JPH02142280A - Ghost elimination device, with noise level calculation circuit used therefor, and television receiver tuner or vtr provided with such ghost elimination device - Google Patents

Abstract

PURPOSE:To reduce time required for eliminating ghost by providing a subtractor obtaining a difference of waveforms of a front edge of a vertical synchronizing signal of different fields and a circuit obtaining mean power of the output of the subtractor, detecting a noise level of an input signal so as to control a noise reducer and an arithmetic circuit in response to the quantity of the noise level. CONSTITUTION:A noise level calculation circuit 102 calculates the noise level from an output waveform of a differentiating waveform 304 and supplies a control signal to the noise reducer 101. A subtractor 123 of a noise level calculation circuit 102 obtains a difference between the waveform of the front ridge of a vertical synchronizing signal and the waveform of the front ridge of the vertical synchronizing signal of one preceding filed outputted from a 1 field delay circuit 122. A mean power calculation circuit 124 calculates the average power of the output of the subtractor 123. Number of times of addition of input waveform in a waveform memory 112 and a divisor N of a divider 113 in the noise reducer 101 are varied with the control signal inputted from a control signal input terminal 111. Thus, the time required for the operation of the noise reducer is reduced and the time required for eliminating ghost is shortened.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a ghost removal tifi used in television receivers and the like, a noise level calculation circuit used therein, and the ghost removal length i! [Relating to a television receiver, chainer or video tape recorder equipped with the following.

[Conventional technology]

In the TV program, if the % wave (hope [)] that comes directly from the transmitting antenna and the electric wave 1 (reflection [) that is reflected from a building etc.) are simultaneously damaged by the damaged antenna, the hope hypothetical by iI! ii) The image is misaligned due to the source and number of reflections. A so-called ghost occurs.

TV show 1 in 1! Ghosts caused by Id are a major cause of deterioration of image quality, and conventional methods have been attempted to eliminate or prevent ghosts using extreme methods. One of them is a ghost removal device using a transversal filter in the video band.

The basic operation of ghost removal by a transversal filter will be explained using FIG. 2.

201 is an input terminal, 202 is a tapped delay line, 205 is a tap amplifier, 204 is an adder, and 205 is a subtracter.

206 is an output terminal, and τ is the tag interval of tapped delay #202, which is determined by the highest frequency frequency component included in the video signal.

When a video signal as shown by α in the figure is input to the input terminal 201, with a delay time of 5τ and an in-phase ghost of the signal @A added to the desired signal, 3 from the input terminal 201i value of the tapped delay spider 202 Let A be the gain of the tap amplifier 2020 connected to the tap of interest (hereinafter referred to as the tag gain). At this time, the output of the IJow device 204 contains b
The subtractor 205 is superior because the ghost complement shown in
By directly correcting the ghost compensation signal from the input signal, a signal from which ghosts have been removed, indicated by O, can be obtained at the output terminal 206.

There are many algorithms that automatically determine the tap gain, but from the standpoint of stability, IJ's・26゜August 1980%ri! 629 true to 655 pages (IEEE.

Transaction 03 Consumer E
electronics Vol, CB, '-26, Arigagozt, 1980, p. 629-p.
Correlation algorithms are commonly used.

(A specific example of a ghost removal device is JP-A-55-1
49525%) FIG. 3 shows a diagram 1022 of an example of a ghost removal device using a correlation method algorithm.

601 is an input terminal, 302 is an AlD carrier circuit, 603 is a transversal filter, 304 is a differentiation circuit, 305
is the noise retether, and 306 is the output 171? memory.

so7+s 8 circuits, soa & zism type memory, 30
9 is a true device, 310 is a tap gain memory, 511 is a D/A switching circuit, and 512 is an output terminal.

A video signal input from an input terminal 501 is converted into a digital signal by an A/D conversion circuit 502, and ghosts are removed by a transversal filter 303.

D/A'434 circuit 311, Ana ag! i
The signal is output from the output terminal 312.

