KR950004467B1 - Image signal time base error detection and correction apparatus - Google Patents

Abstract

The circuit detects and compensates the time base error of the image signal by identifying the variation rate of the time base error in the video disk player or VCR. The circuit includes a selector (404) for selecting one of the outputs of two time base error detectors (402,403) according to the selecting signal (SEL) of a controller (412), an image signal memory (406) for storing the digital image signal of an A/D converter (401), a time base error detector (407), a detector (408) for detecting the variation rate of time axis error in the color burst, a compensator (410) for compensating the time base error according to the output of a variation rate detector (409), and a D/A converter(414).

Description

Time axis error detection and compensation device of video signal

1 is a time axis error compensation circuit diagram of a conventional video signal.

2 (a) to 2 (c) are waveform diagrams for explaining the time axis error according to FIG.

(A) and (b) of FIG. 3 are changes in the time axis error of the horizontal main period.

4 is a time axis error detection and compensation circuit diagram of an image signal of the present invention.

(A) and (b) of FIG. 5 are waveform diagrams of a time-base error change rate using a color burst according to the present invention.

* Explanation of symbols for the main parts of the drawings

401: Anlog / Digital Converter

402: time axis error detection unit using a horizontal synchronization signal

403: time-base error detection unit using color burst signal 404: selection unit

405: light clock generator 406: video signal memory

407: color burst time base error detection unit

408: time axis error change rate detection unit in the color burst

409: 1 horizontal period time axis error change rate detection unit 410: time axis error compensation unit

411: change rate determination unit 412: controller

413: lead, clock generator 414: digital / analog converter

415: adder 416: register

The present invention detects and corrects a time axis error present in the video signal itself and a time axis error caused by an inconsistent rotation speed of the head and the disc in a video recorder such as a video disc player (VDP) or a video cassette recorder (VCR). In particular, it detects the time-base error change rate of one cycle color burst signal in front and rear of the video signal, and compensates the time-axis error existing in one horizontal cycle by making and compensating time-base error in one horizontal cycle using this change rate. The present invention relates to a time axis error detection and compensation device of a video signal.

In the video recorder, time-base error causes bad phenomena such as color bleeding and bleeding, and in severe cases, color disappears and is displayed in black and white screen.

Therefore, the time base error should be as small as possible to obtain a clear and clear picture.

However, a video recorder such as a video disc player or a video cassette tape has a time axis error because the time axis difference between the recorded video signal itself and the rotational speed of the head or the disc are not constant.

Therefore, the circuit for compensating for the time axis error of the conventional video signal is horizontal in the output signal of the analog / digital converter 101 for converting the video analog signal having the time axis error into a digital signal as shown in FIG. A time axis error detection unit 102 using a horizontal synchronization signal that detects a large time axis error by separating the synchronization signal HSYNC, and a color time signal separated from the color burst signal RST present in the output signal of the converter 101. A time axis error extracting unit 103 using a divergence signal for detecting an error, a selecting unit 104 for selecting one of two signals inputted from the time axis error detecting units 102 and 103, and this selecting unit 104 A light clock generator 105 for generating a light clock WCLK varied with a clock proportional to a time axis error of the signal generator; and a video signal memory 10 for storing a digital video signal of the converter 101. 6) and an intra-horizontal error compensator 108 for linearly compensating for the time axis error generated during the horizontal period, and a clock proportional to the compensated time axis error of the error compensator 108 within the horizontal period. A lead clock generator 107 for generating a variable read clock RCLK, a controller 109 for controlling each system, and converting a digital signal compensated linearly with a time axis error into an analog video signal and outputting the analog video signal. It consists of a digital / analog converter 110.

The error compensator 108 in the horizontal period generates a difference between the signals of the registers 1081 and 1082 and the registers 1081 and 1082 for temporarily storing the output signal of the selector 104. A horizontal divider generator 1083, a divider 1084 for dividing the signal of the secondary generator 1083 by 910, and an adder for receiving and adding output signals from the divider 1084 and the register 1087. 1085 and an arithmetic operator 1086 that adds and subtracts an output signal of the registers 1082 and 1087.

In the conventional circuit configured as described above, when the video signal Vi having the time axis error is input to the analog / digital converter 101 as shown in (b) of FIG. 2, the analog / digital converter 10l is written to. (Write) The clock generator 105 receives a constant light clock (WCLK) for one horizontal period, converts an analog signal into a digital signal, and converts an image signal memory 106 and a sykan axis error detection unit 102 using a horizontal synchronous signal. And output to the time axis error detection unit 103 using the color burst signal.

