JP4612201B2 - Color signal demodulator - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a color signal demodulating device, in particular, Y / C (brightness / chromaticity signal) is separated from a synthesized video signal to demodulate the color signal, and a video signal of a VTR (video tape recorder) or the like is color demodulated with high performance. The present invention relates to a color signal demodulator.
[0002]
[Prior art]
A video signal in a color television or the like is transmitted by combining a luminance signal (Y) and a color signal (C). When such a synthesized video (color television) signal is compression-encoded and transmitted with high efficiency, it is important that the color signal is demodulated with high accuracy during demodulation by the receiver. The color signal demodulation method normally performs demodulation using the relationship between the color subcarrier (color subcarrier) frequency fsc and the horizontal synchronization frequency fh of the color television signal. In the case of an NTSC color television, there is a relationship of fsc = 455/2 × fh. Even when Y / C separation and color signal demodulation are performed by digital signal processing, color signal demodulation is performed using these relationships. .
[0003]
A normal television (TV) signal maintains the above relationship, but an NTSC color television signal reproduced from a VTR may not always maintain this relationship. When a video signal is compressed and encoded, it is encoded with high efficiency using the frame correlation of the TV signal. Therefore, the TV signal is sampled with a sampling clock synchronized with the horizontal synchronizing signal. There is a need. For this reason, when the VTR signal is sampled with the sampling clock synchronized with the horizontal synchronization frequency, the phase of the color subcarrier sampled is shifted for each line. In other words, the color subcarrier frequency varies with respect to the horizontal synchronization frequency.
[0004]
This frequency shift is not shifted suddenly for each line, but gradually shifts little by little, so that the phase at the beginning of the line and the phase at the end of the line are shifted. For this reason, if the phase of the color subcarrier is determined from the color burst at the head of the line, the phase of the subcarrier is shifted at the end of the line. There was a problem that the color signal could not be demodulated correctly and color misregistration or color unevenness occurred. Furthermore, if the color signal cannot be demodulated correctly, there is a problem that the predictive coding efficiency of compression coding is lowered.
[0005]
As a prior art of the color signal demodulation method, for example, there is “video signal processing circuit” (hereinafter referred to as the first prior art) in Japanese Patent Laid-Open No. 04-96595, and its configuration is shown in a block diagram in FIG. That is, a synchronizing signal separation circuit 111, a PLL (phase lock loop) circuit 112, an A / D (analog / digital) conversion circuit 113, a video data line memory 114, a color subcarrier signal reproduction circuit 115, a color subcarrier line memory 116, A timing generation circuit 117, a digital comb filter 118, and a color signal demodulation circuit 119 are included.
[0006]
This video signal processing circuit operates as follows. The A / D conversion circuit 113 digitizes the video signal with a sampling clock synchronized with an integral multiple of the horizontal synchronization signal. The color subcarrier signal reproduction circuit 115 obtains the phase difference Q and the amplitude CC of the color burst using a sampling theorem from two consecutive burst signal values. Based on this, a digital color subcarrier signal for one line is generated and stored in the color subcarrier line memory 116. The color signal demodulation circuit 119 performs color demodulation on the color signal (C) data separated by the digital comb filter 118 using the color subcarrier supplied from the color subcarrier line memory 116.
[0007]
Since this processing circuit configuration is mainly intended to demodulate a standard TV signal, it is assumed that fsc and fh have the above relationship and that the phase of the color sub-carrier is constant at the beginning and end. A carrier wave is generated. Therefore, a signal such as a VTR has a drawback that the color signal cannot be demodulated correctly around the end of the line. In addition, since the color subcarrier signal reproduction circuit 115 and the color signal demodulation circuit 119 are separate, this processing circuit configuration increases the circuit scale.
[0008]
Further, there is a “hue correction video signal decoder” (hereinafter referred to as second prior art) disclosed in Japanese Patent Laid-Open No. 8-331582, and FIG. 7 is a block diagram showing the configuration thereof. The decoder (demodulator) includes multipliers 71 and 73, low-pass filters (LPF) 72 and 74, sin and cos (sine, cosine) table 75, phase difference detector 76, hue correction amount register 77, and digital oscillator. 78 and an adder 79. The phase difference detector 75 detects the phase difference from the demodulated U signal and V signal in the burst signal portion. The digital oscillator 76 generates the phase of the color subcarrier having the same phase as that of the color burst from the detected phase difference. The hue correction amount register 78 is set with a hue correction amount, and the correction amount is added to the phase of the subcarrier.