After the output signal of the transversal filter 503 is differentiated by a differentiating circuit 504, the noise is suppressed by a noise reducer 305, and the output [V] is stored in a memory 306. The arithmetic circuit 307 calculates the data p(
Y* ) and! *[data iRi preliminarily stored in the form memory 308] is used to perform the correlation calculation shown in the linear equation.

Zi=α・ΣRkJu寥・, Ykllll Here, α is a positive constant subtractor 309, and the tap gain memo! Stored in 7510
The output signal zi of the arithmetic circuit 507 is subtracted from the old data C1 that is present in the data C1, and a new tap value C! (& denotes the t-th tap amplifier) is obtained.

ldC-CL-No. 2 (2) lrL-tap interest %Ci is supplied to the transversal filter 503. Repeat this action
Finally, the ghost is removed.

Next, the transversal filter 50 is
Differentiate the output qI of 3. Explain about rice.

Generally, ghost detection is performed using the ltU edge of the vertical synchronization signal shown in the fourth peak α. I.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.E.9 Transaction Seal on Conscheme Elec)”
0nyx polynium c.e. 24, 1978
August, page 267 to page 2 CIEEE, T-an
saction on CorLmmrEl
gctvorucz.

Vol, CB=24, August, 197
8, P, 267-P, 271) Therefore, in order to easily detect the delay time and amplitude of the ghost, the front line of the vertical lWj period signal is differentiated to form the pulse waveform shown in Figure 4bK, and the pulse waveform is stored in the reference waveform memory 508K. The correlation with a similar pulse waveform without a ghost (FIG. 4C) that is stored is determined, and a correlation value Zi shown in FIG. 4D is obtained.

Next, the operation of the noise resolver 305 will be explained.

Summary In a ghost removal device using a transpersal filter as shown in Figure 3, if the SN ratio (signal-to-noise ratio) of the input signal is small, the ghost removal performance deteriorates.

(I.E.E.Transaction On)
Concema Electronics, C.I.26,
1980, pages 629 to 636 (I E E E
, Trans, Consumtr Elgct-r
onzcz, CE-26* 1980, p, 6
29 w P, 656) o A specific example of a ghost removal device equipped with a noise resolver is disclosed in Japanese Patent Publication No. 62-22507.
) Therefore, the noise reducer 505K improves the SN ratio of the signal supplied to the output waveform memory 506 to improve ghost removal performance.

@5 Figure K shows a block diagram of a configuration example of the noise reducer 505.

501 is an input terminal, 502 is an adder, 505 is a waveform memo IJ, 504 is a divider, and 505 is an output terminal.

The differential waveform of the leading edge of the vertical synchronization signal shown in Figure 4 is inputted to the input terminal 501, and is compared with the contents of the waveform memory 505 in the adder 502, and the result is stored in the waveform memory 505.
It is stored in 05. When this operation is repeated N times, the waveform memory 505 stores the sum of the differential waveforms of the leading edge of the vertical synchronization signal of N fields. This result is multiplied by 1/v in divider 504 and output from output terminal 505.

One signal component among the waveforms input to the noise reducer 503 has the same waveform for each field. - 10,000, noise has no correlation for each field. Therefore, if the input waveforms of N fields are averaged 7111 times, one signal component will not change, but the noise will be suppressed by 1, #K with the average power. The contrast is determined by the number of additions N of input waveforms. For example, if N=8, the SN ratio improvement amount is t/a
(9dB).

[The problem that the invention attempts to solve]

In the above-mentioned conventional technology, the following problems arose due to the S/N ratio of the input signal.

(1) If the noise reducer 505 is configured as shown in FIG. 5, N
Since it is necessary to add the differential waveform before the vertical synchronization signal of the field, the time required to correct the tap interest percentage increases by N fields.

Therefore, the S/N ratio of the input signal is sufficiently large (i.e., the output of the differential circuit 304 is
Even if the arithmetic circuit 307 can be operated stably even if the signal is supplied to the direct output waveform memory 306, the time required for ghost removal increases as the signal passes through the noise reducer 305.

12) In the noise resolver 305 shown in Figure 5, the number N of additions to the input waveform is fixed, so the SN ratio improvement factor is also fixed. Therefore, the input signal.

