JPH0423680A - Ghost elimination device for television signal - Google Patents

Abstract

PURPOSE:To prevent erroneous ghost elimination from being caused by providing an S/N measuring circuit monitoring the S/N of an input television signal and an S/N discrimination circuit inhibiting the revision of a filter coefficient when the above S/N is deteriorated and initializing a coefficient data when the S/N of the input television signal is restored to the excellent state. CONSTITUTION:An S/N measuring circuit 10 measures the S/N over a prescribed period in a reference signal and gives the S/N data to be measured to an S/N discrimination circuit 11. The S/N discrimination circuit 11 discriminates whether or not the S/N is equivalent to the level in which the random noise component is to be sufficiently suppressed by a summing mean processing by an adding mean circuit 3. When it is discriminated by the S/N discrimination circuit 11 that a noise component is unable to be eliminated by the summing mean processing, the S/N discrimination circuit 11 outputs a hold signal to a coefficient setting circuit 8, and the coefficient setting circuit 8 receiving the hold signal inhibits the generation of a new coefficient data and holds the just preceding coefficient data.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a ghost removal device for television signals used in television receivers, video equipment that handles television signals, and the like.

[Description of Prior Art] An example of a conventional ghost removal device will be described with reference to FIG. 2.

Note that all of the processing in this example is performed digitally.

Reference numeral 1 denotes a transversal filter that receives a television signal as input, corrects waveform distortion, and removes ghost components, and in this example, is constituted by a digital filter with variable filter coefficients. In such a transversal filter, ghost components are removed by setting filter coefficients that are in opposite phase to the ghost components contained in the input signal, for example. The output of the transversal filter 1 becomes an output video signal and is also supplied to the waveform extraction circuit 2. The waveform extraction circuit 2 extracts a predetermined period including a reference signal from a television signal. This reference signal has a pattern that can detect the reference timing in one horizontal scanning period, and an example of a typical reference signal is shown in FIG.

3(a) and 3(b) each show a step-like signal superimposed on a part of one horizontal scanning period, and in the case of FIG. 3(a), for example, the rising edge of the step-like signal In the case of FIG. 3(b), for example, the falling edge of the step signal can be used as the reference timing.

FIG. 3(c) shows a pulsed signal superimposed on the vertical synchronizing signal, and in this case as well, the rising or falling edge can be used as a reference.

In the so-called 8-field sequence method, the above-mentioned reference signal (data) is obtained by eliminating data in a predetermined field, data in a predetermined other field, and data in a predetermined other field. .

The signal of a predetermined period including the specific pattern signal extracted by the waveform extraction circuit 2 is supplied to the averaging circuit 3, which averages noise components randomly generated in the output of the waveform extraction circuit 2. .

A circuit example of the waveform extraction circuit 2 and the averaging circuit 3 is shown in the fourth example.
As shown in the figure.

In FIG. 4, the waveform extraction circuit 2 has a gate circuit 21 that allows an input signal of a predetermined period including a reference signal to pass through when it is given. is controlled on and off. The signal for a predetermined period that has passed through the gate circuit 21 is
The signal is supplied to the memory 32 via the adder 31 of the averaging circuit 3, and the output of the memory 32 is fed back to the adder 31. In other words, the adder 31 adds the supplied input signal to the signal that has already been written into the memory 31 and is delayed for a predetermined time and supplies the result to the memory 32, so the output of the memory 32 is Noise components that exist randomly in the axial direction are averaged out.

After performing this averaging circuit a plurality of times, the gate circuit 33 is opened in response to an averaging result acquisition pulse from the controller, and the averaged signal is taken out.

Returning to FIG. 2 again, the output of the averaging circuit 3 is filtered by the differentiating circuit 4 to remove the influence of the DC component, and then the subtracter 6 as a comparison means converts the output of the averaging circuit 3 into the reference signal (g) generated from the reference waveform generating circuit 5. The output of the subtracter 6 is supplied as an error signal to the magnification setting circuit 7.The magnification setting circuit 7 determines the magnification of the loop, and based on the magnification set here, the coefficient setting circuit 8 Coefficient data for setting filter coefficients of the Transpar Nor filter 1 is generated.

