a composite color television signal according to the method of the present invention.

That is, the Y signal and the C, C2- signals, which are separated as described above and have the spectra shown in FIGS. lead

The sampling circuits 4, 5, and 6 are supplied with sampling pulse signals of frequencies set as shown in FIGS. Conduct sampling.

That is, for the standard color preview signal in Japan, the sampling frequency fs(y) for the luminance signal Y is
≈4.5 MHz, and the sampling frequency fs(C) for the color signals C1 and C2 is set to approximately 1.0 to 1.2 MHz.

Note that the sampling frequency fs(C
) = 1.0 to 1.2 MHz because the upper limit frequency fL of the low-mid range signal component of the luminance signal Y is set to 3.0 MHz, so that the signal frequency band fc of the common part of the luminance signal and chrominance signal is ≈0. This is because the frequency was .6MHz.

Each signal sampled as described above is sent to the PCM circuits 9 and 1.
0.11 plus a quantization step passed through each signal level to perform standard pulse encoding on each signal independently, and furthermore, a differential pulse encoding (DPCM) circuit 12,1
3.14, the bit rate is reduced by encoding only the difference between successive quantized signals in the standard PCM signal, and each of these DPCM output signals is time-division multiplexed in the multiplexing circuit 15, and Transmitted as an efficiency-encoded color television signal.

In the decoding circuit shown in FIG.
8, each DPCM signal for the luminance signal Y and color signal C1°C2 obtained by separating the time-division multiplexed component signals is applied to the DPCM decoding circuits 19, 20, and 21 and decoded, and their output signals are Y, C1, and C2 are led to a color encoder 22, and a color subcarrier signal fsc is added from an input terminal 23 to restore a composite color television signal obtained by frequency multiplexing a luminance signal and a carrier color signal, and is taken out from an output terminal A24. figure.

The Y signal component obtained from the DPCM decoding circuit 19 is the decoded output of the signal sampled at the sampling frequency set as shown in FIG. 3C, so the original signal and the fundamental wave component of the sampled signal are different. The aliasing distortion component caused by frequency interpolation in the superimposed f'L to fL band is processed using a comb-shaped filter constructed in the same manner as described above using a delay line having a delay amount of 1/fh for one horizontal scanning period. Then, only the low-mid-range luminance signal Y component having the frequency band of the upper limit frequency fL is restored.

Further, from the DPCM decoding circuits 20 and 21, color signals C as shown in FIG.
Since C1.I and C2 are obtained as restored outputs, these decoded signals are led to the color encoder 22 together with the above-mentioned low and mid-range luminance signal Y component, and C1. If a color-specific carrier wave signal is added while maintaining the same phase relationship as the C2 signal, and orthogonal modulation is performed using the C1°C2 signal of the decoded output, the output of the color encoder 22 becomes a composite color preview signal as shown in FIG. 2a. Can be restored.

FIG. 5 shows a diagram of FIG.
Similarly, an example of a color signal component spectrum including a high-band luminance signal component spectrum is shown.

In the PAL color television signal, of the two color difference signals B-Y and R-Y, for example, the polarity of the R-Y signal component is alternately inverted in the scanning line sequence. When the RY color difference signal component is demodulated by synchronous detection and the polarity of the demodulated output is alternately inverted in the scanning line sequence, a signal having a spectral distribution as shown in FIG. 5a is obtained.

The dotted line in the figure is the color difference signal spectrum, and the solid line is the luminance signal high frequency component is divided into two orthogonal components, and the frequency is divided into the respective spectral strings of the B-Y color difference signal component and the R-Y color difference signal component. The interpolated spectrum is shown.

In the encoding method of the present invention, as shown in FIG. 2d or FIG. 5a, the color signal component spectrum is sampled as it is, with the frequency interpolation of the spectral sequence of the high-frequency component of the luminance signal. In the process of transmitting high-efficiency encoding using differential PCM, etc., and receiving and decoding it, NTS
When restoring a C system or PAL system color television signal, a wideband brightness signal is restored by adding the low and middle range brightness signal components that have been encoded and transmitted separately.

Figure 5 shows that on the transmission side of PA and L system color television signals, the polarity of the RY color difference signal demodulation output during YC separation is not alternately inverted in the scanning line sequence as in Figure 5a. In addition, since the polarity of the R-Y color difference signal is alternately (+) and (-) in the scanning line sequence, the distribution of the color difference signal spectrum is a solid line between the B-Y color difference signal and the R-Y color difference signal. 1/2 for the luminance signal high-frequency component spectrum sequence shown by
This shows a state in which there is a deviation of h.

Even in such a case, the color difference signal component with the spectral distribution as shown is highly efficiently encoded and processed in the same manner as described above to restore the original signal.

As is clear from the above explanation, YC according to the method of the present invention
Separate high-efficiency coding transmission provides the following remarkable effects.

(1) The luminance signal separates only the low-mid frequency components that do not share the same frequency band as the carrier color signal, and demodulates the carrier color signal into two color signals while still including the luminance signal high-frequency components. In order to code the signal high-frequency components as two low-frequency signals of orthogonal phase with frequency interpolation into the color signal,
Signal processing processes such as YC separation and aliasing distortion removal become extremely simple.

(2) Since the sampling frequency for encoding can be significantly lowered, the encoding bit rate can be significantly reduced, the necessary transmission frequency band can be narrowed, and extremely high-efficiency encoding can be achieved. I can do it.

(3) High-performance, high-efficiency encoding with minimal image quality deterioration by performing extreme frequency band separation even for color television signals for which YC separation is inherently difficult, such as SECAM color television signals. Transmission can be performed easily.

[Brief explanation of the drawing]

Figures 1 a to d are spectral diagrams showing aspects of luminance signal/carrier chrominance signal separation in the conventional method, and Figures 2 a to d are spectral diagrams showing aspects of luminance signal/carrier chrominance signal separation in the method of the present invention. , Fig. 3 a and b to d are spectral diagrams showing the relationship between sampling frequency and signal spectrum in the conventional method and the present invention method, respectively, and Fig. 4 a and b show signals encoded and decoded by the present invention method. FIGS. 5a and 5b are block diagrams showing the processing steps, respectively, and spectral diagrams showing other aspects of color television signal encoding according to the method of the present invention. 1.2, 3... Separation signal input terminal, 4, 5, 6... Sampling circuit, 7, 8... Sampling signal input terminal, 9, 1
0,11...PCM circuit, 12゜13.14...DPC
M circuit, 15...Multiple circuit, 16...Encoded signal output terminal, 17...Encoded signal input terminal, 18...Separation circuit,
19, 20° 21...DPCM decoding circuit, 22...Color encoder, 23...Color subcarrier input end, 24...
Composite color television signal output terminal.

Claims (1)

[Claims] When encoding a color television signal configured by frequency multiplexing one luminance signal and a carrier chrominance signal, a low-mid-range signal component of a portion of the luminance signal that does not share a frequency band with the carrier chrominance signal; Two signal components of the carrier color signal, each including a high frequency signal component of a portion sharing a frequency band with the carrier color signal of the luminance signal, are independently encoded. A high-efficiency encoding method for color preview signals.