JPS5923148B2 - Encoder/decoder for narrowband color television signals - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an encoding/decoding apparatus for composite color television signals such as NTSC and PAL systems.

In a color television signal such as the NTSC system, a carrier color signal (hereinafter referred to as C signal) in which the subcarrier is amplitude-modulated with a color signal is added to the high frequency range of a luminance signal (hereinafter referred to as Y signal), and the spectrum of the Y signal and the spectrum of the C signal. A frequency multiplexed signal format is used so that the signals have an interleaf relationship.

In order to efficiently encode such a composite color television signal, the composite color television signal is converted into baseband signals (for example, in the NTSC system, luminance signal Y, chrominance signal, and Q signal).
), and then performs band compression encoding suitable for each individual signal, which is the so-called separate encoding method, and high-order consonant DPCM, which is a method that directly encodes the baseband signal without demodulating it.
There are direct encoding methods such as orthogonal transform encoding such as Hadamard transform. In the direct encoding method, the composite color television signal is directly encoded, so when sampling the signal, the sampling frequency fs needs to be more than twice the signal band. .5 to 3 times the value (N
In the case of TSC color television signals, it was selected as 9 to 11 MHz). For this reason, the transmission bit rate required to transmit the encoded signal tends to be large because the sampling frequency is high. An object of the present invention is to sample a composite color television signal at a sampling frequency fs lower than twice the subcarrier frequency fsc, that is, lower than the Nyquist sampling frequency (twice the frequency of the signal band). An object of the present invention is to provide an encoding/decoding device that realizes compression encoding. The encoding and decoding apparatus of the present invention has a horizontal scanning frequency such that the sampling frequency fs is lower than twice the subcarrier frequency fsc, and the spectrum of the carrier color signal is folded between the luminance signal vectors. An encoding device controlled to directly sample a composite color television signal at a frequency that is an integral multiple of FH, and a composite color television that performs interpolation from the signal sampled by the encoding device to remove aliasing distortion. and a decoding device that has the function of reproducing signals.

According to the invention, the subcarrier frequency F, is lower than the Nyquist frequency.

It becomes possible to directly sample a composite color television signal at a sampling frequency F, which is lower than the frequency 2f80, which is twice the frequency 2f80, and an efficient encoding/decoding device can be obtained. The case where the signal format is the NTSC system will be explained below using the drawings as an example.

In frequency multiplexed color television signal systems such as NTSC, in order to reduce image quality deterioration due to mutual interference between the Y signal and C signal, there is a The subcarrier frequency F8O has a frequency interleaf relationship with the horizontal scanning frequency FH so that the C signal spectrum is included.

That is, in the case of the NTSC system, F. - Selected as V-dan/FH. Therefore, the distribution of the signal spectrum generally spreads over the entire frequency range, but in a normal video signal, the amplitude of the frequency vector in the high frequency region of the Y signal is considered to be very small. Therefore, suppose that the sampling frequency is set to F8, and sampling is performed at a sampling frequency that is lower than the value width 1j of the temple F8 and the carrier frequency Fsc, that is, F8×2f8c. Therefore, the frequency from 0 to -Ff8 is F
Pass through the vicinity of a frequency that is an integer multiple of H, and the frequency is -Ff
Band-limiting of the Y signal spectrum and the C signal spectrum is performed using a comb-type filter having a characteristic of passing near the middle between frequencies that are integral multiples of FH from s to F8.
The frequency spectrum of the input signal sampled at s
As shown in Figure 1, the frequency spectrum of the input signal is O
From 4-F to -Ffs, the Y signal spectrum is concentrated around frequencies that are integral multiples of FH, and from 4-Fs to Fs, the C signal spectrum is concentrated around frequencies that are integral multiples of FH. Make it. When an input signal (t) having such a frequency spectrum or (ω) is sampled at the sampling frequency F8, the aliasing caused by sampling will be between the aliasing of the Y signal and the spectrum of the C signal, respectively. The sampling frequency Fs is selected so that the folding of the C signal falls between the spectra of the Y signal. That is, it is made to be an integral multiple of the horizontal scanning frequency FlI as shown in the following equation. When the band-limited input signal 9(t) is sampled at such a sampling frequency F, the sampled output signal is x(n)=? (
NT) (however, the period T=1/Fs), the Z transformation X(Z) of the sampled output signal sequence x(n) is as follows.