[0009]
[Problems to be solved by the invention]
In the above-described second prior art decoder, the circuit for obtaining the color subcarrier signal and the color demodulating circuit are configured by one demodulating circuit. Therefore, if this decoder is applied to the above-described second prior art, the configuration is as follows. It will be easy. However, since the phase difference is determined at the head of the line, in the video signal such as the VTR described above with respect to the first prior art, the problem or drawback that the color signal cannot be correctly demodulated around the end of the line cannot be improved.
[0010]
OBJECT OF THE INVENTION
Accordingly, an object of the present invention is to provide a color signal demodulator capable of accurately demodulating a color signal even when demodulating a color television signal such as a VTR.
[0011]
[Means for Solving the Problems]
The color signal demodulating device of the present invention includes a line delay circuit for line delaying a video signal sampled with a sampling clock that is an integral multiple of the horizontal synchronization frequency, a luminance signal (Y) and a carrier color signal from the line delayed signal. Y / C separation circuit for separating (C), a switch circuit for selecting the color burst signal of the digital video signal not delayed in line and the carrier color signal delayed in line, and the color signal selected by this switch circuit as color A color signal demodulating circuit that demodulates using the phase angle of the subcarrier, a phase detector that detects the phase angle of the color burst from the two demodulated color difference signals (U, V), and a deviation between the previous line and the current line A correction circuit for correcting and a phase angle generator for obtaining the phase angle of the subcarrier from the phase corrected by the correction circuit are provided.
[0012]
According to a preferred embodiment of the color signal demodulator of the present invention, the Y / C separation circuit has a separation filter characteristic using a comb filter and a band-pass filter characteristic, and a correlation between lines of a color burst signal. When the correlation is strong, Y / C separation is performed with a comb filter, and when the correlation is weak with a video signal such as a VTR, Y / C separation is performed with a band-pass filter. The line delay circuit and the comb filter delay circuit are shared. As the phase detector, a means for roughly obtaining the phase angle from the two color difference signals, a means for determining whether the state is the initial state or the pull-in state, a means for setting the rough phase value to the integral initial value in the initial state, Integrating means for integrating a value obtained by multiplying the signal by a predetermined coefficient. Judgment is made based on the line correlation of the color burst, and the coefficient value is increased when the color burst phase changes greatly in the video signal such as VTR, and the coefficient value is decreased when the color burst phase of the broadcast signal or the like is stable.
[0013]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, the configuration and operation of a preferred embodiment of a color signal demodulator according to the present invention will be described in detail with reference to the accompanying drawings.
[0014]
FIG. 1 is a block diagram showing the configuration of a preferred embodiment of a color signal demodulator according to the present invention. This color signal demodulator includes an A / D converter 1, a line delay circuit 2, an adaptive Y / C separation circuit 3, a PLL circuit 4, a switch (SW) circuit 5, a color signal demodulator 6, a synchronization separation circuit 7, and a control. The circuit 8, the phase angle generator 9, the phase detector 10, and the correction circuit 11 are configured.
[0015]
Next, the function of each component constituting the color signal demodulator shown in FIG. 1 will be described. The synchronization separation circuit 7 separates the horizontal synchronization signal from the input video (or image) signal and supplies it to the PLL circuit 2. The PLL circuit 2 generates a sampling clock having a frequency equal to N times the frequency fh of the horizontal synchronizing signal. The sampling frequency used in the MPEG2 system for video compression code is 13.5 MHz as the sampling frequency suitable for the PAL signal and the NTSC signal. In this case, N = 858. The control circuit 8 outputs the head of the line, a signal indicating a color burst section, and other control signals and supplies them to the SW circuit 5, the phase angle generator 9 and the correction circuit 11. In order to correct the delay of the adaptive Y / C separation circuit 3, it is necessary to delay the signal from the output of the A / D converter 1 to the SW circuit 5 by the delay time. Omitted.
