JPS63200693A - Reproducing method for digital record of video signal - Google Patents

Abstract

PURPOSE:To scarcely generate deterioration on a display screen when error propagation occurs in the process of forecasting encoding power inversion by an error such as dropout, by recording the output which is the same in phase as a chrominance subcarrier among the output in a forecasting encoder on a same track. CONSTITUTION:A channel split 9 splits the output in the forecasting encoder 3 to arrange the output which corresponds with a picture element of the same phase point as the chrominance subcarrier in the same channel and to generate the error propagation for the chrominance subcarrier at the same phase point. That means, the information on a first channel data string to a fourth channel data string is respectively recorded on tracks 48-51. Therefore when the dropout occurs at one part of the track 49 and a burst error occurs at one part of the second channel data string, the output in a forecasting encoding power inverter 40 errs four picture elements by four every four picture elements. In such a case, as the error is not propagated to the neighboring picture elements of the error picture element, interpolation can be executed by using these neighboring picture elements.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a method for recording and reproducing video signals, and particularly to a method for digitally recording and reproducing composite video signals.

BACKGROUND OF THE INVENTION Many systems have been studied as video signal recording and reproducing devices, and as a result, various systems have been put into practical use. V
TR can be said to be a typical example.

By the way, serious studies have begun to digitize these video signal recording and reproducing devices. Example X 7 For example, various studies are being conducted on digital VTRs.

From now on, we will proceed with the discussion using a digital VTR as an example. One of the biggest challenges toward the practical use of digital VTRs (hereinafter referred to as "DVTRs") is how to secure recording time. In other words, since the frequency bandwidth is extremely expanded by digitizing the video signal, it is necessary to take some measures.

High-efficiency encoding technology can be considered as a means to solve this problem. Predictive coding is a typical example of this, and can be said to be one of the best methods that can significantly reduce the data rate and provide high image quality.

Therefore, the case where the predictive coding method is applied to DVTR will be described below.

FIG. 4 is a block diagram showing a predictive encoder. In the figure, 76 is an input terminal, 77 is a subtracter, 78 is a quantizer, 79 is an adder, 80 is a predictor, 81 is a coder, and 82 is an output terminal.

Input terminal 76 after digitizing the video signal to be recorded
is applied to the subtractor 77 via the subtracter 77. The subtracter 77 subtracts the predicted value created by the predictor 80 from the value supplied to the input terminal 76 to calculate a prediction error. The prediction error is quantized by quantizer 7.
The signal is quantized by 8, and is led to an adder 79 and a coder 81. Adder 79 adds the output value of quantizer 78 and the predicted value and applies the result to predictor 80 . On the other hand, the output value of the quantizer 78 is converted into a predetermined code by a coder 81 and sent to the next stage via an output terminal 82. DV
In the case of TR, the data is recorded on a magnetic tape via a magnetic head after being subjected to error correction encoding and modulation.

FIG. 5 is a pixel diagram for explaining the operation of FIG. 4. In the figure, 83 is the signal waveform of the nth line, 84 is the signal waveform of the (n+1)th line, and 86 to 87 are pixels. This figure shows a case where an NTSC signal is handled as a composite signal, and sampling is performed at a frequency four times that of the color subcarrier. Waveforms 83 and 84
are for emphasizing the carrier color signal components, respectively. For composite signals, creating predicted values based on pixels with the same phase of color subcarriers is superior in terms of compression efficiency.

Considering the moment when pixel 87 is applied to input terminal 76 in FIG. 4, predictor 80 often uses pixels 85 and 86, for example.

Problems to be Solved by the Invention As already described, the output data of the predictive encoder shown in FIG. recorded in During playback, the information recorded on the magnetic tape is read out using a magnetic head, demodulated, subjected to processing such as error correction, and then subjected to predictive encoding and reverse conversion to obtain the original image data. Incidentally, when recording and reproducing information on magnetic tape, particular attention must be paid to the occurrence of dropouts.

Normally, when a dropout occurs, if error correction cannot be performed, the data for that period will only be erroneous, but if predictive coding is employed, image deterioration will increase due to error propagation. In other words, if an error in 6 that remains due to dropout (even if it is not a dropout, the same applies when transmission errors occur continuously), etc. continues for one cycle or more of the color subcarrier, the subsequent image data will be completely lost. This will result in a very serious deterioration in image quality.

