JPH036185A - Video signal digital transmission device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] Industrial Application Field The present invention provides a video signal that can transmit higher quality video at the same transmission rate as before when digitizing and transmitting a video signal. This invention relates to digital transmission equipment.

Conventional technology Conventionally, an NTSC color video signal is transmitted at a subcarrier frequency f.
A digital VTR for broadcasting has been announced that samples at four times the SC and quantizes it to 8 bits for recording and playback. In this case, the recording rate of data recorded on the tape is extremely high, 115 Mbps.

FIG. 8 shows a block diagram of a recording/reproducing system of a digital VTR. In FIG. 8, 17 is an analog video signal input terminal, 18 is a pre-filter, 19 is an A/D converter, 20 is a recording signal processing circuit, 21 is a recording head, 22 is a magnetic tape, 23 is a reproduction head, 24 is a reproduction signal processing circuit, 25
26 is a D/A converter, 26 is a post-filter, and 27 is an analog video signal output terminal.

The NTSC color video signal input to the analog video input terminal 17 is first subjected to a prefilter where frequency components higher than the Nyquist frequency are removed. Then, in the A/D converter 19, the signal is sampled at a frequency four times the subcarrier frequency fsc, quantized to 8 bits, and sent to the next recording signal processing circuit 20. This recording signal processing circuit 20 performs signal processing necessary during recording, such as error correction encoding and channel encoding. The recorded data is transferred to the magnetic tape 22 via the recording head 21.
recorded in During reproduction, the signal reproduced from the magnetic tape 22 is input to the reproduction signal processing circuit 24 via the reproduction head 23, where it is subjected to processing such as channel code decoding and error correction code decoding, and then sent to the D/A converter 25. The signal is returned to the original analog video signal, and finally, after the aliasing component due to sampling is removed in the post-filter 26,
The analog video signal is output from the analog video signal output terminal 27.

Problems to be Solved by the Invention As mentioned above, in broadcasting digital VTRs, video signals are generally quantized at 8 bits, and the signal converted to a -degree digital signal is recorded and played back repeatedly. However, image quality does not deteriorate unless an uncorrectable error occurs. Therefore, in a system that passes through A/D conversion and D/A conversion only once, sufficient image quality should be obtained with 8-bit quantization. However, in the current environment in which digital VTRs are used, not all surrounding broadcasting equipment is necessarily digitized, and it is not uncommon for programs to go through multiple A/D conversions and D/A conversions in the process of production. And the video signal is A/D converted and D/A
Each time the conversion is repeated, the amplitude characteristics of the pre-filter and post-filter, or imperfections in the phase characteristics and the A/D
Image quality gradually deteriorates due to accumulation of quantization noise during conversion. That is, in a system in which A/D conversion and D/A conversion are repeated many times, 8-bit quantization is no longer sufficient, and 9-bit or 10-bit quantization becomes necessary.

However, when a video signal is quantized at 9 bits, the recording data rate is 9 bits compared to when it is quantized at 8 bits.
It becomes 78 times. This means that assuming a videotape cassette with the same recording capacity, the recording time will decrease to 879, and conversely, assuming the same recording time, the length of the tape and the running cost of the tape will decrease. 9
/ This means that it will increase by eight times. Therefore, one possible way to solve this problem is to perform quantization using 9 bits, but reduce the amount of data to 8/9 during recording.

Traditionally, DPC has been used as a method to reduce the data rate.
High-efficiency coding such as M-coding or orthogonal transform coding is well known, but these high-efficiency codings have several problems. The first is the problem of error propagation, and the second is the problem that the manner of image quality deterioration varies depending on the picture pattern of the input video signal. In particular, the second problem is
Despite recording at the recording data rate of 8-bit quantization, the image quality of 8-bit quantization may not be obtained depending on the picture, such as a black and white striped pattern, where the voltage level of the input video signal changes drastically. It means. This is a fatal drawback for digital VTRs for broadcasting, where image quality is of the utmost importance.