If the SN ratio is small, the SN ratio cannot be sufficiently improved, and the arithmetic circuit 307 may malfunction.

For example, if the S/N ratio of the input signal is 50 dB and the S/N ratio is improved by 9tLB by the noise reducer 505, the S/N ratio of the signal supplied to the output waveform memory 306 becomes 59 dB.

From memory, I set the ghost detection limit in the arithmetic circuit 507 to 40
dB, the arithmetic circuit 307 considers all pulse waveforms whose amplitude ratio with the desired signal is less than aodB to be ghosts, and adds tap gain to the arithmetic circuit 307. The purpose of the present invention is to solve the above problem, where the possibility of malfunction increases.
(Noise reducer 30 according to the SN ratio of the 11 human input signal)
Ghost removal device capable of changing the number of additions N of input waveforms in step 5 .

[Means to solve the problem]

The above objective +11 or 12) is to provide a subtracter for determining the difference in the waveforms of the leading edges of the bulk orthogonal J91 multiples of different fields, and a circuit for determining the average power of the output of this subtractor.

Furthermore, (1) a control signal generation circuit is provided which generates a signal for controlling the number of additions N of input waveforms in the noise reducer 305 according to the output of the circuit for calculating the average power, and the output of this control signal generation circuit is used as a noise reducer. 505.

or (2) provide a control signal generation circuit that generates a signal to control the ghost detection limit in the χ calculation circuit 507 according to the output of the circuit for calculating the average power, and output the output of the control signal generation circuit to the calculation circuit. 507. This is achieved by:

[Effect]

Detection of the SN ratio of the human input signal is performed using the above-mentioned [X device and a circuit that calculates the average power by X.

The waveforms of the leading edges of the tone synchronization signals of two different fields are respectively 5(tl + FLl(1!I 13+5
(tl -1-F&ff1(tl +4) where stt+ is the signal component, ll'l sle, (t) is the noise. Also, where T is the average power of the noise, T is the output waveform. This is the period during which the data will be stored.

By subtracting the difference between formula 51 and formula (4) by X, we get
(tl -3, (tl (61) When calculating the average power of equation 61, nl(tl and nR(t) have no correlation, so the fifth term of equation (7) becomes 0.

σ3. =2σ (The output of the circuit for calculating the average power is proportional to the average power of the noise in the input signal.

Therefore, if the waveform input to the subtracter is appropriately normalized, the SiV ratio of the input signal can be obtained as the output of the circuit for calculating the average power.

Next, the operation of the means (11) will be explained.

In the means (1), the control signal is generated so that the SN ratio of the input signal (I) becomes large (when the SN ratio of the input signal Ig becomes large, the number of additions N of the input waveform in the noise reducer 505 becomes small).

Therefore, when the SN ratio of the input signal is large, the number of noses N becomes small, and the noise reducer 30
The time spent watching the action in step 5 can be shortened, and the time spent admiring ghost removal can be shortened. Further, when the SN ratio of the input (NjI) is small, the number of additions N becomes large, and the noise suppressing effect of the noise resolver 505 can be increased.

Next, the operation of the first means will be explained.

In the means t2), when the S/N ratio of the human input signal becomes small, a control signal is generated to lower the ghost detection limit in the arithmetic circuit 507K.

Therefore, when the S/N ratio of the human input signal is small, the arithmetic circuit 307 considers only the pulse waveform with a small field of view ratio with the desired signal as a ghost and calculates the tap gain, so noise may be mistakenly regarded as a ghost. Malfunctions that occur can be prevented.

〔Example〕

Embodiments of the present invention will be explained below using the drawings. 6. FIG. shows 10 tips with the same function.

1Ω1 is a noise rib counter, and 1o2 is a noise level calculation circuit.

The noise level calculation circuit 102 calculates the noise level from the output waveform of the differential waveform 504 (using the methods shown in equations 51 to (8)) and supplies a control signal to the noise reducer 101.

An example of the configuration of the noise level detection circuit 102 is shown in FIG. In the IWI diagram, 121 is an input terminal, 122 is a first field delay circuit, 123 is a subtracter, 124 is an average power calculation circuit, 125 is a control signal generation circuit, and 126 is an output terminal.