As mentioned above, ghosts are caused by the closed loop formed by the transversal filter 1, the waveform extraction circuit 2, the averaging circuit 3, the differentiation circuit 4, the subtracter 6, the reference waveform generation circuit 5, the magnification setting circuit 7, and the coefficient setting circuit 8. The removal circuit can be configured. In other words, the filter coefficients of the transversal filter 1 are successively varied according to the reference signal in the input television signal and the error signal obtained from the reference signal, and closed-loop control is performed so that ghost components are removed. .

An example of the prior art described above is configured with a feed 1<pack loop, but there is also a ghost removal circuit using a feedforward loop in which the waveform extraction circuit 2 is arranged on the input side of the 1 to Lance Barsal filters 1.

[Problem to be solved by the invention] However, the S/N of the input television signal is as shown in Fig. 5 (
When the S/N ratio of the input television signal is poor as shown in (b) of the same figure, a certain amount of noise can be removed by the averaging process.
Since noise cannot be sufficiently suppressed only by the averaging process by the averaging circuit 3, the ghost component contained in the reference signal cannot be detected accurately.

In addition, if pulse noise is mixed into a predetermined period including the reference signal, or if the waveform extraction circuit 2 extracts another signal other than this predetermined period, if the averaging process is performed as it is, components different from the ghost component will be removed as ghost components. It will be considered as such.

In either case, an erroneous error signal will be output from the subtracter 6, and since ghost removal processing is performed in accordance with this error data, this will cause problems in the output video.

The present invention has been made to overcome the drawbacks of such prior art, and it constantly monitors the S/N of an input television signal and prohibits setting of new coefficient data when the S/N deteriorates. At the same time, when the S/N returns to a good state, the coefficient data that determines the filter coefficients of the transversal filter is initialized to prevent erroneous ghost removal and improve output image quality.
The object of the present invention is to provide a television signal ghost removal device that does not interfere with No. 1.4a.

[Means for Solving the Problems] In order to achieve the above object, the present invention includes a transversal filter provided in the input television signal path, and a transversal filter that detects a reference signal for ghost detection from the television signal for a predetermined period. The ghost signal of the television signal is removed by extracting the signal, setting coefficient data according to the error signal obtained by comparing the reference timing in the reference signal and the reference signal, and varying the filter coefficient of the transversal filter. The ghost removal device includes an S/N constant circuit that monitors the S/N of an input television signal, and when the S/N deteriorates, prohibits changing the filter coefficient and The present invention is characterized by comprising an S/N determination circuit that initializes the coefficient data when the S/N of the signal returns to a good state.

[Operation of the Invention] According to the television signal ghost removal device having the above configuration, when the S/N of the input television signal is poor and noise cannot be removed sufficiently even by averaging processing, for example,
Since the filter coefficients cannot be changed completely, deterioration of the output image can be suppressed.

In addition, when the S/N returns to a good state again, the past ghost state and the current ghost state may be different, so by initializing the generated coefficient data, the current ghost Coefficient data can be generated according to the state.

[Example] Hereinafter, an example of the present invention will be described with reference to the drawings. Note that the same parts as those described in the description of the prior art are given the same reference numerals, and the description thereof will be omitted as appropriate.

FIG. 1 is a block diagram showing the configuration of an embodiment of a television signal ghost removal apparatus according to the present invention.

In the figure, the ghost removal device is configured by closed-loop control using a transversal filter 1, a waveform extraction circuit 2, an averaging circuit 3, a differentiation circuit 4, a reference waveform generation circuit 5, a subtracter 6, a magnification setting circuit D7, and a coefficient setting circuit 8. What is done is the same as in the case of the prior art shown in FIG.

In this embodiment, the output of the waveform extraction circuit 2 is supplied not only to the averaging circuit 3 but also to the S/N measurement circuit 10. As shown in FIG. 5(b), the S/N measuring circuit 1o measures the S/N of the reference signal over a predetermined period and supplies the measured S/N data to the S/N determining circuit 11. are doing.

The S/N data at this time is also provided to the averaging circuit 3 so that it can be used as data for determining the number of additions in the averaging process. Based on this data, the averaging circuit 3 determines the number of additions so that if the S/N is good, the number of additions is small, and if the S/N is bad, the number of additions is increased.