Then, the time axis error detection unit 102 using the horizontal synchronization signal separates only the horizontal synchronization signal HSYNC, and then detects the time axis error using the horizontal synchronization signal, and the time axis error detection unit 103 using the color burst signal 103 uses the horizontal burst signal RST. After class separation, we compare the phase with the reference comparison signal and extract finer time-axis error than using the horizontal synchronization signal.

The time-base error extracted in this way is input to the selector 104, which selects one of the input signals according to the selection signal SEL of the controller 109. When the time axis error included in the video signal is large, the output of the time axis error detection unit 102 using the horizontal synchronization signal is selected. When the time axis error is small, the output of the time axis error detection unit 103 using the color burst signal is selected. Accordingly, the signal selected by the selector 104 supplies the light clock WCLK to the analog / digital converter 101 to generate a clock proportional to the time axis error in the light clock generator 101 to attenuate the time axis error in an instant. do.

However, the detected time axis error is almost unchanged at the first horizontal period, but at the end of the 1 horizontal period, time axis error accumulates as shown in (C) of Figure 2, and after this one horizontal period, the horizontal synchronization signal or color burst signal When the time axis error is detected using the square wave, the output WCLK of the light clock generator 105 is changed to the square wave. This means that the image quality deteriorates toward the right end of the screen when viewed on the screen.

Accordingly, in the horizontal period error compensator 108 for removing the one horizontal period error, the time base error (TBE) input from the selection unit 104 every horizontal period is registered in the registers 1081 and 1082. In this case, the time axis error TBE ₍₁₎ of the register 108l is an error one horizontal period later than the time axis error TBE _(1-1) of the register 1082, respectively. Is entered in 1083.

The difference generator 1083 calculates a difference between two input signals, that is, | TBE ₍₁₎ -TBE _(1-1) ), and outputs it to the divider 1084. The divider 1084 inputs the divider. Calculated result by dividing the signal by 910 (△ S = ) Is input to the A terminal of the adder 1085.

Here, dividing by 910 divides the video signal of one horizontal period into 3.58MHz), when sampling at 4x frequency, it becomes 910 digital signals. You may do it every time.

On the other hand, when the adder 1085 adds the calculation result DELTA S of the divider 1084 and the output of the register 1087 for each latch period T2 and adds 910 transitions, TBE (i) -TBE (i- 1) | It becomes a signal of.

In addition, the operator 1086 performs a calculation process for adding or subtracting the result of the register 1087 to the time axis error signal TBE (i-1) of the register 1082 for each latch period T2 to perform a read clock generator ( 107) to change the read clock RCLK.

Therefore, the time clock error is compensated for by the read signal RCLK that changes linearly by ΔS within one horizontal period and the image signal memory 106 that receives the address signal ADD from the controller 109 within one horizontal period. When the digital video signal is output to the digital / analog converter 110, the digital / analog converter 110 converts the digital video signal into an analog signal to compensate for a linear time axis error within one horizontal period.

However, the time axis error present in the video signal does not actually change linearly but in various shapes. For example, in FIG. 3A, the time axis error increases, and the waveform a indicates the horizontal synchronization signal and the color burst signal. Denotes an error detected for every one horizontal period, and the b waveform represents a case of linearly compensating a time axis error in one horizontal period. In addition, the time-base error varies from the interval T (i-1) to T (i) at the beginning, as in the c and d waveforms, and then gradually increases, and gradually increases in the beginning as in the e and f waveforms. There is a case where a large increase later, and the time axis error decreases in (b) of FIG. 3, and the aspect is similar to that of (a) of FIG.

As a result, the time-base error present in the image signal changes in various shapes, so there is a problem in that it cannot be completely compensated for.

In view of such a conventional deficiency, the present invention obtains the rate of change of the time axis error in the color burst signal existing for 8-11 periods at every horizontal period, and then at every one horizontal period of the front and rear. Is a time axis error detection and compensation device of the video signal that can be completely compensated for the time axis error by extracting the shape of the change in the time axis error. It will be described in detail as follows.