The present invention solves these conventional problems,
Even though a 9-bit quantized signal is recorded at a recording data rate of 8-bit quantization, it is possible to obtain image quality equivalent to 9-bit quantization in a considerable portion of the video, and at least 8-bit quantization is possible. The object of the present invention is to provide a video signal digital transmission device that can ensure the image quality of bit quantization.

Means for Solving the Problems In order to solve the above-mentioned problems, the video signal digital transmission apparatus of the present invention includes a sampling means for sampling a video signal, and a sampling means for sampling a video signal, and N bits (N is a positive integer) of the sampled video signal. a first quantization means for quantizing with;
A thinning means for thinning out the samples obtained by the sampling means at a ratio of one sample to M samples (M is a positive integer), and predicting the samples thinned out by the thinning means using surrounding unthinned samples. a prediction means; a subtraction means for calculating a difference value between the sample value of the sample thinned out by the thinning means and the predicted value obtained by the prediction means; M-1)・N) bit [K
is a positive integer that satisfies N>K>(M-1)・N/M); a dynamic range determination means for determining the dynamic range of the difference value; and a second quantization means for determining the dynamic range of the difference value; a first transmission means that transmits the output of the first quantization means for the samples that are not thinned out, and transmits the output of the second quantization means for the samples that are thinned; a second transmission means for transmitting both the sample value of the unsubtracted sample and the difference value between the sample to be thinned out and its predicted value in bits, and a dynamic range of the difference value determined by the dynamic range determination means. It has a switching means for switching to the first transmission means when the range is below a predetermined range, and switching to the second transmission means for other cases.

According to the present invention, with the above-described configuration, the samples of the quantized video signal are thinned out at a predetermined rate, and the quantization values remain unchanged for the samples that are not thinned out, while the quantization values are kept unchanged for the samples that are thinned out. Then, the difference value between the quantized value of that sample and the predicted value from surrounding samples is transmitted.

EXAMPLES Hereinafter, examples of the present invention will be described with reference to the drawings.

FIG. 1 shows a first embodiment of the invention. In FIG. 1, 1 is a video signal input terminal, 2 is a prefilter, 3 is a sampling circuit, 4 is a sampling pulse generator, 5 is a 1/M frequency divider, 6 is a first quantizer, 7.8 is a predictor, 9.10 is a subtractor, 11 is a second quantizer, 1
2 is a dynamic range (hereinafter referred to as D range) determination circuit, 13, 14, and 15 are multiplexers, and 16 is a D range determination information output terminal.

The present invention will be explained below with reference to FIG. First, the NTSC color video signal input to the video signal input terminal 1 is
After frequency components higher than the Nyquist frequency are removed by the prefilter 2, sampling is performed in the sampling circuit 3 using a sampling pulse from a sampling pulse generator 4. The frequency of this sampling pulse is 4 fs synchronized with the burst signal of the input video signal.
It is c.

Next, the sampled video signal is sent to a first quantizer 6.
It is converted into 9-bit digital data. This 8-bit data is input to input A of multiplexer 13 and predictor 8.
．． Each is input to a subtracter 10. The prediction i#1 unit 8 calculates the predicted value of each sample (quantized value) of the input data using highly correlated samples existing in the vicinity of that sample. Then, a difference value between the predicted value obtained by this predictor 8 and the predicted 9-bit quantized value is obtained in a subtracter 10, and then quantized into S bits in a second quantizer 11. and is input to input A of the multiplexer 14. Here, when samples are thinned out from 9-bit digital data at a rate of 1 sample every M samples, S becomes (9-M). On the other hand, 9
The upper 8 bits of the bit quantization value are sent to the multiplexer 13.
Input B.