The subtracter 123 subtracts the difference between the waveform of the leading edge of the vertical synchronization signal supplied to the input terminal 121 and the waveform of the leading edge of the vertical synchronization signal one field before output from the one field delay circuit 122. . The average power calculation circuit 124 calculates the average power of the output of the subtracter 123. The control signal generation circuit 125 generates a control signal according to the output of the average power calculation circuit and outputs it to the output terminal 126.

An example of the configuration of the noise reducer 101 is shown in Figure 11.

In the figure, blocks that use the same signals as in Figure 1M5 indicate blocks with the same function. 111 is a control signal input terminal;
112 is a waveform memo IJ, and 115 is a divider.

In the noise reducer 101 shown in Fig. 811, the waveform memory 1
The number of additions N of the input waveform in 12 and the divider 115
The divisor N in is a 'III# configuration that changes according to the control signal input from the control signal input terminal 111.

FIG. 6 is a block diagram showing an example of implementing 1 in the above means, and blocks with the same numbers as those in FIG. 1 or FIG. 3 have the same functions. 601 is a noise level calculation circuit, and 602 is an arithmetic circuit.

In the figure, the output of a noise level calculation circuit 601 is supplied to an arithmetic circuit 602, and the X calculation circuit 602 is controlled to reduce ghost detection when the S/N ratio of the differentiation circuit 304 is small.

Fig. 7 is a block diagram showing an embodiment for realizing the above means (11 and (2)), and blocks with the same numbers as in Fig. 1 or Fig. 6 are the same bait blocks. 70
1 is a noise level calculation circuit.

In the figure, the output of the noise level calculation circuit 701 is supplied to the noise reducer 101 and the calculation circuit v11602, and when the S/N ratio of the input signal is larger than a certain threshold value, the noise reducer 1010 calculates the number of additions N of the input technique. Reduce the time required to remove ghosts, and when it is small, it takes a lot of calculations lol! 1602 ghost detection limit is lowered [to prevent malfunctions due to noise].

Figure i@8 is a block diagram showing another embodiment for realizing the above means (1). Blocks with the same numbers as in Figure 1 are the same blocks. 6801 is a switch. , 802 , 805 ! 804 is a noise level calculation circuit, and 805 is a divider.

In the ghost removal device shown in the figure, the switch 801 is connected to α* at the time of starting the operation. First, in the adder circuit 802, the waveform arriving from the switch 801 is
Add times. When the addition circuit 802 completes N additions, the switch 801 is connected to billIll, and m
In calculation circuit 805, input waveforms are added N times.

The output of the adder circuit 803 is input to the divider 8o5.

Adder circuit 805 and divider 805 constitute a noise reducer.

The noise level calculation circuit 804 includes addition circuits 802 and 8.
The SN ratio of the input signal is increased from the output of 05, and the adder circuit 8
05 and a divider 805 to control the number of additions N'k 711 of the input waveform of the noise reducer configured by the adder circuit 803 and the divider 805.

FIG. 15 shows an example of the configuration of the noise level calculation circuit 804. In the same figure, 151 and 1!52 are input terminals,
135 is a subtracter, 134 is an average power calculation circuit, and 135 is a control signal generation circuit.

The gX6155 calculates the difference between two signals input from input terminals 151 and 152. Average power calculation circuit 15
4, the average power of the output of the subtracter 155 is calculated, and the control signal generation circuit 155 generates a control signal according to the output of the average power calculation circuit 154, and outputs it to the output layer 156.

First, I will explain why the SN ratio of the input signal can be calculated using the circuit shown in Figure 15.

The waveform input to adder circuits 802 and 805 is 8t1
+ru 1 (tICi 2 1, 2..., 2N) 19
Furthermore, 51tl is expressed as a signal component, and 1itl is represented as noise. Output waveform of adder circuit 802 N S lt)+
ΣnL(jl (No. 103=1) The output waveform of the adder circuit 805 is expressed as N. Waveforms expressed by equations (10) and (11) are input to input terminals 151 and 152 in FIG.