4 is a time axis error detection and compensation circuit diagram of an image signal according to the present invention. As shown therein, an analog / digital signal converting an image signal Vi having a time base error into a digital signal according to a sampling clock is shown. Separating the horizontal synchronization signal from the output of the transducer 401 using a horizontal synchronization signal for detecting a large time axis error and using a color burst signal for separating a minute time axis error by separating the color burst signal A selection unit 404 for selecting one of the two output signals of the time axis error detection unit 403 according to a selection signal of the controller 412, and a light clock VCLK that generates a time clock error of the selection unit 404. A light signal generator 405, a video signal memory 406 for storing the digital video signal of the analog / digital converter 401 according to a light click WC1K, and the analog / digital video signal; A color burst time axis error detection unit 407 for detecting a time axis error for each one cycle of the front and rear of the 8-11 cycles of color bearers in the digital image signal of the digital conversion unit 401, and the color burst time axis A linear link between the time axis error change rate detection unit 408 in the color burst which detects the rate of change of the time axis error obtained by the color burst of the error detector 407, and the horizontal main period time axis error selected and output from the selection unit 404. The time axis error is determined after determining the shape of the time axis error change occurring in one horizontal period according to the detection of the one horizontal period time axis error change rate detector 409 and the first horizontal period time axis error change rate detector 409. Time-base error compensation unit 410 that compensates by varying and time-base error variation rate and time Axis and the size change rate determination part 411 that determines the size of the error rate, the time-base error of the compensated time base error exclusive NOR gate (41 017) of the compensation unit 410. A lead clock generator 413 for compensating for the lead clock RCLK proportional to the compensated time axis error according to the input through the controller 415 and the register 416; The digital-to-analog converter 414 converts the digital signal of the memory 406 into an analog signal, and the time-axis error compensator 410 has a time-base error change rate and a horizontal main period time-base using color burst at different points in time. The comparator 4100 (4101) for receiving the rate of error change and comparing the magnitudes, the comparator (41020) for comparing the time-base error of one horizontal period, and the NOT gate (4103) for receiving the outputs of the comparator (4100) (4101). -4108) and a selector 41013 which receives the change of the time axis error obtained in the one horizontal circumferential period obtained through the AND gates 4109-41012 and selectively outputs it according to a control signal of the controller 412, and various kinds of time axis error change rates. tablet And a coefficient storage memory 414014 for compensating one horizontal circumferential period according to the selected output of the selector 41013, and outputting a value, and an output value of the coefficient storage memory 41014 and a time-base error change rate. A multiplier 40215 for storing the register 41040 and storing the result in the register 40216, and an adder generating a horizontal axis period time axis error by adding the value of the register 40119 to the value of the register 40216 according to the latch signal T5 ( 41018).

If described in detail the effects of the present invention configured as described above.

To completely compensate for the timebase error in one horizontal cycle, first extract the timebase error in one horizontal cycle and then compensate for it using this extracted information. There are 8 different shapes.

Therefore, eight different time axis errors are detected and corrected. The basic principle is explained with reference to FIG. 5 as follows.

The color burst signal in the video signal is composed of approximately 8-11 cycles. The color burst signal is used to obtain a time-base error change rate in the color burst region. As shown in the waveform of FIG. The time axis error equal to E is detected by comparing the phase of the reference signal with respect to the front one cycle waveform C in the color burst, and the time axis error of F is detected with respect to the rear one cycle waveform D similarly to the waveform C.

Using the time axis errors of E and F obtained in this way, the time axis error change rate A in the color burst time is obtained by the following equation.

Time-base error change rate (A) = (One)

Time tl represents the time from waveform C to D.

When the video signal is sampled at a frequency four times the color burst frequency f _RST , the time tl is sampled four times in one color burst period, so that t1 = 4 points × 5 cycles = 20 points.

Also in the waveform (b) of FIG. 5, which is a color burst signal after one horizontal period, as described above, the time-base error change rate (B) is

(2)

Therefore, by comparing the magnitudes of the time-base error (E, F) (I, J) waveforms, it is possible to determine whether the time-base error increases or decreases, that is, whether the time-axis error change rate (A) (B) is positive or negative (-). .

On the other hand, when the time axis error increases, the time axis error rate of change (A) and (B) is compared with the visibility (Ti) of the time axis error rate of change of one horizontal circumferential period, which is the dotted line waveform (b) of FIG. The shape of the waveform at Ti can be seen, and the actual generated error can be found out from the c to f waveforms of FIG. 3A. Time-base error compensation can be made.