The predictive coding method is a method suitable for highly efficient coding of video signals, and is relatively adopted in the field of transmission. However, in the case of DVTR, since the error rate is high and the dropout frequency is also high, the above-mentioned error propagation causes a large deterioration in image quality.

Therefore, the present invention is intended to cause almost no deterioration on the screen even if erroneous blemishes occur due to dropouts or other causes and erroneous blemish propagation occurs during the predictive encoding inverse conversion process.

Means for Solving the Problems In the present invention, in order to solve the above-mentioned problems, among the outputs of the predictive encoder, those having the same phase with respect to the color subcarrier are recorded on the same track. , so that error propagation during reproduction occurs only at the same phase point with respect to the color subcarrier.

As a result, the error propagation is only propagated to specific pixels.
This eliminates the problem of deterioration in image quality compared to conventional methods. Furthermore, since correct values are restored in the vicinity of erroneous pixels due to error propagation, error correction with high image quality is possible.

Example This practical example is shown below together with an example.

FIG. 1 is a block diagram showing one embodiment of the present invention. In the figure, 1 is a video signal input terminal, 2 is an A/D converter, 3 is a predictive encoder, 4 is a subtracter, 5 is a quantizer, 6 is an adder, 7 is a predictor, 8 9 is a coder, 9 is a channel divider, 10 to 13 are error correction encoders, 14 to 17 are modulators, 18 to 21 are magnetic heads, 22 is a magnetic tape, 27
~30 is an equalizer/detector, 31~34 is a demodulator, 35~3
8 is an error correction decoder, 39 is a channel combiner, 40 is a predictive coding inverse transformer, 41 is a coding inverse transformer, 42 is an adder, 43 is a predictor, 44 is a corrector, 45 is a D/ A converter 46 is a video signal output terminal.

A composite signal to be recorded is applied to a predictive encoder 3 via a video signal input terminal 1 and an A/D converter 2. (This predictive encoder 3 is similar to that shown in FIG.
Detailed explanation of 8 to 8 will be omitted. ) Predictive encoder 3
The output of is divided into four data strings by a channel divider 9. The four divided data strings are each sent to an error correction encoder 1.
0 to 13 and modulators 14 to 17, the magnetic head 1
The data are recorded on the magnetic tape 22 via channels 8 to 21. During reproduction, the information recorded on the magnetic tape 22 is taken out via the magnetic heads 23-26, and the information is transmitted to equalizers/detectors 27-3o, demodulators 31-34, and error correction decoders 36-38, respectively. Pass it through. The equalization/detectors 27 to 3o correct deterioration and distortion occurring during the recording/reproducing process and restore the signal to a binary (or multi-level depending on the modulation method) signal. Demodulator 31
.about.34 performs the reverse operation of the modulators 14 to 17 based on the restored binary signal (in some cases, it may be multi-valued). If an error exists in the output data of the demodulators 31 to 34, the error correction decoders 36 to 38 correct the error data within their capabilities. Therefore, if the error is completely corrected, the output data strings of the error correction decoders 36 to 38 will be output to the error correction encoders 10 to 38, respectively.
This is equivalent to 13 input data strings. Error correction decoder 35
38 output data strings are combined by a channel combiner 39 and applied to a predictive encoder inverse transformer 4o. The predictive coding inverse transformer 40 is composed of a coding inverse transformer 41, an adder 42, and a predictor 43, as is well known.

(Detailed explanation of the predictive coding inverse transformer 4o will be omitted.)
The output of the predictive coding inverse transformer 4o is applied to a corrector 44, which interpolates and corrects pixel data in which errors remain.
The /A converter 45 restores the signal to an analog composite signal, and sends it out via the video signal output terminal 46.

Next, the tape pattern in the embodiment shown in FIG. 1 is shown in FIG. In the figure, 47 is a magnetic tape, 48
-61 are tracks. This tape pattern is an example of the helical scan method.
51 are tracks recorded by the magnetic heads 18 to 21, respectively. Naturally, the magnetic heads 23 to 26 play back while tracing the tracks 48 to 51, respectively.