The subtracter 133 calculates the difference between equations (10) and (113) by
6, z=1 t; N+1 = 2NarL (15) However, σ becomes the average power of 1 (tl), and the output of the average power calculation circuit 134 is proportional to the average power of the noise of the artificial power signal.

Furthermore, in the embodiment shown in FIG. 8, the output of the average power calculation circuit 154 is obtained as a larger value than in the case of equation (8), as seen from equation (13), so the S/N ratio is It is possible to improve detection accuracy.

FIG. 9 is a block diagram showing another embodiment for realizing the above means (2), in which blocks with the same numbers as in FIG. 6 or FIG. 8 are designed and have the same functions. 90
1 is an addition circuit, 902 is a noise level calculation circuit, 903
is a divider.

In the figure, a noise level calculation circuit 902 calculates the SN ratio of the input signal using the same method as in FIG. .

Addition circuit 901 is an addition circuit that adds input waveforms N times, and together with divider 905 constitutes a noise reducer.

FIG. 10 is a block diagram showing another embodiment for realizing the above means (1) and (2), and FIG.
Blocks with the same bud number as in the figure are 10 pieces of the same bait.

103 is a noise level calculation circuit.

In the figure, the output of the noise level calculation circuit 103 is added to the adder circuit 805. It is supplied to the divider 805 and the arithmetic sb path 602. When the input signal QSN ratio is larger than a certain threshold 11IL, the time required for ghost removal is reduced by imitating the number of additions N of the input waveform of the blade calculation circuit 803 and the division 11N1 of the divider 805, and when it is small, the calculation circuit 60
Lowering the detection limit of ghosts in step 2 [preventing malfunctions due to noise].

In the ghost removal device according to the present invention described above, ghosts are detected using the leading edge of the vertical synchronization signal.
A similar effect can be obtained by using another signal (for example, the VITS signal) inserted at any position in the ims period.

In addition, the tap gain Ci is modified according to equation (2), but the tap gain C is modified according to the following equation using only the output signal y&.
It is also possible to modify i. (Zero forcing method) , ntw old = CL-yz (14) Also, although the average power calculation circuits 124 and 154 calculate the average power of the input waveform, the same result can be obtained even if the average amplitude of the input waveform is calculated. or can be obtained.

Next, an application example of the ghost removal device according to the present invention will be described.

FIG. 14 shows a ghost removal device fi1100 according to the present invention.
An embodiment of a television receiver equipped with the following is shown.

In the In1 diagram, 201 is an RF input terminal, 202 is a chainer, 203 is an IF detection circuit, 204 is a luminance signal-chrominance signal separation circuit (hereinafter referred to as Y/C separation circuit), 205
206 is a color demodulation circuit, 206 is a color synchronization circuit, 207 is a matrix circuit, and 208 is a display circuit.

The RF multiplied signal input from the RF input terminal 201,
202, IF detection circuit 205K selects a station.

After the configuration, the ghost is removed by the ghost removal device 100. The output signal of the ghost removal fc Kan 100 is Y
/C separation circuit 204, color @! The RGB signals are converted by the adjustment circuit 205, the color synchronization circuit 206, and the matrix circuit 207&C, and then supplied to the display device 208K.

In Fig. 14, an application example is shown as a TV receiver (1 receiver), but the RF input terminal 201,
2. It can also be applied to a television chainer comprising the IF test circuit 203 and the ghost removal device 100.

Figure 15 shows ghost removal according to the present invention! An example of a VTR equipped with [100] is shown. In In2 diagram.

Blocks with the same numbers as in FIG. 14 indicate the same function blocks. 211 is a video signal recording circuit, 212 is a video head, 215 is a video signal reproducing circuit, and 214 is a video output electronic.

The RFq signal inputted from the RF input terminal 201 is converted into a video signal from which ghosts are removed by a chainer 202, an IF detection circuit M205, and a ghost removal circuit 100. This video signal is recorded on the cMi tape by a video signal recording circuit 211 and a video head 212.
The signal read from the magnetic tape is converted into a video signal and supplied to the video output terminal 214.