Predictor 7. Each is input to a subtracter 9. An 8-bit difference value is obtained from the subtracter 9 and is input to input B of the multiplexer 14. In addition, the D range determination circuit 1
2, if the output of the subtracter 10 is SDR, -2(S-
It is determined whether S<SDR<2 (S-th-1) is satisfied. If this is satisfied, multiplexer 1
In 3.14, human power A is selected. The output of multiplexer 13 is input to input A of multiplexer 15, and the output of multiplexer 14 is input to multiplexer 15.
is input to input B of 1/M in multiplexer 15
According to the control of the frequency divider 5, input B is selected for one sample among the M samples, and input A is selected for the remaining M-1 samples. That is, one sample out of M samples is thinned out, a difference value is transmitted for this thinned sample, and the sample value is transmitted as is for CM-1) samples that are not thinned out.

Furthermore, there are two transmission modes depending on the D range of the difference value determined by the subtracter 10. That is, when the D range is S bits or less, input A is selected by multiplexer 13.14, and the difference value of S bits is transmitted for the samples to be thinned out, and the (M-1) samples that are not thinned out are transmitted. For the samples of 9 obtained from the first quantizer 6
A bit sample value is transmitted. In this case mode 1
It is called. On the other hand, if the D range exceeds S bits, multiplexer 13.14 selects input B, and 8-bit sample values are transmitted for (M-1) samples that are not thinned out. For each sample, a difference value calculated using the 8-bit sample value is transmitted in 8 bits.

This case will be referred to as mode 2. This means that in the area where the prediction error is small in the input video signal, 9
This shows that transmission with bit accuracy is possible, and even in areas with large prediction errors, transmission with 8-bit accuracy is possible at worst.

Here, in order to correctly restore the transmitted data, it is necessary to distinguish which of the above two modes has been transmitted for each M sample. To do this, it is necessary to add a 1-bit index representing each mode to each M sample.

Now, if M=5, the increase in transmission rate due to this index addition is 0.2 bits per sample. In order to reduce this increase in transmission rate, it is sufficient to add an index not in units of M samples but in units of PM samples. In this case, the increase will be 1/(P-M) bits, and if P = 2, the increase will be 0.1 bit per sample (in this case, the D range of P difference values will all be S (It will be transmitted in mode 1 only if it is less than a bit).

In this way, if the values of P and M are chosen appropriately, the increase in transmission rate due to index addition is very small, about 4
This means that 9-bit image quality can be transmitted, albeit partially, at a transmission rate of fsc・8 (bps) (this P
- a block of M samples will be called a mode block). Furthermore, as will be explained later, by configuring the predictor appropriately, it is possible to transmit frequency regions containing large amounts of energy in the input video signal or regions where deterioration is easily noticeable visually, using 9 bits. By doing so, you can get a greater effect. Furthermore, by transmitting one sample out of M samples as a difference value, error propagation in this embodiment is at most one sample, which is much superior to conventional high-efficiency encoding.

Hereinafter, this embodiment will be described in more detail with reference to FIG. 3, focusing on the configuration of the predictor. Here M=5. P=2. S=
Set it to 4. FIG. 3(a) shows the configuration of the mode block. In this figure, 2°9+1 is the line number, and 1) (Ln) is the sample value of the nth sample of the 9th line. Furthermore, black circles indicate samples that are thinned out, and white circles indicate samples that are not thinned out. The range surrounded by the broken line represents the mode block.

If the predicted value for sample 1) ($1.n) to be thinned out here is Q(ff, n), this value is calculated from the values of the four samples before and after this sample using the following equation.

q(ff, n): [p(ff, n-1)+p(
ff , n+1))-(p(31, n-2)+p(R
tn+2))/2 sea (1) Sample p
i+1. The predicted value Q(!Q+1°n) of n) is obtained in the same manner. FIG. 2 shows a block diagram of this predictor.

Here 28. 29. 30.degree. 31 is a one-clock delay device, 32.33 is a 1/2 coefficient multiplier, and 34, 35.38 is an adder. Here, the frequency response of this predictor is H(Φ)
Then, H(Φ) is expressed by the following equation.