Letting the frequency spectrum of this output signal sequence x(n) be X(EjOT), the following equation is obtained.

That is, since the sampling frequency F, in the spectral distribution of the sampled output signal x(n) is lower than the Nyquist frequency of the input signal, aliasing occurs, resulting in a frequency spectral distribution as shown in FIG. At this time, since F8=NfH is selected as shown in equation (1), the Y signal spectrum is folded between the C signal spectra, and the C signal spectrum is folded between the Y signal spectra.

Transmitting a composite color television signal x(n) sampled at a sampling frequency Fs having such a spectral distribution,
On the receiving side, the folded component is removed from the sampled signal containing the folded spectrum to reproduce a composite color television signal. In order to obtain the spectral distribution shown in Figure 1 from the signal with the spectral distribution shown in Figure 2, a comb filter with the same characteristics as used on the transmitting side, that is, a frequency of 0 to
Pass the Y signal spectrum up to +F, -Ff8~F
Up to s, aliasing components caused by sampling at F8 can be removed using a filter having a characteristic of passing the C signal spectrum. In this case, the following method is used to configure the comb filter using a digital filter and reproduce a composite color television signal. sampling frequency F
, interpolate the value of O between the signal sequences x(n) to obtain a signal sequence v(n) with a sampling frequency of 2f8, that is, a period of T/1/2f8. . That is, the interpolated received signal sequence v(n) with period T' is expressed by the following equation.

The z-transform of this signal v(n) is as follows.

Therefore, for a signal whose sampling period is T'=1/2f8, the Fourier transform V(EjOT') of (5) is as follows.
In other words, the received signal v(n)● with O interpolated
The frequency spectrum v(EJCt)T) of ' has the same frequency spectrum distribution as in Figure 2, and the period is 2π/T' = 4π
F, but 2π/T=2πF8.

Therefore, the signal v(n) is separated into a Y signal component and a C signal component using a comb filter, and the Y signal component passes through a low-pass filter (hereinafter referred to as L.P. filter) and becomes an aliased component. is removed and the vector distribution is the same as that of human labor.
I get a signal. The C signal component passes through a high-pass filter (hereinafter referred to as H.P. filter) to remove aliasing components and obtain a C signal having the same spectral distribution as the input.
Then, by adding the two obtained signals, we get the spectral distribution Y (composite color television signal y(r) sampled at a periodic interval of Ej(1dT'') as shown in Figure 3).
is obtained.

As is clear from the above description, the principle of the present invention is that the sampling frequency F8 is the subcarrier frequency F.

The frequency band of the input composite color television signal is set to a value smaller than twice that of the horizontal scanning frequency FH and an integral multiple of the horizontal scanning frequency FH, and the frequency band of the input composite color television signal is set from O to -Ff8 for the Y signal, and from -Ff8 to F8 for the C signal. After limiting the band, direct sampling is performed at the sampling frequency of F, and the sampled signal is sent out. On the receiving side, the distortion caused by the sampling of F8 is removed using a comb filter. Thus, it is a method for reproducing composite color television signals. Using such a method, it is possible to directly sample and transmit a band-limited composite color television signal at a frequency below the Nyquist frequency without demodulating it into a baseband signal.

Examples will be described below.

FIG. 4 is a block diagram showing the configuration of the first embodiment of the present invention.

The encoding device 1 converts an input composite color television signal into a PCM (pulse code modulation) signal having a frequency of 2f8, which is twice the sampling frequency F8. Fs is given by equation (1) which satisfies Fs<2f80.