Based on the S/N data supplied from the S/N presizing circuit 10, the S/N judgment circuit 11 determines the S/N level to such an extent that random noise components can be sufficiently suppressed through averaging processing by the averaging circuit 3. It is determined whether it is N or not. The specific S/N level required for removal can be determined in advance from experimental data or the like. Also, the S/N determination circuit 1
1 is also supplied with time data from timer 12,
This is to distinguish between instantaneous S/N deterioration (pulse noise component) and continuous S/N deterioration.

When the S/N determination circuit 11 determines that noise components that cannot be removed by the averaging process are included, the S/N determination circuit 11
N judgment circuit]-1 outputs a bold signal to the coefficient setting circuit 8, and upon receiving the hold signal, the coefficient setting circuit 8 prohibits generation of new coefficient data and holds the immediately previous coefficient data. . In this way, even if an erroneous error signal is output from the reducer 6 due to the noise component contained in the input television signal, the corresponding correction will not be performed, and an adverse effect on the output image will be prevented. be done.

Next, when the S/N returns to a good state, the hold signal is released, and a processing block including the waveform extraction circuit 2, the averaging circuit 3, the differentiating circuit 4, the W reducer 6, and the magnification setting circuit 7 9, outputs a reset signal. Each circuit in the processing block 9 that receives the reset signal is reset so that the coefficient data set by the coefficient setting circuit 8 is initialized. By doing this, the coefficient data will be reset from the time when the S/N becomes good, and will not be affected by past ghost conditions (noise).
Coefficient data matching the current ghost state is generated.

Further, since the S/N determination circuit 11 is configured not to respond to a temporary deterioration of the S/N due to sudden pulse noise due to the action of the timer 12, the processing block 9
A reset signal is not given frequently.

Incidentally, in the above embodiment, the coefficient setting circuit 8 was configured to hold the immediately previous coefficient data by bold (g), but it is also possible to preset predetermined coefficient data and output it. .

Furthermore, it is not necessary that all the circuits in the processing block 9 be reset by the reset signal; in short, it is only necessary to reset the circuits necessary for the coefficient data generated by the coefficient setting circuit 8 to be initialized. Of course.

[Effects of the Invention] As described above, the television signal ghost removal device according to the present invention constantly monitors the S/N of the input television signal and performs ghost removal when the S/N starts to deteriorate. Since the filter coefficients of the transversal filter are prohibited from being changed, even if an erroneous error signal is generated due to a noise component, an adverse effect on the output image is prevented.

Moreover, since the coefficient data is initialized when the S/N returns to a good state, it is possible to prevent the adverse effects of past noise on the generation of coefficient data.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the television signal ghost removal device of the present invention, FIG. 2 is a block diagram showing the configuration of an example of the conventional ghost removal device, and FIG. 3 is a block diagram showing the configuration of an example of the conventional ghost removal device. A waveform diagram showing an example of
A block diagram showing in detail the configuration of a part of FIG.
FIG. 5 is a waveform diagram that does not show an example of the noise component of the waveform of the specific pattern signal superimposed on the television signal. 1...Transversal filter, 2...Waveform extraction circuit, 3...Averaging circuit, 4...Differentiating circuit, 5...
Reference waveform generation circuit, 6... Subtractor, 7... Magnification setting circuit, 8... Coefficient setting circuit, 10... S/N measurement circuit, 11
- S/N judgment circuit, 12 - timer. Patent applicant: Victor Japan Co., Ltd.

Claims (1)

[Claims] A transversal filter is provided in the input television signal path, extracts a signal of a predetermined period including a reference signal for ghost detection from the television signal, and extracts a signal of a predetermined period including a reference signal for ghost detection from the television signal, and A ghost removal device that removes ghosts from a television signal by setting coefficient data according to an error signal obtained by comparing the transversal filter with a reference signal and varying filter coefficients of the transversal filter, the apparatus comprising: an S/N measurement circuit that monitors the S/N of the input television signal, and when the S/N deteriorates, prohibits the change of the filter coefficient, and restores the S/N of the input television signal to a good state; A ghost removal device for a television signal, comprising: an S/N determination circuit that is initialized by the coefficient data when the coefficient data is used.