〔Effect of the invention〕

According to the present invention, the noise level of the input signal can be detected and the noise reducer and the arithmetic circuit can be controlled according to its magnitude.

(11) If the SN ratio of the human input signal is large, the operation time of the noise reducer is shortened.The time required to remove debris and ghosts can be shortened.

(2) When the SN ratio of the input signal is small, it is possible to lower the ghost detection limit and avoid malfunction of the arithmetic circuit.

[Brief explanation of the drawing]

Fig. 1 is a 1072 diagram showing an embodiment of the ghost removal device according to the present invention, Fig. 2 is a block and waveform diagram showing the basic operation of a transversal filter, and Fig. 3 is a conventional example of a ghost removal device using a transversal filter. Figures 1072 and 4 are waveform diagrams showing a ghost detection method using the leading edge of the vertical 111iirlJ91 signal, m5 is a block diagram showing an example of the Issakaki implementation of the noise resolver 1-server, and Figures 6 to 10 Figure 1072 shows another embodiment of the ghost removal device according to the present invention, and Figure In11 shows an example of the structure of the noise reducer in Figure 1.
! 121i is a block diagram showing an example of the configuration of the noise level calculation circuit in Fig. 1% *Fig. 15 is a block diagram showing an example of the configuration of the noise level calculation circuit in Fig. 38. Figure 1072, 15th illustrating an embodiment of 91 television receivers equipped with the device.
The figure is a diagram 1072 showing an embodiment of a video tape recorder equipped with a ghost removal device according to the present invention. [Explanation of symbols] 100...Ghost removal device 101, 15...Noise reducer 102.103
．． d (11,701,804,902--Noise level calculation circuit 5as-) Lanceversal filter 506--Output waveform memory 507.601--Arithmetic circuit 50B--Reference waveform memory 54G...Tap 8 gain Memory 5G! S, 112-Waveform memory 5G4, 115, 805.905-- Depth excavator 801-
・・Switch 802.805.901-1 addition circuit 124.154 ・・Average power calculation circuit 125.155
...Control multiple generation circuit agent Patent attorney Uji Ogawa
Man A Kuwa Zu Main House October 1 = Yoko” - One of the big bosses [closed] Figure 5 Toran and Versal Phil Crab 2 Figure Purple 4 Figure Goth's Branch η Helmet Figure 5 Close to -a 8 of 4-way repeater 11 Figure 1 shows 81 Acnoi large repeater user's 4-row close 7
Figure, heather, starting θ month, the coast sensing device is also 7 packs, 4 rows closed, 6 figures, wooden tongue SB Ha, 2 prongs removed. Figure 12 Figure Depth 1 Figure 1-h18'
To column 1"5 figure 8 figure 1J if run no 4-fold"&Nle calculation times 2 each -,
t-1,6 Intermediate "-----1 Fi+)4 ffilO Figure 14 Figure 14 An embodiment of a television receiver equipped with a ghost removal device 1 according to the present invention 1 Figure I5 Embodiment of VTR with ghost performance based on invention

Claims (1)