The PCM signal sampled at 2f8 is
The signal is in the band from 0 to -h-F8, and the C signal is +F, ~
After being band-limited to a frequency band up to F, the signal is resampled to a PCM signal with a sampling frequency of F8.

Since Fs is controlled so that it satisfies equation (1), the aliasing caused by resampling to the PCM signal of frequency F, occurs between the Y signal spectrum for the C signal, and the Y
The signal can be encoded in such a way that it falls between the C signal spectrum. As is clear from the above description, the encoding device 1 can be realized by simply adding a circuit for controlling the sampling frequency to a normal A/D converter. Obtaining a sampling frequency that is synchronized with the scanning frequency of a television signal can be easily realized using already known techniques. The band-limited composite color television signal converted into a PCM signal with a sampling frequency F8 by the band-limiting circuit 2 is sent out to a transmission path.

On the receiving side, the PC of the sampling frequency Fs received from the transmission line
The M signal is supplied to the interpolation filter circuit 3.

The interpolation filter circuit 3 interpolates sample values with a value of O between each sample value from the PCM signal with a sampling frequency of F8 to obtain a PCM signal with a sampling frequency of 2 fs, and Y up to O-4-Fs.
The signal and the signals from If8 to F8 are passed. By removing the aliasing component caused by sampling at the sampling frequency F8 using a comb filter, the sampling frequency 2
A PCM signal of a band-limited composite color television signal sampled at fs is obtained. The PCM signal with a sampling frequency of 2f8 obtained by the interpolation filter circuit 3 is supplied to the D/A converter 4, where it is converted into an analog composite color television signal. FIG. 5 shows a specific example of the configuration of the band limiting circuit 2 shown in FIG. 4.

(sampling frequency is 2fsf) PCM signal is Y signal. It is supplied to the C signal separation circuit 51 and is separated from the Y signal and C using line correlation.
The Y signal is separated into the L. P. The C signal is sent to the filter circuit 52 as an H.C signal. P. Each signal is sent to a filter circuit 53.

L. P. In the filter circuit 52, the L. P. The Y signal is band-limited by a filter. H. P. Filter circuit 53
In this case, the H. P. The C signal is band-limited by a filter. L. P. The Y signal output of the filter circuit 52 and the H. P. The C signal output of the filter circuit 53 is supplied to an adder 54 and added, resulting in a band-limited 2f, sampled composite color television signal, which is supplied to a resampler 55. The resampler 55 resamples the input signal every other sampling value, thereby converting the 2f8 sampling PCM signal to the F8 sampling PC.
Convert to M signal. That is, the Y signal C signal separation circuit 51
,L. P. Filter circuit 52, H. P. Filter circuit 5
3 and adder 54 constitute a required comb filter.

FIG. 6 shows an example of a specific configuration of the Y signal/C signal separation circuit 51 shown in FIG.

In this embodiment, the separation of the Y signal and the C signal is performed using the correlation between the current sample value and the sample value one scanning line before.
The case where an H-type comb filter is used is shown.
P of sampling frequency 2f8 sent from encoding device 1
The CM signal is sent to the line memory 61, adder 62, subtracter 6
3 respectively. The line memory 61 is a circuit that provides a delay of one scanning line time and can be composed of a storage element such as a shift register. The input signal and the signal delayed by one scanning line time are added by an adder 62, and multiplied by j by a multiplier 64, thereby outputting a Y signal. Further, the subtracter 63 subtracts the input signal delayed by one scanning line time from the input signal, and the multiplier 65 multiplies the input signal by + to obtain the C signal. Multipliers 64 and 65 can be easily configured by shifting each bit of the input signal one digit downward and outputting the resultant signal. L in FIG. P. As an example of a specific configuration of the filter circuit 52, a secondary digital filter shown in FIG. 7 is used to obtain the required L. P. It is constructed by connecting several stages in cascade to obtain filter characteristics.