[0016]
The A / D converter 1 samples a video signal to obtain a digital video signal. The digital video signal is supplied to the line delay circuit 2 and the SW circuit 5. The line delay circuit 2 delays the digital signal by one line. The line-delayed video signal is supplied to the adaptive Y / C separation circuit 3. In order to form a comb filter, a signal without line delay is also supplied to the adaptive Y / C separation circuit 3 at the same time. The adaptive Y / C separation circuit 3 has a two-line comb filter and band-pass filter characteristic, and has a comb filter characteristic when the line correlation is high. When the line correlation is low, the digital video signal is separated into the luminance signal (Y) and the carrier color signal (C) by the bandpass filter characteristics, and the separated carrier color signal (C) is supplied to the SW circuit 5.
[0017]
The SW circuit 5 selects and outputs the color burst signal of the portion of the digital video signal that is not delayed during the burst flag period according to the control signal from the control circuit 8, and is delayed by one line period during the other periods. The separated transport color signal (C) is selected and output. Compared to the carrier color signal (C), the signal in the color burst section is output earlier by one line. The color signal demodulator 6 generates sin (sine) and cos (cosine) values corresponding to the phase angle of the color subcarrier from the phase angle signal. Then, the color signal is multiplied and passed through an LPF (low pass filter) to demodulate the carrier color signal or the color burst signal to obtain two color difference signals (U, V).
[0018]
The phase detector 10 detects the phase of the color burst from both the color difference signals U and V described above, and outputs the correct phase. In the NTSC signal, the color burst is in the negative direction (180 °) of the U axis (BY axis). Therefore, if the subcarriers are in phase, the U (BY) signal in the color burst period should be a negative value and V (RY) should be a substantially zero value. The phase θ of the color difference signals U and V and the color burst is tan (θ) = V / U from the relationship between the trigonometric function and the sample value. If a conversion table for obtaining θ from the values of the color difference signals U and V is provided, the phase value can be obtained from the values of U and V.
[0019]
When a video signal such as a VTR is input, the correction circuit 11 functions to correct this reference phase shift when the phase of the color burst gradually changes for each line. If the subsequent color subcarrier of one line is calculated based only on the phase of the color burst detected at the head of the line, the phase of the color burst becomes discontinuous for each line. Therefore, instead of resetting the color burst value for each head of the line, the intermediate phase of the line is corrected from the phase of the color burst before and after the line. In the middle of the section of one line, the phase is linearly corrected so that it gradually changes for each sample. As the phase of the color burst, when the phase value of the kth line is α01 and the phase value of the (k + 1) th line is α02, the amount of change during one line (N samples) is (α02−α01). Therefore, the magnitude of (α02−α01) / N is corrected for each sample. Therefore, αi = α01 + i (α02−α01) / N is output as the corrected phase αi of the i-th sample.
[0020]
The phase angle generator 9 generates a subcarrier phase angle. In this phase angle generator 9, the phase angle for each sample of the subcarrier is added to the phase of the corrected color burst to obtain the phase angle θi for each sample. From the subcarrier frequency fsc = 455/2 × fh and the sampling frequency fs = N × fh, the angular velocity ω is ω = 2π × fsc / fs. Therefore, the subcarrier phase angle θi at the i-th sample point is given by the following equation.
θi = αi + (π × 455 / N) × i
Since the phase is corrected by the function of the correction circuit 11, the carrier color signal (C) can be demodulated with high quality without any color shift even in a video signal such as a VTR.
[0021]
Next, FIG. 2 is a diagram illustrating the color burst phase detection operation. The signals supplied from the SW circuit 5 are multiplexed by changing the order of the carrier color signal (C) in the video section and the color signal CB in the synchronization section (burst flag section) including the color verse signal. As shown in FIG. 2, the carrier color signal C (k−1) of the (k−1) th line, the color burst signal BF (k + 1) of the (k + 1) th line, the carrier color signal C (k) of the kth line, The color burst BF (k + 2) of the (k + 2) th line is in this order. When the color burst phase of the kth line is α01 and the phase of the color burst of the (k + 1) th line is detected as α02, the correction circuit 11 demodulates the carrier color signal (C) of the kth line as Correction is performed at the correction phase αt.