It(Φ)=2・C05(Φ)-COS(2Φ)
...(2) However, Φ=2xf/4fsc.

The error function E(Φ) is defined by the following equation.

E (Φ): 1-11 (Φ) ・
(3) The error function of the predictor expressed by equation (1) is shown in FIG. 3(b). In this figure f. is the Nyquist frequency, and when the sampling frequency is 4 fsc, it is fN-"2fsc. By the way, since the error function corresponds to the difference value between the sample value of the sample to be thinned out and the predicted value, this figure Lower frequency component and subcarrier frequency fs
It can be seen that the difference value becomes smaller in frequency components near c, and the probability that the difference value can be transmitted using 4 bits increases. This indicates that in the case of an NTSC color video signal, a considerable portion of the video signal can be transmitted with 9-bit image quality, considering that energy is concentrated in the above frequency components. Moreover, these frequency regions correspond to the low-frequency components of the luminance signal and color signal, have the highest visual sensitivity, and are the regions where image quality deterioration is easily noticeable, so the substantial effect is great.

Next, a second embodiment of the predictor will be described using FIG. This predictor is the simplest one (block diagram omitted), and as shown in FIG. 4(a), the predicted value of sample p(Ln) is the sample value p(L n−
4) is used. Here M=2. P=8.3
=7. The error function in this case is also shown in Figure 4(b).
As can be seen, as in the case of the first embodiment, the error is smallest in the low frequency components and subcarrier frequency components where the energy of the video signal is concentrated. However, compared to FIG. 3(b), the error is relatively large. However, in the case of the first embodiment, the number of quantization bits of the difference value was 4 bits, but in the case of the present embodiment, it was 7 bits, which is 8 bits compared to the first embodiment. Since twice as many difference values can be transmitted, a considerable portion can also be transmitted with 9-bit accuracy in this case as well.

Next, a third embodiment of the predictor will be described using FIGS. 5 and 8. Here, M=5. P=2°S=4, which is the same as the predictor in the first embodiment, but the configuration of the predictor is different. That is, in the first embodiment, prediction was performed using four samples before and after the sample to be thinned out, but in this embodiment, prediction is performed using eight samples before and after the sample to be thinned out (as shown in FIG. 6 (a)). Therefore, more accurate prediction can be made. Figure 5 shows a block diagram of this predictor. In this figure, 37 to 44 are 1-clock delay units;
45 to 51 are adders, and 52 to 55 are coefficient units. That is, the predicted value q(Ln) of sample p(Ln) is q(12, n): 14p(ff, n-1)+p(1
1, n+1))+ to 2・(p(ff, n-2Dp(I
I, n+2))+to 3・(p(ff tn-3)+p
(ff , n+3)) + to 4-(p(!Q , n-4)
+p(2, n+4)) ...(4). However, the ratio is 4:0.5 for 1+, 2+, and 3+.

The frequency response H(Φ) of the predictor of this embodiment is expressed by the following equation.

H(Φ)=2・K1−C05(Φ)+2・K2・C05
(2Φ)+2・K3・cos(3Φ)+2・K4・C0
5(4Φ)...(5) As shown in FIG. 6(b), the prediction error of the error function E(Φ) in this case is much smaller than that of the first embodiment. Furthermore, since the number of quantized bits of the difference value is the same at 4 bits, the area that can be transmitted with 9-bit precision increases dramatically.

Now, when comparing this example and the first example, the biggest difference is that in the case of the first example, the samples used for prediction were limited to the samples within the mode block, that is, the samples within the broken line. , in the case of this embodiment, this means that it straddles the mode blocks on both sides. For this reason, in this embodiment, it is necessary to calculate the predicted value using the upper 8 bits of the quantized value of each sample. This is because if even one of the blocks on both sides is in mode 2, only 8-bit sample values can be used during decoding. However, even in this case, the effective range of the difference value becomes somewhat larger due to calculation errors, but if this can be transmitted in 4 bits, 9-bit accuracy can be ensured. A block diagram showing a second embodiment of the present invention using this predictor is shown in FIG.