In addition, when the time axis error decreases, the time axis error change rate (A) and (B) are corrected when comparing the time axis error change rate of 1 horizontal circumferential period (b ′) of FIG. _1i-i ) and (T _i ), you can see the waveform, so you can see whether the actual time axis error is in the third degree (b) waveform c'-f '. Perfect time base compensation is possible.

The principle of the present invention described so far will be described with reference to FIG.

When the zeroing signal Vi including the time axis error is input to the analog / digital converter 401, the analog video signal is output from the analog / digital converter 401 according to the light clock WCLK, which is an output signal of the light clock generator 405. When the signal is converted into a digital signal, the converted video signal is stored in the video signal memory 406 and input to the time axis error detectors 402 and 403 using a horizontal synchronization / color burst signal.

Then, the time axis error detection unit 402 using the synchronous synchronization signal separates the horizontal synchronization signal from the image signal, compares the phase with the reference comparison signal, and detects a relatively large time axis error. The time axis error detection unit 403 uses a color burst signal. In this paper, only the color burst signal is separated from the meditation signal, and the phase comparison with the reference comparison signal detects more detailed time axis error information than when using the horizontal synchronization signal. When the detected time axis error information is input to the input terminals A and B of the selecting unit 404, the output of the selecting unit 404 is output by the output selection signal SEL of the controller 412. . That is, when a large number of time axis errors are included in the video signal, the output selection signal SEL selects an input terminal of the selector 404 to select the output of the time axis error detection unit 402 using the horizontal synchronization signal, If it is less included in the video signal, select the input terminal of the selector 404 to select the output of the time axis error detection unit 403 using the color burst signal, and the light clock generator 405 and one horizontal period time axis Input to the error change rate detector 409.

The light clock generator 405 generates a clock WCLK to compensate the time axis error according to the input signal and supplies the clock clock signal to the analog / digital converter 401 to compensate for the time axis error every one horizontal period. The digital signal is supplied to the memory 406 to store the digitally converted video signal. At this time, the video signal stored in the video signal memory 406 has a constant light clock (WCLK) of the light clock generator 405 for one horizontal period. Since it is fixed, there is almost no timebase error at the beginning of one horizontal period, but the timebase error increases in the second half.

The color burst time axis error detection unit 407 calculates the time axis errors (E, F) and (I, J) included in the color burst signal, as shown in FIG. 5. The analog / digital converter 401 When inputted to the digital image signal register 4070, the controller 412 generates the latch signal Tl when the first and second sampled signals of the C waveform are input in FIG. The generation interval of the latch signal Tl corresponds to 4f _RST , which is four times the frequency of the color burst frequency f _RST , so that the video signals stored in the registers 4070 and 407l have a 90 ° phase difference. Will have Therefore, if the divider 4042 receives the outputs of the registers 4070 and 4041, and divides the output (Sinθ ₁ ) of the register 4070 by the output (Cosθ ₁ ) of the register 407l, tanθ ₁ (= Sinθ ₁ / Cosθ). ₁ ), which includes the time base error.

The reason is because the value of tanθ ₁ when the value of the time base error of tanθ ₁ in the absence of a time-base error to be different, the difference is immediately available for the time-base error. Therefore, required time-base error for the conversion of tanθ ₁ if the time-base error exists for tanθ ₁ there is no time-base error, which is performed in the memory (4073) and the result is the waveform E of FIG. 5 (a).

Also, the time axis error F included in the D waveform of FIG. 5A is obtained by the same operation as described above, and if there is a difference, the latch signals T2 of the registers 4044 and 4075 are (Tl). ) Occurs 20 cycles after the first occurrence, 5 cycles later.

The time axis errors such as (E, F) and (I, J) in FIG. 5 obtained according to this operation are input to the time axis error change rate detection unit 408 in the color burst. Then, the time difference error rate detection unit 408 in the color burst operates to generate A or B, which is the rate of change of the time axis errors E, F, and I, J in FIG.

The comparator 4080 of the time difference error rate detecting unit 408 in the color burst receives the E, F, I, J signal of FIG. 5 which is the time axis error in the color burst from the color burst time axis error detecting unit 407, and its size. Compare and determine whether the time axis error increases or decreases, outputs to the time axis error generator 410 and the selector (41013), and outputs to the exclusive oragate (4081) (4082).

In the adder 4084, which is applied to the A input terminal through the exclusive oar gate 4041 and the note gate 4083, and the signal is input to the B input terminal through the exclusive oin gate 4082, that is, the difference between the incoming signals. To obtain | EF | or | IJ |, the operation is as follows.