[0022]
Next, FIG. 3 is a block diagram showing a detailed configuration example of the color signal demodulator 6 in FIG. The color signal demodulator 6 includes multipliers 30 and 32, LPFs 31 and 33, a cos (cosine) table 34 and a sin (sine) table 35. The color burst signal and the color signal of the carrier color signal (C) supplied to the color signal demodulator 6 are supplied to the multipliers 30 and 32. The subcarrier phase angle signal is supplied to the sin table 35 and the cos table 34. The sin table 35 is a conversion table of sin (θ) with respect to the input phase angle θ, and the output is supplied to the multiplier 32. The cos table 34 is a conversion table of cos (θ) with respect to the input phase angle θ, and the output is supplied to the multiplier 30. The multiplier 30 supplies the multiplication result to the LPF 31. The LPF 31 has a low-pass digital filter characteristic, the high-frequency band is blocked, and outputs a U (BY) color difference signal in the base band. Similarly, the multiplier 32 supplies the multiplication result to the LPF 33, the high frequency is blocked by the LPF 33, and a color difference signal of V (R−Y) is output.
[0023]
Next, FIG. 4 is a block diagram showing detailed configurations of the line delay circuit 2 and the adaptive Y / C separation circuit 3. 1H line memories 41 and 47, coefficients 40, 42 and 48, an adder 43, an SW circuit 44, an adaptive control circuit 49, a band pass filter (BPF) circuit 45 and a subtractor 46. In this embodiment, a configuration example in which the comb filter is a 2H comb filter is shown. Usually, the color signal output from the comb filter is further band-limited by BPF to obtain the carrier color signal (C). Since the video signal such as VTR has no line correlation, the color signal (C) is separated only by the BPF.
[0024]
The line memories 41 and 47 output the input signal with a delay of 1H. Coefficient units 40, 42 and 48 have coefficients k0, k1 and k2, respectively, and output each input multiplied by each coefficient. The coefficients k0, k1, and k2 of the 2H type comb filter have values of -1/4, 2/4, and -1/4, respectively. The adder 43 adds the outputs from the coefficient units 40, 42, and 48 and outputs the result. In a normal NTSC signal, the phase of the color subcarrier is inverted for each line. When the coefficients are -1/4, 2/4, and -1/4, the output of the adder 43 includes a color subcarrier frequency component signal, that is, a carrier color signal component signal that has passed through the comb filter. (C) is output.
[0025]
The SW circuit 44 and the adaptive control circuit 49 are supplied with a 1H-delayed NTSC signal and a carrier color signal (C). The adaptive control circuit 49 compares the color burst signals in the burst flag section. When the difference between the two is obtained and the magnitude of the difference is larger than a certain ratio (preset value) of the color burst size, it is determined that there is no line correlation in the color burst, and the color signal is comb-shaped. Instead of using a filter, the switching control signal w is supplied to the SW circuit 44 so as to be obtained by the BPF circuit 45.
[0026]
The SW circuit 44 selects based on the control signal w and supplies it to the BPF circuit 45. The BPF circuit 45 is a band-pass filter that passes in the vicinity of the subcarrier frequency and passes the carrier color signal (C). The carrier color signal C is supplied to the subtractor 46 and the next stage circuit. The subtractor 46 subtracts the carrier color signal (C) from the NTSC signal and outputs a luminance signal (Y) as an output. As another example, a configuration may be employed in which two BPF circuits 45 are prepared instead of one, and the characteristics are switched between when the comb filter is used and when it is not. The filter characteristics are set so that the pass band is wide in the case of the comb filter, and narrow in the case of only the BPF. The comb filter has a configuration in the case of 2H. In FIG. 4, the line memory 47 and the coefficient 48 are deleted, and the coefficient values k0 and k1 of the coefficient multiplier 40 and the coefficient multiplier 43 are set to -1/2, 1 respectively. By setting to / 2, a 1H comb filter can be configured. In this case, the separation characteristic of the comb filter is slightly lowered, but the circuit scale can be reduced.
[0027]
Next, FIG. 5 shows a detailed configuration of the phase detector 10 shown in FIG. When the circuit for obtaining the phase θ from the values of the color difference signals U and V is configured by a conversion table, if the U signal is 8 bits, the V signal is 8 bits, and the accuracy of θ is 8 bits, the input signal is 8 + 8 = 16 bits. On the other hand, a conversion table for obtaining an 8-bit output is required, and when this is configured with a ROM, the number of words increases. This configuration is an example that can be simplified by a ROM having an 8-bit size instead of a 16-bit size.