On the other hand, FIG. 3 is a pixel diagram. In the same figure, 62 is n
The signal waveform of the th line, 63, the signal waveform of the (n+1)th line, and 64 to 75 each indicate a pixel. (
Since this is a similar diagram to FIG. 6, detailed explanation of each part will be omitted.

) Waveforms 52 and 53 show the case of an NTSC signal, and the sampling is four times the color subcarrier. 1st
Regarding the relationship between the prediction methods of predictors 7 and 43 and pixels in the figure, as already explained, it is effective to use pixels at the same phase point, and focusing on pixel 71, this pixel is similar to pixel 68 and Predicted at 67. Therefore, it is assumed that the predictors 7 and 43 create predicted values using the four previous pixels and the (1 horizontal scanning period+2 pixels) previous pixel.

On the other hand, channel division 9 divides the output of the predictive encoder 3 so that those corresponding to pixels at the same phase point with respect to the color subcarrier become the same channel. That is, pixel 6
4, 68, 62, 67, . . . are the first channel data string, pixels 55, 59, 63°1o vane 68, 72, . . . are the second channel data string, pixels 56. 60, 64, 65. . . . is the third channel data string, pixels 57, 61, 66, 70, . ...
... are divided into channel data strings into the fourth channel, respectively.

The first to fourth channel data strings are separately subjected to error correction encoding and modulation, and then recorded onto the magnetic tape 22 via magnetic heads 18 to 21, respectively. As a result, information regarding the first to fourth channel data strings are recorded in tracks 48 to 51 in FIG. 2, respectively.

Now, a dropout has occurred on part of track 49,
Assume that a burst error occurs in a part of the second channel data string. At this time, errors occur in pixels 55, 59, 63°68, . . . in the output of the predictive coding inverse transformer 4o. In other words, one pixel out of every four pixels will be incorrect. Even in such a case, the wrong pixel 55,
Errors are not propagated to neighboring pixels 59, 63, . . . , and interpolation can be performed with high precision using these neighboring pixels. This interpolation is performed in the corrector 44 of FIG.

Up to 11, the embodiments of the present invention have been described with reference to the drawings. Although the present invention is particularly effective for composite signals such as NTSC signals and PAL signals, it is also possible to interleave predicted pixels on the screen for component signals. You can get the same effect by using it on something.

Furthermore, the present invention is not limited to those using magnetic tape as a recording/reproducing device, but can also be applied to disk-shaped media and optical recording, and even in the case of magnetic tape, it is not limited to helical scanning.

Furthermore, the sampling frequency is not only four times the color subcarrier, but also three times the color subcarrier, and can be applied to sub-Nyquist sampling, and the number of channel divisions is not limited to four, but is suitable for sampling and prediction methods. The number of divisions can be set as follows.

Effects of the Invention It is clear from the above explanation that by using the present invention, even if burst errors occur due to dropouts, etc., and furthermore, the errors propagate in the process of predictive coding and inverse transformation, they will not occur. Errors occur only for each predetermined pixel, and correction can be performed with extremely high precision.

Therefore, error propagation, which is the biggest problem with predictive coding methods, can be substantially overcome, and as a result, a predictive coding method with high compression efficiency can be adopted for DVTH.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a tape pattern diagram for explaining the embodiment of FIG.
3 is a pixel diagram for explaining the embodiment of FIG. 1, FIG. 4 is a block diagram of a predictive encoder for explaining the conventional method, and FIG. 5 is for explaining FIG. 4. This is a pixel diagram for 3... Predictive encoder, 9... Channel divider, 39... Channel combiner, 4o...
. . . Predictive coding inverse transformer, 44 . . . Modifier,
47...Magnetic tape, 48-51...
truck. Name of agent: Patent attorney Toshio Nakao, 1st person, 2nd person
Figure 3 A s-q 63 and σ and σ? Thank you for your understanding.Figure 4Figure 5

Claims (1)

[Claims] A digital video signal that is characterized in that a composite video signal to be recorded is predictively encoded, and of the predictively encoded output, those having the same phase point with respect to a color subcarrier are recorded on the same track on a recording medium. Recording and playback method.