In the operation of E-F

E> F | EF | = EF = E + +1

∴ + 1's two's complement of F

When E <F | FE | = FE = F + +1

∴ +1 is two's complement of E

Here, E> F is a case where the output (A <B) of the comparator 4080 is 0, and E <F is a case where the output (A <B) of the comparator 4080 is l. Same as below.

B input of adder 4084

Input A of adder 4084

That is, when the output (A> B) of the comparator 4080 is 0 ", that is, in the case of E> F, the calculation (3) is required, so the B input of the adder 4084 is inputted as E. The A input of the adder 4084 is inputted with the F value inverted, and the operation of equation (3) is performed, and when the output A <B of the comparator 4080 is "1", that is, E <F In this case, since the calculation of equation (4) is required, the B input of the adder 4084 is inverted and the value of E is input, and the A input of the adder 4084 is input as it is, so that the calculation of the equation (4) is performed. Is done.

On the other hand, the calculation of | I-J | is performed in the same manner as above.

The value of | EF | obtained as described above is input to the divider 4085. When the calculation is performed, this represents the rate of change of the time axis error, which is the A waveform of Fig. 5A. The time-base error change rate thus obtained is stored in the register 4086.

Therefore, the rate of change of the time-base error in the color burst before and after the horizontal period is obtained, respectively.

The time axis error obtained for each horizontal period is input to the one horizontal period time axis error change rate detection unit 409, where b and b 'waveforms are generated in FIG. 3 which is the time axis error change rate for one horizontal period.

A first time-base error of every horizontal period of the three-point degree T _(i-1) and T _(i) is a register (4090) and (4091) are stored as soon as well as T _(i-1 in the horizontal scanning period, a difference generator (4092) ₎ And the time-base error difference at the time point T _(i) are obtained and output to the divider 4093, which divides the input signal by 910.

In this case, 910 denotes the number of times when sampling clocks are sampled at one horizontal period at a frequency four times the color burst signal frequency f _RST .

In addition, the output of the register 4091, which is the time axis error at the time point T _{(i-1) of} FIG. 3, and the output of the register 4090, which is the time axis error at the time point T _(i) , are the two inputs (A, B) of the comparator 40020. Are respectively inputted to the note gate 41110 and the exclusive orifice 4116 in the change rate size determining unit 411 while being used in the time axis error generating unit 410 by comparing the sizes thereof.

The time axis error compensator 410 determines the time axis error change in the horizontal period shown in FIG. 3 and compensates for the time axis error value, which will be described in detail below.

1 horizontal period for outputting the register 4086 of the color burst vs. time axis error rate change rate detector 408 and the horizontal time period error rate rate change rate using the color burst at the time T _{(i) of} FIG. The time axis error change rate detector 409 receives the two time axis error change rates of the divider 4093, and determines the magnitude of the magnitude in the comparator 4100 and uses the color burst at the time T _{(i-1) of FIG.} Two time-axis errors of the divider 4093 of the register 4087 of the error change rate detection unit 408 in the galasturist storing the error change rate, and the horizontal change period time-base error change rate detection unit 409 outputting the horizontal main period error change rate. The change rate is input and the magnitude is compared by the comparator 410l. As a result, the shape of the time-base error of one horizontal main period is initially compared to the slope of b and b 'as shown in c, c', d, and d 'of FIG. Large error If it becomes small while the occurrence, e, e ', f, f' and the first, while the time base error smaller than generating groups discipline of b, b 'as shown, and outputs comparison indicating largely generated later.

The outputs of the comparators 4100 and 4101 are inputted to the notegates 4103, 4104, 4106, 4108, and anggregates 4109, 410l, 4101l, and 4212. Together, the time-period error change rate of one horizontal main period shown in FIG. 3 is determined. The results are summarized as follows.

Therefore, it is possible to distinguish the change in time axis error in one horizontal period into four groups, but each group needs to be divided into two groups. For example, when the AND gate 4109 is "1", c of FIG. Or d, and it is impossible to discriminate in the AND gates 4109-41012.

Therefore, in each group, it is necessary to punish whether it is a c waveform or a d waveform on the basis of a constant size, which is performed by the change rate size discrimination unit 4111.