FIG. 7 shows an example of a specific configuration of a secondary digital filter. The transfer characteristic of this digital filter is expressed by a z function.

Constant A. i. all. a2l, Bll,. b2l is determined by the required transfer characteristics. The input signal is multiplied by AO1 by the multiplier 8 and supplied to the subtracter 802. A subtracter 802 subtracts the output of the multiplier 801 and the output of the adder 807 and supplies the output to a register 803 and an adder 811. The signal delayed by one sampling period in register 803 is supplied to multiplier 804, register 805, and multiplier 808. Shift register 805 delays the input signal by one sampling period and outputs a signal, which output signal is supplied to multipliers 806 and 809. The signal multiplied by Bli in the multiplier 804 and B2l in the multiplier 806
The multiplied signals are added by adder 807 and subtracter 8
Sent to 02. Also, the signal multiplied by the multiplier 808a11 and the signal multiplied by A2i by the multiplier 809 & the adder 81
It is added with 0 and sent to the adder 811. Adder 811 adds the output of subtracter 802 and the output of adder 810 and outputs the result. That is, the input signal input to the multiplier 801 is filtered and appears at the output of the adder 810.

Therefore, by cascading n-stage secondary digital filters, L. P. If the filter circuit 52 is configured, this L. P. The transfer characteristic HL(Z) of the filter is,
constant n, {AOiL{AllL{A2IL{BllL{
B2l} is the required L. P. Each is set to an appropriate value to satisfy the filter characteristics.

H in FIG. P. 1 of the specific configuration of the filter circuit 53
For example, L. P. Similar to the filter circuit 53, a secondary digital filter shown in FIG. 7 is used to obtain the required H. P. It is constructed by connecting several stages in cascade to obtain filter characteristics.

Therefore, by cascading n-stage secondary digital filters, H. P. If the filter circuit 53 is configured, this H. P. Filter transfer characteristics HlI (2S) 4 insects 7
) is the same as that shown by the formula ゛. In other words, it becomes.

Here constants n, {AOi}, {All}, {A2l}
, {Bll}, {B2l} are the required H. P. Each is set to an appropriate value to satisfy the filter characteristics. FIG. 8 shows a specific example of the configuration of the interpolation filter circuit 3 shown in FIG. 4.

The received PCM signal with sampling frequency F8 is sent to one input of switch 1 which switches at sampling frequency 2f8.

The other input is a PCM signal with value 0. 2fs
By switching the switch 101 at the frequency, a 2 fs sampled received signal with a value of 0 interpolated is obtained at the output of the switch 101.

This signal is sent to the Y/C signal separation circuit 102, and is separated into a Y signal containing an aliased component and a C signal using a comb filter that utilizes correlation. The separated Y signal is the L. P.
It is sent to the filter circuit 103 and the frequency is O-1f-F8.
+F8 caused by passing only the signal up to
The Y signal component up to F8 is removed and the 2fs sampling Y consists of the same spectrum as the band-limited original signal Y signal.
Get a signal.

Also separated C signal & turtle H. P. Filter circuit 10
4 and whose frequencies are between +F8 and F8, and removes the C signal component up to O-+F, which is generated by folding, and extracts the signal from the same spectrum as the band-limited original C signal. In other words, whether the filter is configured with an analog filter or a digital filter, the frequency response characteristics of the filter necessary for the band limiting circuit and the interpolation filter circuit are such that the Y signal is passed through up to 0-1 Ff, and the Y signal is passed through 4-F8 to Ff.
, up to the C signal is sufficient as long as it satisfies this frequency characteristic, its configuration method is not particularly limited. Also, the Y signal C signal separation circuit of the embodiment shown in FIG. 6 is a so-called 1H Although the configuration is of a type, it is not limited to this configuration method, and for example, a so-called 2H type configuration using front and rear scanning lines is also possible.

