JPS6032387B2 - Dropout correction method of time axis corrector using pilot method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is intended to prevent a phase shift of a video signal that occurs when a dropout occurs in a time base corrector that uses a pilot signal as a direct clock signal.

Conventional time axis correctors correct time ruts in the reproduced video signal using only the phase information of the burst signal, but for this reason, if the S/N of the burst signal is poor, residual jitter occurs, causing time axis correction. This worsened the convergence of the subsequent video signal.

In addition, in order to correct the time axis fluctuation within one horizontal synchronization period (hereinafter referred to as one line) called velocity error, which occurs due to differences in the angle and depth of contact between the head and the tape, each line The phase change between the burst and the burst one line before it can be stored for one rotation of the head (capacitor memory method), or the image can be delayed by one line and the phase of the burst one line before and the current burst can be compared. (line-by-line method), the phase fluctuation is discretely known for each line, and the fluctuation is first-order approximated to a sawtooth wave, and the clock signal of the A/D converter is the sawtooth wave. It was made by phase modulating the clock signal, but in this case it was a first-order approximation, so it was not possible to correct small fluctuations within one line.

In order to overcome these drawbacks, it is possible to record a pilot signal by adding it to a video signal, extract the pilot signal from the reproduced signal, and create a clock signal for the A/D converter by shaping the waveform. Since there is a one-to-one correspondence between this clock signal and the phase of the video, the velocity error can also be completely corrected.

However, since the clock signal is directly generated, if a dropout occurs and the pilot signal disappears, the clock signal for the memory address counter also disappears, and the memory address counter is delayed by the amount of the missing clock signal.

Therefore, when data is read from the memory, the phase advances by the dropout period from the point where the dropout occurred, and on the video monitor, only the line where the dropout occurred shifts to the left, making it appear as a line.

To prevent this, there is a method of replacing the pilot signal with one from one line before, but there is a possibility that a false clock signal will be output due to switching transients, resulting in malfunction.

The present invention is characterized in that the clock signal of the address counter is generated by an oscillator whose phase is locked to the pilot signal, so that even if the clock signal is lost due to a dropout, the counter of the address generator continues counting. The purpose is to eliminate horizontal displacement of the image.

Another feature of the present invention is that when the pilot signal drops out, dropout compensation is performed even without a dropout pulse.The purpose of this is to compensate for the so-called "medaka noise", which is a small dropout that is difficult to detect. The purpose is to make it possible to replace the signal from one line before.

The figure shows an embodiment of the present invention, where 1 is an m/n multiplier (where m and n are natural numbers, the same applies hereinafter), 2 is a write pulse generator, 3 is a memory, 4 is a locked oscillator,
5 is an A/D converter, and 6 is an address generator.
Further, a is a pilot signal extracted from the reproduced signal, b is a clock signal of the A/○ converter, c is a write pulse, d is a clock signal of a write address generator, e is a dropout pulse, and f is an A/D converted signal. g is the write address, h' is the meeting data of the memory, and i is the reproduced video signal excluding the pilot signal.

Next, this operation will be explained.

The pilot signal a extracted from the reproduced signal is converted to m/n, multiplied by m/n, multiplied by a comparator, etc., and binarized to create a clock signal b for the A/D converter 5. .

From this signal, the generator 2 generates a write pulse c for the memory 3, and at the same time, the locked oscillator 4 generates a clock signal d whose phase is locked to the clock signal of the A/○ converter 5.
make. This clock signal d does not need to have color phase information of the video signal. Color phase information of the video signal is contained only in the clock signal d and the clock signal of the D/A converter (not shown).

Therefore, the lock-in accuracy of the locked oscillator 4 may be poor as long as there is no problem with the operation of the memory.The counter of the address generator 6 is driven by the clock signal d, and the data f is written into the memory 3 at the write address g.

Note that this data f is obtained by converting the video signal i into a digital signal in the A/D converter 5 using the clock signal b. Next, the data h is read out from the memory 3 using a clock signal (not shown) locked to an external subcarrier and subjected to D/A conversion, thereby obtaining a video signal whose time axis has been completely corrected. Next, the operation when dropout occurs will be explained.

When a dropout occurs, the dropout pulse (
A DOC pulse (hereinafter abbreviated as DOC pulse) e enters the generator 2 and inhibits the write pulse c. However, since the address generator 6 is operated by the clock signal of the locked oscillator 4, the address advances. After the dropout ends, the write pulse c is generated again and writing starts, but since the address g is exactly the same address as if the dropout had not occurred, it is written to the memory in the normal state. Further, since no writing is performed during the dropout period, data from one line before remains in the memory 3.

Therefore, when this is read out using an external reference signal, D
Only the portion with the OC pulse is reproduced with a video signal that is replaced with the video signal of one line before.

Next, the operation when there is no DOC pulse and the pilot signal a is dropped will be explained.

In this case, it is exactly the same as above, and since there is no clock signal for the A/D converter, incorrect data will be output as data f, but since the write pulse c stops, it is the same as when the DOC pulse comes. Perform the following actions.

In this embodiment, the pilot signal extracted from the reproduced signal is multiplied by m/n, and the resulting clock signal is immediately used as the clock signal for the A/D converter 5. It may be used with a phase shift.

Further, in the above embodiment, the write pulse of the memory 3 is A/
Although it is generated from the clock signal of the D converter 5, it may also be generated from the clock signal of the address generator 6.

As explained above, according to the present invention, in a time base corrector that directly uses a pilot signal as a clock signal,
It is possible to eliminate screen misalignment that occurs when a dropout occurs, and even when the pilot signal is missing due to the absence of a DOC pulse, the same correction can be made.

[Brief explanation of drawings]

The figure is a system diagram showing one embodiment of the present invention. 1: m/n transmitter, doubler, 2: write pulse generator, 3: memory, 4: locked switch, 5: A/
D converter, 6: Address generator, a: Pilot signal extracted from the reproduced signal, b: A/D converter clock signal, c: Write pulse, d: Write address generator clock signal, e: Dropout pulse,
f: A/D converted data, g: write address, h
: Memory read data.

Claims (1)

[Claims] 1. A/ The clock signal of the D converter changes the frequency of the regenerated pilot signal by m/
It is directly created by multiplying by n times (m, n: natural number) or by phase shifting it, and the memory write address is a write address driven by the output of an oscillator whose phase is locked to the clock signal of the A/D converter. 1. A dropout correction method for a time axis corrector using a pilot method, characterized in that the dropout correction method is made using a built-in address generator.