In the rate-of-change size determination unit 411, a difference between a time-axis error change rate linearly connected to one horizontal circumferential period and a time-axis error change rate obtained in the color burst section is obtained as shown in FIG. 3, b and b '. "By judging whether the value is larger or smaller than the value, it is determined whether the time-base error changes rapidly or slowly.

The comparator 4101 of the time axis error interpolation unit 410 compares the magnitude of the time axis error change rate linearly in one horizontal circumferential period with the magnitude of the time axis error change rate in the color burst. The yen "1" is generated and inputted into the exclusive orifices 4111 and 4112 of the change rate size determining unit 411. The exclusive OA gates 4111 and 4112 generate 0 when the two input signals are the same and 1 when the other input signals are the same. However, when "1" is input to one input terminal as described above, the input signal is input to the other side. Is reversed. Therefore, since the signal input to the B input terminal of the adder 4114 is inverted and the signal input to the A input terminal of the adder 4114 is inverted once again by the note gate 4113, the original signal is input as it is. This is a calculation of (time-base error change rate linearly in one horizontal main period)-(time-base error change rate in color burst), and when the time-base error change rate in color burst is larger, the output of comparator 4101 is "0." "(Time-axis error change rate in color burst)-(time-axis error change rate linearly connected to one horizontal main period) is calculated.

That is, the adder 4114 outputs the difference between the two input signals and inputs it to the comparator 4115 to determine whether the change is abruptly compared with the reference signal "X", and the determined signal is a time axis error interpolation unit ( Input to selector 41013 of 410.

Since a variety of information for determining which of the waveforms of FIG. 3 is input to the input terminal of the selector 40213, a coefficient storage memory storing a value to be compensated for one horizontal circumference period of the waveform of FIG. When 41014 is designated, the coefficient storage memory 41101 outputs a value for compensating for a time axis error that varies by one horizontal period, and inputs it to the multiplier 40215.

The multiplier 40215 then multiplies the time-base rate of linear change by one horizontal circumference linearly with the value of the coefficient storage memory 41014, stores it in the register 40410, and sends it to the adder 4018 to the adder 41100 and registers. The value 41019 adds the value of the register 41016 every T5 periods of the controller 412 to generate a time axis error between one horizontal period, and inputs it to one input terminal of the exclusive NOR gate 41017. The output of the register 4091, which is a time axis error value at the time point T _{(i-1) of} FIG. 3, is input to the B input terminal of the adder 415, and is added to the result of the exclusive no-gate 40410. The noah gate 41017 passes the input value as it is or according to the result of the comparator 41020, or outputs an inverted value, which is the time axis at the time axis error at the time T _(i-1) of FIG. Although the value of the register 41019 of the error interpolation unit 410 needs to be added, in the case of FIG. 3B, the time axis error value at the time of T _(i-1) is subtracted.

Therefore, by the above-mentioned and trapped operation, it is possible to generate a time axis error having the shape of various waveforms shown in FIG. 3, which is a lead clock generator 413 which varies a clock in proportion to a value input through the register 41. The video signal is read from the video signal memory 406 so that the time axis error is perfectly compensated by the video signal memory 406 by varying the clock corresponding to the time axis error generated during one horizontal period. The digital / analog converter 414 is input to convert an analog signal into an output signal.

As described in detail above, in the present invention, when a video signal having a time axis error is input to a video device such as a video disc player or a video tape recorder, the rate of change of the time axis error for one cycle before and after each of the color bursts of 8 to 11 cycles is input. After detecting, using this rate of change, the time-base error in one horizontal cycle is compensated for by making perfect compensation of the time-base error in one horizontal cycle.

Claims (8)