[0028]
The phase detector 10 includes a conversion table 51, a coefficient unit 52, an adder 53, a SW circuit 54, and a register 55. The conversion table 51 receives the upper 4 bits of the U and V signals U4 and V4. The conversion table 51 has a conversion characteristic for obtaining a phase θ from a 4-bit U signal and a 4-bit V signal, and outputs the phase θ for the input U4 and V4 signals with, for example, an 8-bit accuracy. This is supplied to the SW circuit 54. A control signal S for controlling the SW circuit 54 is also output. When the input U signal is negative and the V signal is in the range of -8 to 7, when U4 is negative and V4 is -1 or 0, S = 0 is set as the pull-in state, otherwise it is initialized. As a state, a signal of S = 1 is output and supplied to the SW circuit 54. The V signal is supplied to the coefficient unit 52, and the magnitude is k times (for example, ¼ time), and is supplied to the adder 53. A separate circuit is required, but if the color subcarrier amplitude is small and if it is desired to converge quickly to a phase of 180 °, the value of k is increased.
[0029]
The adder 53 adds the output of the coefficient unit 52 and the output of the register 55 and supplies the result to the SW circuit 54. When the control signal S is in the pull-in state S = 0, the SW circuit 54 selects and outputs the signal from the adder 53 and supplies it to the register 55. When the control signal is initialization S = 1, the value of the phase θ supplied from the conversion table 51 is selected and output and supplied to the register 55. Therefore, when the control signal S is S = 1, the phase θ obtained from U4 and V4 is set and output to the register 55 as an initialization state. When the control signal S is S = 0, a value obtained by multiplying the value of the V signal by k as the pull-in state is integrated by an integrator composed of the adder 53 and the register 55. Integration is performed with a precision of 10 bits. The 10-bit phase value is shown normalized to 360 °.
[0030]
In the register 55, the clock operates in the color burst period in accordance with the control signal from the control circuit 8, and the signal from the SW circuit 54 holds the sample clock and outputs it. The output of the register 55 is output as a phase signal. When the signal U is negative and the signal V is positive, the phase of the color burst is between 90 ° and 180 °, but when the value of the signal V (positive) is integrated, the phase increases and integration occurs. The phase of the value is close to 180 °. At 180 °, the phase of the color burst coincides with the phase of the color subcarrier at 180 °, the value of the signal V becomes 0, the integral value does not change, and therefore the phase remains at the 180 ° position.
[0031]
Since there are about 10 color burst cycles, first, when U and V are roughly phased with the signal values of the upper 4 bits, and the initial value is set, the burst is roughly correct from then on. From this phase, the color signal is demodulated. The values of the signals U and V are values that are determined to be in the pulled-in state. When the control signal S outputs a pull-in state, the pull-in state process is performed from the next time. That is, correction is performed by integration processing, and a value obtained by multiplying the value of the signal V by k by the coefficient unit 52 is added to the value of the register 55, and correction is sequentially performed. At the end of the color burst period, the phase has converged to the correct phase, and with the control signal from the control circuit 8, the integration process stops when the burst period ends, and the obtained phase value is held until the next line. Is done.
[0032]
When the correlation is high, the phase hardly changes from line to line, so the value of k is set sufficiently smaller than 1. When the phase of the color burst changes greatly for each line in a video signal such as a VTR, the value of k is set to a value as large as 1. If the value of k is adaptively switched by inter-line correlation or phase correlation of the color burst signal, more accurate color signal demodulation is possible. Whether or not the color burst signal has a strong correlation between the lines can be determined by whether or not the value of the phase detected by the phase detector 10 varies greatly for each line. It is also possible to control switching of the adaptive Y / C separation circuit 3 using this determination result.
[0033]
In the configuration of FIG. 5, the color burst cycle is 10 cycles, but the phase detector 10 forms a loop, there is a delay due to the LPF, the response due to the correction is delayed, and the color burst cycle in the state of the pull-in process. When the length is short, the pull-in time is insufficient, or when the amplitude is small, the correction value is small and cannot be corrected within the time, and the correct phase cannot be sufficiently pulled. For this reason, the amplitude of the color burst (for example, the magnitude of the signal U) is detected. When the amplitude is small, the value of the signal V also becomes small. Therefore, if the switching control is performed so that the value of the coefficient k is increased so as to increase the correction amount, the pull-in can be made faster, and the color burst is increased. The correct phase can be detected quickly not near the rear edge but near the center.