[Claims] 1. A transversal filter to which an input video signal is supplied, and a waveform of a portion of the output signal of the transversal filter in which a ghost is detected (ghost detection waveform) are extracted and averaged multiple times. A noise reducer that suppresses noise in the ghost detection waveform, a first storage circuit that stores the waveform of the output signal of the noise reducer, and a second storage circuit that stores the reference waveform for detecting ghosts. an arithmetic circuit that calculates the tap gain of the transversal filter from the waveforms stored in the first and second storage circuits; and a circuit that changes the tap gain of the transversal filter according to the output signal of the arithmetic circuit. In the ghost removal device provided, a noise level in the output signal of the transversal filter is calculated, and a control signal is supplied to the noise reducer accordingly, and when the noise level is high, a ghost detection waveform is calculated. 1. A ghost removal device comprising a noise level calculation circuit that controls the noise reducer so as to increase the number of times the ghost detection waveform is extracted, and to decrease the number of times the ghost detection waveform is extracted when the noise level is low. 2. In the ghost removal device according to claim 1, the noise level calculation circuit supplies a control signal to the arithmetic circuit, and when the noise level is high, determines the minimum value of the amplitude of the ghost to be detected (ghost detection limit). 2. A ghost removal device, characterized in that the arithmetic circuit is controlled to increase the detection limit of the ghost and to decrease the detection limit of the ghost when the noise level is low. 5. The ghost removal device according to claim 1, wherein the noise level calculation circuit supplies a control signal to the noise reducer and the arithmetic circuit, and in the noise reducer, when the noise level is lower than a certain threshold value. The ghost detection waveform extraction circuit is controlled to be small in size, and the arithmetic circuit is controlled to increase the detection limit of the ghost when the noise level is higher than the threshold value. removal device. 4. The ghost removal device according to claim 1, 2 or 3, wherein the noise level calculation circuit includes a subtraction circuit that calculates a difference between the ghost detection waveforms in two different fields in a television signal, and an output signal of the subtraction circuit. 1. A ghost removal device comprising: an average power calculation circuit that calculates an average power or an average amplitude; and a control signal generation circuit that generates the control signal in accordance with an output signal of the average power calculation circuit. 5. The ghost removal device according to claim 1, 2 or 3, wherein the noise reducer includes a first addition circuit that adds the ghost detection waveform a plurality of times, and an output signal of the first addition circuit that adds the ghost detection waveform a plurality of times. The noise level calculation circuit is configured with a division circuit that divides, and the noise level calculation circuit is separate from the first addition circuit that adds the output signal of the first addition circuit and the ghost detection waveform the same number of times as the first addition circuit. A ghost removal device, which receives an output signal of a second adder circuit provided as an input, calculates a noise level in an output signal of the transversal filter, and outputs the control signal. 6. The ghost removal device according to claim 5, wherein the noise level calculation circuit includes a subtraction circuit that calculates the difference between the output signals of the first and second addition circuits, and an average power or an average of the output signals of the subtraction circuit. A ghost removal device comprising: an average power calculation circuit that calculates an amplitude; and a control signal generation circuit that generates the control signal according to an output signal of the average power calculation circuit. 7. A subtraction circuit that calculates the difference between ghost detection waveforms in two different fields in a television signal, an average power calculation circuit that calculates the average power or average amplitude of the output signal of the subtraction circuit, and an output of the average power calculation circuit. A noise level calculation circuit comprising: a control signal generation circuit that generates a control signal according to the signal. 8. A first adding circuit that adds the ghost detection waveform a plurality of times; and a first adding circuit that adds the ghost detection waveform a plurality of times;
a second addition circuit that adds a ghost detection waveform at a different point in time to the ghost detection waveform added by the addition circuit;
a subtraction circuit that calculates the difference between the output signals of the first and second addition circuits, an average power calculation circuit that calculates the average power or average amplitude of the output signal of the subtraction circuit, and an output signal of the average power calculation circuit. A noise level calculation circuit comprising: a control signal generation circuit that generates a control signal; 9. A tuning circuit that tunes a television RF signal, an IF detection circuit that converts the output signal of the tuning circuit into a video signal, a circuit that converts the output signal of the IF detection circuit into an RGB signal, and a display device. 6. A television receiver comprising: a ghost removal device according to claim 1, claim 2, claim 3, or claim 5 provided after said IF detection circuit. 10. A television tuner comprising a tuning circuit for tuning a television RF signal and an IF detection circuit for converting an output signal of the tuning circuit into a video signal, in which the I
After the F detection circuit, claim 1 or claim 2 or claim 3
Alternatively, a television tuner comprising the ghost removal device according to claim 5. 11. A channel selection circuit that selects a television RF signal, an IF detection circuit that changes the output signal of the channel selection circuit to a video signal, and a change in the output signal of the IF detection circuit to a signal that is recorded on a magnetic tape. A video tape comprising: a video signal recording circuit that records and reproduces output signals of the video signal recording circuit on a magnetic tape; and a video signal reproduction circuit that converts signals recorded on the magnetic tape into video signals. A video tape recorder, characterized in that the ghost removing device according to claim 1, 2, 3, or 5 is provided after the IF detection circuit.