From the output of the analog / digital converter 401 which converts the video signal Vi having the time axis error into a digital signal according to the sampling clock, a large time axis error and a fine time axis error are obtained by using a horizontal synchronization signal and a color burst signal. One of the two output signals of the time axis error detectors 402 and 403 and the time axis error detectors 402 and 403 using the horizontal synchronization signal and the color burst signal to be detected is selected according to the selection signal SEL of the controller 412. Writes the selection unit 404, the light clock generator 405 for generating a light clock proportional to the time axis error of the selector 404, and the digital video signal of the analog / digital converter 401. Detects the time-base error for one cycle before and after each of the 8-11 cycles of color burst in the video signal memory 406 and the digital video signal of the analog / digital converter 401 stored according to the clock. Is a color burst time axis error detection unit 407, a color burst time axis error change rate detection unit 408 for detecting the rate of change of the time axis error obtained from the color burst of the gala burst time axis error detection unit 407, and the selection. According to the detection of the one horizontal period time axis error change rate detector 409 and the one horizontal period time axis error change rate detector 409 which detects the time axis error change rate linearly connecting the horizontal main period time axis error selected and output from the unit 404. The time axis error compensator 410 determines the shape of the time axis error change generated within one horizontal cycle and then compensates for the variable by varying the time axis error, and the time axis error change rate and color burst within the two horizontal main periods. The compensated time axis error of the change rate size determining unit 411 and the time axis error compensation unit 410 for determining the magnitude of the time axis error change rate A lead clock generator 413 which generates a lead clock RCLK proportional to the compensated time axis error according to the inputs through the exclusive noble gate 41017, the adder 415, and the register 416, and the lead clock A device for detecting and guaranteeing a time axis error of a video signal, which is a digital / analog converter 414 for converting a digital signal of a video signal memory 406 that is variable and compensated according to RCLK to an analog signal. The register of claim 1, wherein the color burst time base error detector 407 stores the digital video signal of the analog / digital converter 40l in accordance with the latch signal Tl (T2) of the controller 412. 4070,4071, (4074,4075), a divider (4072) (4076) and a divider (4072) for receiving the outputs of the registers (4070,4071), (4074,4075) for division operation processing. A time axis error detection and compensation device for an image signal having a memory (4073) (4077) for receiving an output of the (4076) and generating a time axis error corresponding to the input value. The time axis error of the video signal according to claim 2, wherein the registers (40) 4040 and 4041 and 4075 have a phase difference θ ₁ according to the occurrence interval of the latch signal 1 (Tl) (T2). Detection and compensation device. The method of claim 3. And a time interval error detection and compensation device of an image signal determined by a color burst frequency (f _RST ) at which the latch signals (T1) and T2 are generated. The comparator (4080) according to claim 1, wherein the time difference error rate detecting unit (408) in the color burst receives and compares different time axis errors of the color burst time axis error detecting unit (407). Exclusive ore gates 4041 and 4042 that operate on the output of the comparator 4080 (A> B) as one input terminal and receive the time-side error signal as the other input terminal, and a gate The output of the exclusive oragate 4041 reversed through 4083 is inputted through the A input terminal and the adder 4084 which adds and receives the output of the exclusive oragate 4042 as the B input terminal, respectively. And a divider 4085 for dividing the output by the adder 4084 into 20 and a register 4086 and 4087 for storing the clock of the divider 4085 as a latch signal (T3). Time axis error detection and compensation device of video signal. The method of claim 1. The first horizontal period time axis error rate change rate detection unit 409 stores registers 4090 and 4091 for storing time axis errors per horizontal period according to the latch signal T3, and are stored in the registers 4090 and 4091. The time of an image signal including a horizontal period difference generator 4092 for detecting a time axis error difference before and after one horizontal period, and a divider 4093 for dividing the output of the horizontal period difference generator 4092 by the number of samples. Error detection and compensation device. The comparator 4100 of claim 1, wherein the time axis error generating unit 410 receives a time axis error change rate using a color burst and a horizontal main period error rate change rate at different points in time, and compares the magnitudes with each other. 1 horizontal circumferential period obtained through a note gate 4103-4108 and an end gate 4109-41012 receiving the outputs of the comparator 40210 and the comparators 4101 and 4101 for comparing time axis errors. A selector 41013 which receives the change shape of the time axis error and selectively outputs it according to the control signal of the controller 412, and stores various information on the time axis error change rate, and also one horizontal column according to the selected output of the selector 41013. A coefficient storage memory 40410 for outputting a period compensated value, a multiplier 40215 for multiplying the output value of the coefficient storage memory 40410 with a rate of change in the time base error of one horizontal period, and storing the result in a register 40216; A time axis error detection and compensation device for an image signal comprising an adder (41018) for generating a horizontal axis period time axis error by adding a value of the register (41019) to the value of the master (4101) according to the latch signal (T5). The change rate determining unit 411 of claim 1, wherein the rate-of-change size determining unit 411 includes a linear orifice 4111, a note gate 4113, and an exclusive oar gate 4112, and a linear time-period variation rate of linearity. And adder 4114, which calculates the difference between the time-base error change rate found in the color burst and the color burst, and compares the output of the adder 4114 with an arbitrary reference signal X. Time axis error detection and compensation device of the image signal by the comparator (4115) for determining whether the change.