[0034]
The configuration and operation of the preferred embodiment of the color signal demodulation circuit according to the present invention have been described in detail above. However, it should be noted that such embodiments are merely examples of the present invention and do not limit the present invention in any way. Those skilled in the art will readily understand that various modifications and changes can be made according to a specific application without departing from the gist of the present invention.
[0035]
【The invention's effect】
As understood from the above description, according to the color signal demodulation circuit of the present invention, the following remarkable effects in practical use can be obtained. First, by providing a correction circuit, a color signal can be accurately demodulated even with a video signal such as a VTR. Also, the carrier color signal (C) separated and demodulated with high precision can be efficiently encoded and transmitted even when encoded and transmitted. The circuit configuration can be simplified by switching between the line-delayed carrier color signal (C) and the non-delayed color burst signal and demodulating the color signal. By obtaining the initial value of the phase roughly and then obtaining the phase in a configuration in which correction is sequentially performed by integration, the size of the conversion table can be reduced as compared to obtaining the phase using the direct conversion table.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a preferred embodiment of a color signal demodulator according to the present invention.
FIG. 2 is a diagram for explaining multiplexing and correction of a carrier color signal and a burst signal.
3 is a block diagram showing a detailed configuration of a color signal demodulator shown in FIG. 1. FIG.
4 is a block diagram showing a detailed configuration of an adaptive Y / C separation circuit shown in FIG. 1. FIG.
5 is a block diagram showing a detailed configuration of the phase detector shown in FIG. 1. FIG.
FIG. 6 is a block diagram showing a configuration of a conventional phase synchronization apparatus.
FIG. 7 is a block diagram showing a configuration of a conventional phase-locked clock generator.
[Explanation of symbols]
1 A / D converter 2 Line delay circuit 3 Adaptive Y / C separation circuit 4 PLL circuits 5, 44, 54 SW circuit 6 Color signal demodulator 7 Sync separation circuit 8 Control circuit 9 Phase angle generator 10 Phase detector 11 Correction Circuits 30 and 32 Multipliers 31 and 33 LPF
34 cos conversion table 35 sin conversion table 40, 42, 48 Coefficient unit 41, 47 Line memory 43, 53 Adder 49 Adaptive control circuit 45 BPF (band pass filter)
46 Subtractor 51 Conversion table 52 Coefficient unit 55 Register

Claims (5)

A line delay circuit for line-delaying a video signal sampled with a sampling clock that is an integral multiple of the horizontal synchronizing frequency, and Y / that separates a luminance signal (Y) and a carrier color signal (C) from the line-delayed signal. A C separation circuit, a color burst signal of a digital video signal that is not line-delayed, a switch circuit that switches and selects the line-delayed carrier color signal, and the color signal selected by the switch circuit uses a phase angle of a color subcarrier. A color signal demodulating circuit for demodulating, a phase detector for detecting the phase of the color burst from the two demodulated color difference signals (U, V), and the color for each sample of the current line from the color burst phase of the previous line and the current line A correction circuit that corrects a subcarrier phase shift, and a phase angle generator that obtains a subcarrier phase angle from the phase corrected by the correction circuit. A color signal demodulating apparatus characterized by obtaining. The Y / C separation circuit has a separation filter characteristic using a comb filter and a band-pass filter characteristic. The Y / C separation circuit is adaptively determined using a correlation between lines of a color burst signal. 2. The color signal demodulator according to claim 1, wherein Y / C separation is performed by a comb filter, and Y / C separation is performed by the band pass filter when a correlation is weak in a video signal such as a VTR. 2. The color signal demodulator according to claim 1, wherein a line delay circuit and a comb filter delay circuit are shared. As the phase detector, means for roughly determining the phase from the two color difference signals and determining whether the phase is in an initialization state or a pull-in state; in the initialization state, means for setting the coarse phase value to an integral initial value; 2. The color signal demodulating apparatus according to claim 1, further comprising an integrating unit that integrates a value obtained by multiplying the V signal by a predetermined coefficient in the pulled-in state. Judging by the correlation of the line of the color burst, when the color burst phase changes greatly in the video signal such as VTR, the coefficient value is increased, and when the color burst phase of the broadcast signal or the like is stable, 5. The color signal demodulator according to claim 4, wherein the coefficient value is reduced.

Images