Also, L. shown in FIGS. 5 and 8. P. filter circuits 52 and 103, H. P. Although the filter circuits 53 and 104 are constructed by cascading several stages of secondary digital filters in this embodiment, they are not limited to this construction method, and may be constructed using other methods, such as non-recursive filters. It is also possible to configure it with a type digital filter.

Furthermore, the interpolation filter circuit shown in FIG. 8 is not limited to the configuration method of this embodiment, but may be constructed using other methods, for example, the switch 101 can be configured as a Y signal C signal separation circuit 102 and an L signal separation circuit 102. P. Between the filter circuits 103 and the Y signal C signal separation circuit 10
2 and H. P. It is also possible to configure them so that they are respectively placed between the filter circuits 104. FIG. 9 shows the second embodiment of the present invention.
FIG. 2 is a block diagram showing the configuration of an embodiment of the present invention.

This embodiment is a combination of the first embodiment and a band compression encoding method, and the encoding device 1, band limiting circuit 2, interpolation filter circuit 3, and D/A converter 4 are the encoding device shown in FIG. 1, band limit circuit 2, interpolation filter circuit 3, and D/
It has the same function as the A converter 4.

The band compression encoding circuit 5 is a circuit that performs band compression encoding such as DPCM (differential PCM) Hadamard transform on the signal converted into a PCM signal by the encoding device 1.

The signal encoded by the band compression encoding circuit 5 is sent out to a transmission path. The band compression decoding circuit 6 is a circuit that decodes the band compression encoded signal sent from the transmission path, and the decoded signal is processed by the interpolation filter circuit 3 to remove aliasing components and then sent to the D/A converter 4. converted to an analog signal. FIG. 10 shows an example of a specific configuration of the band compression encoding circuit 5 and the band compression decoding circuit 6 shown in FIG. In this embodiment, a case will be described in which high-order DPCM using a high-order consonant function adapted to a color television signal including subcarriers is adopted as a band compression encoding method.

In high-order DPCM, the prediction function PZ) may be any function that efficiently predicts both the luminance signal component and the carrier color signal component. The prediction function P(Z) when the sampling frequency ~3 wave number F8=-FsO is selected as P(ro=αZml+β
Z-3-αβZ8(9) can be considered as an example.

However, α and β are constants, for example, α=0.5, β-1-2-
N (N is a positive integer). The high-order DPCM encoder includes a subtracter 121, a quantizer 122, an adder 123, a prediction filter 124, and a code converter 125. The high-order DPCM decoder includes a code inverter 126, an adder 1
27 and a prediction filter 128.

The input signal to the higher order DPCM encoder is supplied to subtractor 121.

In the subtracter 121, the prediction filter 12 is extracted from the input signal.
The prediction error is calculated by subtracting the prediction signal that is the output of step 4.
The prediction error that is the output of the subtracter 121 is sent to the quantizer 122, and the quantizer 122 quantizes it with a predetermined quantization characteristic, and the quantized signal is sent to the adder 123 and code converter 125. . In the adder 123, the signal quantized by the quantizer 122 and the output signal of the prediction filter are added to calculate a locally decoded signal, and the signal quantized by the prediction filter 1 is added.
Sent to 24. Prediction filter 124H has a transfer characteristic shown in equation (9), calculates a predicted value of the next sampling time from the locally decoded signal, and outputs it as a predicted signal. Code converter 12
5, a quantizer 122 converts the quantized signals into codes corresponding to the Q minimization level, and sends the codes to the transmission path. In the high-order DPCM decoder, the code sent from the transmission path is converted by the code inverse conversion circuit 126 into a signal level corresponding to the quantization level quantized by the quantizer 122, and the adder 1
Send to 27.

The adder 127 adds the signal sent from the code inverse conversion circuit 126 and the prediction signal sent from the prediction filter 128 to obtain a decoded signal. The decoded signal is predicted by prediction filter 1.
28, which calculates and outputs a predicted signal for the next sampling time. Note that the prediction filters 124 and 128 have the same transfer characteristics.