In FIG. 7, 56 is a video signal input terminal, 57 is a prefilter, 58 is a sampling circuit, 59 is a sampling pulse generator, 60 is a 1/M frequency divider, 61 is a quantizer, and 62 is a predictor. , 63 is a subtracter, and 84 is a quantizer 2.
65 is a D range determination circuit, 88, 87, and 68 are multiplexers, and 69 is a D range determination information output terminal. Since the operation is exactly the same as that shown in FIG. 1, the explanation will be omitted, but FIG. This embodiment is different from the first embodiment of the present invention.

As described in detail, if the present invention is used for digital transmission of video signals, it becomes possible to transmit image quality equivalent to 9 bits/sample even on a transmission path that originally has a transmission capacity of only 8 bits/sample. Moreover, unlike conventional high-efficiency coding, there is no problem that the image quality may be less than 8 bits depending on the picture (and error propagation only requires one sample. Therefore, the present invention can be applied to digital VTRs for broadcasting. If applied, it is possible to maintain high image quality even in a situation where A/D conversion and D/A conversion are repeated multiple times while maintaining the same recording rate as the conventional method.

Further, in the embodiment, only the NTSC color video signal was explained as the input signal, but it goes without saying that the present invention can be applied to other composite signals such as PAL and component signals in exactly the same manner.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a first embodiment of the present invention;
FIG. 2 is a block diagram showing the first embodiment of the predictor in the present invention, FIG. 3(a) is a mode block configuration diagram of the predictor, and FIG. 3(b) is the error function characteristic of the predictor. Figure, 4th
Figure (a) is a mode block configuration diagram of the second embodiment of the predictor, Figure 4 (b) is an error function characteristic diagram of the same predictor, and Figure 5 is a block diagram showing the third embodiment of the predictor. Figure, Figure 6 (a
) is a mode block diagram of the predictor, FIG. 6(b) is an error function characteristic diagram of the predictor, FIG. 7 is a block diagram showing the second embodiment of the present invention, and FIG. 8 is a conventional example. FIG. 3... Sampling circuit, 4... Sampling pulse generator, 5... 1/M frequency divider, 6...
- First quantizer, 7, 8...predictor, 9.10
...Subtractor, 11...Second quantizer, 1
2...D range determination circuit, 13.14. 15・
...Multiplexer.

Claims (2)

[Claims] (1) A sampling means that samples a video signal, a first quantization means that quantizes the sampled video signal with N bits (N is a positive integer), and M samples of the samples obtained by the sampling means. (M is a positive integer); a prediction means that predicts the sample to be thinned out by the thinning means using surrounding unthinned samples; a subtraction means for calculating the difference value between the sample value of the sample and the predicted value obtained by the above prediction means;
-1)・N] bit [K is N>K>(M-1)・N/M
a second quantization means that quantizes by a positive integer that satisfies the following; a dynamic range determination means that determines the dynamic range of the difference value; and the first quantization means for the sample that is not thinned out. a first transmission means for transmitting the output of the second quantization means for the sample to be thinned out; If the dynamic range of the difference value judged by the dynamic range judgment means is below a predetermined range, the second transmission means transmits the difference value in K bits, and the first transmission means is switched to the first transmission means, and the other transmission means A video signal digital transmission device characterized in that it has a switching means for switching to the second transmission means. (2) The switching means is configured such that the dynamic range of P difference values among P/M (P is a positive integer) samples is all [M
・The video signal digital transmission device according to claim 1, characterized in that the video signal digital transmission device switches to the first transmission means only when the number is equal to or less than K-(M-1).N] bits, and switches to the second transmission means in other cases. .