FIG. 11 shows the prediction filters 124 and 128 of FIG.
The first part of the concrete configuration is shown below.

In this embodiment, the function P(4) given by the formula 9) is expressed by a non-recursive digital filter.13
1, 132, 133 and 134 are clock frequencies F,
135 is a subtracter, 136 is an adder, and 137 and 138 are multipliers with magnifications α (=0.5) and β (=1-2-N).

Multiplier 137 simply shifts the input signal one digit lower. The multiplier 138 can be easily constructed without subtracting from the input signal a signal obtained by shifting the input signal N digits lower. The operation of each component is a well-known digital filter configuration and can be clearly understood from the diagram, so a description thereof will be omitted.

As is clear from the above explanation, by combining the band compression encoding method with the encoding/decoding device that can encode at a sampling frequency lower than the Nyquist frequency described in the first embodiment, a significant band compression encoding can be achieved. A possible encoding/decoding device can be realized.

Moreover, since the composite color television signal is not demodulated into a baseband signal and all post-encoding processing can be performed using digital logic circuits, highly accurate calculations can be performed.

Furthermore, the band compression encoding method in the second embodiment is also a high-order D
The present invention is not limited to PCM, and may be an orthogonal transform coding method such as Hadamard transform or an interframe coding method.

[Brief explanation of the drawing]

Figure 1 shows the frequency spectral distribution of the band-limited input composite color television signal, Figure 2 shows the frequency spectral distribution of the input signal sampled at F, and Figure 3 shows the frequency spectral distribution of the input signal with aliasing removed. 2f is a diagram showing the frequency spectrum distribution of the decoded signal of sampling. FIG. 4 is a block diagram showing the configuration of the first embodiment of the present invention, FIG. 5 is a block diagram showing an example of a specific configuration of the band limiting circuit, and FIG. 6 is a specific block diagram of the Y signal C signal separation circuit. FIG. 7 is a block diagram showing an example of a specific configuration of a secondary digital filter circuit.
FIG. 8 is a block diagram showing →u of the specific configuration of the interpolation filter circuit, FIG. 9 is a block diagram showing the configuration of the second embodiment of the present invention, and FIG. 10 is a high-order DPCM encoder decoder. FIG. 11 is a block diagram showing the specific structure of the prediction filter. 1... Encoding device, 2... Band limit circuit, 3...
Interpolation filter circuit, 4...D/A converter, 51
...Y signal C signal separation circuit, 52...
L. P. Filter circuit, 53...H. P. Filter circuit, 54... Adder, 55...
Resampler, 61...Line memory, 62...
... Adder, 63 ... Subtractor, 64, 65
. . . Multiplier, 5 . . . Bandwidth compression encoding circuit, 6 . . . Bandwidth compression decoding circuit.

Claims (1)

[Claims] 1. In an encoding/decoding device for a color television signal that is frequency-multiplexed so that a color signal is frequency-interleaved with a luminance signal using a subcarrier with a frequency f_S_C, the sampling frequency f_S is lower than 2f_S_C and 1. an encoding device comprising means for controlling a sampling frequency f_S so that /2f_S is approximately the lower limit or less of a carrier color signal component and a value approximately an integral multiple of a horizontal scanning frequency f_H; The carrier color signal spectrum folded between the luminance signal spectrum centered at an integer multiple of f_H from 0 to 1/2f_S for the sampled signal and the carrier color signal spectrum centered at an integer multiple of f_H from 1/2f_S to f_S The aliasing distortion is removed by sufficiently attenuating the luminance signal spectrum aliased at the point where the frequency becomes 0.
to 1/2f_S and a carrier color signal component whose frequency is from 1/2f_S to f_S. An encoding/decoding device for a narrowband color television signal, characterized in that the encoding is performed at a sampling frequency lower than the Nyquist frequency.